CAPE® Computer Science Syllabus, Specimen Papers, Mark

ComputerScience

ScienceComputer

ComputerScience

rScienceComputer

ComputerScience

CAPE

Computer

Science

SYLLABUS

SPECIMEN PAPER

MARK SCHEME

SUBJECT REPORTS

Macmillan Education

4 Crinan Street, London, N1 9XW

A division of Macmillan Publishers Limited

Companies and representatives throughout the world

www.macmillan-caribbean.com

ISBN 978-0-230-48244-9

) 2017

www.cxc.org

www.cxc-store.com

The author has asserted their right to be identified as the author of this work in accordance with the

First published 2014

This revised edition published November 2017

Permission to copy

The material in this book is copyright. However, the publisher grants permission for copies to be

made without fee. Individuals may make copies for their own use or for use by classes of which they

are in charge; institutions may make copies for use within and by the staff and students of that

institution. For copying in any other circumstances, prior permission in writing must be obtained

from Macmillan Publishers Limited. Under no circumstances may the material in this book be used,

in part or in its entirety, for commercial gain. It must not be sold in any format.

Cover design by Macmillan Publishers Limited and Red Giraffe

AER

CAPE® Computer Science Free Resources

LIST OF CONTENTS

CAPE® Computer Science Syllabus Extract

CAPE® Computer Science Syllabus

CAPE® Computer Science Specimen Papers:

Unit 1 Paper 01

Unit 1 Paper 02

Unit 2 Paper 01

Unit 2 Paper 02

CAPE® Computer Science Mark Schemes:

Unit 1 Paper 01

Unit 1 Paper 02

Unit 2 Paper 01

Unit 2 Paper 02

CAPE® Computer Science Subject Reports:

2004 Subject Report

2005 Subject Report

106

2006 Subject Report

120

2008 Subject Report Rest of Caribbean

136

2009 Subject Report

145

2010 Subject Report

155

2011 Subject Report

165

2012 Subject Report

181

2013 Subject Report

193

2014 Subject Report

205

2015 Subject Report

219

234

2016 Subject Report

Computer Science

Computer science is the study of the theoretical foundations of information and

computation and their implementation and application in computer systems. The CAPE

Computer Science Syllabus provides persons with advanced knowledge, skills and

attitudes to enable them to understand the uses and the impact of computer technologies,

and to use the technology to create new computer applications for all areas of human

activity. This syllabus provides opportunity for the acquisition of knowledge, skills and

attitudes as preparation for further studies in Computer Science and the world of work.

This syllabus consists of two Units, each comprising three Modules.

Unit 1: Fundamentals of Computer Science

Module 1 – Computer Architecture and Organisation

Module 2 – Problem-Solving with Computers

Module 3 – Programming

Unit 2: Further Topics in Computer Science

Module 1 – Data Structures

Module 2 – Software Engineering

Module 3 – Operating Systems and Computer Networks

CARIBBEAN EXAMINATIONS COUNCIL

Caribbean Advanced Proficiency Examination

CAPE

COMPUTER SCIENCE

SYLLABUS

Effective for examinations from May/June 2009

Published by the Caribbean Examinations Council

transmitted in any form, or by any means electronic, photocopying, recording or otherwise without prior

permission of the author or publisher.

Correspondence related to the syllabus should be addressed to:

The Pro-Registrar

Caribbean Examinations Council

Caenwood Centre

37 Arnold Road, Kingston 5, Jamaica, W.I.

Telephone: (876) 630-5200

Facsimile Number: (876) 967-4972

E-mail address: cxcwzo@cxc.org

Website: www.cxc.org

The Garrison, St Michael BB14038, Barbados

CXC A19/U2/08

Contents

RATIONALE ..................................................................................................................................... 1

AIMS .................................................................................................................................................. 2

SKILLS AND ABILITIES TO BE ASSESSED .................................................................................... 2

PRE-REQUISITES OF THE SYLLABUS ........................................................................................... 4

STRUCTURE OF THE SYLLABUS .................................................................................................. 4

UNI T 1 : F UNDA M E N T A L S OF C OM P UT E R S C I E N C E

MO D U L E 1 : CO M P U T E R AR C H I T E CT U R E AN D O R G AN I S AT I O N ......... 5

MO D U L E 2 : P R O B L E M - SO L VI N G W I T H C O M P U T E R S ....................................... 8

MO D U L E 3 : P R O G R A MMI N G .......................................................................................... 11

UNI T 2 : F UR T H E R T OP I C S I N C OM P UT E R S C I E NC E

MO D U L E 1 : DA T A S T R UC T U R E S ................................................................................... 1 3

MO D U L E 2 : S O F T W A R E E N G I N E E R I N G ................................................................... 15

MO D U L E 3 : O P E R A T I N G S Y S T E MS A N D C O MP U T E R N E T WO R K S .......... 19

OUTLINE OF ASSESSMENT ............................................................................................................ 23

REGULATIONS FOR PRIVATE CANDIDATES ............................................................................. 35

REGULATIONS FOR RESIT CANDIDATES .................................................................................. 35

ASSESSMENT GRID ......................................................................................................................... 36

LOGIC SYMBOLS ............................................................................................................................. 36

GLOSSARY ........................................................................................................................................ 37

CXC A19/U2/08

This document CXC A19/U2/08 replaces CXC A19/U2/03 issued in 2003.

Please note that the syllabus was revised and amendments are indicated by italics.

Revised 2008

Please check the website, www.cxc.org for updates on CXC’s syllabuses.

CXC A19/U2/08

Introduction

he Caribbean Advanced Proficiency Examination (CAPE) is designed to provide certification of the

academic, vocational and technical achievement of students in the Caribbean who, having completed

a minimum of five years of secondary education, wish to further their studies. The examinations address

the skills and knowledge acquired by students under a flexible and articulated system where subjects are

organised in 1-Unit or 2-Unit courses with each Unit containing three Modules. Subjects examined under

CAPE may be studied concurrently or singly.

The Caribbean Examinations Council offers three types of certification. The first is the award of a certificate

showing each CAPE Unit completed. The second is the CAPE diploma, awarded to candidates who have

satisfactorily completed at least six Units, including Caribbean Studies. The third is the CAPE Associate Degree,

awarded for the satisfactory completion of a prescribed cluster of seven CAPE Units including Caribbean Studies

and Communication Studies. For the CAPE diploma and the CAPE Associate Degree, candidates must complete

the cluster of required Units within a maximum period of five years.

Recognised educational institutions presenting candidates for CAPE Associate Degree in one of the nine categories

must, on registering these candidates at the start of the qualifying year, have them confirm in the required form, the

Associate Degree they wish to be awarded. Candidates will not be awarded any possible alternatives for which they

did not apply.

CXC A19/U2/08

Computer Science Syllabus

◆

RATIONALE

he widespread application of Computer Science, as embodied in the tools and techniques for gathering,

manipulating, analysing and disseminating information, made possible because of dramatic

improvements in computer and telecommunications technologies, has significantly changed society. A

large proportion of business transactions is performed over computer networks. Multi-media computers have

had a significant impact on the way in which people learn and on the way they seek entertainment. Moreover,

the increased integration of computer and telecommunications technology, exemplified by the Internet

and

associated technologies, has led to an increased globalisation of the world economy. Computer Science,

including the Internet, has significantly changed personal communication, commerce and the way in which

academic research is conducted. M

oreover, continuing developments in this field, including the increased use of

mobile networks and the further improvement and decreasing cost of computer hardware, mean that the world

has not seen the last of these changes.

The increasing importance of computer-based applications provides an important economic opportunity for

the region. In recognition of this, a number of regional governments have made the provision of information

services, including computer programming and software engineering, an important element in their economic

development plans.

However, in order for the Caribbean to become an integral part of this new world and to take advantage of

the economic opportunities it offers, citizens need to be able to use existing computer-based systems and to

create and maintain them. The latter requires a solid foundation in Computer Science. Thus, Caribbean

students need to acquire advanced knowledge, skills and attitudes to enable them to understand the uses and

the impact of computer technologies, and to use the technology to create new computer applications for all

areas of human activity. The syllabus is intended primarily for people who want to pursue a professional

career in Computer Science or related disciplines and provides the opportunity for the acquisition of relevant

knowledge, skills and attitudes as preparation for further studies in Computer Science and the world of work.

CXC A19/U2/08

◆

AIMS

The syllabus aims to:

1. develop a range of cognitive skills, including critical thinking skills;

2. develop an understanding of the components, the architecture and the organisation of a computer system;

3. equip students with the knowledge necessary to make informed decisions about the selection of components of

computer systems;

4. develop an understanding of the problem-solving process;

5. equip students with skills to create algorithms to solve problems;

6. develop skills to write correct programs to solve problems;

7. develop an understanding of the concepts of software engineering;

8. provide students with an understanding of abstract data types and their usefulness for manipulating data;

9. develop skills in using essential tools and techniques in system development;

10. develop an appreciation for the characteristics of operating systems and their applications;

11. develop an understanding of how computer networks can be used to connect computers together, regardless of

distance;

12. equip students with skills to design simple computer networks.

◆

SKILLS AND ABILITIES TO BE ASSESSED

The skills that students are expected to have developed on completion of this syllabus have been grouped

under three headings:

(i) Knowledge and Comprehension;

(ii) Application and Analysis;

(iii) Synthesis and Evaluation.

CXC A19/U2/08

Knowledge and Comprehension

The ability to:

- identify, recall, and grasp the meaning of basic facts, concepts and principles;

- select appropriate ideas, match, compare and cite examples of facts, concepts, and principles in

familiar situations;

explain phenomena in terms of generally applicable principles.

Application and Analysis

The ability to:

use facts, concepts, principles and procedures in unfamiliar situations;

transform data accurately and appropriately and use common characteristics as a basis for

classification;

identify and recognise the component parts of a whole and interpret the relationships between those

parts;

identify causal factors and show how they interact with each other; infer, predict and draw

conclusions;

recognise the limitations and assumptions of data gathered in an attempt to solve a problem.

Synthesis and Evaluation

The ability to:

- make reasoned judgements and recommendations based on the value of ideas and information and

their implications;

use the computer and computer-based tools to solve problems;

justify the appropriate application of techniques of problem-solving;

- select, justify and apply appropriate techniques and principles to develop data structures and

application programs for the solution of a problem.

CXC A19/U2/08

◆

PRE-REQUISITES OF THE SYLLABUS

Any person with a good grasp of the Caribbean Secondary Education Certificate (CSEC) Information

Technology or Mathematics syllabuses, or their equivalent, should be able to pursue the course of study defined

by this syllabus. However, successful participation in the course of study will also depend on the possession of

good verbal and written communication skills.

◆

STRUCTURE OF THE SYLLABUS

This syllabus consists of two Units comprising three Modules each of 50 hours. Although the Units are

independent of each other, together they provide a comprehensive introduction to the field of Computer

ience.

UNI T 1 : FU N D A M E N T A L S O F C O M PU T E R S C I E N C E

Module 1 - COMPUTER ARCHITECTURE AND ORGANISATION

Module 2 - PROBLEM-SOLVING WITH COMPUTERS

Module 3 - PROGRAMMING

UNI T 2 : F UR T H E R T OP I C S I N C OM P UT E R S C I E NC E

Module 1 - DATA STRUCTURES

Module 2 - SOFTWARE ENGINEERING

Module 3 - OPERATING SYSTEMS AND COMPUTER NETWORKS

CXC A19/U2/08

◆

UNIT 1: FUNDAMENTALS OF COMPUTER SCIENCE

MODULE 1: COMPUTER ARCHITECTURE AND ORGANISATION

GENERAL OBJECTIVES

On completion of this Module, students should:

1. understand the workings of the components of computer-based systems;

2. develop an appreciation of the functional components of the computer system, including the characteristics,

performance and interactions.

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

1. state the purpose of the main components of a

computer system;

2. describe the basic building blocks of a computer;

3. explain how data is represented in a computer

system;

Ha r d w a r e C o m p o n e n t s

Input/output devices: port connectivity; speed; quality of

output; specialised devices.

Memory types: ROM; RAM; EPROM; EEPROM.

Memory features: speed; size; word size; volatility.

Storage devices: capacity, access speed, access method,

portability.

Security: surge protectors, voltage regulators,

Uninterruptible Power Supplies (UPS).

Types of computers: supercomputer, mainframe,

microcomputer, Laptop, PDA.

Co m p u t e r Ar c h i t e c t u r e

Truth tables (refer to symbols on page 36).

Logic gates; Flip flops; registers; counters; multiplexors;

encoders, decoders.

Da t a R e p r e s e n t a t i o n

Bits; bytes; fixed (signed magnitude, ones and twos

complement) and floating point (sign, mantissa and

exponent) numbers and character representation; number

bases.

CXC A19/U2/08

UNIT 1

MODULE 1: COMPUTER ARCHITECTURE AND ORGANISATION (cont’d)

SPECIFIC OBJECTIVES

Students should be able to:

CONTENT

4. describe the main characteristics of a processor. Co m p u t e r O r g a n i s a t i o n

CPU components (ALU, CU, Registers), instruction

format (addresses per instruction, fixed length vs variable

length), types (data manipulation control and

input/output) and sets; instruction fetch, decode and

execute.

Clock speed, cache memory.

Suggested Teaching and Learning Activities

To facilitate students’ attainment of the objectives of this Module, teachers are advised to engage students in the teaching

and learning activities listed below.

1. Site visits to computer sales companies to view the various components of a computer system.

2. Invite computer professionals to talk to students on topics relating to the components of a computer

system.

3. Divide class into groups and each group asked to conduct research on a topic related to the

components of a computer system. Each group will then be required to present a report to the class.

Students should be encouraged to gather updated information from various sources such as the

Internet, current computer magazines, books and by interviewing computer p

rofessionals.

4. View interactive video tapes and Compact Disc, together with training materials on the components

of a computer system.

5. Provide students with opportunities to talk to the class on a topic relating to the components of a

computer system. Teacher will assign topics to students.

CXC A19/U2/08

UNIT 1

MODULE 1: COMPUTER ARCHITECTURE AND ORGANISATION (cont’d)

RESOURCES

Bradley, R. Understanding Computer Science for Advanced Level, London: Stanley

Thornes, 2005.

Heathcote, P. A Level Computing, London: Letts, 2005.

Parsons, J. and Oja, D. Computer Concepts, Albany, New York: International Thomson

Publishing Company, 2004.

Shelly, G., Cashman, T. and

Vermaat, M.

Discovering Computers 2008, Boston: Thomson Course Technology, 2008.

CXC A19/U2/08

UNIT 1

MODULE 2: PROBLEM-SOLVING WITH COMPUTERS

GENERAL OBJECTIVES

On completion of this Module, students should:

1. understand the problem-solving process;

2. appreciate the role and importance of algorithms in the problem-solving process;

3. understand the process of developing algorithms.

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

1. explain the concept of problem-solving; Definition of problem-solving.

2. describe the stages in the problem-solving process; Problem definition; problem analysis; identify and evaluate

possible solutions; select and justify the optimal solutions;

implementation and review.

3. explain the concept of an algorithm; Definition; algorithm as a problem-solving strategy; its role

and importance in the problem-solving process.

4. identify the necessary properties of algorithms

that are well designed;

A general solution to the problem, clearly defined and

unambiguous steps, finite number of steps, and flow of

control from one process to another.

5. identify ways of representing algorithms; Inclusion of narrative, flow charts and pseudocode.

6. explain constructs used in structured

programming;

Input and output statements.

Control Structures:

Sequencing; Selection; Iteration or repetition (bounded, for

example, fixed number of iterations and unbounded, for

example, sentinel control); Assignment statement.

7. interpret algorithms from case problems; Determination of output and correctness of a given

algorithm (the algorithm may be expressed in narrative,

flow charts or pseudocode).

CXC A19/U2/08

UNIT 1

MODULE 2: PROBLEM-SOLVING WITH COMPUTERS (cont’d)

SPECIFIC OBJECTIVES

Students should be able to:

CONTENT

8. correct algorithms from case problems; Determination of whether an algorithm achieves its stated

objective and if not provision of the correct algorithm.

9. develop algorithms from case problems;

10. explain the need for developing the logic of a

computer program.

Algorithms as logically sequenced instructions.

Suggested Teaching and Learning Activities

To facilitate students’ attainment of the objectives of this Module, teachers are advised to engage students in the teaching

and learning activities listed below.

1. Engage students in a discussion leading to the definition of a problem. The activity should be geared to the

students communicating their perspective of a problem. The teacher should then give feedback on those

perspectives by identifying problems in different scenarios.

2. Encourage students to have an appreciation that not every problem can be solved using the computer. From a list

of problems, the students should distinguish between problems that can be solved by using a computer and those

which cannot be solved using the computer.

3. Give a set of scenarios in which there are either opportunities or problems encountered by an entity such as an

organisation or a school. Students are required to (a) identify a problem, (b) formulate a problem statement, (c)

suggest two possible solutions, and (d) recommend one of the solutions and justify the choice.

4. Use algorithms to solve simple tasks, for example, to compute the sum of a set of numbers.

5. Use different program constructs in developing algorithms.

RESOURCES

Bradley, R. Understanding Computer Science for Advanced Level, London: Stanley

Thornes, 2005.

Heathcote, P. A Level Computing, London: Letts, 2005.

CXC A19/U2/08

UNIT 1

MODULE 2: PROBLEM-SOLVING WITH COMPUTERS (cont’d)

Kendall, K. and Kendall, J. Systems Analysis and Design, New Jersey: Prentice Hall, 2007.

Parsons, J. and Oja, D. Computer Concepts, Albany, New York: International Thomson

Publishing Company, 2004.

Shelly, G., Ashman, T. and Vermaat,

Discovering Computers 2008, Boston: Thomson Course Technology,

2008.

CXC A19/U2/08

UNIT 1

MODULE 3: PROGRAMMING

GENERAL OBJECTIVES

On completion of this Module, students should:

1. appreciate the need for different programming languages and program translation;

2. develop the ability to implement solutions to problems using a programming language.

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

1. identify the characteristics of different

programming paradigms;

Procedural or Imperative, Object-oriented, Functional and

Declarative and others (for example, Aspect and Scripting).

2. explain the need for different programming

languages;

Appropriateness to application (web application, games,

formula translation, application for mobile devices).

3. explain how assemblers, compilers, virtual

machines and interpreters are involved in the

execution of High-level programming languages;

Stages in the translation process: lexical analysis; syntax

analysis; semantic analysis; intermediate code generation;

code optimization; code generation.

Role of preprocessors; linkers.

4. assign values to declared variables; Declare variables using appropriate names. Use

appropriate and primitive data types (integer, float,

double, char and enumerated).

5. use input and output statements; Input data into variables; output formatted data from

variables; print headings.

6. choose appropriate conditional and iterative

constructs;

7. use conditional and iterative control constructs;

8. use arrays in programs; Read data into arrays, output data from arrays,

manipulate or modify data in arrays. Character arrays

(strings).

CXC A19/U2/08

UNIT 1

MODULE 3: PROGRAMMING (cont’d)

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

9. apply the techniques of structured decomposition

to reorganise a program into smaller pieces;

Write simple functions; programs should be clear,

orthogonal (small blocks of code) and simple.

10. implement algorithms to solve a given problem; Write, test and debug programs; syntax and semantic

errors; use of range tests and desk checks; code debugging

strategies (trace tables, use of ‘watches’ to examine the

values of variables).

11. use records as a means of grouping related

information;

The concept of ‘struct’ in C.

12. use text files to store data and records; File operations: open, close, read, write, append.

13. develop good programming style. White space (proper spacing), indentation, appropriate

comments.

CAN D I D AT E S W I L L B E AS S E S S E D I N P R O CE D U R AL ‘ C’ O N L Y .

Suggested Teaching and Learning Activities

To facilitate students’ attainment of the objectives of this Module, teachers are advised to engage students in the teaching

and learning activities listed below.

1. Critique previously written programs focusing, for example, on the use of structure, constructs, comments,

indentation, variable names and error handling.

2. Divide class into groups and ask each group to conduct research on a topic related to the implementation of

different data structures with respect to performance. Each group will then be required to present a report to the

class.

3. Develop test cases to illustrate the importance of testing.

4. Divide students into groups to research different languages, paradigms and translators and to examine the

weaknesses and strengths of each language, paradigm and translator.

RESOURCES

Shelly, G., Ashman, T. and Vermaat,

Discovering Computers 2008, Boston: Thomson Course Technology,

2008.

CXC A19/U2/08

◆

UNIT 2: FURTHER TOPICS IN COMPUTER SCIENCE

MODULE 1: DATA STRUCTURES

GENERAL OBJECTIVES

On completion of this Module, students should:

1. appreciate the use of abstract data types (ADTs) in the efficient manipulation of data;

2. understand basic algorithms for sorting and searching.

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

1. describe the concept of abstract data types

(ADTs);

2. distinguish among ADTs; Stacks (LIFO), queues (FIFO), singly linked list

(INSERT and DELETE): definition, structure and

operation.

3. perform basic operations on standard ADTs

using diagrams and algorithms;

Stacks: Push, Pop, Empty, Full.

Queues: ENQUEUE, DEQUEUE.

4. implement basic ADTs using one-dimensional

arrays;

Write programs to implement Stacks, Queues.

5. describe searching and sorting algorithms using

one-dimensional arrays;

Linear search; binary search; simple selection sort; bubble

sort.

6. implement searching and sorting algorithms. Linear search; binary search; simple selection sort; bubble

sort.

Suggested Teaching and Learning Activities

To facilitate students’ attainment of the objectives of this Module, teachers are advised to engage students in the teaching

and learning activities listed below.

1. Use scenarios to illustrate the application of Abstract Data Types.

2. Make reference to real-life situations that demonstrate ADTs in action; for example, adding and removing plates

from a stack of plates; customers in a queue (line) in a bank.

CXC A19/U2/08

UNIT 2

MODULE 1: DATA STRUCTURES (cont’d)

RESOURCES

Heathcote, P. A Level Computing, London: Letts, 2005.

Kendall, K. and Kendall, J. Systems Analysis and Design, New Jersey: Prentice Hall Inc., 2005.

Shelly, G., Ashman, T. and Vermaat,

Discovering Computers 2008, Boston: Thomson Course

Technology, 2008.

Sommerville, I. Software Engineering, Harlow: Addison Wesley, 2006.

CXC A19/U2/08

UNIT 2

MODULE 2: SOFTWARE ENGINEERING

GENERAL OBJECTIVES

On completion of this Module, students should:

1. understand the phases of the software development life cycle;

2. have an appreciation for the methods, processes, tools and techniques used in software engineering.

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

1. explain the reasons for a structured

approach to the software development

process;

Increased dependence of many organisations on their

computer systems.

Crises in previous developments: for example,

increasing costs of software development;

dissatisfaction of users and management with the

quality and suitability of software; increasing length

and complexity of the software.

Requirements for standard interfaces, both to users

and to other software.

Need for tighter control and management of process;

visibility of the process; risk management.

Importance of the need for the involvement of end

users and management.

2. explain the attributes of a well-engineered

software product;

Properties of well-engineered software:

maintainability; dependability; efficiency; usability;

portability; availability of appropriate documentation.

3. identify different generic software process

models and examine their strengths and

weaknesses;

Phases of the Software Development Life Cycle.

Life Cycle Models: waterfall approach; evolutionary

development including rapid prototyping; fountain

approach; formal transformation; reuse-oriented

approach.

CXC A19/U2/08

UNIT 2

MODULE 2: SOFTWARE ENGINEERING (cont'd)

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

4. outline the main activities, tools, techniques

and deliverables of the analysis phase;

Requirements and Specification Process: feasibility study;

requirements analysis.

Tools and Techniques: Interviews, questionnaires,

observations, review internal documents, prototyping, Data

Flow Models (Data Flow Diagrams) and their use to

document the flow of information: use of symbols to depict

data stores, process, data flows and external entities; Data

Dictionaries; Semantic Data Models (Entity-Relationship

Diagrams), Object Models;

Computer Aided Software Engineering (CASE) tools.

Deliverables: requirements specification (feasibility report,

functional and non-functional specification).

5. outline the main activities, tools, techniques and

deliverables of the design phase;

Design process: architectural design; interface design; data

structure design; algorithm design.

Tools and techniques: Structure charts, HIPO chart,

CASE tools.

Design Methods: top-down, bottom-up, system structuring

(sub-systems, modules, programs); Design Strategies:

functional versus objected-oriented.

Guidelines for screens, reports, user interfaces.

Deliverables: system architecture, design specification.

6. outline the main activities, tools, techniques and

deliverables of the implementation phase;

Coding process.

7. outline the main activities, tools, techniques and

deliverables of the validation phase;

Need for the testing process, test plans; software inspection;

software testing (unit inspection, acceptance test, test case

design).

8. outline the main activities, tools, techniques and

deliverables of the evolution phase.

CXC A19/U2/08

UNIT 2

MODULE 2: SOFTWARE ENGINEERING (cont'd)

Suggested Teaching and Learning Activities

To facilitate students’ attainment of the objectives of this Module, teachers are advised to engage students in the teaching

and learning activities listed below.

1. Identify organisations that use custom-built software applications. Students should be divided into

groups and asked to interview both users and management of these organisations to determine the

following:

(i) methodology (Life Cycle model used);

(ii) problems encountered during the development of the application(s);

(iii) level of user involvement;

(iv) lessons learned;

(v) what steps could have been done differently and why;

(vi) other relevant considerations.

Students can present their findings to the class.

2. Divide students into groups to research various Life

Cycle models, tools and techniques used during

the analysis and design phases. Students should report on their findings, including the strengths and

weaknesses of models, tools and techniques.

3. Invite professionals involved in developing software to make presentations to students to give them

additional perspectives on issues relevant to the topics. The professionals should be encouraged to

bring samples of deliverables.

4. Identify a ‘problem’ and engage students in developing a simple system which could solve the

problem.

5. Present ‘poorly-designed” screen layouts, data structures, reports and user interfaces and ask students

to critique them, for exam

ple, focusing on the appropriate use of font type and size; colours, spacing,

labelling or instructions, ease of use.

CXC A19/U2/08

UNIT 2

MODULE 2: SOFTWARE ENGINEERING (cont'd)

RESOURCES

Bradley, R. Understanding Computer Science for Advanced Level, London: Stanley

Thornes, 2005.

Heathcote, P. A Level Computing, London: Letts, 2005.

Parsons, J. and Oja, D. Computer Concepts, Albany, New York: International Thomson

Publishing Company, 2004.

Shelly, G., Ashman, T. and Vermaat,

Discovering Computers 2008, Boston: Thomson Course Technology,

2008.

Sommerville, I. Software Engineering, Essex: Pearson Educational Limited, 2006.

CXC A19/U2/08

UNIT 2

MODULE 3: OPERATING SYSTEMS AND COMPUTER NETWORKS

GENERAL OBJECTIVES

On completion of this Module, students should:

1. understand the functions of operating systems;

2. develop an appreciation for networking technology and applications.

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

1. explain the main functions of operating systems; Resource manager; interface.

2. describe how operating systems have evolved

over time from primitive batch systems to

sophisticated multi-user systems;

History of operating system development.

3. describe the functions of operating systems; Operating system functions:

Bo o t s t r a p p r o c e s s

Pr o c e s s M a n a g e m e n t

Definition

Process states: Running, Ready, Blocked.

Definition

How the interrupt mechanism works

Types of interrupt:

- interrupt generated by the running process

- Input/Output Interrupt

- External Interrupt

- Restart Interrupt

Deadlock (what is Deadlock)

The pro

cess control block (process descriptor)

Scheduling Algorithms Pre-emptive (Shortest-Job-First (SJF),

round robin) and Non-pre-emptive (FCFS), Shortest-Job-

First (SJF)

(explain the concepts only)

CXC A19/U2/08

UNIT 2

MODULE 3: OPERATING SYSTEMS AND COMPUTER NETWORKS (cont’d)

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

Me m o r y Ma n a g e m e n t

Virtual Memory, paging, thrashing

Fi l e M a n a g e m e n t

Directories/Folders, Files

Se c ur i t y ( o f f i l e s )

User IDs, Passwords, Lockwords, Access control list, file

encryption, file compression

Activity logs

In terf a ce (user)

Types of interfaces: Menu, command prompt, GUI and

the manipulation of the interface

De v i c e M a n a g e me n t

Device drivers

Interrupt handling (PCB)

Input/output control

Peripheral control, polling Buffering, Spooling.

Ne t w o r k i n g

Network management (user accounts, access logs)

Networking Protocols (TCP/IP)

4. distinguish among networked, client-server, and

distributed;

Ne t w o r k A r c h i t e c t u r e : Ethernet, FDDI.

Ne t w o r k t o p o l o g y : Star, Ring, Bus, Hybrid.

Ne t w o r k d e v i c e s : Modems, switches, routers, bridges,

network interface cards (NIC), hubs.

Tr a n s m i s s i o n M e d i a : wired (twisted pair, fiber-optics,

coaxial); wireless (satellite, microwave)

IEE1394 (Firewire) and cable connectors using diagrams.

Pr o t o c o l :

Transmission Control Protocol/Internet Protocol

(TCP/IP), File Transfer Protocol (FTP), Hypertext

CXC A19/U2/08

UNIT 2

MODULE 3: OPERATING SYSTEMS AND COMPUTER NETWORKS (cont’d)

SPECIFIC OBJECTIVES CONTENT

Students should be able to:

Transfer Protocol (HTTP); Hypertext Transfer Protocol

Secure Sockets Layer (HTTPS); IEEE802.11a/b;

IEEE802.16g; characteristics of Voice Over Internet

Protocol;

Open System Interconnection (OSI) model.

Ac c e s s M e t h o d s f o r m o b i l e n e t w o r k s : CDMA,

TDMA, GSM, GPRS

Ne t w o r k i n g c o n s i d e r a t i o n :

cost, security, management, expandability,

interconnectivity, wired vs wireless

Ne t w o r k C o n f i g u r a t i o n :

Types: Multi-user; client server, centralised vs. distributed

system, peer to peer.

Ne t w o r k S e c u r i t y :

Firewalls

design simple networks.

Use diagrams to design networks.

Suggested Teaching and Learning Activity

To facilitate students’ attainment of the objectives of this Module, teachers are advised to engage students in the teaching

and learning activity below.

Divide class into groups and each group asked to conduct research on the functions of one type of operating system with

respect to convenience, efficiency, and the ability to evolve. Each group will then be required to present a report to the

class. Students should be encouraged to gather updated information from various sources such as the Internet, current

computer magazines, books and by interviewing computer professionals.

CXC A19/U2/08

UNIT 2

MODULE 3: OPERATING SYSTEMS AND COMPUTER NETWORKS (cont’d)

RESOURCES

Ritchie, C. Operating Systems Incorporating UNIX and Windows, London:

Letts Educational, 2003.

Shelly, G., Ashman, T. and

Vermaat, M.

Discovering Computers 2008, Boston: Thomson Course

Technology, 2008.

CXC A19/U2/08

◆

OUTLINE OF ASSESSMENT

Each Unit of the syllabus will be assessed separately. The scheme of assessment for each Unit will be the

same. Candidate’s performance on each Unit will be reported as an overall grade and a grade on each Module

of the Unit. The assessment will comprise two components, one external and one internal.

EXTERNAL ASSESSMENT (80%)

Paper 01

(1 hour 30 minutes)

Forty-five multiple-choice items, fifteen (15) from each Module.

Each item is worth 1 mark.

(30%)

Paper 02

(2 hours 30 minutes)

Six questions, two from each Module. Candidates will be

expected to answer all questions.

(50%)

INTERNAL ASSESSMENT (20%)

Paper 03A

The Internal Assessment for each unit is compulsory.

Un i t 1 : F u n d a me n t a l s o f C o m p u te r S c ie n c e

Candidates are expected to choose a problem for which a software solution is appropriate and create an algorithm for the

solution using sequencing, selection, assignments, iteration (bounded and unbounded). They should represent their

algorithms using any combination of narrative, flow charts and pseudocode. Candidates are expected to implement the

algorithm in C using arrays with no less than five functions and create a test plan.

Un i t 2 : F u r t h e r T o p i c s i n C o m p u t e r Sc i e nc e

Candidates are expected to choose a problem for which a software solution exists and then develop the software using

software engineering techniques. In particular, they are expected to demonstrate the tools and techniques used in the analysis

of the software to be developed. They are then expected to design, code, and test their software using appropriate

techniques.

Paper 03B

Private candidates are required to write an Alternative Paper to the Internal Assessment Paper.

CXC A19/U2/08

MODERATION OF INTERNAL ASSESSMENT

An Internal Assessment Record Sheet will be sent each year to schools submitting students for the examinations.

All Internal Assessment Record Sheets must be submitted to CXC by May 31 of each year of the examination. A sample

of assignments will be requested by CXC for moderation purposes. These samples will be re-assessed by CXC Examiners

who moderate the Internal Assessment. Teachers’ marks may be adjusted as a result of moderation. The Examiners’

comments will be sent to teachers.

Copies of the students’ submissions must be retained by the school until three months after publication by CXC of the

examination results.

ASSESSMENT DETAILS

External Assessment

Paper 01 and Paper 02

The external assessment consists of two written papers. They are externally set and externally assessed. Together they

contribute 80% of the final mark.

P ap e r 0 1 (1 h o u r 3 0 m inu t e s)

Co m p o s i t i o n o f t h e P a p e r

The paper will consist of forty-five (45) multiple-choice items, fifteen (15) from each Module. All

questions are compulsory and knowledge of the entire Syllabus is expected. The paper will assess the

candidate’s knowledge across the breadth of the Syllabus.

M a r k A l l o c a t i o n

The paper is worth 45 marks, with each question being allocated 1 mark.

3 .

Qu e s t i o n T y p e

Questions may be presented using diagrams, data, graphs, prose or other stimulus material.

CXC A19/U2/08

P ap e r 0 2 ( 2 h o u r s 3 0 m inu t e s )

Co m p o s i t i o n o f P a p e r

This paper consists of six questions, two from each Module, arranged in three sections. Candidates are required

to do all questions in each section.

M a r k A l l o c a t i o n

This paper is worth 150 marks, each question is worth 25 marks.

Qu e s t i o n T y p e

Each question may present a situation related to a specific topic in the syllabus and consists of three or four sub-

questions. The required responses to a sub-question may range in length.

A w a r d o f m a r k s

Marks will be awarded for knowledge and comprehension, application and analysis and synthesis and

evaluation.

Internal Assessment (20% of Total Assessment)

Internal Assessment is an integral part of student assessment in the course covered by this syllabus. It is

intended to assist students in acquiring certain knowledge, skills and attitudes that are associated with the

subject. The activities for the Internal Assessment are linked to the syllabus and should form part of the

learning activities to enable the student to achieve the objectives of the syllabus.

During the course of study for the subject, students obtain marks for the competence they develop and

demonstrate in undertaking their Internal Assessment assignments. These marks contribute to the final marks

and grades that are awarded to students for their performance in the examination.

The guidelines provided in this syllabus for selecting appropriate tasks are intended to assist teachers and

students in selecting assignments that are valid for the purpose of Internal Assessment. The guidelines provided

for the assessment of these assignments are intended to assist teachers in awarding marks that are reliable

estimates of the achievement of students in the Internal Assessment component of the course. In order to

ensure that the scores awarded by the teachers are not out of line with the CXC standards, the Council

undertakes the moderation of a sample of the Internal

Assessment assignments marked by each teacher.

Internal Assessment provides an opportunity to individualise a part of the curriculum to meet the needs of

students. It facilitates feedback to the student at various stages of the experience. This helps to build the self-

confidence of students as they proceed with their studies. Internal Assessment also facilitates the

development of critical skills and abilities emphasised by this CAPE subject and enhances the validity of the

examination on which candidate performance is reported. Internal Assessment, therefore, makes a significant

and unique contribution to both the developme

nt of relevant skills and the testing and rewarding of students

for the development of those skills.

CXC A19/U2/08

The Caribbean Examinations Council seeks to ensure that the Internal Assessment scores are valid and

reliable estimates of accomplishment. The guidelines provided in this syllabus are intended to assist in doing

so.

Each candidate’s total Internal Assessment mark for any Unit should be divided in three and allocated to each

Module equally.

Fractional marks should not be awarded. Wherever the Unit mark is not divisible by three, then

(a) when the remainder is 1 mark, it should be allocated to Module 1

(b) when the remainder is 2, one of the marks should be a

llocated to Module 2 and the other mark to

Module 3.

Paper 03A

UNIT 1: Fundamentals of Computer Science

1. The aims of the project are to:

(i) develop candidate’s personal insights into the fundaments of Computer science;

(ii) provide opportunities for all candidates to show, with confidence, that they have mastered

the syllabus.

2. Requirements

Each candidate is expected to choose a problem for which a software solution is appropriate and

create algorithms for the solution using sequencing, selection, assignments, and iteration (bounded

and unbounded). They should represent their algorithms using narrative format and either flow

charts or pseudocode. Candidates are expected to implement their algorithms as C programs using

arrays with no less than five functions and using at least one file. They must also create a test plan

that is used to test their implemented programs for correctness.

3. Integration of Project into the course

(i) The activities related to Project work should be integrated into the course so as to enable

candidates to learn and practise the skills of undertaking a successful project.

(ii) Some time in class should be allocated for general discussion of project work. For example,

discussion of how data should be collected, how data should

be analysed and how data

should be presented.

(iii) Class time should also be allocated for discussion between teacher and student, and student

and student.

CXC A19/U2/08

4. Management of Project

(i) Planning

An early start to planning project work is highly recommended and the schedule of the dates

for submission should be developed by teachers and candidates.

(ii) Length

The length of the report of the project should be between 1500 and 2000 words excluding

diagrams, graphs, tables and bibliographies.

(iii) Guidance

Each candidate should know the requirements of the project and its assessment process.

Although candidates may consult with resource persons besides the teacher the candidates

submission should be his or her own work.

Candidates are not expected to work on their own. The teacher is expected to give

appropriate guidance at all stages of project work, for example, chapters to read, alternative

procedures to follow and other sources of information.

(iv) Authenticity

Teachers are required to ensure that all projects are the candidates’ work.

A recommended procedure is to:

(a) engage candidates in discussion;

(b) ask candidates to describe procedures used and summarise findings either orally or

written;

CXC A19/U2/08

ASSESSMENT CRITERIA FOR THE PROJECT

General

It is recommended that candidates be provided with an assessment criteria before commencing the project.

(i) The following aspects of the project will be assessed:

(a) Definition of problem;

(b) Narrative and flow charts or pseudocode;

(d) Testing and presentation;

(e) Communication of Information.

(ii) For each component, the aim is to find the level of achievement reached by the candidate.

(iii) For each component, only whole numbers should be awarded.

(iv) It is recommended that the assessment criteria be available to candidates at all times.

CXC A19/U2/08

CRITERIA FOR MARKING INTERNAL ASSESSMENT PROJECT

The project will be graded out of a total of 60 marks and marks will be allocated to each task as outlined

below. Candidates will be awarded marks for communicating information in a logical way using correct grammar. These

marks are awarded under Task 5.0 below.

1. Definition of Problem [4]

•

Complete and accurate description of the problem

•

Partial description of the problem

3-4

1-2

2. Narrative and Flow Charts or Pseudocode [15]

•

Algorithms expressed in narrative format (4)

Narrative is an accurate description of a solution

Narrative is a partially correct description of a solution

3-4

1-2

•

Algorithms expressed as flow charts or pseudocode (6)

Flow chart/Pseudocode is logical, easy to follow and is an accurate

description of a solution using the appropriate symbols or algorithmic

structures

Flow chart/Pseudocode is organised, easy to follow for the most part, and

is a clear description of a solution using the appropriate symbols or

algorithmic structures

Flow chart/Pseudocode is not well organised, and description of solution

lacks clarity

5-6

3-4

1-2

•

Demonstration of structured programming concepts (5)

Program displays excellent use of structured programming concepts

Program displays good use of structured programming concepts

Program displays limited use of structured programming concepts

3-4

1-2

3. Coding of Program [25]

•

Structured decomposition using functions (6)

Overall, program comprises functions as independent units

Program comprises most functions as independent units

Program comprises some functions as independent units

5-6

3-4

1-2

•

Use of appropriate data structures (6)

Data structure chosen were appropriate for the problem definition

Data structure chosen were reasonable but not appropriate

Data structure chosen were inappropriate

5-6

3-4

1-2

•

Demonstration of the concept of structured programming (3)

Evidence of looping structures 3

Evidence of conditional statements 2

Evidence of other structures (for example assignment, input, output) 1

•

Appropriate programming style and documentation (4)

Appropriate document in significant areas

Standard indentation of code

3-4

1-2

CXC A19/U2/08

•

Evidence that code matches algorithm (4)

Code matches sequencing of algorithm 4

Code matches MOST of the sequencing of algorithm 3

Code matches SOME of the sequencing of algorithm 2

Sequencing of code inconsistent with algorithm 1

•

Evidence of file manipulation (2)

Correct file types used, for example, text, binary, sequential, random 2

File used appropriately 1

4. Testing and presentation [11]

•

Test Plan

Test Plan with exhaustive data set

Test Plan with acceptable data set

Test Plan with minimal data set

•

Test Results

Normal input giving correct results

Extreme input giving correct results or appropriate error message

Erroneous input (for example, text when number required) giving

appropriate error message

Incomplete input giving appropriate message

•

Overall presentation

Appropriate cover page

Use of table of contents

Sequencing in document easy to follow

(3)

(5)

3-4

(3)

5. Communication of Information [5]

•

Communicates information in a logical way using correct grammar and

appropriate jargon ALL of the time

•

Communicates information in a logical way using correct grammar and

appropriate jargon MOST of the time

•

Communicates information in a logical way using correct grammar and

appropriate jargon MOST of the time

4-5

2-3

TOTAL 60

UNIT 2: Further Topics in Computer Science

1. The aims of the project are to:

(i) develop candidate’s personal insights into further topics in Computer Science;

(ii) provide opportunities for all candidates to show, with confidence, that they have mastered

the syllabus.

CXC A19/U2/08

2. Requirements

Each candidate is expected to choose a problem for which a software solution exists and then develop the software

using software engineering techniques. In particular, the candidate is expected to demonstrate appropriate choice

of the tools and techniques used in the analysis of the software to be developed. They are then expected to design,

code, and test their software using appropriate techniques.

The following are examples of the kinds of projects that students can develop for the Internal Assessment:

(i) simple process scheduler for an operating system;

(ii) vehicle parking system to allocate spaces to vehicles in a parking lot;

(iii) system to manage a CD/DVD collection;

(iv) student registration system to keep track of student information, course grades and registration details.

3. Integration of Project into the course

(i) The activities related to Project work should be integrated into the course so as to enable

candidates to learn and practise the skills of undertaking a successful project.

(ii) Some time in class should be allocated for general discussion of project work. For example,

discussion of how data should be collected, how data should be analysed and how data

should be presented.

(iii) Class time should also be allocated for discussion between teacher and student, and student

and student.

4. Management of Project

(i) Planning

An early start to planning project work is highly recommended and the schedule of the dates

for submission should be developed by teachers and candidates.

(ii) Length

The length of the report of the project should be between 1500 and 2000 words excluding

diagrams, graphs, tables and bibliographies.

(iii) Guidance

Each candidate should know the requirements of the project and its assessment process.

Although candidates may consult with resource persons besides the teacher the candidates

submission should be his or her own work.

CXC A19/U2/08

Candidates are not expected to work on their own. The teacher is expected to give

appropriate guidance at all stages of project work, for example, chapters to read, alternative

procedures to follow and other sources of information.

(iv) Authenticity

Teachers are required to ensure that all projects are the candidates’ work.

A recommended procedure is to:

(a) engage candidates in discussion;

(b) ask candidates to describe procedures used and summarise findings either orally or

written;

ASSESSMENT CRITERIA FOR THE PROJECT

General

It is recommended that candidates be provided with an assessment criteria before commencing the project.

(i) The following aspects of the project will be assessed:

(a) Specification of requirements;

(b) Design Specification;

(d) Communication of Information.

(v) For each component, the aim is to find the level of achievement reached by the candidate.

(vi) For each component, only whole numbers should be awarded.

(vii) It is recommended that the assessment criteria be available to candidates at all times.

CXC A19/U2/08

CRITERIA FOR MARKING INTERNAL ASSESSMENT

Candidates will be awarded a total of 5 marks for communicating information in a logical way using correct

grammar. The marks are awarded as shown in the mark scheme below.

1. Specification of requirements [25]

•

Definition of problem (5)

Complete accurate description of the problem

Generally accurate description for the problem

Partially accurate description for the problem

Weak description for the problem

1-2

•

Techniques of analysis used (5)

Sound and relevant techniques used

Mostly sound and relevant techniques

Techniques were partially sound and relevance was limited

3-4

1-2

•

Use of Data Flow diagrams and E-R diagrams (9)

Data Flow Diagrams (DFD)

Context Level

Complete and accurate diagram of all relevant entities, data flows.

Accurate diagram of most relevant entities, data flows.

Accurate diagram of few relevant entities, data flows.

(3)

•

Level 1 Diagram

Complete and accurate diagram of all relevant processes, data flows and

major data stores

Accurate diagram of most relevant processes, data flows and major data

stores

Accurate diagram of few relevant processes, data flows and major data

stores

(3)

•

Entity Relation Diagram (ERD)

Complete and accurate diagram of all relevant entities and relationships.

Accurate diagram of most relevant entities and relationships.

Accurate diagram of few relevant entities and relationships

(3)

•

Functional and non-functional requirements (6)

Functional Requirements

Complete and accurate description of all requirements

Complete and accurate description of most requirements

Complete and accurate description of few requirements

(3)

CXC A19/U2/08

•

Non Functional Requirements

Complete and accurate description of all requirements

Accurate description of most requirements

Accurate description of some requirements

(3)

2. Design Specification [14]

•

System structuring (4)

Complete and accurate diagram of all processes 4

Accurate diagram of most processes 3

Accurate diagram of some processes 2

Accurate diagram of few processes 1

•

User interface design (2)

Thorough analysis and appropriate justification of interface design 2

Partial analysis and justification of interface design 1

•

Report design (2)

Appropriate and well implemented 2

Generally appropriate and satisfactorily implemented 1

•

Algorithm design (3)

Appropriate and well implemented algorithm design 3

Generally appropriate algorithm design 2

General understanding of algorithm design 1

•

Choice of appropriate data structures (3)

Appropriate and well implemented 3

Generally appropriate 2

Partially appropriate and implementation was limited 1

3. Coding and Testing [15]

•

Code achieves functionality

Code achieved functionality (documentation, error trapping, correct

output, usability and reporting)

Code achieved some functionality (documentation, error trapping,

correct output, usability and reporting)

Functionality was limited

(5)

3-4

1-2

•

Code corresponds to design

Code achieves all the design specifications

Code achieves most of the design specifications

Code achieves few of the design specifications

(5)

3-4

1-2

•

Test plans

Test Plan with exhaustive data set

Test Plan with acceptable data set

Test Plan with minimal data set

(5)

3-4

1-2

CXC A19/U2/08

4. Communication and Presentation [6]

•

Communicates information in a logical way using correct grammar and

appropriate jargon ALL of the time

•

Communicates information in a logical way using correct grammar and

appropriate jargon MOST of the time

•

Communicates information in a logical way using correct grammar and

appropriate jargon SOME of the time

5-6

3-4

1-2

TOTAL 60

◆

REGULATIONS FOR PRIVATE CANDIDATES

Candidates who are registered privately will be required to sit Paper 01, Paper 02 and Paper 03B. Paper 03B will test

the student’s acquisition of the skills in the same areas of the syllabus identified for the internal assessment. Consequently,

candidates are advised to undertake a project similar to the project that the school candidates would normally complete

and submit for internal assessment. It should be noted that private candidates would not be required to submit a project

document.

◆

REGULATIONS FOR RESIT CANDIDATES

Resit candidates must rewrite Papers 01 and 02 of the examination for the year in which they re-register. Resit candidates

may elect not to repeat the Internal Assessment component provided they rewrite the examination no later than two years

following their first attempt.

Resit candidates must be entered through a school, approved educational institution, or the Local Registrar’s Office.

CXC A19/U2/08

◆

ASSESSMENT GRID

The Assessment Grid for each Unit showing marks assigned to each paper and to Modules, and the

percentage contribution of each paper to the total scores.

Papers Module 1 Module 2 Module 3 Total (%)

External Assessment

(15)

(45)

(30)

Paper 01

Multiple Choice

(1 hour 30 minutes)

Weighting

Paper 02

Essay

(2 hours 30 minutes)

150

(50)

Internal Assessment

Paper 03A

Paper 03B

(1 hour 30 minutes)

(20)

Total

100

300

(100)

◆

LOGIC SYMBOLS

p, q, r propositions

∧

conjunction

∨

(inclusive) disjunction

∼

negation

→

conditionality

↔

bi-conditionality

implication

⇔

equivalence

AND gate

OR gate

NOT gate

T, 1 true

F, false

CXC A19/U2/08

◆

GLOSSARY

WORD DEFINITION/MEANING NOTES

analyse examine in detail

annotate add a brief note to a label Simple phrase or a few words only.

apply use knowledge/principles to solve problems Make inferences/conclusions.

assess present reasons for the importance of particular

structures, relationships or processes

Compare the advantages and

disadvantages or the merits and

demerits of a particular structure,

relationship or process.

calculate arrive at the solution to a numerical problem Steps should be shown; units must be

included.

classify divide into groups according to observable

characteristics

comment state opinion or view with supporting reasons

compare state similarities and differences An explanation of the significance of

each similarity and difference stated

may be required for comparisons

which are other than structural.

construct use a specific format to make and/or draw a graph,

histogram, pie chart or other representation using

data or material provided or drawn from practical

investigations, build (for example, a model), draw

scale diagram

Such representations should normally

bear a title, appropriate headings and

legend.

deduce make a logical connection between two or more

pieces of information; use data to arrive at a

concl

usion

define state concisely the meaning of a word or term This should include the defining

equation/formula where relevant.

demonstrate show; direct attention to...

derive to deduce, determine or extract from data by a set

of logical steps some relationship, formula or result

This relationship etc., may be general

or specific.

CXC A19/U2/08

WORD

DEFINITION /MEANING

NOTES

describe provide detailed factual information of the

appearance or arrangement of a specific structure or

a sequence of a specific process

Description may be in words,

drawings or diagrams or any

appropriate combination. Drawings

or diagrams should be annotated to

show appropriate detail where

necessary.

determine find the value of a physical quantity

design plan and present with appropriate practical detail Where hypotheses are stated or when

tests are to be conducted, possible

outcomes should be clearly stated

and/or the way in which data will be

analyzed and presented.

develop expand or elaborate an idea or argument with

supporting reasons

diagram simplified representation showing the relationship

between components.

differentiate/ state or explain briefly those

distinguish differences between or among items

(between/ which can be used to define the

among) items or place them into separate categories.

discuss present reasoned argument; consider points both

for and against; explain the relative merits of a case

draw make a line representation from specimens or

apparatus which shows an accurate relation

between the parts

In the case of drawings from

specimens, the magnification must

always be s

tated.

estimate make an approximate quantitative judgement

evaluate weigh evidence and make judgements based on

given criteria

The use of logical supporting reasons

for a particular point of view is more

important than the view held; usually

both sides of an argument should be

considered.

explain give reasons based on recall; account for

find locate a feature or obtain as from a graph

CXC A19/U2/08

WORD

DEFINITION /MEANING

NOTES

formulate devise a hypothesis

identify name or point out specific components or features

illustrate show clearly by using appropriate examples or

diagrams, sketches

interpret explain the meaning of

justify explain the correctness of

investigate use simple systematic procedures to observe, record

data and draw logical conclusions

label add names to identify structures or parts indicated

by pointers

list itemise without detail

measure take accurate quantitative readings using

appropriate instruments

name give only the name of No additional information is

required.

note write down observations

observe pay attention to details which characterise a

specimen, reaction or change taking place; to

examine and note scientifically

Observations may involve all the

senses and/or extensions of them but

would normally exclude the sense of

taste.

outline give basic steps only

plan prepare to conduct an investigation

predict use information provided to arrive at a likely

conclusion or suggest a possible outcome

record write an accurate description of the full range of

observations made during a given procedure

This includes the values for any

variable being investigated; where

appropriate, recorded data may be

depicted in graphs, histograms or

tables.

CXC A19/U2/08

WORD

DEFINITION /MEANING

NOTES

relate show connections between; explain how one set of

facts or data depend on others or are determined by

them

sketch make a simple freehand diagram showing relevant

proportions and any important details

state provide factual information in concise terms

outlining explanations

suggest offer an explanation deduced from information

provided or previous knowledge. (... a hypothesis;

provide a generalization which offers a likely

explanation for a set of data or observations.)

No correct or incorrect solution is

presumed but suggestions must be

acceptable within the limits of

scientific knowledge.

test to find out, following set procedures

Wes t e r n Z o n e O f f i c e

200 8 / 06/02

CXC A19/U2/08

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

ADVANCED PROFICIENCY EXAMINATION

SPECIMEN PAPER

MULTIPLE CHOICE QUESTIONS

FOR

COMPUTER SCIENCE - UNIT 1

Paper 01

READ THE FOLLOWING DIRECTIONS CAREFULLY

Each item in this test has four suggested answers lettered (A), (B), (C), (D). Read each item you are

about to answer and decide which choice is best.

Sample Item

Which of the following is a function of process management in an operating system?

Sample Answer

(A)

Repairing disk errors

(B)

Scheduling

A B C D

(D) Partitioning

The best answer to this item is “Scheduling”, so answer space (B) has been shaded.

There are 21 items in this specimen paper. However, the Paper 01 test consists of 45 items.

You will have 120 minutes to answer them.

02115010/SPEC 2008

- 2 -

GO ON TO THE NEXT PAGE

02115010/SPEC 2008

Which of the following descriptions BEST

suits the function of Cache memory?

(A) It gives the CPU more rapid access

to data.

(B) It increases data transfer rate

between computer and printer.

hard disk.

(D) It stores the operating system when

the computer “boots”.

Which of the following devices would help

prevent data loss in the event of an electrical

outage?

(A) Power strip

(B) Surge protector

(D) Voltage regulator

A computer’s word size is the

(A) length of an instruction

(B) maximum number of characters in a

typed word

memory

(D) number of bits that the CPU can

manipulate at one time

. The 8-bit two’s complement representation

of the decimal numeral -39 is:

(A) 00100110

(B) 10100101

(D) 11100110

The 8-bit Sign & Magnitude representation

of the decimal numeral -25 is:

(A) 10011001

(B) 10100101

(D) 11100110

What is the purpose of the program counter

in a microcomputer?

(A) To determine how many programs

can be opened at one time

(B) To determine the sequence in which

the program instructions are to be

executed.

instruction executed

(D) To keep a count of the number of

instructions in memory

Which of the followingis NOT true of a flip-

flop?

(A) It has two inputs.

(B) It has two outputs.

(D) It can function as a 1-bit memory.

. The list I - VI represent steps in the

computer-based problem solving process.

I. Analyze the problem

II. Code the solution

III. Define the problem

IV. Develop an algorithm

V. Maintain the program

VI. Test and debug the program

Which of the following represents the

correct sequence of steps in the computer-

based problem-solving process?

(A) I, II, III, IV, V, VI

(B) III, I, IV, II, VI, V

(D) III, I, VI, II, V, VI

- 3 -

GO ON TO THE NEXT PAGE

02115010/SPEC 2008

Items 9 - 10 refer to the following algorithm.

Num = 5

For I = 1 to Num do

Print I *2

EndFor

. This algorithm is an example of

(A) unbounded iteration

(B) recursion

(D) bounded iteration

10.

What is the output of the algorithm?

(A) 1 1

(B) 1 2 3 4 5

(D) I*2 I*2 I*2 I*2 I*2

11.

What is the technical term for the graphical

representation of an algorithm?

(A) Flowchart

(B) HIPO chart

(D) Pseudocode

. Which of the following are properties of a

well-designed algorithm?

I. A finite number of steps

II. Flow of control from one process to

the next

III Correct syntax

IV. Ambiguous instructions

(A) I and II only

(B) I and III only

(D) II, III and IV only

. Which of the following statements are true

about programs and algorithms?

I. Every program uses algorithms.

II. All programs are algorithms.

III. Algorithms are derived from

programs.

(A) I and II only

(B) I and III only

(D) I, II and III

. What are the values of p and r after

execution of the following algorithm?

p = 8

q = 4

r = 8

if (p > q) AND (q > r)

p = + 1

else

r = r - 1

end if

(A) p = 7, r = 8

(B) p = 9, r = 7

(D) p = 8, r = 7

. What is the error in the following C program

which is intended to print the string constant

‘C programming is fun’?

#include <stdio.h>

{printf(“C programming is fun”); }

(A) The brackets ( and ) should be on

different lines.

(B) The function main is missing.

different lines.

(D) Single quotation marks ‘ and ‘ should

be used.

02115010/SPEC 2008

- 4 -

IF YOU FINISH BEFORE TIME IS CALLED, CHECK YOUR WORK ON THIS TEST.

. Programming languages may be classified by

Generation and Paradigm among other things.

Which of the following is NOT an example

of a programming language paradigm?

(A) Imperative

(B) Functional

(D) Declarative

Item

refers to the following algorithm.

#include <stdio.h>

main ( )

{

int a, b, sum;

a = 10;

b = 20;

sum = a + b;

printf(“The sum of the numbers is

%d/n”, sum);

}

. The program line: sum = a + b is an example

(A) a function call

(B) a C operator

(D) the C assignment

. What output would you expect from the

following program?

#include <stdio.h>

main ( )

{

printf(“One——”);

printf(“Two——”);

printf(“Three\n”);

}

(A) One——Two——Three

(B) One——, Two——, Three

Two——

Three\n

(D) One——, Two——, Three\n

. Which of the following programs translates

all program instructions at one time and

produces a stand-alone object program that

can be executed on its own?

(A) An interpreter

(B) A compiler

(D) A generator

. The compilation process can be broken up

into THREE main stages. The CORRECT

order of the stages is

(A) syntax analysis, lexical analysis, code

generation

(B) lexical analysis, code generation,

syntax analysis

code generation

(D) code generation, lexical analysis,

syntax analysis

. Consider the following partial declaration in

#include <stdio.h>

#define SIZE

main ( )

{

int k;

int table [SIZE];

Which of the following lines of code will read

FIVE integers from the standard input?

(A) for (k = 0; k < SIZE; k++)

printf (“%d”, table [k]);

(B) for (k = 0; k < SIZE; k--)

printf (“%d”, table [k]);

Scanf (“%d”, table [k]);

(D) for (k = 0; k < SIZE; k++)

scanf (“%d”, table [k]);

02215010 MS/SPEC/2008

FORM SPEC 02215010/SPEC/2008

CARIBBEAN EXAMINATIONS COUNCIL

ADVANCED PROFICIENCY EXAMINATION

COMPUTER SCIENCE

SPECIMEN PAPER

UNIT 1 - FUNDAMETALS OF COMPUTER SCIENCE

PAPER 02

2 ½ hours

INSTRUCTIONS TO CANDIDATES

1. DO NOT open this examination paper until instructed to do so.

2. Answer ALL questions from the THREE sections

02215010 MS/SPEC/2008

SECTION A

MODULE 1: COMPUTER ARCHITECTURE AND ORGANISATION

Answer BOTH questions.

1. (a) Differentiate between EACH of the following pairs as they pertain to

computers and information systems:

(i) ROM and EPROM [2 marks]

(ii) EPROM and EEPROM [2 marks]

(iii) MAN and LAN [2 marks]

(iv) Mainframe computer and microcomputer [2 marks]

(v) Workstation and supercomputer [2 marks]

(vi) Cache memory and RAM [2 marks]

(b) State how EACH of the following devices can help protect and preserve the

resources of a computer:

(i) Surge protector [4 marks]

(ii) (Uninterruptible Power Supply) UPS [2 marks]

(i) The eight-bit signed-magnitude representation of +10 using sign +

magnitude. [2 marks]

(iii) The decimal of 00001101 (binary). [1 mark]

(d) Given the number 8.312 x 10

in standard form:

(i) Identify the exponent [1 mark]

(ii) Identify the base or radix [1 mark]

(e) Determine if the result of 0011 + 1100 can be stored as a four bit binary

number.

[2 marks]

Total 25 marks

02215010 MS/SPEC/2008

2. (a) Briefly explain what is meant by the term ‘truth table’. [2 marks]

(b) Name the THREE primary logic gates. [3 marks]

EACH gate and give the corresponding two input truth table for the gate.

[6 marks]

(d) Draw the minimum logic diagram to show how these primary logic gates can

be used to build the Exclusive – OR. [4 marks]

(e) (i) Name the TWO major components of the Central Processing Unit

(CPU). [2 marks]

(ii) For EACH of the components named in (e) (i) above briefly describe

its function. [4 marks]

(iii) Explain what is meant by EACH of the following terms:

(a) Word length [2 marks]

(b) Clock speed [2 marks]

TOTAL 25 marks

SECTION B

MODULE 2: PROBLEM-SOLVING WITH COMPUTERS

Answer BOTH questions

3. (a) Define the term ‘algorithm’. [2 marks]

(b) Structured algorithms and programs are designed using three basic control

constructs. Name and briefly describe EACH of these constructs.

[6 marks]

sections that fit together depending on the type of problem to be fixed.

(i) List FOUR stages of the problem solving process. [4 marks]

(ii) Select THREE of the stages listed in (c) (i) and describe EACH stage.

[9 marks]

02215010 MS/SPEC/2008

(d) The following algorithm is supposed to print the squares on the numbers

between 1 and 100 (inclusive). However, it is not generating the derived

output. Find and correct the problem in the algorithm.

Start = 1

x = 1

while start c = 100

print start * start

x = x + 1

endwhile [4 marks]

TOTAL 25 marks

4. (a) A soft-drink manufacturer sells five soft-drink flavours: bananas, cherry,

mango, orange and pineapple. The company knows that banana and pineapple

are the two best-selling flavours and is carrying out a poll to determine which

of these two flavours is the more popular among its customers

Write an algorithm to find the MORE popular flavour and the NUMBER of

votes it obtained. Assume that 100 customers participate in the poll and that

there is NO tie. [15 marks]

(b) Trace through the execution of the following algorithm and draw the output in

your answer booklet exactly as would be generated by the algorithm. You

should carefully note the following;

 printSpaces(n) prints n spaces from the current cursor position

 print continues output on the current line form the current cursor position

 println causes output on the current line to be terminated, and subsequent

output begins on a new line.

set SIZE = 10

for i = 1 to SIZE

print (“*”)

endfor

println ()

for i = 1 to 3

print (“*”)

printSpaces (SIZE – 2)

printlin (“*”)

endfor

for i = 1 to SIZE

print (“*”)

endfor

println () [10 marks]

TOTAL 25 marks

02215010 MS/SPEC/2008

SECTION C

MODULE 3: PROGRAMMING

Answer BOTH questions from this section

5 (a) You are employed as a programmer by a software house. You want to learn a

new programming language at a nearby community college. You would like

the company to pay the cost of the course. Unfortunately, your supervisor

sees no reason for the company to pay for the course. His argument is that

you already are well versed in one programming language, say PL. He sees no

reason for you to learn another. However, he has suggested that you write to

the managing director to get approval to do this course. What arguments

could you use to convince the managing director that knowledge of a single

programming language is not sufficient? [10 marks]

(b) Explain the concept of ‘structured programming.’ [5 marks]

[4 marks]

(d) Assume that you have been given a library implementing a list of positive

integers. The library supplies you with the following functions:

is_empty

first_element

rest

All functions take a list as input. The first function returns true if the list is

empty (contains no elements) and false otherwise; the second function returns

the first element in the list (assuming of course that the list is not empty); the

third function returns a list consisting of all the elements of the input list

except for the first element. Thus, if we pass the list [1 2 3 4] into the function

rest, then we get the list [2 3 4] as output.

Write an iterative function to count the number of elements in the list.

[6 marks]

Total 25 marks

02215010 MS/SPEC/2008

6. (a) Describe FOUR major stages in the compilation process. [12 marks]

(b) An integer array, P, contains m positive integer values. The array is sorted in

ascending order. An integer array, Q, contains n positive integer values. This

array is also sorted in ascending order. Write code in procedural C to merge

the integers in P with the integers in Q. The result of the merge is placed into

another integer array, R, such that the integers in R are in ascending order.

Show ALL relevant declarations. Assume that P and Q are already loaded

with integers. [13 marks]

TOTAL 25 marks

END OF TEST

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

ADVANCED PROFICIENCY EXAMINATION

SPECIMEN PAPER

MULTIPLE CHOICE QUESTIONS

FOR

COMPUTER SCIENCE - UNIT 2

Paper 01

READ THE FOLLOWING DIRECTIONS CAREFULLY

Each item in this test has four suggested answers lettered (A), (B), (C), (D). Read each item you are

about to answer and decide which choice is best.

Sample Item

Which of the following statements is true about objects and classes?

(A) Objects may contain only methods, while

Sample Answer

classes

can have both variables and methods

(B)

Classes may only contain methods, while

A B C D

objects

can have both variables and methods

(D) An object is an instance of a class

The best answer to this item is “An object is an instance of a class”, so answer space (D) has been

shaded.

There are 21 items in this specimen paper. However, the Paper 01 test consists of 45 items.

You will have 120 minutes to answer them.



02215010/SPEC 2008

- 2 -

GO ON TO THE NEXT PAGE

02215010/SPEC2008

. An ADT is

(A) the logical properties of data

(B) a combination of data types within a

record

associated operations

(D) a list of variables

. Which ADT is BEST described as a LIFO

structure?

(A) A queue

(B) A variable

(D) A pointer

Item

refers to the following queue of jobs

waiting to be printed.

Front Rear

What would the queue look like when

three

jobs have been printed and

two new

jobs, Job6

and Job7 have joined the queue?

(A)

Front Rear

(B)

Front Rear

(C)

Front Rear

(D)

Front Rear

Consider a circular Q with a pointer

front

that references the location of the front of

the queue. Assuming that the locations range

from 0 to 4, what is the value of

front

after

performing the following operations on the

queue Q?

enqueue

dequeue

enqueue

dequeue

enqueue

(A) 0

(B) 1

(D) 3

Item

refers to the following pseudocode.

x = 1;

i = 1;

while (x <10)

begin

x = 2x;

i = i + 1;

end;

. What is the value of i at the end of the

pseudocode?

(A) 4

(B) 5

(D) 7

. Which of the following combinations is a stack

operation?

(A) Enqueue; push;

(B) Dequeue; pop;

(D) Push; pop;

Job1

Job2

Job3

Job4

Job5

Job3

Job4

Job5

Job6

Job7

Job4

Job5

Job6

Job7

Job6

Job1

Job2

Job7

Job6

Job1

Job2

Job3

- 3 -

GO ON TO THE NEXT PAGE

02215010/SPEC 2008

. Which of the following is TRUE of the

waterfall approach to the software

development process?

I. Cascades from one phase of the

process to another.

II. The following stage should not start

before the previous has finished.

III. The following stage starts before the

previous stage has finished.

IV. The project is partitioned into distinct

stages before the phases begin.

V. This approach caters to changing

user requirements.

(A) I, II and IV only

(B) I, II and V only

(D) III, IV and V only

. The software process model defined by its

approach of refining an initial system based on

customer input is the

(A) waterfall approach

(B) evolutionary development

(D) reuse-oriented approach

. What are the TWO phases of the

requirements engineering process (software

specification)?

(A) Feasibility study and user and systems

requirements documentation

(B) Feasbility study and production of

interface specification

structure specification

(D) Sub-system testing and maintenance

. Which of the following system models depict

data transformation as data is processed?

(A) Data dictionaries

(B) Entity-relationship diagrams

(D) Data flow diagrams

. The binary search is BEST described as

(A) a search of only a section of a given

list of elements

(B) a search which takes a divide and

conquer approach

short list

(D) a search of un-order elements

. Which of the following terms represents types

of software maintenance?

I. Adaptive

II. Corrective

III. Defective

IV. Perfective

(A) I, II and III only

(B) I, II and IV only

(D) II, III and IV only

. The MAIN goal of component testing is to

(A) expose faults in the components

(B) demonstrate that the system meets

its requirements

that the system meets

requirements

(D) test how one component interacts

with others in the system

. Which of the following is NOT a tool in a

software testing workbench?

(A) Test data generator

(B) Report generator

(D) Test manager

02215010/SPEC 2008

- 4 -

IF YOU FINISH BEFORE TIME IS CALLED, CHECK YOUR WORK ON THIS TEST.

. Which transmission media is BEST suited for

transfer of large volumes of data over

continents?

(A) Fibre optic cables

(B) Microwave

(D) Satellite

. Which of the following sequences of terms

are the MOST related to each other?

(A) Coaxial cables, electronic pulses,

strands of glass

(B) Fiber optic cable, light pulses, LAN

telephone

(D) Twisted pair cable, fast and noise

resistant, voice and data

transmission

. Which of the following BEST represents

the benefits of ALL types of network?

(A) Sharing of peripheral devices, sharing

of programs and data, better

communication, increased access

to database

(B) Sharing of peripheral devices,

sharing of programs and data,

communication over a wide

geographical area, increased

access to databases

sharing of programs and data,

improved data security, increased

access to databases

(D) Sharing of programs and data

communication over a wide

geographical area, improved data

security, increased access to

databases

18.

What is the name of the device that uses the

half-duplex method of transmission?

(A) Television

(B) Radio

(D)

Telephone

. Accessing a website via a web browser is

an example of which type of network

configuration?

(A) Peer-to-peer

(B) Client-server

(D) Server-to-server

. The use of a switched line for data

communications offers the user

(A) greater flexibility than does the use

of a leased line

(B) the ability to connect to other

computers without a modem

private line

(D) access to more facilities than the use

of dial-up

Item

refers to the following statements.

I. Smaller computer systems

communicate with one another

through the host.

II. A signal is broadcast to all the nodes

but only the destination node

responds to the signal.

III. A disadvantage of this topology is, if

connection is broken, the entire

network stops working.

. To which network topologies does EACH of

the above statements refer?

(A) I - STAR II - RING III - BUS

(B) I - RING II - STAR III - BUS

(D) I - STAR II - BUS III - RING

02215010 MS/SPEC/2008

FORM SPEC 02215010/SPEC/2008

CARIBBEAN EXAMINATIONS COUNCIL

ADVANCED PROFICIENCY EXAMINATION

COMPUTER SCIENCE

SPECIMEN PAPER

UNIT 2 – FURTHER TOPICS IN COMPUTER SCIENCE

PAPER 02

2½ hours

INSTRUCTIONS TO CANDIDATES

1. DO NOT open this examination paper until instructed to do so.

2. Answer ALL questions from the THREE sections.

02215010 MS/SPEC/2008

SECTION A

DATA STRUCTURES

Answer BOTH questions.

(a) Define an ‘Abstract Data Type (ADT)’.

[2 marks]

(b) (i) Identify ONE primary characteristic of an Abstract Data Types.

[1 mark]

(ii) Identify ONE specific characteristic of a stack.

[1 mark]

(iii) Identify ONE specific characteristic of a queue.

[1 mark]

which performs the popping operation of a stack. The

stack is

represented by a one dimensional integer array of 100 elements called

stack, which must be passed to the function as a parameter. An integer

value called number, which identifies (the top number of elements) in

the stack must also be passed to the function as a parameter. In

popping the stack, the function should replace the emptied array

location with the value of “-99-

” and return the value which was

popped.

[10 marks]

(d) Write a function called dequeue in the programming language C,

which performs the dequeue operation of a queue. The queue is

represented by a one dimensional integer array of 100 elements called

queue, which must be passed to the function as a parameter. The

function should replace the emptied array location with the value of “-

99” and return the value which was dequeued. The empty location of

the array should be set to “-99”.

[10 marks]

Total 25 marks

02215010 MS/SPEC/2008

2. (a)

(b)

(c)

Explain the principles of EACH of the following techniques.

(i) Bubble sort

(ii) Binary search

Outline in detail, using narrative, an algorithm for bubble sorting

integers into ascending order.

A linear or sequential search involves the examination of EACH

element in an array. This can be very slow if the desired element is

near the end of the list. A better method is to make use of a binary

search. This involves storing the data in order as a binary tree.

Illustrate, with a diagram, how it is possible to search and find any

element in a list of 1000 by examining at most only ten of the

elements.

[3 marks]

[10 marks]

[9 marks]

Total 25 marks

02215010 MS/SPEC/2008

SECTION B

SOFTWARE ENGINEERING

Answer BOTH questions

(a) (i) Name FOUR attributes of a well-engineered

software product and describe EACH of them.

(ii) For EACH of the FOUR attributes identified in (a)

(i) briefly explain its importance to the software.

[8 marks]

(b) Identify THREE weaknesses of an unstructured approach to

software development.

[3 marks]

[2 marks]

(d) Develop a context level diagram for a DVD rental store

which engages in the following activities: rents DVDs to its

members, receives returned DVDs and purchases new

DVDs from suppliers. The sales cler

k provides monthly

reports to management.

[12 marks]

Total 25 marks

(a) Explain the terms ‘software process’ and ‘software process

model’.

[2 marks]

(b) The Waterfall model and Evolutionary development are two

of many generic process models.

Name and briefly describe FOUR fundamental activities

which are common to ALL process models.

[8 marks]

function and requirements’.

[4 marks]

(d) Various types of tests are carried out during the validation

phase of software development.

Briefly describe EACH of the following types of tests.

(i) Unit

(ii) System

(iii) Acceptance.

[6 marks]

(e) List FIVE items which should be included in the structure of

a test plan.

[5 marks]

Total 25 marks

02215010 MS/SPEC/2008

SECTION C

OPERATING SYSTEMS AND COMPUTER NETWORKS

Answer BOTH questions

(a) Distinguish between ‘intranet’ and ‘extranet’.

[3 marks]

(b) Describe ‘TCP’ and ‘IP’ as they relate to the TCP/IP protocol.

[3 marks]

for computer communication.

[6 marks]

(ii) Describe the functions of EACH of the bottom THREE

layers of the model drawn in 5 (c) (i) above.

[6 marks]

(d) Explain using an example how deadlock may arise in an

operating system.

[3 marks]

(e) A program called exec1 is being executed by an operating

system before it can be completed. It is interrupted by a small

program exec2, which contains four instructions.

(i) What determines which of exec1 or exec2 will be run?

[1 mark]

(ii) Assume exec2 is allowed to run. Explain what is done

in the operating system to ensure that exec1 can be

properly executed later.

[3 marks]

Total 25 marks

6. (a) Explain in detail how an operating system can execute THREE

processes at the same time.

[8 marks]

(b) Describe the characteristics of EACH of the following transmission

media.

(i) Coaxial cable

[3 marks]

(ii) Twisted pair

[3 marks]

network diagnostic tools.

[4 marks]

(d) Explain TWO reasons why compression utilities are useful.

[4 marks]

(e) Describe ‘radio waves’ as a transmission media.

[3 marks]

Total 25 marks

END OF TEST

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

CARIBBEAN ADVANCED PROFICIENCY

EXAMINATION

COMPUTER SCIENCE - Unit 1

SPECIMEN PAPER 2008

Item Number

Key

MODULE 1

1 A

2 C

3 D

4 C

5 A

6 B

7 A

MODULE 2

8 B

9 D

10 C

11 C

12 A

13 A

14 D

MODULE 3

15 B

16 C

17 D

18 A

19 B

20 C

FORM SPEC 02115010/SPEC/2008

CARIBBEAN EXAMINATIONS COUNCIL

ADVANCED PROFICIENCY EXAMINATION

COMPUTER SCIENCE

SPECIMEN PAPER

UNIT 1

PAPER 02

MARK SCHEME

02115010 MS/SPEC/2008

Question 1

(a) (i) Instructions in ROM (1) cannot be changed, but the contents of EPROM can be

erased using special equipment (1).

[2 marks]

(ii) EPROM chips must be removed from the computer to be changed but EEPROM

chips can be changed by using special electrical impulses within the computer (1).

EPROM normally has to b

e removed from a computer before it can be

reprogrammed while EEPROM can be reprogrammed without being removed

from the computer (1).

[2 marks]

(iii) MAN is a network that spans a city area (1) and LAN is a network that spans a

small area like a building. (1)

[2 marks]

(iv) Main frame computers are bigger in size (1), more expensive, have much more

memory than microcomputers. (1)

[2 marks]

(v) A supercomputer is much more powerful than a workstation computer. (1) Which

can handle less complex tasks (1).

[2 marks]

(vi) Cache memory is much faster than RAM (1) and consequently more expensive (1)

while RAM is similar to cache memory.

[2 marks]

(b) (i) Surge protectors protects the computer (1) equipment from surges in power (1).

The protector allows multiple components (1) to be plugged into

one power

outlet. (1)

[4 marks]

(ii) UPS – propose is to convert power and store it in a battery OR runs on battery if

the power goes off, to give users time to shut down the computer equipment.

[2 marks]

eight digits

correct answer 00001010

[1 mark]

(ii) The decimal equivalent of 00001101 is + 13

[1 mark]

(d) (i) exponent: 5

[1 mark]

(ii) base or radix: 10

[1 mark]

(e) 0011

1100

1111 Result

[1 mark]

Yes it can be stored as a from digit binary number

[1 mark]

Module 1 – Specific Objectives: 1, 3

02115010 MS/SPEC/2008

Question 2

(a) A truth shows the output (1) to various combinations of logic input (1)

[2 marks]

(b) Three primary topic gates are

NOT

AND

[3 marks]

B C

AND Gate

A B C

OR Gate

A B C

A C

NOT Gate

1 mark for each symbol

2 marks for correct truth table

1 mark for partial completion of truth table

[6 marks]

02115010 MS/SPEC/2008

Question 2 cont’d

(d)

[4 marks]

(e) (i) Control Unit (CU) (1) and Arithmetic and Logic Unit (ALU). (1)

(ii) The CU controls the movement of data (1) and instructions (1) into and out of the

CPU and controls the operation of the ALU (1).

The ALU does the actual computation or processing of data (1). It performs all

arithmetic and logic operations (1).

(iii) (a) Word length – the number of bits the CUPCAN process at any one time

[2 marks]

(b) Check speed – an internal of time in the processor

[2 marks]

Module 1 – Specific Objective: 2

02115010 MS/SPEC/2008

Question 3

(a) An algorithm is a sequence of unambiguous steps (1) for solving a problem (1).

[2 marks]

(b) Sequential (1): Steps are performed in a strictly sequential manner, each step being

executed exactly once (1).

Selection (1): One of several alternative actions is selected and executed. (1)

Repetition (1): One or more steps is performed repeatedly (1).

[6 marks]

(2) Problem Analysis.

(3) Generating possible Solutions.

(4) Analyzing the Solutions.

(5) Selecting and Justify Solutions.

Any four 4 marks

The process is only a guide for problem solving. It is useful to have a structure to follow

to make sure that nothing is overlooked. Nothing here is likely to be brand new to

anyone, but it is the pure acknowledgement and reminding of the process that can help the

problems to be solved.

(ii) 1. Problem Definition

The process for solving a problem will initially involve defining the problem (1)

you want to solve. You need to decide what you want to achieve and write it

down. The first part of the process not only involves writing down the problem

to solve, but also checking that you are answering the right problem (1). It is a

check-step to ensure that you do not answer a side issue or only solve the part of

the problem that is most easy to solve. People use the most immediate solution

to the first problem definition that they find without spending time checking the

problem is the right one to answer (1).

2. Problem Analysis

This step involves determining what are the inputs (1)

, what processing is

involved (1) and what are the outputs (1). The next step in this process is used

to check where we are and what is the current situation. Also determine how

the data and information or results will be stored temporarily or for future use.

3. Generating possible Solutions

In this stage you concentrate on generating many solutions (1) and not evaluate

them at all (1)

. Very often an idea, which would have been discarded

immediately, when evaluated properly, can be developed into a superb solution.

At this stage, you should not pre-judge any potential solutions (1) but should

treat each idea as a new idea in its own right and worthy of consideration.

02115010 MS/SPEC/2008

Question 3 cont’d

This section of the problem solving process is where you investigate the various

factors about each of the potential solutions (1). You note down the good and

bad points and other things which are relevant to each solution (1). Even at this

stage you are not evaluating the solution because if you do so then you could

decide not to write down the valid good points about it because overall you

think it will not work. However you might discover that by writing down its

advantages that it has a totally unique advantage (1).

5. Selecting and Justify Solutions

This is the section where you look through the various influencing factors (1)

for each possible and decide which solutions to keep (1) and which to disregard

(1). You look at the solution as a whole and use your judgement as to whether

to use the solution or not. Sometimes pure facts and figures dictate which ideas

will work and which will not.

(d) (i) Infinite loop (1): The number 1 is printed repeatedly (1)

(ii) Replace the statement y = y+1 with x = x+1 (1)

Total 25 marks

Module 2 – Specific Objectives: 2, 3, 6, 8

02115010 MS/SPEC/2008

Question 4

(a)

{ bCount = pCount 0

[1 mark]

for j = 1 to 100 do

input choice

if choice = ‘banana’ then

bCount = bCount + 1

else

pCount = pCount + 1 {Assumes all choices are valid}

endif

endfor

[1 mark]

if bCount > pCount then

print “The most popular flavour is banana.”

print “Votes:”, bCount

else

print “The most popular flavour is pineapple.”

print “Votes:”, pCount

endif

[1 mark]

Total 10 marks

(b)

Output generated by the algorithm is as follows:

* * * * * * * * * *

* *

* * * * * * * * * *

If the above diagram is drawn exactly as shown:

If there are variations, use the following mark scheme:

All of top line:

Some of top line:

All of left side of rectangle:

Some of left side of rectangle:

All of right side of rectangle:

Some of right side of rectangle:

All of bottom line:

Some of bottom line:

Top line properly connected to lower part:

Bottom line properly connected to upper part:

[2 marks]

[1 mark]

[2 marks]

[1 mark]

[2 marks]

[1 mark]

[2 marks]

[1 mark]

[2 marks]

[1 mark]

Total 10 marks

Module 3 – Specific Objectives: 7, 9

02115010 MS/SPEC/2008

Question 5

(a) The development of a program essentially involves two tasks

(i) Program design – designing a solution (1)

(ii) Coding – writing code in a programming language (1)

There are several strategies used to develop a design, each strategy required that the

programming language used to code the program support particular features (1).

Therefore knowing just one language locks the programmer into using particular

strategies (1). This is undesirable since the strategy selected to design the program

should be based on the type of programming problem rather than the eventual language

of implementation (1).

Also, once the program is designed a programmer should be able to choose among a

ariety of languages since even within a particular class of languages, individual

language differs. The programmer should be able to choose the most appropriate

language, that is the language that provides:

(i) Ease of programming particular solution (available predefined data structures

and subroutines) (1)

(ii) Run-time efficiency – some languages are inherently more efficient than others

(1)

(iii) Portability (1)

(iv) Ease of debugging (1)

(v) Maintainability (1)

[10 marks]

(b) Structured programming is an approach to programming with can give higher

productivity (1) and which can produce programs of high quality which are easy to test,

debug, modify and maintain. (1) The approach emphasizes modular development of a

program, that is, a large problem is divided into smaller subdivided and so on. (1) For

each subproblem, a module is developed to solve it. (1) The approach recommends the

use of a few simple control structures in specifying algorithms, also GOTO’s should not

be used so that program would be easier to understand and debug. (1)

[5 marks]

(ii) Modules can be kept in a library and be reused in other programs (1)

(iii) Programs are easier to build, read, debug and maintain (1)

(iii) The modular approach means that several programmers can each work on

separate module, thus shortening the total development time of a large project

(1).

[4 marks]

02115010 MS/SPEC/2008

Question 5 cont’d

(d) Int number_of_elements (List I)

{

int ret = 0

while (!is_empty (I))

{

return ++;

I = rest(I);

}

return ret;

} }

[1 mark]

[6 marks]

Module 3 – Specific Objectives: 1, 9

02115010 MS/SPEC/2008

Question 6

(a)

Lexical analysis (1) - Lexical analysis scans (1) the characters of the

same program from left to right (1) and builds

the actual symbols of the program – integers,

identifiers, reserved words, etc. (1)

Semantic analysis (1)- Takes constructs and checks them for semantic

correctness (1), and stores the necessary

information (1) about the constructs in the

symbol table. (1)

Code optimization (1)-

Reduces execution time (1) for project by

looking at how the code (1) could execute more

efficiently. (1)

Code generation (1) - Actual translation of the internal source program

(1) into assembly language (1) or machine

language. It is the most detailed part of the

compilation. (1)

1 mark for EACH stage

2 marks for EACH explanation

[12 marks]

02115010 MS/SPEC/2008

Question 6 cont’d

(b)

#include <stdio.h>

void main ( )

{ int a [ ] = {2, 3, 4, 4, 4, 6, 7, 8, 10, 12};

int b [ ] = {1, 3, 5, 10, 11, 13, 14, 15};

int c [25];

int i = 0;

int m = 10, n = 8;

int j = 0, k = -1;

while (i < m && j < n)

{ k++;

if (a[i] < b [j])

{ c [k] = a [i];

i++;

}

else { c [k] = b [j];

j ++;

}

int t;

if ( i = = m ) //end of array a

for (t = j; t < n; t++)

{ k++;

c [k] = b [t];

}

else

if ( j = = n ) // end of array b

for (t = i; t < m; t++)

{ k++;

c [k] = a [t];

}

[1 mark]

TOTAL 13 marks

Module 3 – Specific Objectives: 3, 8

Item Number

Key

MODULE 1

1 C

2 C

3 A

4 C

5 B

6 D

7 B

MODULE 2

8 B

9 A

10 C

11 A

12 B

13 A

14 D

MODULE 3

15 D

16 C

17 A

18 C

19 B

20 A

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

CARIBBEAN ADVANCED PROFICIENCY

EXAMINATION

COMPUTER SCIENCE - Unit 2

SPECIMEN PAPER 2008

02215010 MS/SPEC/2008

FORM TP 02215010/SPEC/2008

CARIBBEAN EXAMINATIONS COUNCIL

ADVANCED PROFICIENCY EXAMINATION

COMPUTER SCIENCE

SPECIMEN PAPER

UNIT 2

PAPER 02

MARK SCHEME

02215010 MS/SPEC/2008

Question 1

(a) The combination of a data structure and a group of functions or procedures

(1) designed to manipulate the data within the data structure (1).

[2 mark]

(b) (i) An ADT must possess the following characteristics:

1. the facility to create a container to hold the data;

2. the facility to add a new element to the container;

the facility to remove/delete an element which satisfies some

criterion form the container;

4. the facility to find an element which satisfies some criterion within

the container;

5. the facility to destroy the container when it is no longer required.

[Any one 1 mark]

(ii) a stack

The specific

operation which defines a stack is the concept of

“Last-in-first-out”. (1) Data elements which have been added to the

container are removed from the container in the reversed order (1)

to which they were inserted. (1) Typically, this ADT is identified

by a variable called “the head” (1) which identifies the location of

the most recent element (1) added to the container. The functions

of “push” and “pop”

are typically used to describe the process of

inserting and deleting elements respectively, from the stack.

[Any one 1 mark]

(ii) a queue

The specific operation which defines a queue is a concept of “Fist-

in-first-out”. (1) Data elements which have been added

(1) to the

container are removed from the container in exactly the same order

(1) in which they were inserted. Typically, this ADT is identified

by a variable called “the head” (1) which identifies the location of

the longest existing element in the container, (1)

and another

variable called “tail”

which identifies the most recent element

added to the container. (1)

The functions of “enqueue” and

“dequeue” are typically used to

describe the process of inserting

and deleting elements respectively, (1)

from the queue. In the

queue, data elements are inserted through the tail and removed from

the head. (1)

[Any one 1 mark]

02215010 MS/SPEC/2008

Question 1 continued

(d)

int popStack(int stack[], int number)

{

int value;

1 mark for data type of return value

1 mark for passing array parameters

1 mark for int parameter number

/* Check for error situation first */

if (number < 1)

{

1 mark for testing stacks empty

printf(“The stack is already emty”);

return -1;

}

else if (number > 99)

{

1 mark for return message

printf(“Elements size is beyond the stack size”);

return -1;

1 mark for testing

1 mark for message

}

else /*Stack can be popped */

{

value = stack[number -1];

stack[number-1] = -99;

number = number -1;

return value;

}

1 mark for assignment

1 mark for replacing value with -99

1 mark for returning value

02215010 MS/SPEC/2008

Question 1 continued

(e)

int dequeue(int queue[])

{

int value;

1 mark for data type of value to be returned

1 mark for passing array as parameter

/* Check for error situation first */

if (queue [0] = = -99)

{

1 mark for test condition

printf(“The queue is already empty”);

return -1;

}

else /* Queue can be dequeued */

{

value = queue [0];

/*Shift all the elements one place up */

for (int x = 0; queue [x] !=-99 && x < 100;x++)

{

3 marks for conditional

loop

queue [x] = queue [ x + 1];

}

return value;

}

1 mark for assignment

1 mark for return value

Module 1 – Specific Objectives: 2, 3, 4

02215010 MS/SPEC/2008

Question 2

(a) (i) A sorting procedure that works by repeatedly stepping through the list of

items to be sorted (1), comparing two items at a time and swapping them

if they are in the wrong order (1). The pass through the list is repeated

until no swaps are needed, which means the list is sorted (1).

(ii) A search technique for finding a particular value in a sorted list (1). The

desired item is compared to the data in the middle of the list (1). The half

that contains the data is then compared in the middle, and so on, either

until the key is located or a small enough group is isolated to be

sequentially searched (1).

(b) An example using N(i), where i is the position of the ith element in the list.

(i) Start with the first pair of numbers in the list (1) and compare them (1),

that is, compare N(i) with N(i+1).

(ii) If N(i) greater than N(i+1), then swap (1). That is, N(i) becomes N(i+1)

and N(i+1) becomes N(i). If a swap occurs, make note of it,that is, set

flag to indicate swap occurred (1).

(iii)Take next pair of numbers, that is, increment i (1) and repeat steps (i)

and (ii), until the last pair if number in list have been compared (1).

(iv) If a swap took place, that is, the flag was set, then clear flag (1) and

repeat the entire procedure (1).

(v) If no swap occurred during any iteration (1), then end (1).

Search:

1000

1st

500

2nd

250

3rd

125

4th

5th

6th

7th

8th

9th

10th

(3 marks)

(10 marks)

[1 mark for each correct search, no more than 9 marks]

(3 x 2 marks = 6 marks)

Module 1 – Specific Objectives: 5, 6

02215010 MS/SPEC/2008

Question 3

(a) Maintainability: Software should evolve to meet the changing needs

of customers.

Dependability: Software should be reliable, secure and safe.

Efficiency: Software should not make wasteful use of system

resources.

Usability: Software should have appropriate user interface and

adequate documentation.

[2 x 4 = 8]

(b) Software is likely to be of poor quality and unsuitable.

Software development cost is likely to be higher that expected.

Software development time is increased.

[3 x 1 = 3]

Allows efficient code to be developed

Can be adapted to modify functions and to add new functionality.

[Any 2 x 1= 2]

(d)

Return date

Payment

Member ID Return Date

DVD No. Receipt

Returned DVD

Monthly report

New DVD Details

1 for each labeled entity

1 for each labeled data flow

Module 2 – Specific Objectives: 2, 4

[12 marks]

Customer

Supplier

Management

DVD

Rental

System

02215010 MS/SPEC/2008

Question 4

(a) A software process is a structured set of activities required to develop a

software system. (1)

A software process model is an abstract representation of a software

process. (1)

[2 marks]

(b) Software specification:

The process of establishing what services are required and the constraints

on the system’s operation and development.

Software design:

The process of converting the system specification into an executable

system.

Software validation:

Software is checked to ensure that it does what the customer wants.

Software evolution:

Software is modified to adapt to changing customer needs.

1 mark for Name

1 mark for explanation [2 x 4]

[8 marks]

Statements of services the system should provide, (1)

how the system

should react to particular inputs and how the syst

em should behave in

particular situations. (1)

Non-functional requirements

The constraints on the services or functions (1) offered by the system such

as timing constraints, constraints on the development process and

standards. (1)

(d) (i) Unit testing: Individual components (1) are tested

independently. (1)

(ii) System testing: Testing of the system as a whole. (1) Testing

of emergent properties is particularly

important. (1)

(iii) Acceptance testing: Testing with customer data to check that the

system meets the customer’s needs. (1)

(e) - The testing process

- Requirements traceability

- Tested items

- Testing schedule

- Test Record procedures

- Hardware and Software requirements

- Constrains

[4 marks]

[6 marks]

[5 marks]

[Any 5 x 1 mark each]

Module 2 – Specific Objectives: 3, 7

02215010 MS/SPEC/2008

Question 5

(a) (i) An intranet a

llows persons within an organisation to access

information (1). Where as an extranet is an intranet that has been

extended (1) to allow selected external organizations to access

information (1).

[3 marks]

(b) TCP is a reliable transport protocol (1) with retransmissions etc. (1).

Where as the IP is a network layer protocol which can be unreliable (1).

[3 marks]

One mark for each layer, correctly positioned

[6 marks]

(ii) Network: routing packets (1) congestion control (1).

Data Link: error control (1). Communication between 2 nodes (1).

Physical: bits->cable signals(1). How much volts is a 0/1 (1).

[6 marks]

(d)

Process A needs a resource that process B is using (1) while process B

needs a resource that process A is using (1). Neither process can proceed

(1) until the desired resource is obtained, resulting in a deadlock (1).

A->B

[3 marks]

(e) (i) Priorities of each program (1).

[1 mark]

(ii) The state of exec1 is saved (1) before switching to exec2 (1). After

exec2

completes (1) exec1's state is restored and execution continues(1).

[3 marks]

Total 25 marks

Module 3 – Specific Objectives: 3, 4

Application

Presentation

Session

Transport

Network

Data Link

Physical

02215010 MS/SPEC/2008

Question 6

(a) (i) Time-slicing in a round-

robin fashion is normally used (1). If

process A is running initially, it is pre-empted (1)

and its state

changed and saved (1). Then process B can be loaded (1) and run

and so on (1). When process

A’s turn comes again, its state and

information is restored (1) and it is executed for a short time again

(1). This continues until all processes are completed (1).

[8 marks]

(b) (i) Coaxial cable – has a bandwidth that exceeds 100 Mbps (1). Its

typical use is for transmitting cable TV signals (

1). Coaxial cable

has a copper wire core (1).

[3 marks]

(ii) Twisted pair – this cable usually contains 4 wires (1). Each wire is

coated in plastic, so the copper wires do not come in direct contact

with each other (1). It is available as unshielded twisted pair and

shielded twisted pair (1).

[3 marks]

(iii) Fiber-optic cables – consist of a bundle of extremely thin tubes of

glass (1). Each tube is much thinner than a human hair. Data is

sent using optical signals (1). These cables are not prone to

electromagnetic interference (1).

[3 marks]

can help locate the location of the fault (1).

[2 marks]

(2)

Diagnostic tools can help detect malfunctioning hardware on the

network (1), e.g. network interface cards. (1)

[2 marks]

(d) Compressed files occupy less space (1), so more files can actually be stored

on your hard disk (1).

[2 marks]

Because compressed files occupy less space (1),. They can be transmitted

more quickly across a network than uncompressed files (1).

[2 marks]

(e) Radio waves are omni-

directional, i.e. they spread out in multiple

directions when they are transmitted (1). These signals can cover very long

distances (1). For radio broadcast, you need a transmitter to send the

signals and a receiver to accept the broadcast signal (1).

[3 marks]

Total 25 marks

Module 3 – Specific Objectives: 3, 4

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2004

COMPUTER SCIENCE

St Michael, Barbados

- 2 -

COMPUTER SCIENCE

CARIBBEAN ADVENCED PROFICIENCY EXAMINATIONS

MAY/JUNE 2004

INTRODUCTION

This is the fourth year of open examinations for the Unit-one course comprising three

papers and the third year for the Unit-two course. Papers 01 and 02 for both units

were examined externally by CXC while Paper 03 was examined internally by the

teacher and moderated by CXC.

Paper 01 consisted of three sections, each corresponding to a module of the sylla-

bus. There were five compulsory short-response questions within each section, for

which a maximum of 50 marks could be obtained, for a total of 150 marks.

There were six questions of equal weighting in Paper 02, two questions per module

and candidates were expected to answer one question from each module. These

extended-response questions required more depth of understanding than the ques-

tions in Paper 01.

The individual contributions of Papers 01, 02 and 03 to the examination remained 50

per cent, 30 per cent and 20 per cent respectively.

GENERAL COMMENTS

Performance in Unit 1 suggests that teachers and candidates are completing the sylla-

bus, and as such candidates are for the most part performing well.

Examiners continue to be very concerned about candidates’ performance in Unit 2; in

particular Modules 2 and 3 where programming knowledge is tested. Candidates

continue to find it extremely difficult to write even simple programs, especially in the

object-oriented languages.

However, it appears that teachers are not completing the syllabus for Unit 2, and

therefore, candidates are coming to the examination having not acquired all the neces-

sary skills. It seems that candidates are not getting the necessary practice during

preparation for the examination. Perhaps this is so because, in many cases, they are

coping with learning a new programming language, whilst simultaneously developing

a system in order to satisfy the requirements for the internal assessment. For Unit 2,

teachers are reminded to have several practical sessions with students before the final

exam.

- 3 -

Candidates should be encouraged, as part of their examination technique, to read

questions carefully and to respond with sufficient detail to commensurate with the

marks indicated.

DETAILED COMMENTS

UNIT 1

PAPER 01

SECTION A – Components of Computer Systems

Question 1

This question tested candidates’ knowledge of terms relating to computer hardware.

In general, part (a) was poorly answered. Many candidates described clock speed as

time taken to perform a task. Few candidates were able to mention the function of

cache memory. The main strong point in the responses was the ability of most

candidates to correctly identify two problems that the UPS can prevent. In a few

cases candidates listed features of the UPS as problems that it can prevent.

Question 2

This question examined candidates’ understanding of the resource management func-

tions of an operating system. Candidate response was generally weak. There were

no strong points in the responses. Of particular concern was the candidate’s lack of

knowledge of deadlock, interrupt and channel bandwidth. A bit more depth of cover-

age of this area of the syllabus is recommended so that candidates can express

themselves more clearly.

Question 3

This question tested candidates’ knowledge of the reasons why operating system

(OS) software needs to be kept up-to-date and the services offered by disk manage-

ment software.

The average mark for this question was 50 per cent. Most candidates failed to give

correct responses to Part (a). They seemed not to recognize that updates are really

additions or changes to the original software. Updates are needed to fix errors due to

undetected bugs; add new features to the OS; provide new code that can perform

tasks more efficiently. There were generally good responses to Part (b). Generally,

- 4 -

candidates were able to correctly identify the services offered by disk management

software.

Question 4

This question tested candidates’ understanding of the functions of components in a

network. Candidate responses were generally good. The diagrams and explanations

relative to a hub in a network were quite good. The meaning of ‘data encryption’ was

well explained, including the method and reasons. Candidates understood the need

for confidentiality and security. Items that are normally encrypted were correctly

identified.

Question 5

This question examined candidates’ knowledge and understanding of transmission

media and network topologies. The overall performance on this question was fairly

good. Parts (a) and (b) (i) were well answered. Candidates seemed not to under-

stand what was required for Part (b) (ii). The main weak point was that many candi-

dates’ descriptions of the FDDI topology were not reflected in the diagrams that they

drew.

SECTION B – Application of Computers

Question 6

This question was designed to test candidates’ understanding of the role of comput-

ers within an organization. The majority of candidates seemed to have misinterpreted

the question. They treated the application areas of the business as Word Processing,

Spreadsheets and Database Management. Consequently, responses to Part (b) were

generally meaningless.

Question 7

This question tested candidates’ knowledge of the various types of information sys-

tems. This question was generally well done. Part (a) was well known by most

candidates. However a few candidates did not mention at which level in the organiza-

tion the information system was used. The main weaknesses were in Part (b). Few

candidates were able to recognize that an ‘executive information system’ is similar to

a ‘decision support system’.

- 5 -

Question 8

This question tested candidates’ knowledge of issues relating to telecommuting, pi-

racy and data security. Part (a) was not answered well by a lot of candidates. Most

candidates mistook telecommuting for telemarketing and wrote about buying and

selling. Part (b) was well done as most candidates knew the implications of pirating

software and also the use of firewalls and encryption.

Question 9

This question examined the different types of data security and the strategies for

ensuring data security. This seemed to be a popular topic with candidates. The

question was very well done. The security controls were well identified and ex-

plained.

Question 10

This question tested candidates’ knowledge of the benefits and drawbacks of using a

computer-based system in an organization. This question was generally well-an-

swered. Most candidates were able to give good responses. Candidates should be

encouraged to use examples, descriptions and explanations that are required at this

level.

SECTION C – Computer-based Problem Solving

Question 11

This question assessed candidates’ knowledge of the use of a computer-based tool

in the solution to a real-life problem. The question also tested knowledge of informa-

tion sources. It was answered very well. The only weak point was that a few candi-

dates did not know the features of the tools that were useful in performing the given

task.

Question 12

This question tested candidates’ knowledge of the characteristics of information

sources. The responses were for the most part reasonable. Candidates were able to

correctly identify and discuss characteristics such as currency, relevance, cultural

context, bias, accuracy etc. In part (b), even though candidates knew differences

between ‘book and newspaper’, ‘CD-ROM and DVD-ROM’, those differences did

not relate to information. Surprisingly many could not state similarities such as: book

- 6 -

and newspaper – both use a printed medium CD-ROM and DVD-ROM – both

based on optical technology; both can store at least 650 MB.

Question 13

This question tested candidates’ knowledge and understanding of issues related to

online/internet publishing. Part (a) was poorly done. Candidates failed to recognize

that the publisher could derive revenue by letting customers pay a fee to subscribe

and then each customer gets a username and password. Parts (b) and (c) were fairly

well done. Candidates were able to correctly describe advantages and disadvantages

to the customer of having the book available online.

Question 14

This question tested candidates’ knowledge and understanding of the selection of

computer-based tools that are best suited to a given situation. The question was

generally well done. In both Parts (a) and (b), the candidates demonstrated that they

were familiar with the tools and that they knew the appropriateness of the chosen tool

for the task to be performed. In Part (c) there was some doubt as to the precise

functions of the built-in grammar checker. For example some did not know that a

grammar checker also provided the meaning of words.

Question 15

This question tested candidates’ knowledge and understanding of issues relating to

different types of information sources. This question was very well done. Candi-

dates were easily able to identify information sources and suggest why they were

suitable choices. The difficulty was with Part (c) where the discussion on the disad-

vantages was limited and most candidates received 2 of the allotted 4 marks for that

section.

PAPER 02

SECTION A – Components of Computer Systems

Question 1

This question examined candidates’ knowledge and understanding of different types

of networks, network configurations and protocols. Generally Part (a) was answered

fairly well. Some good diagrams were drawn. However, in some cases candidates’

- 7 -

description of the function of the Virtual Private Network (VPN) was limited. They

omitted the fact that the VPN devices at the branches translated between the packet

format of the Internet and the packet format of the networks at the branches. Part (b)

was well known by most of the candidates. Part (c) was poorly done. Many candi-

dates could not show how the peer-to-peer and client/server configurations could co-

exist on the same network. Most candidates got the OSI model correct. The weaker

candidates were able to draw at most 4 layers in the correct order. The main weak

points in the responses were the major cost associated with VPN and the client/server

– peer-to-peer diagram.

Responses to this type of question can improve with a better understanding of net-

work types, especially VPN and VAN.

Question 2

This question was designed to test candidates’ knowledge of network components

and their understanding of the factors that have to be considered when installing

networks. This question was poorly done. Most of the candidates got less than 30

percent of the marks allocated. For Part (a) very few could identify the factors to be

considered. Even fewer understood what was involved in providing simultaneous

Internet access to a number of computers in a network. For Part (b) again only a few

candidates correctly identified component A as a gateway and component B as a

hub. The majority of candidates obtained the marks for Part (c).

SECTION B – Application of Computers

Question 3

This question tested candidates’ knowledge and understanding of data security is-

sues/problems, and their understanding of expert systems. Responses were gener-

ally good. Part (a) was well answered. Candidates demonstrated a good understand-

ing of the data security problems - and the strategies to deal with those problems -

which a company may face when sensitive data is held on a computer. In response to

Part (b), candidates’ definitions of an expert system were not clear. For Part (c),

some gave advantages of expert systems such as: hospital can save money by hiring

fewer doctors. Those who attempted Part (d), mentioned only one disadvantage,

namely that the expert system could misdiagnose and possibly lead to the death of a

patient. Another disadvantage is that the system is difficult to keep up-to-date with all

recent medical findings.

- 8 -

Question 4

This question tested candidates’ knowledge of the use of robots in industry, and

issues relating to the use of Automatic Teller Machines (ATM) in banks. The ques-

tion was for the most part well done. Most candidates were familiar with the use of

the robotics, particularly in a vehicle assembly plant. Part (c) was also well done.

Candidates demonstrated a good understanding of the ATM. The question intended

to ask for one technique that the bank could use to protect a customer’s account

from disclosure while using the ATM. However, many candidates interpreted the

question to be requiring one technique that the customer can use. Part (a) was the

weak point. Candidates did not understand what an expert system was and therefore

could not properly answer this part. On the whole this was a very popular question.

SECTION C – Computer-Based Problem Solving

Question 5

This question tested candidates’ knowledge and understanding of the stages in the

problem-solving process, and their ability to describe types and characteristics of

information sources. Very few candidates answered this question to the satisfaction

of the examiners. Those who attempted Part (a), succeeded only in listing the stages

in the System Development Life Cycle, but were unsure of the deliverables of each

stage. Part (b) was fairly well done. The responses to Part (c) showed that candi-

dates had ideas of what was requited but had difficulty expressing their ideas clearly.

Question 6

This question examined candidates’ knowledge of the features of modern productiv-

ity tools, and their suitability to certain tasks. Overall performance in this question

was fair. Part (a) was well done with most of the candidates obtaining more than 75

percent of the allotted marks. Many candidates did seem to understand the question.

Many wrote of preventing paper wastage by reducing font size and margins. In Parts

ity for the given task.

-9 -

UNIT 2

PAPER 01

SECTION A – Software and System Development

Question 1

This question tested candidates’ knowledge and understanding of the need for a

structured approach to software development and the activities that are common to

all software development processes. Both sections of the question proved difficult

for the candidates. Part (a) was poorly done. Many candidates wrote the stages of

the System Development Life Cycle. Even though most candidates performed better

in Part (b), the responses given were for the most part verbatim from the recom-

mended texts.

Question 2

This question examined candidates’ knowledge of the attributes of well-engineered

software, and the effect that lack of such attributes will have on the software. This

question was fairly well done by most candidates. Some candidates, however, cor-

rectly identified the attributes of well-engineered software, but gave incorrect descrip-

tions. Responses to Part (b) were in many cases generalized and were not related to

the attributes stated in Part (a). A comparison chart or attribute table may be instruc-

tive in assisting candidates with this area of the syllabus.

Question 3

This question assessed candidates’ knowledge of functional and non-functional re-

quirements and the techniques used in the analysis phase to obtain these require-

ments. This question was very well done.

Question 4

This question tested candidates’ knowledge and understanding of the use of Data

Flow Diagrams (DFD) during analysis. Part (a) was well done. In Part (b) many

candidates copied the symbols as given in the question, but were unable to identify

the data flows to complete the DFD. More exposure to DFDs should be provided.

-10 -

Question 5

This question examined candidates’ knowledge and understanding of the activities,

tools, techniques and deliverables of the design phase of software development.

Most candidates were able to list the activities of the design phase, but had difficulty

in providing meaningful explanations. For example many candidates did not include

key terms in their responses such as a HIPO chart emphasises the ‘relationships’

among the elements of a system. Most mentioned only that the HIPO chart showed

input, processing and output. Most of the examples used did not reflect a system,

but rather a component of a system. Candidates should have more exposure to

practical examples of the use of HIPO charts and teachers should ensure that proper

annotations are included.

SECTION B – Programming Languages

Question 6

This question assessed candidates’ understanding of the term ‘algorithm’ and their

ability to write a simple algorithm. Most candidates were able to explain what was

meant by the term ‘algorithm’. However, many had difficulty writing the required

‘algorithm’ for Part (b). Candidates need to be given more practice in the writing of

‘algorithms’.

Question 7

This question tested the candidates’ knowledge of the characteristics of program-

ming languages, and the ways in which programming languages differ. The question

was generally well done, however, many candidates attempted to give too much detail

when the question asked for ‘brief’ descriptions. In their description of the object-

oriented programming paradigm, many candidates omitted the fact that the programs

consist of a set of communicating objects.

Question 8

This question examined the candidates’ knowledge and understanding of program-

ming constructs and tested their ability to describe the function of the constructs

using examples in each case. Most of the candidates were able to describe the

function of the construct and provide appropriate examples. However, surprisingly,

they could not identify the constructs by their names, for example, ‘selection’, ‘itera-

tion’, ‘sequencing’.

-11-

Question 9

This question tested the candidates’ understanding of the factors that have to be

considered when choosing a programming language for a given application. The

question was generally well done. The candidates’ main strong points were in identi-

fying the factors to be considered. However, their explanations were far from ad-

equate. Part (b) was poorly done - only a few candidates demonstrated any under-

standing of the functional programming paradigm.

Question 10

This question tested candidates’ understanding of concepts associated with the ob-

ject-oriented and procedural paradigms. Many candidates had a general idea of the

conceptual building blocks of object-oriented design, but only a few demonstrated

real mastery. Almost all candidates were able to explain the term ‘recursion’ with

respect to the procedural paradigm. Part (c) was generally well done.

SECTION C – Program Development

Question 11

This question tested candidates’ ability to draw the layout of a simple Help system

using a Graphical User Interface (GUI), and to identify events that the help system

should anticipate when the user interacts with the system. In Part (a) most candidates

had difficulty identifying the events and it was clear they were not sure what an event

was. For Part (b) it was widely known by candidates that a diagram had to be drawn.

However, many candidates did not properly identify the GUI objects in the layout. In

some cases candidates thought that designing a layout meant writing a description or

writing code.

Question 12

This question tested candidates’ knowledge and understanding of abstract data types

(ADT) and operations on ADTs. Parts (a) and (c) were answered well. However,

Part (b) was poorly answered. Generally, candidates demonstrated little knowledge

of ADT operations.

Question 13

This question assessed candidates’ knowledge of the tools typically available in a

programming environment. This question was generally poorly answered. Candi-

-12 -

dates were unable to identify or describe editor features. Some even confused this

development tool with a newspaper editor. In Part (b), candidates were able to state

the tools but could not describe them.

Question 14

This question examined candidates’ knowledge and understanding of the activities of

the implementation phase of software development. This question was fairly well

done. Most candidates knew what code libraries were and the purpose of user

documentation. Some candidates however did not describe the advantages of using

code libraries during the coding process even though they knew what they were.

Many candidates did not realize that the testing required was unit testing, system

testing or dry-run testing.

Question 15

This question tested the candidates’ ability to write code in an object-oriented pro-

gramming language to implement a ‘Square’ class. Many candidates attempted this

question, but the responses were generally very poor. The candidates had particular

difficulty with the writing of constructors. In addition, many candidates did not know

which methods did and did not require parameters.

PAPER 02

SECTION A – Software and Systems Development

Question 1

This question tested candidates’ understanding of the software development life cycle

models and the tools and techniques of the analysis phase of software development.

The overall response to the question was for the most part satisfactory. There were

many correct responses to Parts (b) (i) and (c). In Part (a) some candidate confused

the word ‘features’ with ‘steps’ and so most of them simply listed the steps in the

‘waterfall’ approach. Parts (b) (ii), (b) (iii) and (d) were poorly answered. Even

though it is clear that candidates are familiar with evolutionary development, they are

unable to discuss the pros and cons of one development approach viz a viz another.

The responses to Part (d) expose the need for more practice in the use of data flow

models to document the flow of data.

-13 -

Question 2

This question examined candidates’ ability to differentiate between: ‘top down’ and

‘bottom up’ design strategies; functional and object-oriented design strategies; and

to identify the features of good software design. In Part (c), an Entity-Relationship

Diagram (ERD) was needed to capture given information. Candidate responses indi-

cated that this was a less popular question but the responses were generally satisfac-

tory. Some of the strong points in this question were the ability of the candidates to

correctly draw the ERD, and their understanding of the differences among the vari-

ous design strategies. The weak point was that many candidate responses to Part (b)

(i) showed that they were unclear about what was required and consequently they

were unable to answer Part (b) (ii). It seems this was a first exposure to this type of

question.

SECTION B – Programming Languages

Question 3

This question tested candidates’ understanding of the major stages in the compilation

process and their ability to compare the procedural and object-oriented paradigms.

Candidates were also required to write recursive and non-recursive functions that

involved finding the factorial of a non-negative integer. Parts (a) and (b) were gener-

ally well done. However, in Part (a) the candidates did not explain the stages in

sufficient detail. For Parts (c) and (d) few candidates were able to formulate a solu-

tion. The majority simply did not know how to approach the problem. More time

has to be spent on the development of algorithms in the problem solving process.

Question 4

This question assessed candidates’ ability to identify and use programming con-

structs, features and techniques in both the object-oriented and procedural para-

digms. Candidates’ strength was demonstrated in their ability toidentify objects in

an application, state, with type, instance variables for each object and use examples

to differentiate between bounded and unbounded iteration.

Some candidates thought that the use of the word ‘specify’ in Part (d) meant that they

were to write code. The responses to Part (g) indicated that the candidates were

guessing at the role of the linker in a programming language. The candidates per-

formed generally well in this question.

-14 -

SECTION C – Program Development

Question 5

This question tested candidates’ ability to develop an object-oriented application to

keep track of course and student information for a university. Candidates’ under-

standing of abstract data types (ADT) and their operations was also tested. This was

a very popular question among candidates, but the responses were generally poor.

The only strong point was the ability of candidates to declare the class attributes in

Part (a) and to identify the appropriate ADT and operations in Part (b). Writing code

for the class – constructor and methods proved very difficult as candidates did not

understand what was required. In the classroom greater attention needs to be paid to

the syntax of the language which the candidates are expected to use in the examina-

tion.

Question 6

This question tested candidates’ knowledge of the operations of the queue abstract

data type (ADT), and their ability to write code to implement the queue ADT as a

class. Few candidates attempted this question. Those that did attempt it were able

only to describe the function of each of the operations. Writing code proved ex-

tremely difficult for the candidates. For the most part there was a clear lack of

understanding of object-oriented language programming. Candidates’ performance

on programming questions can be improved only with practice in the classroom.

PAPER 03

INTERNAL ASSESSMENT

GENERAL COMMENTS

Candidates performed well on the Internal Assessment. For the most part the projects

areas were well chosen, covering a variety of areas and the reports were generally of

good quality. Candidates did best when the projects were based on real situations

which they had experienced.

-15 -

SPECIFIC COMMENTS

There are some concerns which the moderators have with the samples submitted.

Unit 1

The main areas of concern are

· The awarding of marks for work not submitted. In one instance, the teacher

awarded full marks for Networking when the candidate did not present any

work on networking.

· The Data Analysis and Organizational Impact sections of the report were

poorly represented in the project submitted.

Teachers (especially new teachers) are reminded that they should be fully familiar

with the requirements and composition of the Internal Assessment and are to be

guided by the criteria set down in the syllabus.

Unit 2

· Some of the samples submitted were not of the standard expected of candi-

dates at the CAPE level.

· All section of the report must me submitted since moderators must be able to

verify the marks awarded to each candidate.

· Testing of projects was often totally inadequate. Sample output from screens,

must be included as part of the documentation to verify that the program is

working.

· In many cases the user documentation was inadequate.

· The written report is expected to reflect the work done from the beginning of

the project, not simply to concentrate on the finished product.

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATIONS

MAY/JUNE 2005

COMPUTER SCIENCE

St Michael Barbados

COMPUTER SCIENCE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATIONS

MAY/JUNE 2005

INTRODUCTION

This is the fifth year of open examinations for Unit 1 and the fourth year for Unit 2.

There were three examination papers in both units, namely, Paper 01, Paper 02, and

Paper 03. In each unit, Paper 01 and Paper 02 were examined externally by CXC

while Paper 03, the Internal Assessment, was examined internally by teachers and

moderated by CXC.

In each unit, Paper 01 consisted of short-answer questions that were designed to test

candidates’ breadth of coverage of the syllabus. On the other hand, Paper 02

consisted of essay-type questions that were designed to test their depth of

understanding of the syllabus. Thus, candidates were expected to show deeper insight

and understanding of the topics examined in Paper 02.

In each unit, Paper 01 consisted of three sections, each one corresponding to a

module of the syllabus. There were five compulsory questions within each section,

and each question carried 10 marks. The maximum amount of marks that could be

obtained in Paper 01 was therefore 150.

Similarly, Paper 02 of each unit consisted of three sections, each one corresponding

to a module of the syllabus. There were two questions within each section, and each

question carried 30 marks. Candidates were expected to answer one question from

each section. The maximum amount of marks that could be obtained in Paper 02 was

therefore 90.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade

remained 50 percent, 30 percent, and 20 percent, respectively.

GENERAL COMMENTS

In general, performance on both units has continued to improve, compared to

previous years. Performance in Unit 1 suggests that teachers and candidates are

completing the syllabus and are becoming more familiar with what is required.

Even though the performance in Unit 2 has improved, there is still concern about the

level of programming ability being demonstrated in Sections B and C of both written

papers. Candidates continue to find it extremely difficult to write even simple

programs, especially in the object-oriented programming languages.

A major problem detected in the responses to questions in both Units that require the

writing of prose is that of candidates’ inability to clearly express their answers.

Candidates need to understand what is required in a question and should learn to

distinguish between key words such as “state”, “list”, “explain”, “discuss”,

“describe” and “identify”. Many candidates lose marks by making a simple

statement on a topic and not discussing the topic in sufficient detail. Other candidates

also need to learn to label their responses appropriately and not lump together the

answers to the different parts of a question in an essay. Since this is an advanced

level examination, candidates are also advised of the importance of handwriting and

the need for maturity in their responses.

Finally, candidates are encouraged, as part of their examination technique, to read

questions carefully before answering, and to respond with sufficient detail that is

commensurate with the marks indicated in the question.

DETAILED COMMENTS

UNIT 1

PAPER 01

SECTION A – Components of Computer Systems

Question 1

This question examined candidates’ knowledge of server computers in a computer

network and their ability to distinguish between two different types of cabling used in

a computer network. Most candidates seemed to be quite knowledgeable of the role

of a server in a client/server network and the question was well answered by most of

them. In Part (a)(ii), candidates were generally unable to name specific types of

servers typically found in a computer network (e.g., printer server and mail server).

Some candidates mistakenly believed that peer-to-peer, client-server, hub, and

routers were types of servers and went on to explain the purpose of these terms.

Question 2

This question tested candidates’ understanding of operating system concepts such as

memory protection, scheduling, and file management. It also required candidates to

identify two applications in which a supercomputer can be used. The responses

indicated that candidates had a very limited understanding of the concepts of memory

protection and scheduling algorithms. Some candidates even wrote that memory

protection refers to physically protecting the hard drive. However, the majority of

candidates gave good responses for Parts (a) and (d).

Question 3

This question tested candidates’ understanding of the data and processing

requirements that should be considered before purchasing a computer system. This

question was not done well by the majority of the candidates. Data requirements

include considering the volume of data to be stored, sharing data, and securing data.

Processing requirements include considering the speed and type of processing.

Question 4

This question examined candidates’ knowledge and understanding of local area

networks and different types of network configurations such as distributed, peer-to-

peer, and centralised. The overall performance on this question was quite good, with

an average of almost 6. However, a number of candidates confused the term

‘configuration’ with ‘topology’ and ‘network type’, and gave answers for network

topologies and network types. In Part (b), while it was clear that candidates

understood what is a local area network, their diagrams were badly drawn.

Question 5

This question tested candidates’ knowledge and understanding of various concepts

related to computer memory, such as ‘volatility’, ‘memory size’ and ‘word size’. It

also required candidates to describe one specialised feature of a mouse as an input

device. However, it is felt that some candidates could have obtained more marks if

they had not misinterpreted Part (a)(i) to be “What is volatile memory”. ‘Volatility of

memory’ refers to the degree that data in memory can be lost easily, for example,

when electricity goes. Also, in Part (b), instead of describing a specialised feature of

a mouse such as a scroll button, some candidates incorrectly gave examples of

specialised input devices while others described the purpose of a mouse.

SECTION B – Application of Computers

Question 6

This question tested candidates’ understanding of computer applications in an

organization as well as the implications of computer automation. Most candidates

were able to outline at least one benefit and one negative consequence of using

computers in an organization.

Question 7

This question tested candidates’ knowledge and understanding of various types of

information systems such as transaction processing systems (TPSs) and data

warehouses. Candidates were also required to discuss what happens during data

mining. As with similar questions in the past, most candidates answered the question

poorly and the average mark was less than 4. Some candidates were unable to give an

actual example of a TPS though they correctly named the inputs and outputs of the

system. Others had difficulty in distinguishing between the inputs/outputs of a

system and the input/output devices for that system. Using automatic teller machines

as an example, some candidates incorrectly identified keyboard/mouse and printer/

monitor as inputs/outputs of the system, respectively. The responses for Part (b) were

generally vague, demonstrating candidates’ limited knowledge of data warehouses

and data mining. For example, a common response was that a data warehouse is a

location where data is stored.

Question 8

This question required candidates to demonstrate their understanding of special-

purpose computer systems, specifically, expert systems and embedded systems.

Many candidates gave good examples of expert systems and correctly identified the

inputs/outputs of these systems, although some had problems identifying inputs/

outputs as explained in Question 7 above. However, the inference engine was

generally omitted in describing an expert system in Part (a). Some candidates also

confused expert systems with embedded systems. Many candidates incorrectly stated

that an embedded system is a piece of software that is added to another larger piece

of software such as an operating system. Candidates were expected to state that an

embedded system is a computer system placed inside other products to add features

and capabilities that are computer-based (e.g., an embedded system is used inside a

microwave oven).

Question 9

This question tested candidates’ understanding of the concepts of telecommuting and

videoconferencing and was answered fairly well. Most candidates were able to

correctly explain how communication is achieved in videoconferencing. However,

many candidates confused telecommuting with teleconferencing and telemarketing.

Telecommuting refers to a work arrangement where employees work away from their

offices using personal computers (e.g., at home), communicating with their offices

using email and other forms of communication technology. The advantages of

telecommuting and videoconferencing were also not clearly stated by many

candidates.

Question 10

This question tested candidates’ understanding of data security and how to protect

data in an organisation, both from damage and from unauthorised access. In Part (a),

most candidates identified two ways in which data could be accidentally damaged but

did not state exactly what damage was caused. However, in Part (b), most candidates

were knowledgeable about the techniques that could be used to protect data from

unauthorised access.

SECTION C – Computer-based Problem Solving

Question 11

This question examined candidates’ knowledge of using different computer-based

tools to solve the real-life problems of generating a report on the performance of

candidates in the CSEC Information Technology examinations and maintaining

student information at a school on a day-to-day basis. This question was well

answered by most candidates. . In both Part (a) and Part (b), candidates were able to

correctly identify appropriate software tools (word processor and spreadsheet, and

database management systems, respectively). However, several candidates did not

clearly state the features of these tools that made them appropriate for the task at

hand. As indicated in the question, it is important for candidates to link the features

of the tools to the task that must be performed.

Question 12

This question examined candidates’ knowledge of using a software tool to solve the

real-life problem of making a presentation of a company’s sales performance.

Candidates were also expected to understand the kinds of problems that could affect

the use of the tool selected and how to solve these problems. This was a fairly easy

question and it is surprising that candidates’ performance was not much better than it

was. The obvious software tool for the problem was presentation software such as

Microsoft Powerpoint. However, some candidates stated that spreadsheet software

could be used and this was also accepted. Similar to Question 11, candidates were

often unable to state three features of the tool that made it appropriate for the task.

Several candidates gave simplistic responses for Part (b) such as the software and

presentation data must be stored on a computer. It should be noted that the questions

in this section deal with Computer-based Problem Solving, so the need for

computers, software, and data is clearly understood. More appropriate responses on

potential problems that can occur when using the software tool are that the vice-

president may not know how to use the software (i.e., training is required) or that the

presentation file may become damaged for some reason (requiring backup measures

to be put in place).

Question 13

This question tested candidates’ knowledge and understanding of information

sources and their ability to select appropriate knowledge sources for a given

information need. The question was well answered by the majority of the candidates.

Several candidates obtained full marks in this question. However, given its

straightforward nature, a higher average was expected.

Question 14

This question tested candidates’ knowledge and understanding of the characteristics

of information (with an emphasis on reliability) as well as their ability to analyse a

situation to determine the quality of information that can be obtained. It was

generally answered satisfactorily with many candidates losing marks for not

developing their responses or repeating a source of un/reliable information already

given. Also, the majority of candidates were unable to furnish an example in Part (a)

(i) when they explained the term ‘reliability of information’. In Part (b), the reasons

given for not accepting the information were often quite poor or underdeveloped.

Candidates were expected to give responses such as the receptionist had no authority

to give out the information requested and may also give inaccurate information since

this was not his/her job area.

Question 15

This question tested candidates’ knowledge and understanding of a digital library as

an information source. It was clear that many candidates had limited understanding

of a digital library, and many of them opted to answer the question from the point of

view of a CD. A digital library is a database containing documents such as journal

and magazine articles in digital form that can be accessed by users and downloaded

to their own computers. Candidates also gave a general list of advantages and

disadvantages of a digital library, without attempting to elaborate on the points

stated. This may be attributed to their limited understanding of a digital library.

PAPER 02

SECTION A – Components of Computer Systems

Question 1

This question tested candidates’ in-depth knowledge and understanding of a number

of hardware-related and network-related concepts covered in the unit. Generally,

candidates were able to differentiate between ROM and EPROM, MAN and LAN,

and between workstation and supercomputer. However, some displayed weaknesses

in comparing a hub and router and in comparing EPROM and EEPROM. The

characteristics of the transmission media given were also well described. However,

weak responses were obtained for Part (c), when describing network-related

problems that can be solved using diagnostic tools. Many candidates also seemed

unfamiliar with compression utilities. Their discussion of radio waves as a

transmission medium was often poor.

Question 2

This question tested candidates’ understanding of issues related to purchasing a

mouse, the role of the user interface in an operating system, and various network-

related concepts such as microwave transmission and the FDDI and ring network

topologies. Candidates gave very good responses on the issues related to purchasing

a mouse, but the role of the user interface in an operating system was not clearly

explained. Also, most candidates were unaware that FDDI is composed of two rings

of fibre-optic cable. This caused them to lose marks in the other sections of Part (d).

Candidates generally gave good answers for how data is transmitted in a ring

network, but often failed to mention the presence and purpose of the token.

SECTION B – Application of Computers

Question 3

This question tested candidates’ understanding of the changing information needs of

an organization over time as well as their understanding of the historical changes in

computer applications over time. Very few candidates (less than 5%) chose to answer

this question. From the responses obtained, it seemed that few candidates understood

what was being asked by the question and their responses were vague and off the

point. Even though candidates were able to identify changes in storage and

processing capabilities, they did not relate these changes to the specific needs of the

business, such as handling an increased number of customers, employees, products,

etc. In Part (c)(i), candidates were generally unable to specify the information

processing activities that would take place at the operational and strategic levels of

the organization. Candidates need to be able to better distinguish between the

operational and strategic aspects of information processing in an organization.

However, it is encouraging to note that in Part (c)(ii), several candidates were able to

correctly identify transaction processing systems, management information systems,

and other systems that could be used at the organization.

Question 4

This question tested candidates’ in-depth understanding of data security and how to

protect data in an organisation, both from unforeseen circumstances such as

hurricanes and fire, and from unauthorised access. More than 95% of the candidates

attempted this question and all parts of this question were generally answered well. In

particular, there were very good answers for Part (a)(i) when candidates were asked

to discuss two means by which unauthorised access to data could occur in the bank.

SECTION C – Computer-based Problem Solving

Question 5

This question tested candidates’ knowledge and understanding of using different

computer-based tools to solve a real-life problem of attracting investors to a

company. Only about 30% of the candidates chose to answer this question instead of

Question 6. Candidates were able to correctly identify software tools and explain

how these could be used in the company’s venture to attract investors. However,

marks were often lost because an appropriate discussion was not given.

Question 6

This question tested candidates’ knowledge and understanding of information

sources and evaluative criteria that can be used to determine the validity of an

information source. It also tested candidates’ understanding of the problem solving

process required for developing a computer-based solution for a business problem.

About 70% of the candidates opted to answer this question. Part (a)(i) was answered

well by most candidates who correctly identified four other sources of information

that could have been used to confirm the information obtained. Part (b) was also

answered well by most candidates who recognised that many different kinds of

information could be obtained from the Internet. However, candidates were weak in

their understanding of evaluative criteria for information, such as currency and

authorship. Weak responses were also obtained for Part (c) when describing the four

stages of the problem solving process.

UNIT 2

PAPER 01

Section A – Software and Software Development

Question 1

Candidates answered the question generally well although many of them did not fully

explain their answers. Most candidates had some concept of the answers required

and most stated responses either exactly as required or provided synonyms for the

terms required. A few candidates interpreted the term ‘development’ as ’design’.

Question 2

This question was poorly answered. One strong point was that candidates knew the

steps for the waterfall model. Some candidates did not understand what was required

for Part (a) and simply stated the phases. Generally, candidates had a poor

understanding of the waterfall model when compared to the evolutionary model of

software development. Those who understood the models emphasized the wrong

points. For Part (c), some candidates needed to understand that stating an advantage/

disadvantage is not enough; it must also be explained.

Question 3

This question was answered fairly well. There were many good responses for Part

(a). Some candidates confused the feasibility study with the steps of the development

process.

Question 4

This question was answered fairly well by most of the candidates. Generally, the

data flow diagrams were well done. However, misconceptions on functional/non-

functional requirements were present.

Question 5

This question was generally done satisfactorily. Candidates had a strong

understanding of user-interface design principles. There were some misinterpretation

and confusing arguments in some of the responses of the candidates.

Section B – Programming Languages

Question 6

There was a major misinterpretation of this question by candidates. Some candidates

focused on the systems development life cycle rather than the characteristics of a

program. Generally, the candidates who scored between 6 to 8 (highest) had a good

knowledge of the concepts and theory being examined. A few candidates

misinterpreted the question as the “stages of a compiled program” and the “software

life cycle”.

Question 7

Generally, the performance on this question was good and there were several cases

where candidates got full marks. Most candidates knew how to describe imperative

languages. However, many candidates were unable to describe declarative languages.

For Part (b), some candidates thought that defining a top-down approach was

specifying an advantage.

Question 8

Overall responses for this question were very good but in some cases, groups of

candidates performed very poorly. In many cases candidates achieved full marks.

Candidates did very well at distinguishing between 2GLs and 3GLs. Many

candidates did not relate their explanations of the factors given (such as runtime /

portability / maintainability) to the issue of choice of programming language. In the

context of choice of programming language, the responses were generally vague.

Many candidates could explain the factors given (in isolation), but displayed

confusion when attempting to say how the factors influenced the choice of a

programming language.

Question 9

The performance on this question was fair. Logical thinking by candidate was

generally not displayed. Loops were in the wrong place and many candidates had the

correct logic, but lost marks when printing the required value. The drawing of flow

charts instead of writing an algorithm was one of the weak points. The use of an

array to store the values instead of the use of two variables would have also been a

correct approach to this question. Candidates need to read the question to determine

what is required. More practice on the logical flow of a solution should be done.

Practical coding would benefit both candidates and teachers.

Question 10

This question was not done very well. Many candidates were able to define the term

algorithm. However, some candidates could not distinguish between bounded and

unbounded iteration and it was clear that they were not familiar with the terms. Also,

many candidates did not have a good grasp of classes and objects as well as private

and public methods.

SECTION C – Program Development

Question 11

The responses in this question were fairly good. Part (a) and Part (b)(i) were

answered fairly well. However, Part (c) was challenging to candidates. Many

candidates were able to identify the user interface objects and describe the purpose of

these objects quite well except for fonts. Most identified the event in Part (b)(i)

correctly. In Part (b)(ii) many candidates wrote algorithms based on procedural

programming. Some candidates actually wrote programming code for Part (b)(ii), but

an algorithm was required. Future candidates are encouraged to gain practice in

writing algorithms for event-driven programming problems.

Question 12

This question was not well done by most candidates. The candidates who knew the

material scored higher marks and the remainder performed badly. Some candidates

drew the stack and showed the contents or used trace tables. Some candidates did not

show the contents of the stack during each pass of the loop and just wrote the output.

Many candidates did not seem to understand the purpose of the “else” statement. In

the end, many could not figure out the purpose of the algorithm. Future candidates

and teachers are encouraged to gain practice in tracing and understanding algorithms

by using more practical examples of ADTs.

Question 13

The responses to this question were generally not of a high quality. In many cases, it

was clear that candidates did not know the proper syntax for programming statements

and tended to give answers in pseudocode. Candidates need to obtain more practical

experience in programming to answer the questions in this section.

Question 14

The responses to this question were generally good. The strong point of this question

was candidates’ ability to correctly list four tools typically provided in a

programming environment. However, the explanation of how these tools are used

was often poor, with few candidates being able to differentiate between a tracer and a

stepper. DFDs, HIPO charts etc. were common responses. It is suggested that

candidates make more active use of the tools in a programming environment so that

they can answer questions of this nature.

Question 15

The responses to this question were fairly good. Most candidates were able to list

two types of documentation and describe them properly. More than 50% were able to

get at least 1 out of 2 for Part (a)(i). Some candidates gave the same explanation for

unit testing and system testing. The key difference between unit and system testing

needs to be understood by candidates. A good response should have included a

phrase like “integration of modules” when describing system testing. More practical

examples should be used in demonstrating to candidates how system testing and unit

testing is done.

PAPER 02

Section A – Software and Software Development

Question 1

The question was fairly well answered by candidates. Candidates comfortable with

data flow diagrams (DFDs) gained most of the marks for this question. Most

candidates were able to identify the customer entity in the DFD. Some candidates

were able to explain what is an entity-relationship diagram (ERD) but were unable to

differentiate between an ERD and a data dictionary. Candidates often confused data

flow diagrams with ERDs and flow charts. Hence they were unable to give good

answers for Part (b). Candidates were also unable to give a good example to explain

what is a data dictionary. Future candidates need to gain more practice in drawing

DFDs. The distinction between DFDs and other types of diagrams such as ERDs

should also be carefully made and their respective symbols clearly distinguished.

Question 2

In Part (a), most candidates were unable to properly explain the use of CASE Tools.

Part (c) of the question was answered well by many candidates. The strong points of

candidates in this question were in identifying the attributes and relationships in Part

(b), defining object-oriented design in Part (c), and listing the design process

activities in Part (c). . Candidates also need to know the proper symbols to use when

drawing an entity-relationship diagram (ERD) and the features of function-oriented

design. Future candidates also need to gain more practice in drawing ERDs and

understanding the differences between other types of diagrams such as data flow

diagrams.

Section B – Programming Languages

Question 3

Candidates demonstrated a poor grasp of several basic concepts in computer

programming and the only strong point observed was in their explanations for Part

(a), on generations of programming languages. Most candidates had no

understanding of the concept of recursion and how to write a recursive algorithm.

Candidates should spend some more time on the topic of recursion and in

understanding control structures (if-else, loops).

Question 4

The responses to this question were generally good. The concept of an object,

instance variables, and classes were understood by most of the candidates. Parts (b)

and (c) were generally answered well; however, Part (c) had some poor answers. In

Part (a), some candidates were unable to identify the accessor and modification

methods required, so they just wrote any methods that came to mind. Candidates also

demonstrated weakness in explaining the concept of encapsulation. In Part (a)(iv),

they also had some problems in drawing the class diagram for the Plane object. In the

future candidates should practice drawing class diagrams listing the attributes, and

methods of a class.

SECTION C – Program Development

Question 5

This question was poorly answered, especially where ADT operations were required.

The best-answered portion of the question was Part (a), which required candidates to

identify different types of GUI controls for the user interface. Candidates need to

develop application and synthesis skills to tackle the problems in this section. Many

candidates are weak at problem solving and algorithm design. To perform well in this

section, future candidates need to improve in these areas.

Question 6

This question was generally not answered well. However, there seemed to be an

improvement in the quality of responses to this type of question, compared to the

examinations of 2004. Some candidates are still not able to write proper

programming statements. In Part (a), many candidates did not understand what

“constructors” and “accessors” were. In Part (b), the “addStudent” method was

generally poorly done, with incorrect code being written. Part (c) was generally well

done except for section (iii), which involved adding Student objects to a Faculty

object. This operation required the Faculty object to have a data structure such as an

array that could be use to hold the Student objects. The “addStudent” method is then

implemented by inserting the Student object into the array at an appropriate position

(e.g., the end of the array).

PAPER 03

INTERNAL ASSESSMENT

GENERAL COMMENTS

The performance on the Internal Assessment was generally good. However,

compared to the performance in 2004, there was a decline in both units, especially in

Unit 2. It was observed that the projects submitted by candidates were deficient in

various aspects and were still being awarded high marks by teachers. There were also

a number of cases where it was clear that candidates did not know exactly what was

required of them in the Internal Assessment and consequently, obtained very low

marks. Better performances on the Internal Assessment should lead to better overall

performances on both units as well as better performances on the other papers.

Candidates need to maximize the opportunity to get higher marks on the Internal

Assessment. Teachers also need to become more closely involved in the supervision

of the projects. This should lead to an improvement in candidates’ performance on

the Internal Assessment, solidifying their understanding of the theoretical aspects of

the units, which in turn, will lead to an overall improvement in performance.

Generally, most candidates chose appropriate topics for the Internal Assessment. The

topics chosen were relevant to the level of the candidates and the specific objectives

of the respective syllabuses. The treatment of the topics by the candidates was

adequate. A small percentage was comprehensive though some tended to be

superficial. The reports were also generally well presented and teachers complied

with requirements such as ensuring that there was a cover page for each project and

entering the marks on the required form.

DETAILED COMMENTS

Unit 1

Some of the projects submitted demonstrated below average performance in a

number of areas such as description of context, purpose of study and solution

process. In many cases, the responses to these areas and others defined in the

syllabus were non-existent, vague, or poorly expressed. Often, marks were awarded

to areas that were not specified in the syllabus and in those instances teachers were

not adhering to the mark scheme provided. In other cases, full marks were awarded

even though the defined criteria were completely absent.

In the problem identification section, it was clear that a number of candidates did not

understand the term “external entities and processes”. The description of the current

system and procedures also needed improvement. Candidates took the purpose of

their study straight from items described in the syllabus. However, these items need

to be tailored to their specific projects. Finally, candidates need to be reminded that

there is a word limit for their projects.

Unit 2

There were a few cases where candidates used the old Unit 2 syllabus to prepare their

Internal Assessment. As a result, irrelevant information was submitted. Similar to

Unit 1, marks were often awarded in a manner that was not consistent with what is

defined in the syllabus.

From the moderation of the Internal Assessment, it is clear that many candidates do

not understand how to write programs in a programming language. In some cases,

candidates wrote pseudocode where program code was required and this was

awarded full marks. In most cases, the functionality of the program written was

poorly described and there were no screen shots of the working system displayed in

the reports (perhaps indicating the inability of the program to work in the first place).

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2006

COMPUTER SCIENCE

St Michael Barbados

COMPUTER SCIENCE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2006

INTRODUCTION

This is the sixth year of open examinations for Unit 1 and the fifth year for Unit 2. There were three

examination papers in both units, namely, Paper 01, Paper 02, and Paper 03. In each unit, Paper 01

and Paper 02 were examined externally by CXC while Paper 03, the Internal Assessment, was exam-

ined internally by teachers and moderated by CXC.

In each unit, Paper 01 consisted of short-answer questions that were designed to test candidates’ breadth

of coverage of the syllabus. On the other hand, Paper 02 consisted of essay-type questions that were

designed to test their depth of understanding of the syllabus. Thus, candidates were expected to show

deeper insight and understanding of the topics examined in Paper 02.

In each unit, Paper 01 consisted of three sections, each one corresponding to a module of the syllabus.

There were five compulsory questions within each section, and each question carried 10 marks. The

maximum amount of marks that could be obtained in Paper 01 was therefore 150.

Similarly, Paper 02 of each unit consisted of three sections, each one corresponding to a module of the

syllabus. There were two questions within each section, and each question carried 30 marks. Candidates

were expected to answer one question from each section. The maximum amount of marks that could be

obtained in Paper 02 was therefore 90.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade remained 50 per cent,

30 per cent, and 20 per cent, respectively.

GENERAL COMMENTS

In general, performance on both units of the syllabus has continued to improve. In Unit 1, more than half of

the candidates are now obtaining grades I to III. In Unit 2, more than 40 per cent of the candidates are

now obtaining grades I to III. However, the performance on the School Based Assessment should have

been better. Better performances on the SBA should lead to better overall performances on both units as

well as better performances on the theory papers. Candidates need to maximize the opportunity to get

higher marks on the SBA.

Even though the performance in Unit 2 has improved, there is still concern about the level of programming

ability being demonstrated in Sections B and C of both written papers. Candidates continue to find it

extremely difficult to write even simple programs, in both the imperative and object-oriented programming

languages. Teachers are encouraged to have several programming labs and exercises done with the

candidates.

- 2 -

- 3 -

As recommended last year, candidates are encouraged, as part of their examination technique, to read

questions carefully before answering, and to respond with sufficient detail that is commensurate with the

marks indicated in the question.

GENERAL COMMENTS

UNIT 1

In general, performance on both units of the syllabus has continued to improve. In Unit 1, more than half of

the candidates are now obtaining grades I to III. In Unit 2, more than 40 per cent of the candidates are

now obtaining grades I to III. However, the performance on the School Based Assessment should have

been better. Better performances on the SBA should lead to better overall performances on both units as

well as better performances on the theory papers. Candidates need to maximize the opportunity to get

higher marks on the SBA.

Even though the performance in Unit 2 has improved, there is still concern about the level of programming

ability being demonstrated in Sections B and C of both written papers. Candidates continue to find it

extremely difficult to write even simple programs, in both the imperative and object-oriented programming

languages. Teachers are encouraged to have several programming labs and exercises done with the

candidates.

As recommended last year, candidates are encouraged, as part of their examination technique, to read

questions carefully before answering, and to respond with sufficient detail that is commensurate with the

marks indicated in the question.

DETAILED COMMENTS

UNIT 1

PAPER 01

SECTION A – Components of Computer Systems

Question 1

This question tested how a UPS could help protect a computer (part (a)), the concept of port connectivity

(part (b)) and the use of satellites as a transmission medium (part (c)). Several candidates obtained full

marks in this question. However, many candidates ignored the marks available for part (a) and just wrote

one point. A good response for part (a) would have mentioned that a UPS gives users time to close

applications and save their work. Part (b) was generally well answered with many candidates saying that

port connectivity had to do with advances in port design and cable types. For part (c), some candidates

failed to mention that satellite signals propagated to space and then to earth.

Question 2

This question tested candidates’ knowledge of the instruction set of a computer, client/server software and

peer-to-peer networks. Some candidates thought that the instruction set was the programming language

statements of a particular language e.g. Pascal. For part b(iii), some candidates did not know that the client

- 4 -

makes a request which the server attempts to service in the specific applications chosen. For part 2(a), a

good response was ‘the instruction set of a computer refers to the complete set of instructions which are

hard-wired into the micro-processor and which defines the set of instructions the computer will understand

and be able to execute’.

Question 3

Question 3 looked at features of disk management software, memory management in an operating system,

multitasking and multiprocessing. Responses for this question were generally good and many candidates

scored full marks. Weaker candidates mixed up ‘multitasking’ and ‘multiprocessing’ in part (c).

Question 4

This question required the candidates to know about features provided by network operating system

software as well as two utilities typically present in system software. The question attracted some of the

best responses in the examination. Candidates displayed a sound knowledge of most features provided

by network operating system software. Weaker candidates could explain either none or only one of the

utilities typically present in system software.

Question 5

In this question, candidates had to compare extranets to the Internet, explain the term ‘capacity of a

storage device’, explain ‘access speed of a storage device’ and explain ‘portability’. Except for parts a(i)

and a(ii), most of the other parts were generally handled well. In part (a), some candidates confused

‘intranet’ with ‘extranet‘. Part b(i) which referred to the capacity of a storage device was well answered,

with some candidates giving values of actual device capacities e.g. 100 MB.

SECTION B – Application of Computers

Question 6

In this question, candidates had to demonstrate knowledge of expert systems, data warehouses and data

mining. This question was generally poorly done. In Part (a), most candidates did not know what an

expert system was. An expert system stores knowledge of a limited domain in the form of rules and makes

inferences based on those rules, similar to a human expert. Also, many could not describe data warehouses.

However, Part b(ii) was answered better than Part (b)(i).

Question 7

This question examined monitoring and control systems, user interfaces and suitability of a computer

application for an organization. Most candidates were not familiar with monitoring and control systems as

tested in Part (a). Monitoring and control systems use computer technology to track/monitor processes

and take certain courses of action if problems occur. They are typically present in many industrial applications.

Parts (b) and (c) were well-answered.

Question 8

This question tested candidates’ knowledge of data security, system back-up and real time updating of

files. Most candidates answered Part (a) fairly well; however some did not describe, but rather just listed,

- 5 -

the main points. There are logical and physical approaches to data security. In the logical approach,

programs can have password protected access. In the physical approach, computers can be placed in

enclosed/protected areas. In Part (b), many candidates explained system back up but did not give an

example. An example could be a secretary backing-up her word-processing files on a file server at the end

of the day. Part (c) was done well by most candidates.

Question 9

This question started by asking how firewalls and encryption could protect an organization’s data from

unauthorized access. The other parts required a suggestion of ways in which computers could affect a

financial organization and the distinctions between batch and on-line processing. Responses for this question

were generally of a poor quality although some candidates scored full marks. Part (a) was generally

answered fairly well. One candidate stated that firewalls ‘helped protect an organization’s data from

unauthorized access’. However, this was already stated in the question and gained no marks. Part (c)

contained important types of processing used in information technology but many did not know the meaning

of the terms. Batch processing is where transactions are gathered together into a collection and are used to

update master files at a later time. Online processing is where transactions are executed and changes are

made to master files almost immediately.

Question 10

This question examined telecommuting and the ways in which computers help us entertain ourselves and

facilitate communication. In Part (a), many candidates mixed up telecommuting and teleconferencing.

Telecommuting is where employees can work from one location and communicate their work to the main

office at another location using computer and telecommunication technologies. Parts (b) and (c) were

well-answered.

SECTION C – Computer-based Problem Solving

Question 11

Part (a) of this question asked for similarities and differences between information stored on CD-ROM

and in a digital library. Part (b) pertained to sources of information relevant to monitoring the progress of a

hurricane. Part (a) was generally well answered with many candidates scoring full marks. For Part b(i),

approximately half of the responses stated that newspapers were a good source due to the fact that it was

updated regularly. However, it should be noted that newspapers are not updated regularly enough to

monitor a hurricane. Parts (b)(ii) and b(iii) were generally well answered.

Question 12

This question was about information and characteristics of information sources related to making a

financial decision. The businessman got his initial information from an anonymous email. Part (a) was not

answered very well as candidates did not incorporate certain key words in their responses which should

have related to information characteristics.

For Part (b), many candidates listed sources rather than listing and explaining them.

- 6 -

Question 13

This question looked at data sources and their characteristics. This question was a fairly simple question,

but many candidates answered it poorly. Questionnaires were a good source of information in this case.

Part (b) attracted some fairly good responses.

Question 14

For this question, candidates had to distinguish between DVD and CD ROMs as information sources.

Part (b) asked about sources of information based on a scenario. Overall, this question was answered well

and some candidates scored full marks. Part (a) was well answered. In Part (b)(i), some candidates used

sources that were not relevant to Francis’ situation. Three good sources of information for Francis were

newspapers, magazines and the Internet.

Question 15

This question was based on software tools appropriate for different tasks performed by a toy store at

Christmas. For Part (a), most candidates did not give a very suitable response and hardly anyone obtained

full marks. Spreadsheet was a good response for Part (a). In Part (b), most candidates were able to

identify the software tool, but there were hardly any reasonable responses for precautions which should be

taken. One precaution was to keep back-up hard-copies of the presentation.

PAPER 02

SECTION A – Components of Computer Systems

Question 1

Part (a) required candidates to differentiate between several pairs of computer terms. Part (b) examined

how an operating system could execute three processes at the same time. Part (c) tested candidates’

knowledge of the data requirements and communication requirements of a package.

a) Generally, candidates did not differentiate between the two terms given. Many candidates defined

each term separately without highlighting what made them different. For Part (a) (iii), a good

response was ‘cache memory is much faster than RAM and also more expensive’.

b) Clearly, most candidates knew that the operating system could handle and execute many processes

at the same time. However, only brief explanations of how this was done were given by most

candidates. A good response should have contained a discussion of round-robin allocation of

time-slices, preemption of processes, interrupts and saving and reloading of process states.

c) Some candidates seemed confused in Part (i) and spoke about qualities of information. They did

not read the scenario carefully and answer based on the situation presented. Many also spoke

about “database attributes”. More focus should have been placed on the fact that different

departments were involved.

- 7 -

Question 2

Part (a) of this question presented a diagram and candidates were asked to identify the network topology

illustrated. Part (b) asked for two advantages of fibre-optic cable over coaxial cable, Part (c) asked about

the OSI model for computer communication and Part (d) looked at data transmission in bus and ring

networks. Overall, this question was answered very well in most instances. For Part (c), some candidates

did not focus on the key functions of the various layers of the OSI model. For example, the main role of the

network layer is to route packets although there are other subordinate functions.

SECTION B – Application of Computers

Question 3

Part (a) required candidates to use examples to distinguish between data loss and data corruption. Part (b)

examined levels of management and the associated information systems. Part (c) asked about precautions

to minimize data loss in the event of a hurricane. Overall, the question was generally well answered although

it was not a popular question. In Part (b)(ii), many candidates knew of the different types of information

systems but had problems relating them to the levels of decision making. Parts (a), (b)(i) and b(iii) were

fairly well done as were Parts 3(c)(i) and 3(c)(ii). Some weak responses for 3(c)(ii) were UPS, surge

protectors and covering computers with plastic bags. One precaution was ‘to move data to a safe storage

location in the building: the safe storage location could be a sturdy cabinet that was waterproof and located

in a sturdy room’.

Question 4

This question looked at techniques to deter unauthorized access to a web based system, software piracy,

ways in which an organization could benefit from automation and three ways in which computer applications

have affected the job market. The question was very popular. Parts (a) and (b) were satisfactorily done.

Many responses for Parts (c) and (d) were similar. For Part (c), some responses were too simple and

involved expressions like “more work done”, “less salaries to pay out” etcetera. There was a clear

indication that some candidates misinterpreted Part (d). Also, positive and negative ways in which

computers affect the job market should have been considered for

Part (d). For example, new jobs have been created, for example,. software developer and computer

applications have caused displacement of a number of clerical employees in some organizations.

SECTION C – Computer-based Problem Solving

Question 5

Part (a) of this question presented a scenario and candidates had to describe three other sources of

information to confirm information found on a web site. A discussion of evaluative criteria for acceptance

or rejection of information was then required. Part (b) asked for the benefits to an organization of using the

Internet as an information source and part (c) required a discussion of three characteristics of information

sources. Part a(i) was fairly well answered. Good responses for part a(i) included valid information from

other related sources, Web sites, experts in the financial sectors, articles in reviewed magazines, newspaper

articles and television programmes. Part a(ii) was badly done. Very few candidates were able to identify all

- 8 -

three evaluative criteria correctly. Most candidates did not correctly relate their explanations to the stated

criteria. Many candidates did not focus on “using the Internet as an information source”, but rather discussed

the use of the Internet as a way to promote their business. Candidates should pay attention to what the

question is specifically asking and read questions carefully. Part (c) was not well done. Many candidates

were not clear about the definition for each term and hence were not able to obtain all three marks for each

concept.

Question 6

The majority of the candidates attempted this question which was based on a University scenario that

required the use of software tools for different tasks. Part a(i) was generally well-answered. A minority of

candidates did not obtain the mark due to lack of familiarity with the names of appropriate software tools.

For example, some cited Microsoft ® Excel as an example of a database software tool. Part a(ii) was

fairly well answered although many candidates failed to explain their answers sufficiently in order to obtain

full marks. Candidates need to pay attention to specific technical terms as it relates to the particular software

tool. Parts b(i) (ii) had similar problems to Parts a (i) and a (ii). In Part (c), a minority of candidates

obtained all six marks. Most candidates did not include in their response, the ‘retrieval of information from

the database and spread sheet programs’, into the presentation. Many of them simply stated that Microsoft

® PowerPoint could be used. Part (d) was not well-answered. Many candidates cited word-processors

as opposed to desktop publishing software as the most appropriate tool for producing the newsletter and

brochures. Many of them were not able to effectively explain the reasons for their selection.

UNIT 2

PAPER 01

Section A – Software and Software Development

Question 1

This question tested candidates’ knowledge and understanding of the waterfall model, which is one of the

most widely known models of a software development process. There were good answers to Part (a).

However, many candidates did not properly explain that a software development process is a set of

activities and associated results which lead to the production of a software product, saying instead that it is

a sequence of steps. In Part (b), some candidates did not specify the names of the stages of the waterfall

model or specified the names incorrectly. Poor explanations were sometimes given for the activities that

take place in the three stages, especially for the design stage. Part (c) was generally answered well, though

some candidates indicated that a weakness of the waterfall model is that it is time consuming and costly,

instead of mentioning the inflexibility inherent in the model.

Question 2

This question tested candidates’ knowledge and understanding of two information gathering techniques

that can be used during requirements analysis, namely, interviews and questionnaires. It also required

candidates to determine which one was more appropriate for the scenario given. There were good answers

to all three parts of this question and valid reasons were given for choosing one technique over the other in

the situation given. However, there were cases where candidates stated that interviews allowed one to

have face-to-face or direct encounter with stakeholders but did not indicate why this was valuable to the

- 9 -

information gathering process. Some candidates also failed to state that responses to interview questions

are documented and that responses to questions in a questionnaire are collected and analyzed.

Question 3

This question tested candidates’ knowledge and understanding of Computer Aided Software Engineering

(CASE) tools. The question was generally answered well by candidates though some did not answer the

question at all. However, in Part (a), some candidates failed to mention that CASE tools are software

products. Some failed to mention that it is used to support or automate different aspects of the software

development process. In Part (b), some candidates did not identify the stages of the software process

where CASE tools can be used. Few candidates mentioned pertinent terms such as “graphical system

models” and “technical and user documentation”. In Part (c), some candidates incorrectly stated that one

disadvantage of a CASE tool is that software developers would become less skilled by using the tool.

Question 4

This question tested candidates’ knowledge and understanding of architectural design which is an important

design process activity. This question was poorly answered by most candidates, indicating inadequate

preparation or understanding of this aspect of the syllabus. In Part (a), few candidates were able to

explain that system structuring in the breaking down of the system into subsystems and some even confused

this term with the software development life cycle. In Part (b), many candidates stated that modular

decomposition is the breaking down of the system. However, this is partially correct; modular decomposition

is the decomposition of sub-systems into modules. In Part (c), candidates were unable to distinguish

between a sub-system and a module. Some of them incorrectly stated that a sub-system performs a single

task. The responses to Part (d) were generally poor and candidates failed to mention that outputs include

graphical representations of system models and diagrams showing the system structuring and modular

decomposition.

Question 5

This question tested candidates’ knowledge of user interface design guidelines and guidelines for designing

reports. There were fairly good responses to Part (a) of this question. However, guidelines such as “user

familiarity”, “consistency”, and “user guidance and feedback” were rarely mentioned. Also, instead of

stating “recoverability” as the ability to gracefully recover from errors, candidates used a less appropriate

term, “robustness”. The responses to Part (b) were not satisfactory. Guidelines for report design such as

using meaningful titles, presenting relevant information, and balancing the layout of the information were

generally not described. Instead, candidates described the information that the reports should contain and

how they should be prepared.

Section B – Programming Languages

Question 6

This question tested candidates’ understanding of various object-oriented concepts and their ability to

represent a class in a diagram. Candidates were required to explain the concept of inheritance and to

describe how message passing takes place among objects. Part (a) was answered well by most candidates

though data types for the instance variables and the “()” after method names were often omitted. In Part

(b), candidates seemed to understand how message passing took place but did not clearly explain the

process. Some incorrectly indicated that the Customer object would call the Account object using its

- 10 -

withdraw() method as follows: A.withdraw(). Some candidates missed the point of the question completely

and talked about the physical process of going to the bank to withdraw/deposit money using ATMs and

bank slips. Part (c) was not answered well by most candidates and no response gained full marks. It

should be noted that inheritance is when a class assumes the instance variables and methods of another

class, perhaps adding its own instance variables and methods.

Question 7

This question tested candidates’ knowledge and understanding of several fundamental object-oriented

concepts. The answers to Part (a) were generally correct though some of the candidates were unable to

properly distinguish between an object and a class. A class is like a template that describes the behaviour

and attributes of a set of similar objects while an object is an instance of a particular class. Part (b) was not

answered well and many candidates did not explain that encapsulation is the technique of hiding the internal

implementation details of an object from external objects. For Part (c) few candidates explained that

public instance variables can be accessed directly by external objects. Thus, if these variables change in

the future, all the external objects would have to be changed since they were not shielded from the changes

by encapsulation.

Question 8

This question tested candidates’ knowledge and understanding of programming languages and the use of

compilers and interpreters. Part (a) was answered well and most candidates were able to classify

programming languages in terms of level such as high-level versus low-level. Some gave examples in cases

where they could not identify specific terms and these were accepted. In Part (b), most candidates only

gained half the marks since they only stated one reason, namely that compiled code executes faster than

interpreted code. Another reason is that compiled code can be distributed and executed without need for

the compiler, unlike interpreted code which always requires the presence of the interpreter. Part (c) was

generally answered well though some candidates lost marks by missing out one or more criteria for selecting

a programming language.

Question 9

This question tested candidates’ ability to trace through an algorithm and to determine the output that it

generates. The majority of candidates gained only partial marks for this question since they only had part

of the output correct. Many candidates had no idea how to trace through the algorithm and simply rewrote

the algorithm in their responses. However, a few candidates gained full marks by correctly drawing the

shape that would be generated by the algorithm. The following is the expected output of the algorithm:

**********

* *

**********

Please note that full marks were awarded for alternative output based on a different valid interpretation of

one of the functions given.

- 11 -

Question 10

This question tested candidates’ ability to write algorithms using bounded and unbounded iteration.

Part (a) was answered well by many candidates. However, many candidates only obtained partial

marks for Part (b). Several candidates attempted to use the modulus function but used it incorrectly.

Generally, the responses by candidates in this question indicate that they are unable to write simple

algorithms using the basic programming control structures of sequence, selection, and repetition. Nu-

merous responses indicate that candidates do not know how to write a while loop or how to specify

the conditions of a while loop or an if statement. Candidates have difficulty in integrating the different

programming structures into a working algorithm. They also need to appreciate the difference between a

for loop and a while loop.

SECTION C – Program Development

Question 11

This question tested candidates’ knowledge of graphical user interface objects and their ability to choose

appropriate user interface objects to solve a given problem. The question was answered well by most

candidates, many of them getting full marks. While there were a few candidates who designed the dialog

box with inappropriate graphical user interface objects, the majority designed the dialog box consisting of

the following objects:

• Fonts that were used to create labels and the window title

• Text field to enter the first number, labelled with a label object

• Text field to enter the second number, labelled with a label object

• Text field to display the larger of the two numbers, labelled with a label object

• A Button that causes the program to display the larger of the two numbers

• A Window to contain all the other objects

Question 12

This question was designed to test candidates’ knowledge and understanding of the queue abstract data

type (ADT). The queue ADT removes elements from the top of the queue using a dequeue() operation

while elements are added to the end of the queue using an enqueue(Object) operation. The question was

generally answered badly by most candidates with Part (a) and Part (b) being answered better than Part

(a). Two ADT operations for a queue are enqueue (Object) and dequeue(). An algorithm to remove the

elements of Q1 and place them in Q2 is given below:

while not Q1.isEmpty() do

set element to Q1.dequeue()

Q2.enqueue(element)

end while

In the future candidates are encouraged to gain practice in writing algorithms that employ ADTs such as

queues, stacks, and linked lists.

- 12 -

Question 13

This question was designed to test candidates’ ability to write the source code for a class in an object-

oriented programming language. It required candidates to declare three instance variables of different data

types and to code two methods with simple functionality. Candidates were also expected to write a

constructor for the class given. The majority of candidates performed very badly in this question indicating

that most of them are not getting adequate exposure to practical object-oriented programming. Nevertheless,

a few candidates performed quite well and obtained high marks in this question. A number of candidates

mixed and matched language features from different programming languages. In the future, candidates

should indicate which language they are using and use only the features of this language in their responses.

Question 14

This question tested candidates’ knowledge of tools that can be used to simplify the process of creating

computer programs. The tracer was given as an example. Other tools include editor, stepper, linker, and

debugger. Most candidates answered this question reasonably well. However, some candidates identified

software engineering tools although from the example of the tracer, it was clear that programming tools

were required.

Question 15

This question was designed to test candidates’ knowledge and understanding of the stack abstract data

type (ADT). Candidates were required to find the sum of every other element in a stack using only stack

ADT operations. Like Question 12 involving the queue ADT, most candidates answered this question very

poorly. However, there were a few rare responses that specified a correct algorithm such as the following:

set sum to 0

while not S.isEmpty() do

set element to S.pop()

add element to sum

if not S.isEmpty() then

S.pop()

end if

end while

A few candidates attempted to define the stack ADT operations in their responses to this question. However,

the question required candidates to use the ADT operations to solve the problem given. It was also clear

that many candidates had great difficulty in writing algorithms involving the fundamental programming

constructs of sequence, selection, and iteration. If candidates do not master these concepts, it will be

impossible for them to perform well in the questions involving the writing of algorithms or programming

code.

PAPER 02

Section A – Software and Software Development

Question 1

This question was designed to test candidates’ in-depth knowledge and understanding of two software

process models, evolutionary development and the reuse-oriented approach. The question also tested

- 13 -

candidates’ ability to draw a level-0 data flow diagram from a given narrative. The question was answered

fairly satisfactorily by most candidates; however, there is considerable room for improvement in answering

this type of question. For example, in the data flow diagram, symbols were often mixed up, indicating that

candidates did not fully understand the notation used in data flow diagrams or they lacked experience in

applying the notation. The numbers of the processes were often omitted and the data flows were generally

incorrect especially with respect to their direction or description of data.

In Parts (b) and (c), candidates demonstrated a good understanding of the evolutionary approach and the

reuse-oriented approach for developing software. However, the weaknesses of the evolutionary approach

were not properly discussed in Part (b). Similarly, in Part (c), the discussion of the reuse-oriented approach

was often not as thorough as was expected. Candidates are reminded that in answering the questions of

Paper 2, they are expected to give detailed responses to questions having keywords such as “discuss” or

“describe”.

Question 2

This question was designed to test candidates’ in-depth knowledge and understanding of tools that are

typically used in the analysis phase of software development. In particular, it focused on entity-relationship

modeling and the use of data dictionaries to formally define data in an organization. Part (a)(i) of the

question was fairly well answered by most candidates who attempted this question. Few candidates

described the attributes of entities as being a component of the model and some incorrectly stated that the

cardinality of relationships is a main component of the model. Part (a)(ii) was poorly done by most candidates.

Sources of information to draw an entity-relationship model include computer displays, reports, data flow

diagrams, and business documents such as forms. The performance in Part (a)(iii) was fair. Most candidates

were able to identify at least three entities from the scenario given. Relationships between entities were

generally drawn correctly with a straight line connecting the entities; however, in some cases, no name was

given to the relationship.

Part (b) was generally well done by most candidates. They correctly explained that a data dictionary is like

a database or repository of data that is important to the organization (that is, a database of metadata).

Many candidates obtained full marks for correctly describing the contents of an entry in the data dictionary.

Question 3

This question was designed to test candidates’ knowledge and understanding of programming languages

and the steps of the object-oriented process. It also tested candidates’ ability to write and use recursive

functions. Part (a) was badly done by most candidates. Most of them were unable to write the function

properly with the necessary arguments or call the function with the right parameters and capture the value

returned. They were also unable to return the value of -1 if the search was unsuccessful. Candidates need

to improve their skills in defining and calling functions with the appropriate parameters. They also need to

understand how to design a recursive function.

Parts (b) and (c) were generally well done. However, in Part (b), there were candidates who described the

stages of the general software process instead of the specific stages of the object-oriented process. Also,

even though good answers were given for the maintainability aspect of Part (c), many candidates failed to

mention that the ripple effect of errors is minimized in a maintainable program and that new changes would

not usually cause existing code to malfunction.

- 14 -

Question 4

This question was designed to test candidates’ knowledge of various aspects of programming languages.

It also tested candidates’ ability to write procedural algorithms involving repetition statements (such as for

loops) and simple data structures such as arrays. The majority of candidates who answered this question

omitted Parts (a) and (b), indicating their inability to write algorithms to solve a given problem. Those who

attempted Parts (a) and (b) incorrectly specified the for loop. They were also unable to iterate through the

array (by using the loop index variable) to move or swap the elements. Part (c) was bookwork and was

fairly well done by most candidates. However, in Part (d), a large number of candidates described the

software development life cycle instead of describing the hierarchical decomposition of programs into

modules and the use of the fundamental control structures of sequence, selection, and iteration. Finally, in

Part (e), it was expected that candidates would describe the use of the compiler or interpreter to translate

a program and the subsequent execution of the compiled or interpreted code on the computer.

Question 5

This question required candidates to identify the graphical user interface (GUI) objects that would be used

in a simple program as well as to identify the GUI events that the program needs to anticipate and explain

how these events would be handled when they occur. In Part (a), candidates were able to identify

the GUI objects that would be used, but many of them had difficulty explaining the purpose of the different

GUI objects. In Part (b), several candidates were unable to draw an appropriate GUI with a pleasing

layout that one would expect from a modern GUI application. A few candidates even drew several GUI

objects without placing them on a window. In Part (c), many candidates were unable to describe two

events that the program should anticipate. One event that the program should anticipate is the user clicking

on a “Calculate” (or similar) button when he/she is ready to perform the interest calculation. Another event

is the user pressing the “Return” or “Enter” key in one of the text fields.

In general, Parts (d) and (e) were badly done and many candidates did not answer Part (f) at all. The code

written in these Parts was of a very poor quality, highlighting candidates’ inability to write code for fairly

straightforward problems. In Part (d), most candidates were unable to write the code to populate L with

the required numbers. In Part (e), many candidates did not retrieve the data entered by the user in the text

fields before doing the calculation of simple interest. For example, in Java, a getText() method would be

called to obtain the data entered in the different text fields and this data would then be converted to the

required type such as int or double before calculating the simple interest. In Part (f), after calculating the

monthly installment, a setText() method (in Java) would be used to put the value calculated in the relevant

text field.

Question 6

This question tested candidates’ ability to write code for an object-oriented program. Candidates were

required to write the code for a Product class with given instance variables and methods. They were also

required to write code to create instances of the Product class and to store these instances in an appropriate

data structure such as an ArrayList or Vector (in Java). Many candidates were unable to correctly write

the code for the Product class. Some candidates did not know how to write an accessor, a method that

simply returns the value of an instance variable. Many candidates were unable to correctly reference the

Product objects created in Part (b) and put them into the data structure specified. An iterator would then

be used to access the Product objects in the data structure. By invoking the reorder() method on each

Product, it can be known whether a certain product has to be reordered. In Part (c), candidates were

expected to describe how the technique of inheritance could be used to create a subclass of the Product

- 15 -

class. This subclass would automatically inherit the instance variables and methods of Product. Inheritance

allows additional instance variables to be declared and it also allows the reorder() method to be implemented

differently in the subclass.

PAPER 03

INTERNAL ASSESSMENT

GENERAL COMMENTS

The quality of the Internal Assessments for Unit I in the 2006 examinations was much better than last year.

In Unit 2, most candidates made at least a fair attempt at the Internal Assessment. A few schools used the

old syllabus and this presented some problems.

Generally, most candidates chose appropriate topics for the Internal Assessment. The topics chosen were

relevant to the level of the candidates and the specific objectives of the respective syllabuses.

One centre submitted samples that were based on the same scenario for a batch of candidates. Teachers

should note that each candidate should generally use a different scenario from others for their projects.

DETAILED COMMENTS

Unit 1

Many candidates did not identify a problem in the ‘Problem Identification/Situational Analysis’ section but

rather in the ‘Description of the current system and procedures’ section. Candidates are asked to desist

from including information meant for one section in another section. Teachers should carefully apply the

marking scheme when marking candidates’ internal assessment reports. They were many instances where

full marks were awarded to sections that were not present in the report, for example, some candidates

omitted the ‘Problem Identification/Situational Analysis’ section and still obtained full marks in that section.

Description of the purpose of the study

In this area, many candidates did not address the purpose of the study properly especially with regard to

the guidelines given in the syllabus, for. example, response times, volume of data to be stored and processed,

and tolerance levels. Descriptions of tolerance levels were particularly weak.

Solution Process

Many candidates used data collection such as interviews and questionnaires but most did not analyze the

responses from the questionnaires.

Analysis of Solution

o Hardware Specification: Many candidates stated instead of describing the systems specifications,

for example, instead of describing hardware components, some candidates simply provided a

listing and many teachers awarded full marks.

- 16 -

o Software Specification: Many candidates simply listed examples of various types of software and

provided no descriptions. Network layout diagrams were generally missing.

o Evaluation of Solution: Some candidates attempted to justify their choice of networking components/

hardware/software absolutely rather than relative to other possibilities. In the case of networking,

a good response was that a bus network is generally cheaper to set up than a star. Justification of

compatibility of network, hardware and software was virtually non-existent.

o Presentation: Diagrams were fairly ok.

Organizational Impact

o Enhancements: Much research was not done on other software that existed on the market. Too

many candidates used Microsoft Access® for databases rather than software with a “proven”

track-record for the specific functions required.

o Negative Impact: Fairly well-done

o Identification of strategies for addressing issues: While some candidates were able to address this

area fairly well, the ‘public relations’ and ‘advertising’ sections were either misunderstood or not

attempted at all.

o Justification: Some candidates seemed to have been oblivious to the fact that they needed to justify

strategies used in the section above.

o Presentation: Many candidates did either an inadequate executive summary or none at all. Quite a

few failed to produce a proper contents page and some manually numbered the pages with a pen.

o Communication of information: Candidates need to proof-read their work since there were

grammatical errors in many reports.

RECOMMENDATION

Internal Assessment reports should be typed in black ink using a suitable font with double line spacing.

Unit 2

Most parts of the project had at least fair attempts. The ‘System Models’ section was not properly done.

It should be noted that the ‘System Models’ section of the internal assessment is an important area in

Computer Science. This section shows candidates’ ability to apply rules to a given situation to create

models for analysis. In many cases the rules for drawing entity-relationship diagrams and data flow diagrams

were violated. Many context diagrams included data stores, which is a major error. Some candidates

mixed up level 1 and context data flow diagrams. It is also clear that some candidates have not been using

the recommended texts from the syllabus. Projects were generally well-presented except for a few cases

where the overall project presentation could be improved by the use of double spacing and the use of

black (not colored) ink for text. In most cases, candidates were unable to give appropriate responses for

the non-functional requirements of the project. This information can be found in books such as the Software

Engineering text by I. Sommerville.

RECOMMENDATION

Teachers need to provide more guidance for candidates as they work through the SBA. For example,

teachers can teach the material for the various parts of the SBA and then see how the candidates apply

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2008

COMPUTER SCIENCE

(REGION EXCLUDING TRINIDAD AND TOBAGO)

St Michael Barbados

COMPUTER SCIENCE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATIONS

MAY/JUNE 2008

INTRODUCTION

This is the eighth year of open examinations for Unit 1 and the fifth year for Unit 2. There were three

examination papers in both units, namely, Paper 01, Paper 02, and Paper 03. In each unit, Paper 01 and

Paper 02 were examined externally by CXC while Paper 03, the Internal Assessment, was examined by

teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple choice questions that were designed to test candidates‟

breadth of coverage of the syllabus. On the other hand, Paper 02 consisted of essay-type questions that

were designed to test their depth of understanding of the syllabus. Thus, candidates were expected to

show deeper insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade remained 50 per cent,

30 per cent, and 20 per cent, respectively.

GENERAL COMMENTS

In general, performance on both units of the syllabus has continued to improve. In Unit 1, 80 per cent of

the candidates obtained grades I to III. In Unit 2, 70 per cent of the candidates obtained grades I to III.

However, the performance on the School Based Assessment should have been better. Better

performances on the SBA would lead to better overall performances on both units as well as better

performances on the theory papers. Candidates need to maximize the opportunity to get higher marks on

the SBA.

Even though the performance in Unit 2 has improved, there is still concern about the level of

programming ability being demonstrated in Sections B and C of both written papers. Candidates continue

to find it extremely difficult to write even simple programs, in both the imperative and object-oriented

programming languages. Teachers are encouraged to have several programming labs and exercises done

with the candidates.

As recommended last year, candidates are encouraged, as part of their examination technique, to read

questions carefully before answering, and to respond with sufficient detail that is commensurate with the

marks indicated in the question.

DETAILED COMMENTS

UNIT 1

PAPER 01

SECTION A – Components of Computer Systems

Question 1

This question tested candidates‟ knowledge of extranets, intranets, ROM, RAM, deadlocks and system

software.

For Part (a) most candidates obtained only half of the allotted marks. They understood the difference

between internet and extranet but they did not adequately differentiate between the two.

For Part (b) while the candidates were able to identify the importance of ROM and RAM, they confused

both and as a result they mostly obtained half the mark. They did not adequately identify the software

associated with each type.

In Part (c) most candidates obtained more than half the mark. Most answers showed that the student

applied their knowledge appropriately and gave several responses that demonstrated a higher level of

thinking.

Part (d) was generally well done but in some cases they were suggesting categories of software instead of

giving precise examples.

Part (e) was poorly done. Candidates were unable to explain exactly what constituted a deadlock in

computer terms. A possible explanation could have been that a deadlock refers to specific conditions

when two or more processes are waiting for each other to release a resource. Candidates did not provide

adequate examples.

Part (c) (ii) was poorly done. Candidates had a tendency to give definitions of an interrupt but were

unable to explain how an interrupt is handled whenever it occurs in the operating system of a computer.

A typical solution involves stating that the operating system saves the state of the current process, in

execution, stores information related to the new process, runs the process and then restores the original

Part (c) was attempted by the majority of but many only obtained two marks out of four. A number of

candidates provided file management utilities as well, and hence did not obtain full marks. A good

response would mention fusing bad disk cluster errors and defragmenting.

Question 2

This question tested candidates‟ knowledge of

 The role of the OSI model during the transmission of a text file from one computer to the next

 HTTP, FTP, client/server and peer-to-peer as it relates to the internet

Part (a) was poorly done. Candidates were able to produce a diagram of the OSI model but showed little

understanding of the workings of the model. Candidates were unable to explain the functions of each

layer.

For Part (b), candidates were required to differentiate between „HTTP‟ and „FTP‟. Most candidates

wrote what they stand for instead of giving their functions.

A few candidates understood the concept of the approaches that can be used to share files over the

internet for both client/server and Peer-to-Peer. Most candidates seemed not to understand that the

client/server and Peer-to-Peer are two different network architectures and that the terms are not

interchangeable.

Few candidates were able to explain that distributed network configurations are fault tolerant and that

they have extensive processing power. Most confused the distributed network configuration with the

centralised configuration.

SECTION B - Application of Computers

Question 3

Part (a) of this question presented a scenario and candidates had to discuss economic and social

implications of using vending machines in a business. Part (b) of this question also presented a scenario

and candidates had to suggest advantages of using computer applications to store customer records.

For Part (a) (i) and (ii). These parts were generally fairly well done. However, a number of candidates

seemed not to know the difference between economic implications and social implications. Most

candidates did not “Discuss” the implication.

For Part (b)(i) Most candidates were able to respond adequately to this part.

For Part (b)(ii) The majority of candidates were able to identify advantages of computer applications but

did not go on to discuss these advantages and so were unable to gain full marks. The two main

advantages identified were quick retrieval of data and easy boot up of data.

Responses to part (c) varied due to the fact that it was subjective. As a result, most candidates

comfortably gained marks.

Almost all responses to part (d) identified two ways to control unauthorised disclosure but most were

lacking when it came to the discussion.

The majority of candidates did not gain marks for part d (ii) because they failed to identify an additional

strategy other than password or what was identified in d (i). A possible strategy is the use of remote

biometric scan e.g. retina scan with remote validation.

Question 4

For Part (a) Candidates were required to explain what is meant by real-time updating of files in a

computer application and outline one situation where this approach is necessary. This was well done by

most of the candidates. However:

- Some candidates did explain what real-time updating of files is but failed to give any

example where this approach to updating is necessary.

- Some candidates‟ definitions of real-time file update were incorrect. However, the examples

given were correct and therefore candidates were awarded some of the marks.

- There were few candidates who did not understand the concept by the reflection of their

answers which were totally inappropriate or incorrect.

For Part (b) candidates were required to explain the meaning of data loss, data corruption and

unauthorised access as it relates to problems with the computer system at the retail store.

Most candidates did not explain what unauthorised access is, however, some candidates had problems

differentiating data loss from data corruption and simply gave answers such as “Data loss as it suggests is

loss of data in a computer system.”

For Part (b) (ii) in each of the problems, candidates were required to suggest one way in which the

computer consultant may have determined the existence of that problem. For data corruption if it could

have been determined if the contents of a file were changed.

Some candidates redefined data loss, data corruption and unauthorised access, instead of showing how

the consultant might have determined the existence of that problem.

For Part (c) (i) Candidates were required to describe two strategies that can be used to back-up critical

data in a government organization that is located in a country in a region prone to hurricanes.

Most candidates just stated the two strategies and did not describe them and therefore could only gain a

fraction of the marks awarded.

In Part (c) As a problem solving exercise, candidates were required to determine the steps required to be

taken after a hurricane in order to restore the computer systems and data. A vast number of candidates

made no reference to the problem solving steps but demonstrated their understanding of the step by the

solution they provided.

SETTING - Computer based Problem Solving

Question 5

The majority of candidates scored between 20 and 30 marks. Part (a) was generally well done. Parts (b)

and (c) sought to test candidates‟ appreciation of the qualities and characteristics of information and

information sources; these parts were reasonably well done. For part (a) possible responses could have

included:

(i) It is much more convenient for readers to access articles from their office or home computers

instead of having to wait on the mail or go to the library.

Readers have access to a wider range of articles at a very reasonable price compares to paper

subscriptions or library subscriptions which are limited by cost.

(ii) Managing subscription and distributing articles is more convenient with digital technology.

Articles are placed in the library and readers only have to download articles from the library.

Publishers reduce time and cost by not having the traditional long delays of publishing. The articles are

more attractive to read since they are less out-of-date by the time they are published online.

Question 6

This question tested candidates‟ depth of knowledge of database and spreadsheet applications software,

architecture and usages.

Candidates would have had to read widely and pay attention to formulae to answer the question. Many

candidates seemed to have relied on the auto sum function and were not current with the formulae.

UNIT 2

PAPER 02

SECTION A – Software and Software Development

Question 1

Part (a) of this question tested candidates‟ ability to draw a data flow diagram (DFD) and their

understanding of the different features of the rules governing the creation of DFD‟s. Part (b) tested

candidates‟ knowledge of the properties of well-engineered software. Part(c) tested candidates‟

knowledge of the advantages and disadvantages of evolutionary development over the waterfall

approach.

A significant number of candidates attempted part (a) of the question but many candidates were not able

to differentiate between a context diagram and a level-o diagram.

A context diagram is an overview of flow of information to and from the external entities to the system.

The context diagram contains no data stores and only one process which is the system.

The level-o diagram contains more details and is a more detailed view of the context diagram. Here the

flow of information between the external entities, data stores and processes are shown.

Candidates also had problems correctly labelling data flows where data flows were labeled using verbs

rather than nouns and adjectives.

Many candidates seemed to have been exposed to only one data flow model and therefore incorrectly

identified the error in the diagrams as being an incorrect symbol rather than the presence of a process in

Figure and the absence of input to the process in Figure 1.

In Parts (b) most candidates did a very good job at discussing the two properties of well-organised

software.

For Part (c) answers given by candidates, for one advantage and one disadvantage of evolutionary

development over the waterfall approach were satisfactorily answered. However, some candidates

ignored the word discuss

and therefore were not able to gain full marks as they did not explain their

points.

Question 2

In Part (a)(i), most candidates were unable to identify the two sources of information for drawing an

ERD.

In Part (a)(ii), most candidates were able to construct a proper ERD. In some instances they were unable

to establish proper cardinality and relationships. This was rare however. Few candidates drew a DFD

instead of an ERD but were still able to identify the entities.

Part (b) which tested candidates knowledge of graphical over interface, was fairly well done by most

candidates. However, there were instances where candidates tended not to answer the question within

the context of the given scenario. Instead, they attempted to discuss general features of a user interface.

In Part (c), most candidates obtained 2-3 marks out of 6. This was due to the fact that they often gave

three reasons why users and management needed to be involved in the development of a software

product, which amounted to the same point. They mainly discussed the point that “managers and users

are able to specify their needs”, in different ways.

SECTION B – Programming Languages

Question 3

Part (a) tested candidates‟ ability to write on algorithm to reverse the order of the content of an array.

Many candidates were not able to gain full marks, due to the size of the dataset. Candidates were able to

give a specific solution rather than a general solution. They were not able to swap the elements from

location 5 to 9 correctly.

Part (b) was also poorly answered. Candidates were unable to manipulate the indices of the given array.

This followed the candidates in arranging even numbers to the front of the array and adding numbers to

the back of the array. They were unable to write a proper statement determining whether an integer is

odd or even.

Part (c) which tested candidates‟ knowledge of First and Second generation programming languages was

generally well answered.

Part (d) was also well answered, candidates were able to identify control constructs in structured

programming but were unable to give proper examples.

Question 4

This question tested the candidates‟ ability to:

(a) manipulate predefined functions to produce desired output.

(b) manipulate a list of values to produce the largest values using recursion.

Part (a) was generally well done. Most candidates did not assign values to the parameters and output

string.

Part (b) posed a great deal of difficulty for most candidates. Candidates focused on programming

languages rather than what the applications created would be used for (cell phone and desktop computer).

In Part (c), most candidates did not attempt this part of the question. Responses reviewed showed limited

programming skills and lack of knowledge of recursive functions.

Part (d) was generally well done by the candidates, but most were not able to fully explain how objects

communicate by working methods on other objects to which they have no reference. A possible solution

for part d (ii) is:

Objects communicate in an object oriented program by message passing. If an object, A, of a certain class

needs to communicate with an object, B, of another class, it first obtains a reference to B. A then

communicates with B by sending messages.

SECTION C – Program Development

Question 5

Part (a) of the question tested candidates‟ knowledge of graphical user interface objects. Candidates for

the most part were able to identify correctly the different objects that were used, but many of them had

difficulty explaining the purpose of the different GUI objects.

In Part (b) and (c), many candidates demonstrated that they were not aware of what an event is in Event

Drives Programming. A satisfactory answer would include description of clicking the calculate button to

trigger tax calculation, clicking the exit button to trigger exiting of the window or pressing the enter key

in a text field. Simply typing words in a text field is not considered an event as the API would take care

of this.

In part (d) (i) many candidates were able to distinguish between unit testing and system testing.

For Part d (ii) candidates were unable to outline the tests that could be performed to unit test the Account

class. One of the test that candidates could have given is:

Create a new account with a certain balance. Invoke the get balance method right afterwards and ensure

that the amount returned is the same as that used to create the account.

Question 6

Part (a) tested candidates‟ ability to write source code for a class in object oriented programming

language, it also required candidates to declare instance variables of different data types and code two

methods with simple functionality. Many candidates were unable to correctly write the code for

Employee class. Some candidates did not know how to write an accessor, a method that simply returns

the value of an instance variable.

In Part (b) candidates were expected to write a fragment of a code to perform a given test. The majority

of candidates performed poorly in this question indicating that most of them are not getting adequate

exposure to practical object – oriented programming.

Part (c) required candidates to explain how the Manger class could be derived from the Employee class.

This was poorly done by candidates. Candidates could have stated that the easiest way to create the

Manager class is to reuse the functionality of the Employee class by using inheritance to automatically

have the instance variables and methods of Employee. In the subclass, additional instance variables such

as secretary are simply declared as instance variables in the subclass. Additional functionality is

implemented as new methods in the subclass, perhaps by overriding methods such as raise Salary.

PAPER 03

INTERNAL ASSESSMENT

GENERAL COMMENTS

The performance on the Internal Assessment was generally good. The projects submitted by some

candidates were deficient in various aspects and were still being awarded high marks by teacher.

Teachers need to become more closely involved in the supervision of the projects.

Generally, most candidates chose appropriate topics for the Internal Assessment. The topics chosen were

relevant to the level of the candidates‟ ability and the specific objectives of the syllabus. The treatment

of the topics by candidates was adequate. A small percentage was comprehensive though some tended to

be superficial. The reports were also generally well presented.

DETAILED COMMENTS

INTERNAL ASSESMENT

Candidates should make more of an effort to follow the layout given in the syllabus. This allows the

candidates to clearly identify which parts of the SBA are being responded to. Teachers need to pay

particular attention to the symbols that are used in the respective diagrams.

Teachers also need to ensure that the candidate‟s code is printed and included in the SBA response.

The use of an object oriented language in a non object-oriented fashion should be avoided. Marks are

awarded for the use of appropriate classes and class method, thus the use of an object-oriented style of

programming.

Few candidates are confusing and mixing functional and non functional requirements.

Any soft copy that is submitted should be given in CD format. Avoid using 3 ½ floppy disks.

There were a few cases where candidates used the old Unit 2 syllabus to prepare their internal

assessment. As a result, irrelevant information was submitted.

In some cases, the functionality of the program written was poorly described and there were no screen

shots of the working system producing stated functionality displayed in the reports. The scope of the

programs was too large, thus the stated functionality seldom matched the actual program produced.

Teachers should encourage their candidates to produce programs that focus on simpler more specific

problems from the start, allowing candidates to realistically produce what is proposed in the early

documentation.

RECOMMENDATIONS

(i) A workshop review of the 2008 examination should be done in each school to assess and detail

the concerns and challenges of candidates with particular topics.

(ii) Schools need to carefully compare the new information technology syllabus with the computer

science syllabus and decide which of these subjects to offer based on the resources they have

available and the interests of students. Schools are advised not to offer computer science if

adequate teaching resources are not available.

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2009

COMPUTER SCIENCE

St Michael Barbados

COMPUTER SCIENCE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2009

INTRODUCTION

The revised syllabus was followed this year for the first time. There were three examination papers in

both Units 1 and 2, namely, Paper 01, Paper 02, and Paper 03. In each unit, Paper 01 and Paper 02

were examined externally by CXC while Paper 03, the Internal Assessment, was examined by

teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple-choice questions that were designed to test candidates‟

breadth of coverage of the syllabus. Paper 02 consisted of essay-type questions that were designed to

test their depth of understanding of the syllabus. Thus, candidates were expected to show deeper

insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50

per cent, and 20 per cent, respectively.

GENERAL COMMENTS

In Unit 1, eighty-three per cent of the candidates obtained Grades I to IV. In Unit 2, sixty-four per

cent of the candidates obtained Grades I to IV. However, the performance on the School Based

Assessment should have been better. Better performances on the SBA should lead to better overall

performances on both units as well as enhanced performances on the theory papers. Candidates need

to maximize the opportunity to get higher marks on the SBA.

Overall, there is concern about the level of programming ability being demonstrated in both Units.

Candidates continue to find it extremely difficult to write even simple programs. Teachers are

encouraged to have several programming labs and exercises done with the candidates.

Candidates are also encouraged, as part of their examination technique, to read questions carefully

before answering, and to respond with sufficient detail that is commensurate with the marks indicated

for the question.

DETAILED COMMENTS

UNIT 1

Fundamentals of Computer Science

Paper 01

The performance on the forty-five multiple choice items on this paper produced a mean of 53 out of

90 with scores ranging between zero to eighty-six.

Paper 02

Section A – Computer Architecture and Organisation

Question 1

This question tested candidates‟ knowledge of logic gates, truth tables, multiplexers, and internal

representation.

Part (a) was fairly well done. However, some candidates failed to include inputs and outputs for the

gates. There was also some confusion between AND as well as OR gates. There were some

incomplete truth tables and misunderstanding of the NOT gate as well.

Part (b) was poorly done with only a few candidates being able to show the necessary combination of

gates. Some gave truth tables instead of the circuit diagram required.

Part (c) was not attempted by many candidates. Teachers need to spend more time on multiplexer

design.

In Part (d) (i), some candidates showed a lack of understanding of the binary place values. Part (d)

(ii) was fairly well done but some had problems completing the addition. Many did not respond to

whether the result could actually be stored. Signed magnitude given in Part (d) (iii) was hardly

understood. More practice is needed with these types of numerical questions.

Part (d) (iv) was poorly done and most candidates were unable to get the one‟s complement so that the

subsequent two‟s complement representation was also incorrect.

Question 2

This question tested candidates‟ knowledge of word size, cache memory, clock speed, ROM and

RAM.

Part (a) tested candidates‟ knowledge of word size, cache memory and clock speed. Most candidates

were able to explain cache memory, but word size and clock speed were not generally well explained.

These are concepts that should be reviewed. Additional definitions can be obtained via the Internet.

In Part (b), candidates were asked to distinguish between the computer memory concepts ROM and

RAM, access speed and access method and volatility and capacity. The ROM and RAM distinction

was well answered by most candidates. The access speed and access method distinction as well as the

volatility/capacity distinction were not well answered. However, many candidates were able to

indicate that capacity involved storage space.

In Part (c) (i), most candidates did not identify the „instruction set‟ as a collection of different

instructions that the CPU could execute. Many candidates did not know that the „instruction format‟

involved the layout of the instruction into fields corresponding to the constituent elements of the

instruction.

Part (c) (ii) was fairly well done. The main problem was that some candidates gave examples instead

of stating the types of instructions included in the instruction set.

In Part (c) (iii), some candidates neglected to show the OPCODE as part of the 2-address instruction.

For Part (c) (iv), no candidate scored full marks. Many candidates omitted the storage part of the

cycle. Few candidates paid attention to the use of direct addressing.

Section B - Problem Solving with Computers

Question 3

This question tested candidates‟ knowledge of algorithms.

Most candidates attempted Part (a) but some were unable to give proper definitions. Some key words

omitted included „unambiguous‟, „precise‟ and „logical‟.

Part (b) was fairly well done, however some candidates gave examples rather than listing the actual

constructs.

Part (c) (i) was based on an algorithm that was given to the candidates. This part was poorly done.

Most candidates had difficulty in identifying the concepts „dry runs‟ and „trace tables‟.

Part (c) (ii) was fairly well done. Some identified the correct line but were unable to correct the

errors.

Part (d) was attempted by many candidates. Many candidates did not identify the correct symbols for

flowcharts. There was also some mix-up in the logical sequence of instructions. Some did not show

the looping in the flowchart as required for the „while‟ construct. Candidates need to have more

practice in moving from algorithm to flowcharts and flowcharts to algorithm. One suggestion is to

use partial flowcharts of the correct response and ask candidates to fill in the missing parts during

class exercises.

Question 4

Part (a) asked candidates to discuss what the „Identifying and evaluating possible solutions‟ stage of

problem solving would involve for BuyLo. Part (b) required candidates to trace through the execution

of a given algorithm and draw the output. Part (c) required candidates to write an algorithm that used

repetition to find the sum of all multiples of 7 between 14 (inclusive) and 126 (inclusive).

In Part (a), most candidates listed Buy Lo‟s actions through all of the stages of the problem solving

process, instead of addressing just the stage identified in the question. Some candidates elaborated on

what actions (generally) take place during this stage, but did not give the associated examples

necessary to be awarded full marks for this question. As a result, candidates responded poorly to this

question overall.

In Part (b), candidates generally responded well and most earned more than half the marks.

In Part (c), candidates generally responded well and most of the candidates who attempted this

question got at least 3 marks. Different approaches were also taken towards the solution which

usually resulted in a correct response. However, some candidates left out “print” statements at the

end of their response.

Section C – Programming

Question 5

The question tested candidates‟ knowledge of the translation process and their ability to write code

using if-then-else and loop constructs. Generally, all parts of this question were poorly done.

In Part (a), only a few candidates were able to describe the steps involved in the lexical and semantic

analysis stages. In describing the lexical analysis stage, candidates mentioned that tokens are

„derived‟ but did not adequately explain how these tokens are derived. Many confused syntactic and

semantic analysis.

In Part (b), most candidates omitted the signature of the function (there was no indication of the return

type or list of arguments) and prompted the user to enter the two values to be compared, rather than

pass them as parameters. A significant number of candidates were unaware of the syntax associated

with the if-then-else construct.

In Part (c), candidates did not recognize that the use of loops was required. Some candidates were

unable to work with files; their attempts to open, read from and write to the file were syntactically

incorrect. Very few candidates used the correct data type (float) to store the average.

Question 6

Part (a) asked candidates to describe declarative, imperative and scripting programming paradigms.

Part (b) required candidates to distinguish between syntax and semantics. Part (c) tested candidates‟

ability to implement loops and update records. This question was poorly done by most candidates.

In Part (a), while most candidates were able to describe the imperative paradigm, very few were able

to do so for declarative and even less for scripting, with many of them suggesting that scripting is

documentation.

In Part (b), many candidates did not distinguish between the two terms. Most candidates described

syntax and some semantics but they did not go on to say what makes them different.

In Part (c), many candidates opted not to respond. Most of the candidates who attempted to respond

were able to prompt for and read the input data as well as calculate the pay. Beyond that, some

candidates attempted to implement loops for reading data for one employee and used if statements for

updating the required fields. Some also did not update the required fields but simply assigned the new

values to a variable. Most candidates were able to print the required information although not always

in the required format.

UNIT 2

Further Topics in Computer Science

Paper 01

The performance on the forty-five multiple choice items on this paper produced a mean of 50 out of

90 with scores ranging between zero to eighty-four. Candidates need to familiarize themselves with

Network Architecture, Client Servers and protocol, in particular IEEE802.11 a/b and IEEE802.16g.

Paper 02

Section A – Data Structures

Question 1

This question tested ADTs and associated operations. Overall, it was not well answered.

For Part (a) (i), most candidates misinterpreted what was required and most responses focused on

discussing the THREE stack operations without paying attention to „static computer storage‟.

Part (a) (ii) was not well answered. The function prototype was missing in many cases. Very often,

the top of the stack was not defined.

In Part (b), many candidates were able to give the correct output for the first three iterations of the for

loop. Some did not know how to draw a proper diagram of a stack.

In Part (c), many candidates provided general responses that did not address the question given.

Some explained the LIFO concept instead of indicating how

the stack could be used to determine if a

string is a palindrome. Determining whether a string is a palindrome is a popular exercise which

candidates should be familiar with.

Question 2

This question tested the candidates‟ knowledge of two search methods: linear search and binary

search.

This question was poorly done by most candidates.

In Part (a), some candidates had a basic idea of linear search but had problems expressing the

response as a function. There was regularly a lack of knowledge of the C programming language.

Many candidates could not show return type, parameters, function prototypes or return a value from

the function.

In Part (b), most candidates had no knowledge of the binary search method. Some candidates

attempted this question but most of them had the wrong concept of calculating the midpoint. Some

used the number of elements divided by two, rather than using „low‟ and „high‟.

Section B – Software Engineering

Question 3

Part (a) (i) had fairly good responses with most candidates obtaining maximum marks. Some

candidates incorrectly focused on dealing with the phases of the SDLC rather than discussing

prototyping and the frequent adjustments made to the prototype by the users in order to develop a

final product.

Part (a) (ii) was satisfactorily answered by most candidates with approximately 80 per cent of the

candidates being able to give at least one problem of using the evolutionary development approach.

The majority of responses dealt with the costs associated with this approach, but other acceptable

responses included process visibility, poor structure and special tools and techniques required for

rapid prototyping.

In Part (b), the data flow diagram was poorly drawn with many candidates confusing a DFD with an

ERD. Many candidates used the incorrect symbol for process, file and entity. Some candidates gave

a context diagram rather than a more detailed diagram as was expected by the examiners. Candidates

are therefore advised to only give a context diagram where explicitly indicated by the question,

otherwise give a detailed diagram inclusive of files, entities and the necessary processes. Many

candidates had DFD with data flows not labeled or labeled but direction of flow not shown. Exercises

are required to help candidates improve on the drawing of DFDs.

Part (c) was poorly answered by candidates, with many candidates clearly not knowing what a CASE

tool is, and its advantages and disadvantages. Many candidates misinterpreted 3 (c) (i) as asking

about the advantages of CASE tools rather than the ways in which a CASE tool can be used in

software development.

Question 4

Part (a) (i) was a knowledge based question that tested to see how much candidates know on

functional requirements. Most of the candidates were able to describe „how the system behaves‟.

However, candidates were still unable to identify the services the software should provide and how

the system should react to particular inputs.

In Part (a) (ii), few candidates were able to give complete explanations to this part of the question.

Most candidates were able to write about fact-finding techniques or feasibility studies but were not

able to also mention examination of documents.

Part (b) tested the candidate‟s ability to draw an ERD based on a particular scenario. Some

candidates were able to construct the ERD. What seemed challenging was differentiating between

entity and attributes. Some candidates were actually drawing processes which are found in DFD

instead of entities. A few candidates however managed to construct a good ERD.

Part (c) dealt with testing based on a scenario. Some candidates ignored the scenario and spoke about

general testing procedures. Most were able to score a few marks.

Section C – Operating Systems and Computer Networks

Question 5

Part (a) (i) was fairly well answered as candidates were able to recognise that user accounts can help

track user behaviours; use of username and password for authorized users to protect unauthorized

user entry into the system, and to identify each user.

Part (a) (ii) was well answered. Candidates who understood the answer to Part (i) were able to

distinguish access logs from files that recorded users on the network.

Part (b) (i) was also well answered. Most candidates gained marks in this part.

Part (b) (ii) was poorly answered, most of the candidates had no idea how two out of the four layers

transmitted a file.

Part (c) was not well answered. Candidates generally identified IEEE 802.11b as some wireless

standard, giving responses such as the use for satellite, and Wi-Fi. Candidates could not demonstrate

properly in a topology how data is passed to and from nodes using a wireless medium such as a

wireless router and how to depict this properly in a diagram.

Part (d) was also poorly answered. Candidates were able to identify what the acronyms CDMA and

TDMA stand for, but were unable to outline the difference between the two access methods. Many

gave incorrect responses relating it to mobile phones. Candidates could not link the idea of different

ways of accessing the network by the users on transmission channel, frequency range and spectrum.

Question 6

In Part (a), candidates were not clear about the concepts of device drivers although they may actually

use device drivers everyday. Most knew that it was some sort of interface between the OS and the

device, but few stated that it was a translator/convertor of instructions for the device to understand.

Overall - fairly well done.

In Part (b), most candidates knew that a hybrid network was a combination of two or more network

topologies, but a few stated only „two or more networks‟ and did not state topologies. It was well

done by most candidates

In Part (c) (i), not many candidates knew the proper purpose of spooling. Most only stated that the

document would be put into a buffer/queue, but few stated further points, for example, that the printer

would take control of the printing process from there on. Overall - fairly well done.

In Part (c) (ii), most candidates knew that spooling helps the computer to function efficiently by not

using up resources, but few stated that this was as a result of data being stored elsewhere in a buffer.

Overall - fairly well done.

Part (d) (i), most candidates knew that the menu interface was easier to use due to a list of options, but

not many stated that this would make it easier for novice users. Overall - fairly well done.

In Part (d) (ii) most candidates knew that the command interface would function faster due to

instructions being typed in directly, but few stated that this would only be beneficial to users who

were knowledgeable or versed in using the interface. Overall - fairly well done.

In Part (e), not many candidates were able to correctly identify two interrupts. Most gave graphic

card or video card and RAM problems which would affect game installation but not during game

play. Most failed to state I/O interrupts and external interrupts, as well as most did not give proper

explanations of the interrupts. Overall – not well done.

In Part (f), most candidates were able to state that paging used virtual memory from the hard disk and

that some form of swapping between RAM and virtual memory takes place, but few were able to

properly describe the process of paging. Not well done by most candidates.

Paper 03 - Internal Assessment

GENERAL COMMENTS

In general, performance of the candidates was good. However, there were still some inconsistencies

arising from candidates and teachers not paying attention to the details of the new syllabus.

Teachers need to work closely with candidates on the requirements for the new syllabus, as well

as the specific mark scheme given in the syllabus. Teachers must avoid using old mark schemes

when a syllabus has been revised.

Internal Assessment

UNIT 1

DETAILED COMMENTS

Problem Definitions

Problem definitions were not well done in some samples. The provision of a brief context and clear

ideas of how the problems were manifested in the organization, along with supporting evidence were

often not shown. Many candidates provided extensive backgrounds of the organization instead of the

problem description. Most candidates did not pay attention to the requirements when they were

writing their problem definition.

Narrative and Flowcharts or Pseudo-code

The narrative description of the algorithms was not well interpreted by some candidates. Narratives

were supposed to describe what was designed in the flowchart or pseudo-code algorithm. Some

algorithms were not properly designed. Candidates seemed to know the structures, but some were

unable to use them to produce good flowchart and pseudo code algorithms. Candidates need to pay

attention to correctly designing flowchart algorithms. Teacher practice in this area is recommended.

Coding

The majority of the candidates attempted either flowchart or pseudo code algorithms in the internal

assessment, however, some of the programs designed did not match the algorithms.

Candidates were asked to write programs using procedural C only. Some candidates chose to use

such languages as JAVA, PASCAL, C++ or Visual Basic. This is deviating from the aim of the

syllabus. It is imperative that teachers and candidates pay close attention to the syllabus to avoid

being penalized during moderation. Candidates are also advised to print code from the compiler

directly and not from a word processor.

UNIT 2

GENERAL COMMENTS

In general performance throughout was good. However, there were a few candidates who used the

old syllabus instead of the one prescribed by CXC effective for examination May/June 2009.

Candidates were required to write programs using procedural C only. Some candidates chose to use

languages such as JAVA, PASCAL, C++ or Visual Basic. This is deviating from the aim of the

syllabus. It is imperative that teachers and candidates pay close attention to the syllabus prescribed in

order to achieve its goal.

DETAILED COMMENTS

Problem Definitions

Some candidates focused on providing background information and description of the organization

instead of concentrating on the requirements (See syllabus Page 31).

Techniques of Analysis

Most candidates were able to name the techniques of data collection and describe how each was

performed. However, some failed to give relevant ones.

Data-Flow and E-R Diagrams

Incorrect symbols were often used within these diagrams. As a result, candidates were unable to

produce relevant diagrams. In a few cases, the diagrams did not correctly represent solutions to the

problems identified.

Functional and Non-Functional Requirements

Most candidates were able to correctly identify functional and non-functional requirements of the

system. However, a few candidates used „Hardware and Software requirements‟ (for example,

Processor speeds or Operating Systems) for this section which was incorrect.

System Structuring

Most candidates produced a system structure but failed to give ones that were relevant to the project

they pursued.

User Interface Design

Most interfaces were relevant. Many candidates were able to correctly state the type of interface they

would implement but rarely stated the appropriate justification for its use.

Algorithm Design

Some algorithms were not properly designed. Some candidates seemed to know the structures but

some were unable to use them to produce good flowchart and pseudo-code algorithms. Candidates

need to pay attention to correctly designing flowchart algorithms. Use of symbols should be

practised.

Coding

Some used languages other than C. Some samples were submitted without printed programming code

and/or screen shots to verify program functionality and a softcopy was submitted in its place.

Generally, this part was well done.

Candidates are advised to print code from the compiler directly and not from a word processor.

Testing

Testing usually focused on normal data and tended not to test abnormal or extreme cases.

Recommendations

Each school should review the syllabus in order to assess and detail the concerns and challenges with

particular topics. Schools need to network with each other to utilize the resources available to achieve

the aims of the syllabus.

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2010

COMPUTER SCIENCE

St Michael Barbados

GENERAL COMMENTS

This was the second year for which the revised syllabus was followed. There were three examination

papers in each of Units 1 and 2, namely, Paper 01, Paper 02 and Paper 03. In each unit, Paper 01 and

Paper 02 were examined externally by CXC while Paper 03, the Internal Assessment, was examined by

teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple-choice questions that were designed to test candidates’

breadth of coverage of the syllabus. Paper 02 consisted of essay-type questions that were designed to test

candidates’ depth of understanding of the syllabus. Thus, candidates were expected to show deeper

insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50 per

cent, and 20 per cent, respectively.

In Unit 1, 73 per cent of the candidates obtained Grades I–V. In Unit 2, 92 per cent of the candidates

obtained Grades I–V. Performance on the School Based Assessment for Unit 1 was roughly the same as

in 2009. However, performance on the School Based Assessment for Unit 2 increased by more than 10

per cent compared with 2009.

Performance on questions involving computing programming continues to be poor; however, there have

been small improvements in certain types of programming questions. It is clear that many candidates are

still not getting the experience they need in writing and testing real computer programs on their own.

Of particular concern is the fact that performance on other types of questions such as those on Computer

Architecture and Organization has also been very poor. This indicates a lack of adequate preparation on

all aspects of the syllabus.

DETAILED COMMENTS

UNIT 1

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of 51 out of 90 with scores

ranging from 16 to 86.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

This question examined candidates’ knowledge of logic gates, truth tables, binary counters, line decoders,

and the internal representation of data on a computer.

Overall, the question was poorly done. It is clear that candidates did not master the basics of computer

architecture and organization. Some candidates ignored Part (d) which involved the representation of

data on a computer. Several others did not attempt the questions on truth tables. This is cause for

concern since the representation of data on a computer and truth tables are the most basic and

straightforward items in the syllabus.

In Part (a) of the question, most of the candidates correctly identified the logic gates and gave the required

truth tables. However, some candidates mixed up the gates or identified the gates as NAND and/or NOR

gates.

Part (b) was poorly done by most candidates. Of the candidates who attempted the question, about 98 per

cent of the responses were incorrect. It should be noted that the binary counter goes through a prescribed

set of states (0–15) upon application of an input pulse such as a clock pulse. After the counter reaches 15,

it goes back to 0.

A fair attempt was made at Part (c) (i). However, most candidates did not label the block diagram

properly. In Part (c) (ii), candidates demonstrated a clear lack of knowledge of the truth table of a 2-to-4

line decoder. Part (c) (iii) was also poorly done. Most candidates did not attempt this part. Of those who

attempted this part, the majority of them gave incorrect responses.

Part (d) (i) was fairly well done. In Part (d) (ii), most candidates were able to perform the calculation but

several of them had difficulty in explaining why the result of the calculation could be stored as a 4-bit

binary number. Most candidates attempted Part (d) (iii); however, their calculations were incorrect

resulting in incorrect responses. Part (d) (iv) was well done indicating that candidates have a very good

grasp of two’s complement representation.

Question 2

This question tested candidates’ knowledge of storage devices, cache memory, instruction set, instruction

formats and instruction cycle. It was misinterpreted by most candidates and answers were vague.

In Part (a), most candidates gave qualitative responses such as moderate, low and high when comparing

the capacities of ROM, RAM, hard disk and CD-RW. Similar responses were given when comparing the

access speeds of the devices. Very few candidates gave a quantitative response in terms of numeric

capacities and placing them in order (ascending/descending). Also, instead of comparing the devices

based on access speed, many candidates compared the devices based on access method.

Part (b) was poorly done. Most candidates were unable to explain how cache memory works and to

correctly describe one benefit of its use. In Part (c), candidates focused mostly on the activities of the

instruction cycle. Most of them were unable to explain what the instruction set of a CPU is, and to

describe types of instructions and typical instruction formats.

Section B – Problem Solving with Computers

Question 3

This question tested candidates’ knowledge of the basic control structures used in computer programming

and the use of these control structures in a given algorithm. The question also examined candidates’

knowledge of representing algorithms using graphical techniques such as flow charts and their ability to

understand and modify an algorithm based on given specifications. Overall, the question was not done

very well and it appeared that many candidates misinterpreted what was required.

In Part (a), some candidates were unable to identify the three basic control structures of sequence,

selection and repetition. Some candidates identified them as input, output and processing. This is cause

for concern since the three control structures are the building blocks of computer programs which account

for roughly half of the Computer Science syllabus. Most of the candidates misunderstood Part (b) and

attempted to explain the control structures without making reference to the algorithm given, as instructed

in the question.

In Part (c), some candidates interpreted graphical representation to mean drawing images of actual

bananas, pineapples etc. It was expected that candidates would draw a suitably labelled flow chart

corresponding exactly to the algorithm specified. One problem with the flow charts given was that the

flow lines between components were often missing. In some instances, incorrect symbols were used.

Many candidates wrote for j = 1 to 100 in the decision box, which was incorrect; it should have been, is j

equal to 100? The loop is achieved by an arrow returning to a previous point in the flow chart.

The responses that were given in Part (d) demonstrated that candidates did not have a very good

understanding of tracing through the logic of a fairly straightforward algorithm expressed in pseudocode.

As a result, they were not able to make the required modifications to the algorithm resulting in this part

being done badly by most candidates.

Question 4

This question tested candidates’ ability to trace through a given algorithm to determine its output and to

develop an algorithm from a given specification. The question also examined candidates’ knowledge of

the Implementation and Review stage of the problem solving process. Overall, this question was poorly

done.

In Part (a), candidates were able to trace through most of the algorithm. However, many of them were

unable to produce the correct diagram of an arrow pointing upwards. In general, Part (b) was done fairly

well. Most of the candidates who attempted the question scored at least three marks. This is a good sign

since it indicates that candidates’ problem-solving abilities are improving.

In Part (c), it was clear that many candidates did not read the question thoroughly. They focused on the

actual question and ignored the preamble. Hence, they did not answer the question within the context of

the video club. As a result, many of the candidates achieved low scores on this part. Also, some

candidates described the entire problem-solving process which was not required.

Section C – Programming

Question 5

This question tested candidates’ ability to take a specification of a program and to write a C program

which achieves the desired functionality. It also examined candidates’ knowledge of programming

language features and the appropriateness of programming languages for different types of applications.

Part (a) of the question was generally answered poorly by candidates. Most of their responses described

types of programming languages and the generations of programming languages. Few candidates gave

the correct response that mobile devices have limited graphical capabilities and processing power so the

programming language will offer less graphical features and smaller libraries.

Part (b) was fairly well done. Most of the candidates were able to correctly differentiate between a

character variable and a string variable. Part (c) was poorly done by most of the candidates. Many of the

responses demonstrated poor programming skills and an inability to manipulate files in the C

programming language. Some candidates attempted to answer the question using arrays. However, this

was unnecessary since the data for each employee could be read and processed in a while loop without

saving the data on all the employees.

Question 6

This question tested candidates’ ability to write functions in the C programming language. It also

examined their understanding of what constitutes good programming style.

In Part (a), the majority of candidates were able to correctly identify spacing and indentation as two ways

of improving programming style. The use of consistent case (for example, lowercase) when naming

variables is another way to improve programming style.

For Part (b), most candidates were able to explain the purpose of a struct in C. However, in many of the

responses, the idea of a record was not mentioned.

Responses to Part (c) demonstrated a clear lack of knowledge of modular programming in C using

functions. Many candidates did not write the function prototype correctly (return type, name of function,

followed by parameter list). The code given in their responses often did not have return statements and

many of them were not able to call the functions properly; some candidates even specified the types of the

variables when making the function call.

UNIT 2

Paper 01 – Multiple Choice

Performance on the 45 multiple choice items on this paper produced a mean of 54 out of 90 with scores

ranging from 22 to 84.

Paper 02 – Essay Questions

Section A – Data Structures

Question 1

This question tested candidates’ knowledge of abstract data types and their implementation. It also

examined candidates’ knowledge of manipulating abstract data types to achieve a desired outcome. Most

candidates attempted this question. However, the quality of the responses was generally poor.

In Part (a), most candidates had a general idea of what an abstract data type (ADT) is but most of them

could not define it as a specification of a set of data and the operations that can be performed on the data.

In Part (b) (i), some candidates confused the singly linked list ADT with other abstract data types such as

a queue or a stack. The ADT operation could have been insertFirst (LinkedList, data) or insertLast

(LinkedList, data) or something similar. In Part (b) (ii), most candidates responded well. However, a few

candidates omitted the pointers to indicate the beginning and end of the list.

In Part (c) (i), most candidates did not realize that the computer storage they were asked to describe was

simply the data structures associated with the implementation of a queue (an array and several integer

variables to keep track of the beginning and end of the list).

Part (c) (ii) was generally well done. Part (d) was attempted by most of the candidates. However, some of

the candidates did not demonstrate the skills required to reverse the elements of the queue using the stack

for temporary storage.

In Part (e), most candidates described the attributes of either a stack or a queue. However, few candidates

were able to explain that a stack can contain an embedded linked list and thus, operations such as pop()

can be implemented by calling the linked list operation, deleteFront().

Question 2

This question tested candidates’ knowledge of sorting and searching algorithms. It was generally

answered poorly by most candidates. However, there were very good responses from a few candidates.

In Part (a) (i), most candidates did not give a detailed description of how the selection sort algorithm

works. It was important to specify how the minimum element from 0 to 9 would first be found and

swapped with the element at position 0. It was also important to specify that succeeding iterations would

go from 1 to 9, 2 to 9 and so on, until no more elements are out of place, that is, when the 8 element has

been put in its correct position.

In Part (a) (ii), most candidates skipped the second pass in which no swap was made. However, this is a

valid pass even though the array remains the same as it was at the end of the first pass.

The responses to Part (b) were of a poor quality. The question required candidates to write a program that

performs a linear search. Many candidates incorrectly put the ‘if not found’ test within the for loop,

instead of the outside. This would cause the message to be printed every time the element in the array is

different from target.

In Part (c), explanations of binary search were given but this was not required. Candidates were expected

to answer that the binary search should be used when the elements of the array are sorted in ascending or

descending order.

Section B – Software Engineering

Question 3

This question tested candidates’ understanding of various software engineering concepts as well as their

ability to draw data flow diagrams corresponding to a given narrative. It was done fairly well by most

candidates.

In Part (a), many candidates were able to correctly discuss the attributes of well-engineered software.

However, some of them were unable to list the names of the attributes correctly, for example, reliability

was often used instead of dependability as one of the attributes. Also, many candidates listed portability

as an attribute but were unable to say what it was.

Most of the candidates correctly discussed the need for user involvement in the software development

process in Part (b). However, several of them failed to differentiate between the role of users and

managers in the development process. In many cases, the manager’s role was simply omitted.

In Part (c), many candidates were not able to clearly explain why a software system needed to be

upgraded. The expected response was that the needs of a business change all the time since the business

environment is changing and thus the software must be able to adapt to accommodate these changes.

In Part (d), a few candidates did not know which symbols to use when drawing a data flow diagram. At

times, entity-relationship diagram symbols were used. Also, several candidates gave a context diagram

instead of the Level-0 diagram requested.

Question 4

This question tested candidates’ knowledge of the feasibility study and the requirements specification

document in software engineering. It also tested candidates’ ability to draw an entity-relationship diagram

from a narrative and to specify ways in which a piece of program code can be tested. The question was

well-attempted. However, it was poorly answered.

In Part (a), most candidates identified economic and technical feasibility as reasons for undertaking the

feasibility study. Few candidates recognized that it may be possible for user needs to be satisfied with

existing software and hardware.

Many candidates were able to gain most of their marks in Part (b) of the question. However, several of

them omitted the identification of the primary key, for example, underlining the employee number in

some notations.

Many candidates did not seem to have read Part (c) of the question properly and so did not relate their

responses to the parameters of the function lsearch. Many candidates simply described general forms of

testing and did not apply the principles of testing to the function given. For example, one test is to call

lsearch with an array of integers which does not contain the given key; if lsearch is working correctly, it

should return -1.

Section C – Operating Systems and Computer Networks

Question 5

This question tested candidates’ knowledge of the OSI model for computer communication and various

operating system concepts such as page fault and process states. It also tested candidates’ knowledge of

GPRS. It was generally poorly attempted by most candidates, most of whom obtained under 10 marks,

with a significant number obtaining 0–5 marks.

In Part (a), most candidates were able to correctly identify the seven layers of the OSI level. However, a

few candidates mixed up the transport and network layers. Some candidates also labelled the layers in the

reverse order which resulted in them obtaining fewer marks. Part (a) (ii) was poorly done. Approximately

5 per cent of the candidates were able to correctly describe the purpose of the three layers.

In Part (b), only a few candidates were able to accurately define a page fault and describe how it was

handled by the operating system. Most candidates described the paging process and did not answer the

question. Many of the responses incorrectly stated that a page fault occurs when there is an error on the

page. The correct answer is that a page fault occurs when a program tries to access a page that is not

currently mapped to RAM.

In Part (c), candidates generally confused GPRS with GPS and consequently answered the question

incorrectly. GPRS is a technology used to connect wirelessly to the Internet and can be used by point-of-

sale devices or other devices needing wireless connectivity to the Internet. Part (d) was fairly well-

attempted. However, many candidates did not clearly describe the steps involved when a process moves

from the running state to the ready state.

Question 6

This question tested candidates understanding of various networking and operating system concepts. It

was attempted by most candidates. Responses were fair with average scores between 11 and 15 marks. A

few candidates got 21 – 25 marks.

Parts (a) to (g) and (k) (iii) were well done by most candidates. Parts (h), (i), (j), k (i) and k (ii) were not

answered well by most candidates. They were unable to provide proper explanations which indicate the

need for greater depth of knowledge in these areas, for example, client-server versus peer-to-peer

configurations.

Based on the responses obtained for this question, it is clear that candidates need to become more familiar

with the technical terms used in the syllabus.

GENERAL COMMENTS

Paper 03 - Internal Assessment

In general, performance on the School-Based Assessment was very good. The performance on Unit 1 was

roughly the same as in 2009. However, the performance on Unit 2 increased by more than 10 per cent

compared with 2009.

DETAILED COMMENTS

UNIT 1

Some candidates wrote their programs using separate program files which were saved and printed

individually rather than using separate functions within the same program. There were candidates who did

not provide printed evidence for the headings which were included in their submissions, for example, a

heading stating 'PSEUDOCODE' often had no printout showing an algorithm

in pseudocode.

Some candidates included trace tables which were not required. Teachers need to pay careful attention to

the requirements of the syllabus. Other candidates used data flow diagrams instead of narratives for

describing algorithms. There were also candidates who utilised one main function and no function

decomposition was seen in some samples.

Some candidates submitted a pseudocode which was identical to their programs. A pseudocode should be

produced before the program is written and will generally look different from the source code of the

program. In some samples, pseudocode algorithms were not written using correct logic, for example,

variables controlling iterations were not incremented and while statements were not closed with Endwhile.

There were candidates who submitted the same pseudocode, program code and test data. Subheadings

were often missing in some of the documents submitted and the word limit was exceeded in some

instances. Flow charts in many samples made use of system symbols rather than the normal flowchart

symbols and some test plans did not use erroneous and extreme data for testing. Screen images (screen

shots) to illustrate the program output were not shown in many samples. These are important for verifying

the output of the program.

The problem statements in some samples were either very trivial or too detailed. Candidates are reminded

that the source code for programs must be printed from the C development environment and that a

narrative must be given which is related to the problem solution.

Unit 1 samples often included entity-relationship diagrams and data flow diagrams; however, these are

only relevant to Unit 2.

UNIT 2

Some candidates were unable to differentiate between functional and non-functional requirements.

Functional requirements are features the system must provide. Non-functional requirements are

constraints that apply to the system, for example, it must perform a transaction in less than two seconds.

Attributes of entities were not included in the entity-relationship diagram by most candidates the entity-

relationship diagrams and data flow diagrams were often inconsistent with the problem definition.

It should be noted that user technical manuals are not required and that background information and

abstract information should be kept to a minimum.

Some problem definitions were often too wordy, and, in some instances, made no sense in the context of

the system developed.

Candidates are reminded that all design diagrams must be drawn electronically, for example, system

structuring and user interface design. Programs must be written in C code and printed output must be

done from the C development environment.

When preparing their reports, candidates should use subheadings which are specified in the syllabus. The

ordering of the headings should also be followed.

RECOMMENDATIONS

Candidates are encouraged, as part of their examination technique, to read questions carefully before

answering, and to respond with sufficient detail that is commensurate with the marks indicated for each

question. Candidates need to answer questions in the context given instead of simply regurgitating notes.

Past-paper questions should be carefully studied to understand how syllabus items can be tested.

Teachers need to prepare students adequately for the three modules of each unit. Algorithm Development,

Computer Programming, and Data Structures account for three of the six modules of the Computer

Science syllabus so teachers need to cover these modules in more detail, scheduling as many practical

sessions as possible. This can take the form of

 classroom discussions geared to solving specific problems or developing algorithms to solve

given problems

 laboratory exercises involving the writing and testing of computer programs and experimentation

with ‘buggy’ programs

 using the Internet to download tutorials on programming and sample code to experiment with.

The poor performance by candidates on the Computer Architecture and Organization module can be

handled by adequate coverage of the syllabus items. This module does not involve computer

programming. Candidates need more practice with numerical problems such as those involving decimal to

binary conversion and vice versa, as well as problems involving one’s and two’s complement.

With respect to the School-Based Assessment, teachers need to provide better feedback to studentss. The

School-Based Assessment should be viewed as a process with well-defined milestones. As students

submit work for these milestones, teachers need to provide adequate guidance, correcting errors made and

indicating clearly what is required for the next milestone.

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2011

COMPUTER SCIENCE

St Michael Barbados

GENERAL COMMENTS

This was the third year in which the revised syllabus was examined. There were three examination papers

in each of Units 1 and 2, namely, Paper 01, Paper 02 and Paper 03. In each unit, Paper 01 and Paper 02

were examined externally by CXC while Paper 03, the School-Based Assessment (SBA), was examined

by teachers and moderated by CXC.

In each unit, Paper 01 consisted of 45 multiple-choice questions that were designed to test candidates‟

breadth of understanding of the syllabus. Paper 02 consisted of six essay-type questions that were

designed to test candidates‟ depth of understanding of the syllabus. Thus, candidates were expected to

show deeper insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50 per

cent, and 20 per cent, respectively.

In Unit 1, 87 per cent of the candidates obtained Grades I to V, a 12 per cent improvement over 2010. In

Unit 2, 87 per cent of the candidates obtained Grades I to V, a five per cent decline compared with 2010.

In general, performance on the SBA was very good. In Unit 1, there was a slight decline in the

performance on the SBA compared to 2010. The average mark was 68 per cent compared to 70 per cent

in 2010. However, there was a significant decline in the performance on the School Based Assessment in

Unit 2 compared with 2010. The average mark was 57 per cent compared with 66 percent in 2010.

Nevertheless, in both units, the average mark was an improvement on the corresponding 2009 average.

Performance on questions involving computing programming continues to be poor. However, there was a

reasonable improvement in scores for most of the programming questions in Unit 1 compared with 2010.

Performance on the programming questions in Unit 2 was roughly the same as in 2010. Despite the

improvements noted, it is clear that many candidates still need to get more experience in writing and

testing real computer programs on their own.

Of particular concern is the fact that performance on other types of questions such as those on computer

architecture and organization has also been very poor. This indicates a lack of adequate preparation on all

aspects of the syllabus.

DETAILED COMMENTS

UNIT 1 – Fundamentals of Computer Science

Paper 01- Multiple Choice Questions

Performance on the 45 multiple-choice items on this paper produced a mean of 54 out of 90 with scores

ranging from 18 to 86.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

This question examined candidates‟ knowledge of truth tables, logic gates, circuit diagrams, flip-flops,

multiplexors, decoders, and the internal representation of data on a computer. It also required candidates

to apply their understanding of a decoder to solve a simple problem.

Overall performance on this question was poor and the majority of candidates obtained less than 10

marks.

In Part (a), only a few candidates were able to correctly convert the truth table to a circuit diagram. Some

candidates even used secondary logic gates (NOR, NAND) in their responses which were unnecessary. In

Part (b), most candidates performed well and got full marks or at least half of the marks allocated for this

question.

In Part (c), many candidates stated that a flip-flop can be used to store one bit of data. However, only a

few of them went on to fully explain what is a flip-flop and hence did not obtain full marks. Many

candidates obtained full marks for their explanation of the multiplexor.

Part (d) (i) was generally well done. Most candidates gave a correct diagram for the decoder; however,

some candidates did not label the diagram and a few attempted to draw a circuit diagram. The

performance on Part (d) (ii) was fair. A few candidates drew circuit diagrams instead of using the block

diagram of the decoder. Most candidates did not explicitly state that there are two inputs and that the

bulbs have to be connected to the outputs. Also, some candidates did not explain how the two inputs were

connected to the four outputs (the bulbs).

Part (e) was fairly well done. The majority of candidates were able to convert the decimal value to binary,

but a number of them had difficulty finding the two‟s complement. Part (f) was poorly done. Many

candidates simply converted the binary representation to its decimal value. Some of them correctly

converted the 3-bit exponent to decimal but were unable to use this value to correctly complete the

calculation. A few candidates assumed that the decimal point was to the right (00101.) rather than the left

(.00101).

There is cause for concern with candidates‟ performance on this question, particularly on Parts (e) and (f)

since the representation of data on a computer is one of the most basic and straightforward items in the

syllabus.

Question 2

This question tested candidates‟ knowledge of registers, Central Processing Unit components, storage

devices, cache memory, instruction formats and addressing modes.

The question was answered fairly well and there was an improvement in responses compared with to

those given for a similar question in 2010.

In Part (a), candidates‟ were generally weak in their understanding of registers found in the CPU.

Candidates were expected to mention registers such as the MAR, MBR, PC, AC and IR.

The responses to Part (b) were generally fair. However, many candidates only compared main memory

and the two given storage devices in terms of capacity or access speed alone. They thus lost half of the

marks for this question.

Part (c) was generally well done since this is a topic well understood by most candidates. The Arithmetic

and Logic Unit (ALU) and Control Unit (CU) were popular responses for this question. However, the

function of the components stated (e.g. the CU) was not properly stated in many responses.

The responses to Part (d) were generally fair. However, many candidates could not clearly explain how

cache memory increases the efficiency of data retrieval (by keeping frequently accessed portions of main

memory in the cache).

The responses to Part (e) were generally poor. Candidates need to spend more time understanding the

various types of addressing modes specified in the syllabus.

Section B – Problem Solving with Computers

Question 3

This question tested candidates‟ knowledge of the stages in the problem-solving process and their skills in

constructing a flow chart from an algorithm as well as tracing through and identifying errors in an

algorithm. The question also examined candidates‟ ability to understand and modify an algorithm based

on given specifications.

Overall, the question was well done and it appeared that many candidates understood what was required.

The mean performance on this question was the best of all the questions in Paper 02.

In Part (a), most candidates were able to identify three stages in the problem-solving process. The stages

were sometimes incorrectly given as the stages of the systems development life cycle. Candidates were

often unable to clearly describe each stage, particularly the analysis stage (which seeks to determine how

the problem manifests itself).

Most of the candidates gave good responses for Part (b). However, some of them drew incorrect symbols

to represent input/output and processing steps and some were uncertain about how to represent a loop in

the flow chart. In a few cases, programming code was included with the symbols. Candidates need to

spend more time constructing flow charts and should make sure that they understand the purpose of each

symbol used in a flow chart.

In Part (c), candidates were able to correctly trace through and identify the lines of the algorithm

containing the errors. However, many were unable to correct the errors. It is recommended that candidates

trace through an algorithm after modifying it to ensure its correctness.

Question 4

This question tested candidates‟ ability to trace through a given algorithm to determine its output and to

develop an algorithm from a given specification. The question also examined candidates‟ knowledge of

the properties of a well-designed algorithm.

In Part (a), many candidates were unable to identify three properties of a well-designed algorithm. This

was rather unfortunate since it involved straightforward recall of a syllabus item.

In Part (b), many candidates were unable to trace through the while loop correctly and generate the

triangular shape of asterisks. In addition, there was some confusion between the use of print and println.

However, the question clearly stated that println did the same thing as print except that subsequent output

started on a new line. Only a few candidates were able to correctly calculate the values for y on each

repetition of the while loop. This indicates a weakness in basic arithmetic skills. Candidates need to spend

more time tracing through algorithms since this is a fundamental skill in computer science.

In Part (c), some candidates were unable to write correct algorithms in terms of looping constructs,

accumulating the sum, and finding multiples of 11. It was clear from their responses that many candidates

were unaware that a while loop can be bounded or had a misconception that a while loop is always

unbounded. A loop is bounded if the number of iterations is known beforehand or can be calculated

beforehand. It is sometimes possible to calculate the number of iterations of a while loop before it is

executed; thus, in these situations, the while loop is bounded.

Section C – Programming

Question 5

This question tested candidates‟ knowledge of the stages of the translation process and of the advantages

of using a modular approach in programming. Candidates were required to write a C function which took

two parameters, one of which was an integer array. The question also tested candidates‟ ability to take a

specification of a program and write a C program which achieves the desired functionality. The program

required candidates to know how to read and write from text files.

Part (a) was answered fairly well by candidates. Most of them were able to list the stages of the

translation process.

Part (b) was fairly done. However, a few candidates confused the term modular with the modulus

operation. Also, several candidates were vague or overly general in their responses.

Part (c) was poorly answered. Many candidates gave a program in their response rather than a function

and lost marks because of this. It is fairly straightforward to find the sum of the numbers in an integer

array and the code for this functionality should have been placed in the body of the function. In addition,

candidates were expected to specify the parameter list and return type of the function as well as return the

sum of the values after it was calculated inside the function.

Part (d) was also poorly answered. Candidates seemed to lack skills of analysis to understand the

functionality that the program was required to provide. For those who were able to write a reasonable

program, it was clear that most of them were unable to manipulate text files. Many of the candidates did

not adhere to the syntax of the C language, particularly in the use of semicolons, ampersands, format

specifiers and appropriate quotation marks. Candidates need to spend more time writing programs from a

given specification using the programming language features outlined in the syllabus.

Question 6

This question tested candidates‟ understanding of good programming style such as the use of white space

and proper indentation. It also tested their understanding of the term debugging. Candidates were required

to write a program to determine what type of triangle is present given three integers representing the sides

of a triangle. The key component of the solution consisted of an if-then-else-if statement. Finally,

candidates were required to trace through a segment of C code to determine its output and to modify the C

code to give a different set of output.

In Part (a), the majority of candidates did not obtain full marks since they did not clearly explain white

space.

In Part (b), most candidates were able to define the term debugging using their own words. The concept

of debugging seems to be well understood among candidates.

In Part (c), most candidates had a general concept of indentation being associated with the appearance or

presentation of a program. However, the basic idea of maintenance was not present in their responses.

In Part (d), the responses indicated that candidates lacked the ability to properly sequence code in order to

achieve the required results. Thus, for example, the check for equilateral triangle should have preceded

the check for isosceles triangle. Many candidates were unable to maintain the use of correct syntax

throughout their programs. Program code often lacked logical operators in the condition part of the nested

if statement and several candidates opted to use a series of if statements to avoid the use of a nested if

statement. However, this resulted in incorrect code since one if statement would flow into another one. It

should also be mentioned that some candidates gave their responses in C++. Candidates need to use the C

language in the present Computer Science syllabus.

In Part (e), a number of candidates lacked the ability to trace through the for loop so they were unable to

generate the output. Some responses ignored the inner for loop indicating that these candidates did not

understand nested loops. Of those who correctly traced the required output, only a few were able adjust

the loop to generate the required output in Part (e) (ii). Again, candidates need to spend more time writing

and tracing through programs in order to answer questions like these.

Paper 03 – School-Based Assessment

In general, performance on the Internal Assessment was very good. In Unit 1, there was a slight decline in

the performance on the Internal Assessment compared to 2010. The average mark was sixty-eight per

cent compared to seventy per cent in 2010. However, there was a significant decline in the performance

on the Internal Assessment in Unit 2 compared to 2010. The average mark was fifty-seven per cent

compared to sixty-six percent in 2010. Nevertheless, in both units, the average mark was an improvement

on the corresponding 2009 average.

Many students ` included documentation which was not required for the Unit 1 SBA including structure

charts, Gantt charts, data-flow diagrams, questionnaires and entity-relationship diagrams. Many of the

projects contained flowchart algorithms; however, system flowchart symbols were incorrectly used

instead of programming flowchart symbols.

There were many instances where full marks were awarded for problem descriptions that were either

superficial or overly detailed.

Many students did not provide enough detail in the narratives when describing a solution to the problem

identified.

Generally, the presentation of pseudocode algorithms showed improvement in logic compared with

previous years and were also easy to follow for many of the projects. However, many students provided a

pseudocode algorithm that was identical to their C programs. In some projects, the algorithms were

written using incorrect logic. For example, variables were often not initialized before being used and

repetition statements did not show variables being incremented correctly.

Students often provided programs containing logical steps that were not included in their algorithms.

Some students made use of alternative programming languages such as C++ and Pascal for writing their

programs. Schools and students are reminded that C is the language of the Computer Science syllabus.

Programs were often not printed directly from the C development environment and this made it difficult

to properly verify the documentation and modularization of the source code. Many students did not make

use of comments and did not indent their C code.

Teachers should ensure that programming projects sufficiently cover items in the syllabus such as the use

of arrays and record structures as well as the manipulation of external data files.

Trace tables were used in many instances but these were not required for program testing. Program testing

by many students often did not show the use of normal, extreme and erroneous data. The test results were

often described but not shown by means of screen images. The screen images are required to verify that

the program produces the output as described in the report.

Candidate projects should make use of the section headings given in the Criteria for Marking the Internal

Assessment in the syllabus. Many projects often did not include any section headings.

The use of GOTO statements should not be encouraged in structured programming.

Storage media such as CDs should not be included with the submission of the samples.

Care should be taken when preparing the SBA document to ensure the following:

 The table of contents is accurate

 Pages are numbered

 Text size, font style and line spacing are legible for reading and moderation.

Teachers should also ensure that each sample has a completed CSCI1-5 form detailing the marks awarded

for the assessment criteria.

UNIT 2 – Further Topics in Computer Science

Paper 01

Performance on the 45 multiple-choice items on this paper produced a mean of 58 out of 90 with

candidates‟ scores ranging from 18 to 86.

Paper 02 – Essay Questions

Section A – Data Structures

Question 1

This question tested candidates‟ knowledge of abstract data types (ADTs) and their implementation. It

required candidates to use the Stack ADT to remove the element at the bottom of a stack. The question

also examined candidates‟ knowledge of manipulating ADTs to achieve a desired outcome. In addition,

the question tested candidates‟ understanding of the implementation of a circular queue as well as their

understanding of the bubble sort algorithm.

Part (a) required candidates to define the term abstract data type (ADT) and to explain how the Stack

ADT is implemented. Most candidates had some knowledge of the ADT but were unable to define it

properly. In Part (a)(ii), many candidates listed the operations of the Stack ADT but did not explain how

the stack would be implemented in code (e.g. by using an array and a top pointer and manipulating these

data items in the push and pop operations).

In Part (b), many candidates did not recognize that this was a problem to be solved using the operations of

the Stack ADT (e.g., push, pop, isEmpty). Consequently, they wrote unnecessary code to manipulate the

internal data of the stack. Thus, they ended up writing code to push and pop the elements from the stack

which was not required.

Performance on Part (c) was generally poor. It was clear that many candidates did not understand how a

circular queue operates. Some candidates were able to perform the insertions correctly. However, when

an element was deleted, many candidates simply shifted the remaining elements one position to the left.

In some responses, the size of the array increased or decreased when elements were inserted or deleted,

respectively. It should be mentioned that a circular queue operates on an array of fixed size by

maintaining pointers to the first and last elements in the queue. Elements are not shifted when inserting or

deleting another element.

The quality of the responses for Part (d) was also generally poor. It seems that the majority of candidates

did not have a clear understanding of the bubble sort algorithm. Several candidates described the first pass

correctly but did not go on to explain what happens after the first pass. They simply concluded their

responses by saying that the algorithm is repeated until the entire array is in sorted order. But, getting the

array in sorted order is the purpose of the sort algorithm in the first place. Some candidates wrote C code

in their responses. However, the question did not require the writing of code.

Question 2

This question tested candidates‟ knowledge and understanding of linked lists as well as their ability to

implement the selection sort algorithm in C. The question also required candidates to demonstrate their

understanding of what happens in the linear and binary search algorithms.

In Part (a), most of the linked list diagrams drawn were incorrect. Many candidates gave tables instead of

the standard representation of a linked list (a set of nodes pointing to each other with a special pointer to

the top of the list and a terminator symbol at the last node). Of those who drew a diagram containing

nodes with each one pointing to the next node in the linked list, many did not did not use a terminating

symbol at the last node in the list. In terms of the explanation, many candidates described advantages of

the storage characteristics of a linked list instead of simply describing its main features.

In Part (b), the implementation of the selection sort algorithm was poorly done. Candidates demonstrated

weaknesses in storing values in the array at the beginning of the algorithm, in writing the nested for loops

properly (especially with regard to the initial and terminal values of the loop variables), in swapping

elements in the array; they generally did not know how the selection sort algorithm works.

The responses for Part (c) were generally good. However, many candidates did not clearly state the

conditions under which the linear search algorithm would terminate. Several candidates also did not

properly describe what happens in each execution of the loop in the binary search algorithm.

This question required a significant number of narrative responses from candidates. The level of English

was very poor and it was often difficult to decipher what a candidate was trying to say. Many candidates

did not answer the questions directly, but tended to write down everything they knew on the specific

topic. The level of programming was also very poor indicating that candidates have not developed their

programming skills sufficiently in Unit 1 to master the syllabus items in this module.

Section B – Software Engineering

Question 3

This question tested candidate‟s understanding of various software engineering concepts as well as their

ability to draw a Level-0 dataflow diagram corresponding to a given narrative.

In Part (a), candidates were required to explain the waterfall approach to system development. Candidates

were expected to identify the stages of the waterfall approach and explain how one stage fed into the

subsequent stage. However, very few candidates associated the waterfall model with the systems

development life cycle (SDLC) even though this is one of the most popular approaches for the SDLC.

The minority of candidates who made the link did not identify the phases of the waterfall approach.

In Part (b), candidates were required to describe four tasks which must be performed during the design

phase of systems development (e.g. architectural design, interface design, data structure design and

algorithm design). Only a few of the candidates who attempted this question were able to describe these

tasks.

Part (c) required candidates to draw a Level-0 dataflow diagram for a given scenario involving a mail

order company. Most candidates attempted this question. However, it seems that many of them were not

equipped with “drawing tools”. Consequently, many of the diagrams were drawn free hand and ended up

being very ugly and unsightly for advanced proficiency candidates. Also, on many occasions, the standard

symbols for drawing data flow diagrams were not used.

Question 4

This question tested candidates‟ knowledge of the techniques used for fact-finding during the analysis

phase of systems development as well as the deliverables of the analysis phase. It also tested candidates‟

understanding of the use of CASE tools in software development as well as their ability to use entity-

relationship models for data modelling.

Part (a) (i) was well done overall. Most candidates were able to identify three fact-finding techniques such

as interviews, observation and questionnaires. In contrast, Part (a) (ii) was poorly done by most

candidates. It was clear that many candidates did not understand the term deliverables and as a result only

a small percentage were able to accurately state two deliverables.

In Part (b), the responses were generally poor. It was clear that most candidates did not know what a

CASE tool is and consequently, their responses for both parts of the question were poor.

Part (c) was fairly well done overall. However, many candidates did not use the correct symbol to denote

an attribute of an entity and consequently lost marks for drawing the attributes. Also, candidates had

difficulty accurately identifying the many-to-many relationships in the scenario (e.g. a lecturer is hired by

many departments and a department hires many lecturers). Many candidates drew lines connecting

entities (to represent relationships) but did not label the lines or give the cardinality of the relationships.

Several candidates also used non-standard notations to represent cardinality. In the future, candidates are

advised to use a standard for their entity-relationship models (such as the Unified Modelling Language or

the more dated Chen‟s notation).

Section C – Operating Systems and Computer Networks

Question 5

This question tested candidates‟ knowledge and understanding of various networking concepts such as

network media, network configuration, networking devices, transmission media and access to data and

resources on a network

Some parts of the question were answered fairly well by many candidates. However, the responses to the

other parts were often very poor.

Part (a) was not answered well by most candidates. Most of them were unable to correctly identify and

explain a difference between an analogue signal and a digital signal. Some rather interesting responses

stated that an analogue signal is understood by computers while a digital signal is not.

Most candidates answered Part (b) correctly. However, some candidates gave incorrect answers such as

telephone, radio and fax machines. This suggests that these candidates may not be familiar with the

concept of transmission media.

In Part (c), some candidates gave good responses by explaining the layout of a client-server network with

the aid of a labelled diagram. However, some candidates did not understand what the client-server model

is and instead gave answers related to the peer-to-peer model or to network topology such as star or ring.

Part (d) (i) tested candidates‟ knowledge of network connectivity devices such as bridges, repeaters,

routers, and gateways. Some candidates described network types instead of network devices, while a few

described devices that are used in a network such as UPS, server. A few candidates also described a

network topology such as star, ring, or bus. This suggested that candidates did not know what network

connectivity devices were or did not know the difference between network topology and network devices.

Part (d) (ii) tested candidates‟ ability to apply knowledge of transmission media to a specific situation.

Responses from candidates revealed that some of them did not know what transmission media is and were

thus unable to answer the question correctly. Some candidates even gave responses like cell-phones,

server and pager.

Part (d) (iii) tested candidates‟ knowledge of the different ways that users may be granted access to

resources on a network. Many candidates confused the granting of access to resources on the network

with preventing access to a network and incorrectly gave responses such as encrypting files and using

firewalls. Some candidates also stated that access can be granted by using the Internet, an intranet or an

extranet.

Question 6

This question tested candidates‟ understanding of various networking and operating system concepts.

Performance was generally fair.

In Part (a), candidates were often unable to properly or clearly distinguish between a batch processing

system and a multi-user system. For example, some candidates stated that in a batch processing system,

only one user used the system at a time. The examples provided in Part (a) (ii) and Part (a) (iii) were often

not representative enough of the type of system.

The responses for Part (b) were generally good with most candidates giving answers such as passwords

and encryption.

The responses for Part (c) were very poor. It is clear that most candidates do not understand the concept

of an interrupt and the interrupt processing mechanism.

The responses for Part (d) were also poor. Many candidates did not recognize that the diagram

represented a deadlock situation and gave interesting answers such as „hold ups‟, „event interrupt‟, „round

robin‟ or even „parallel processing‟.

In Part (e), many candidates were unable to completely describe how the round-robin scheduling

algorithm works. It also seemed that weaker candidates had no idea of what was involved in the

algorithm.

PAPER 03 – School-Based Assessment

For the Unit 2 SBA, students are expected to choose a problem for which a software solution exists and

then develop the software using software engineering techniques. In particular, they are expected to

demonstrate the tools and techniques used in the analysis of the software to be developed. They are then

expected to design, code and test their software using appropriate techniques.

Performance on the SBA was generally fair. It was observed that the projects submitted by Students were

deficient in various aspects and were still being awarded high marks. Teachers also need to become more

closely involved in the supervision of the projects.

Generally, most students chose appropriate topics for the SBA. The topics chosen were relevant to the

level of the students and the specific objectives of the syllabus. Diagrams were generally done well;

however, they were very inconsistent and did not flow (as they should) from one to the other. The

treatment of the topics by students was adequate. A small percentage was comprehensive though some

tended to be superficial. The reports were also generally well presented.

Students should put more effort in following the layout of the SBA given in the syllabus. This enables

students to clearly identify which parts of the SBA are being addressed by which sections of the report.

Teachers should instruct students to use the requirements in the mark scheme as sub-headings within their

reports. This ensures that students will have a section corresponding to each requirement.

Teachers need to advise students to restrict the scope of their SBA Reducing the scope of the project or

focusing on a single aspect will give students a better chance of completing what is set out in their

problem definition.

SBA reports should be well secured and bounded to avoid the mixing up or loss of parts of the reports

submitted.

Teachers should not assume that marks are given for things not printed. Only visual/printed elements can

be marked so anything not included in printed form within the SBA will not receive marks. Teachers also

need to ensure that each candidate's code is printed and included in the SBA report. A soft copy of a

candidate‟s code is not accepted.

Teachers should ensure that students clearly state the problem being tackled in the problem definition.

Simply stating that a system is manual or paper-based is insufficient. A manual or paper-based system is

not in itself a problem, as stated by many students. The fact that it may be time consuming, error prone,

tedious and so on are problems that may exist in a particular manual or paper-based system. Not all

manual or paper-based systems suffer from these problems; thus, students should identify the problems

clearly in order to be awarded full marks.

In describing the techniques used for analysis of the problem, many students did not say why the

techniques were relevant.

It should be noted that a few teachers and students continue to confuse the terms functional requirements

and non-functional requirements. Special attention should be paid to the listing of functional requirements

as this was generally done poorly.

Students need to extend their system structuring diagram beyond the elements of the main menu in order

to be awarded full marks for this requirement.

Students seldom included justifications for their choice of user interface. Also, some of them were

confused as to the difference between menu-driven and command-driven interfaces.

Students in general left out the report design; teachers should ensure that this is included and not assume

that marks will be awarded based on screen shots.

Test plans were not done well in general. Students should include an exhaustive set of data to be tested

which includes all forms of data entry. Many students only tested the menu selections. Expected results

and actual results obtained from testing should be included in tests. Tests should also include normal,

abnormal and extreme data.

In most cases, the functionality of the program written was poorly described and there were no screen

shots of the working program producing the functionality described in the reports. The scope of the

programs was often too large; as a result, the stated functionality seldom matched the actual programs

submitted. Teachers should encourage students to develop programs that focus on simpler, more specific

problems from the start. This would allow students to realistically develop what is proposed early in the

documentation.

There were a few cases where students used the old Unit 2 syllabus to prepare their SBA. As a result,

irrelevant information was submitted. Teachers should ensure they are using the syllabus marked

Effective for examinations from May/June 2009.

RECOMMENDATIONS

Students are encouraged, as part of their examination technique, to read questions carefully before

answering, and to respond with sufficient detail that is commensurate with the marks indicated for each

question. Students need to answer questions in the context given instead of simply regurgitating notes.

Past-paper questions should be carefully studied to understand how syllabus items can be tested.

Teachers need to prepare students adequately for the three modules of each unit. Algorithm Development,

Computer Programming and Data Structures account for three of the six modules of the Computer

Science syllabus so teachers need to cover these modules in more detail, scheduling as many practical

sessions as possible. This can take the form of

 classroom discussions geared to solving specific problems or developing algorithms to solve

given problems

 laboratory exercises involving the writing and testing of computer programs and experimentation

with “buggy” programs

 using the Internet to download tutorials on programming and sample code to experiment with.

The poor performance by students on the Computer Architecture and Organization module can be

handled by adequate coverage of the syllabus items. This module does not involve computer

programming. Students need more practice with numerical problems such as those involving decimal to

binary conversion and vice versa, as well as problems involving one‟s and two‟s complement.

With respect to the SBA, teachers need to provide better feedback to students. The SBA should be viewed

as a process with well-defined milestones. As students submit work for these milestones, teachers need to

provide adequate guidance, correcting errors made and indicating clearly what is required for the next

milestone. It should also be mentioned that the SBA provides an invaluable opportunity to gain hands-on

experience with the items in the Computer Science syllabus which ultimately leads to better performance

on the written papers. Thus, the SBA should be viewed not as a goal in itself but as a framework for

mastering the contents of the syllabus.

CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION®

MAY/JUNE 2012

COMPUTER SCIENCE

St Michael Barbados

GENERAL COMMENTS

This was the fourth year for which the revised syllabus was followed. There were three examination

papers in each of Units 1 and 2, namely, Paper 01, Paper 02 and Paper 03. In each unit, Papers 01 and

Paper 02 were examined externally by CXC while Paper 03, the School-Based Assessment, was

examined by teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple-choice questions that were designed to test candidates‟

breadth of coverage of the syllabus. Paper 02 consisted of essay-type questions that were designed to test

candidates‟ depth of understanding of the syllabus. Thus, candidates were expected to show deeper

insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50 per

cent and 20 per cent, respectively.

DETAILED COMMENTS

UNIT 1 – FUNDAMENTALS OF COMPUTER SCIENCE

Paper 01 – Multiple Choice Questions

Performance on the 45 multiple-choice items on this paper produced a mean of 61 out of 90 with scores

ranging between 20 and 90.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

This question tested candidates' knowledge of truth tables, logic gates, circuit diagrams, multiplexors, and

the internal representation of data on a computer. It also required candidates to apply their understanding

of a multiplexor to solve a simple problem.

For Part (a) (i), most candidates were awarded the marks for the input column combinations, however

many of them failed to correctly work out the values for the output, F.

In Part (a) (ii), many candidates did not demonstrate a clear understanding of logic gates. Many of them

confused the AND and OR gates, or did not understand how to place the NOT gates in the circuit design.

For Part (b) (i), some candidates were able to draw the block structure with four inputs and one output.

However, most candidates did not include the two selection lines and sufficient labelling.

There were many non-responses for Part (b) (ii). This part was generally poorly answered. While some

candidates were able to explain how a multiplexer works generally, they were unable to apply that

knowledge to the specific use in the security system.

Part (c) (i) was generally answered well. Many candidates were able to correctly identify the largest and

smallest integers as 7 and -7 respectively.

Many candidates who attempted Part (c) (ii) were able to recognize the sign bit as negative. Few were

able to convert the binary string representing the mantissa to its decimal equivalent. Even fewer

candidates were able to correctly identify the decimal equivalent of the 5-bit mantissa.

Question 2

This question tested candidates‟ knowledge of instruction set, instruction types, instruction formats, cache

memory and storage devices. Overall, the question was answered poorly with a modal score of 6.

In Part (a) (i), candidates were able to define the instruction set but most of them failed to mention that

the operation of the CPU is determined by the instructions it executes.

For Part (a) (ii), many candidates were unable to specifically identify the types of instructions (that is,

data processing, data storage, data movement and control) however, accurate explanations were given for

the aforementioned. In some instances, candidates misinterpreted the question and identified the

instruction cycle (that is, fetch, decode, execute) rather than the types of instructions.

The responses for Part (a) (iii) were generally good but some candidates omitted the fact that a copy of

frequently accessed RAM is stored in cache memory.

Part (b) (i) was well done as the majority of candidates were able to differentiate between the opcode and

the operand of an instruction.

Part (b) (ii) was answered poorly. Many candidates gave no response. From the responses given, a few

candidates were able to identify that the result of the operation of a one-address instruction format is

stored in the accumulator. Candidates were also able to identify that there were two operands in a two-

address instruction format.

The majority of candidates misinterpreted Part (c) (i). Very few candidates were able to differentiate

between the access method of a hard disk (moving arm over a spinning disk) and RAM (electrical

pathways). They were however able to correctly state that RAM had a faster access speed than a hard

disk.

Part (c) (ii) was well done by the majority of candidates.

Section B – Problem Solving with Computers

Question 3

This question tested candidates‟ knowledge of algorithms, iterations within algorithms, and writing an

algorithm based on the narrative of a given problem.

Most candidates scored between 13 and 17 for this question.

Part (a) tested candidates‟ knowledge of the role of algorithms in problem solving. All of the candidates

attempted this part and were able to identify the role of algorithms in the problem solving process. Some

candidates, however, were giving the definition of an algorithm instead of its role.

Most candidates attempted Parts (b) (i) and (ii) and showed some knowledge of iterations. Many of them

could not justify their responses as they related to which algorithm illustrated bounded iteration and

which illustrated unbounded iteration. Few candidates were able to differentiate between bounded and

unbounded iterations.

Part (c) tested candidates' knowledge of the use of sequencing, selection and iteration in programming.

Candidates did not understand the need to initialize the variables needed for the cumulated totals. Some

candidates used a WHILE instead of the FOR statement but were unable to give accurate limits for the

condition. Almost all candidates used the IF statement to select the correct option, however the condition

was not clearly written. The calculation for the cumulative totals for each colour was fairly well done in

most instances. The totalwas calculated and printed well by most candidates.

Part (d) tested candidates‟ knowledge of the use of iteration and selection in programming. Most

candidates did not understand how to intialize variables and as result they scored zero. They also did not

use the limits well in the WHILE loop and failed to increment the counter variable. Candidates did not

use the modulus well, instead they simply divided in most instances.

Question 4

This question tested candidates‟ ability to represent an algorithm as a flow chart. It also examined their

skills in tracing through an algorithm to determine its output.

Part (a) was fairly well done. Most candidates gave good responses. However, some of them drew

incorrect symbols to represent initialization of variables. Some were uncertain about how to represent a

loop in the flow chart. In a few cases, programming code was included with the symbols. The flow of

logic was also not clearly shown for in many instances arrows were left out. More importance should be

placed on constructing flow charts and candidates should make sure that they understand the purpose of

each symbol used in a flow chart.

Part (b) was generally poorly answered. Many candidates were able to list the values of the variable

„sum‟ and output the answer, but they did not state what would be printed by the algorithm as was

required. Additionally, candidates do not appear to understand how a condition is tested in order to

terminate a loop. Candidates continued to test the loop even after the value 5 was entered, thereby

generating output for the value 4 in the test data. While they recognized that the values of 5 and 7 in the

test data would not be executed by the loop, they did not appear to understand that the value 5 would have

ended the loop.

Part (c) was also fairly well done. Most candidates were able to trace through the algorithm. However,

many of them were unable to produce the triangular shape consisting of $, + and &. In addition, there

was some confusion with the use of “println”, with some candidates placing the cursor on a new line and

then placing the output on the new line. In general, most candidates who attempted scored at least five

marks.

Responses to Parts (b) and (c) indicate that more emphasis needs to be paid to the tracing of algorithms,

with special attention being given to the behaviour of loops.

Section C – Programming

Question 5

This question tested candidates‟ ability to write simple functions; manipulate files, create, read from, and

write to files; and their knowledge of the stages in the translation process.

Part (a) was fairly well done. Candidates were asked to describe three of the stages of the program

translation process. Some candidates were able to fully answer this part while others were at least able to

correctly identify three stages. However, several candidates listed either the stages in the problem-solving

process (problem definition, analysis, etc.) or steps in the instruction cycle (fetch, decode, transfer, etc.).

Part (b) was generally well done. Candidates were required to write a C function to calculate 2

given n.

Many candidates were able to obtain the majority of marks allocated but failed to state the relevant

assumptions.

Part (c) was poorly done. Candidates were required to write a C function to write data to a file and then

read the data stored, manipulate it and display the result. Most candidates were only able to establish the

file pointer and open the file in the correct mode. Many were unable to properly form the loop to write to

or read from the file.

Question 6

This question tested candidates‟ understanding of what constitutes good programming style. Candidates

were also required to write C code to create and manipulate records consisting of different data types. The

question also tested candidates' ability to take a specification of a problem and write a C program using an

array which achieves the desired functionality. The key component of the solution was the if-then-else

construct.

Part (a) tested the basic understanding of good programming techniques which candidates should have

learnt at the CSEC level. The fact that so few candidates were able to obtain full marks for this question

clearly indicates a lack of knowledge and understanding of the basic concepts in programming. While

many candidates were able to identify three aspects of good programming style, few were able to offer a

fair explanation as to why each was indicative of good programming.

Again being able to recognize what the question is asking in an important examination strategy that

candidates seemed to lack. Candidates need to pay attention to key operational words which indicate the

level of skill expected. Since candidates were asked to explain, simply identifying three ways would not

gain them full marks.

Those who answered the question were able to identify the use of indentation, documentation,

modularization, use of white spaces and meaningful variable names as techniques of good programming.

Part (b) tested candidates‟ ability to declare a record structure in C and to store and manipulate data in

that structure.

The majority of candidates were unable to make the distinction between a record declaration tested in Part

(b) (i) and the variable declaration tested in Part (b) (ii). As a result, few candidates were able to obtain

full marks for this part.

Most candidates demonstrated knowledge of the Struct declaration in C but failed to use proper

punctuation – semicolon and curly brackets.

Declaring variables for record structures and storing data in those variables also proved to be challenging

for most candidates. Most candidates declared the variables as type int or char rather than the record

structure productRec.

Part (b) (iv) required candidates to exchange the values stored in two records. The logic to perform this

operation should have been acquired at the CSEC level. Again most candidates demonstrated a lack of

understanding of this logic. As a result, this part of the question was not answered by most candidates or

the answer written was not the most efficient code. Only a few candidates, therefore, were able to obtain

maximum marks for this part.

Most candidates who answered the question copied the data „field to field‟ rather than „record to record.‟

This of course cost them valuable time in the exam and resulted in code that was long and cumbersome

rather than simple and elegant.

For Part (c), many candidates were able to write C code that produced correct output. The efficiency of

the solutions provided, however, needs to be improved. Very few candidates demonstrated an

understanding of how to test multiple conditions in one IF statement. Instead most produced a nested-If

statement.

While most candidates met the requirements for printing the number of occurrences of the vowel „A‟ and

the presence of a given vowel, few included code which tested for the absence of a vowel.

Paper 03 – School-Based Assessment (SBA)

UNIT 1

Students must be commended for their creativity when choosing an SBA project. Although most topics

were recycled, students were able to be innovative in their delivery.

Many students were able to accurately describe the problem complete with the description of the current

system and to give examples of what takes place when the problems arise. Some projects only included

what activities the proposed system would perform without completely describing the current system and

the problems which we are to be solved using software.

Narratives were fairly well written in most samples. However, some samples did not provide a

completely accurate description of the algorithm. Pseudocode algorithms were generally well done.

Samples which included flow charts in some instances incorrectly used system flow chart symbols and in

others, diagrams were poorly presented and were difficult to follow. In the many samples using

pseudocode algorithms, some were clearly modified copies of the programming source code. Students

should be encouraged to develop their algorithms independent of the programming code.

It was quite evident that students were comfortable with procedural C programming language. This year‟s

programs were logically written and properly decomposed. Nevertheless, some students disregarded the

SBA requirement of using procedural C and used C++. A few samples did not make use of the key data

structures (struct, files and arrays). Students must remember to print their source code directly from the

compiler as transferring to a word processor changes the spacing of the code. Generally, most students

did not include adequate comments at key areas of their code.

This year, the majority of samples provided a suitable range of test data, but a few students did not

include all four testing criteria (normal, extreme, erroneous and incomplete) in their test plan. Test results

must include actual screen shots of the working program and testing must be done using the test data

outlined in the test plan.

Generally, this year‟s SBA projects were well presented. Teachers need to ensure that students use the

headings outlined in the Criteria For Marking Internal Assessment Project in the syllabus, as well as

follow the order in which these heading occurs. They should also ensure that students check that the

numbering of their table of contents corresponds to the numbering in the document.

UNIT 2 – FURTHER TOPICS IN COMPUTER SCIENCE

Paper 01 – Multiple Choice

The performance on the 45 multiple-choice items on this paper produced a mean of 57 out of 90 with

scores ranging between 14 and 90.

Paper 02 – Essay Questions

Section A – Data Structures

Question 1

This question tested candidates‟ knowledge of abstract data types and their implementation. It also

examined candidates‟ knowledge of manipulating abstract data types to achieve a desired outcome. The

question required candidates to manipulate a Stack ADT, a Linked List ADT, and a Queue ADT. It was

attempted by more than 95 per cent of the candidates, however approximately 20 per cent gave

satisfactory responses.

Part (a) (i) was poorly done. Candidates identified operations used with the Stack ADT (push, pop) but

they did not answer the question by stating the difference between the Stack ADT and the C

implementation of the Stack.

Part (a) (ii) was generally poorly done. Weaker candidates did not know the meaning of the word

variable and listed operations push and pop in their answers.

Part (a) (iii) was poorly done. Candidates knew how to use the operations but could not write C code to

implement them. Some candidates mixed up the conditions for underflow and overflow.

Part (b) was, surprisingly, poorly done. Most candidates identified the top and end of the linked list but

placed the elements in the same order as given (indicating 43 as the top and 25 as the end)

Part (c) was generally well done. Candidates ignored the last statement in the question and used top, rear,

front and other variables to write the algorithm.

Question 2

This question required that candidates show an understanding of how to manipulate a one-dimensional

array structure using the C programming language. The question also tested candidates‟ knowledge of

searching and sorting algorithms.

Part (a) required candidates to write C programming code in their responses, however many candidates

provided pseudocode solutions. Generally, candidates knew that they had to use a loop to traverse the

array; however, the condition that determined if the mark was within the range given was either poorly

done or left out.

Part (b) required candidates to describe how the binary search algorithm may be used to locate an item

that was in the list and one that was not in the list. Most candidates understood the general concept of the

algorithm but there were two common mistakes. Firstly, when locating the „middle‟ location the majority

of candidates „rounded up‟ instead of truncating and using the integer section, that is given 4.5 candidates

used 5 instead of 4. The other common error occurred in the final „divide and conquer‟ for an item that

was not in the list. Candidates were unable to give an accurate description that determines when the item

was not present.

Part (c) required candidates to draw an array after three passes of the selection sort algorithm. This part

was fairly well done. However, a noticeable number of candidates sorted the entire array.

Section B – Software Engineering

Question 3

This question required candidates to construct a level-0 data flow diagram (DFD), and to demonstrate an

understanding of the evolutionary and waterfall approaches to software development.

For Part (a), candidates were required to draw a level-0 DFD to depict the scenario of a registration

system in a university. The DFD was poorly developed with many candidates confusing a DFD with an

entity-relationship diagram (ERD). Many candidates did not use the correct symbols for the process, file

and entity. Some candidates gave a context diagram rather than a more detailed diagram. Many

candidates had DFD with data flows not labelled or labelled but with the direction of flow not shown.

Exercises are required to help candidates improve on the drawing of DFDs.

Part (b) was poorly answered by candidates. Many candidates seemed to have studied the waterfall

approach, and gave advantages and disadvantages of that approach.

In Part (c), candidates were required to describe four phases in the waterfall approach to software

development (that is, analysis, design, coding and testing, maintenance). This part was generally well

done. Most candidates who attempted this question were able to describe four phases.

Part (d) was satisfactorily answered by most candidates. Many cnadidates were able to give one reason

for the involvement of the user in the software development process.

Question 4

This question tested candidates' knowledge of entity relationship models and their ability to draw the

model from a given narrative of a system. The question also tested the candidates' knowledge of tests

involved in binary search.

For Part (a) (i), most candidates were able to correctly identify the components of an entity-relationship

diagram but in many cases they were not able to properly describe them, particularly the entity

component.

In Part (a) (ii), candidates were generally able to draw the correct entity with only very few using the

wrong symbol. Most candidates were also able to name the correct relationships. However, many were

not able to represent the correct relationship cardinalities.

It was noted that with the exception of a few candidates those who used the Chen notations (1:1, 1:m,

m:n) did not know the correct symbols for many to many , that is, m:n.

For Part (b), a large number of candidates were not able to describe a suitable test. Many described

various aspects of unit testing. In cases where candidates described suitable tests for the binary search,

most were only able to describe two of the three tests. Very few candidates were able to describe a

suitable third test.

Section C – Operating Systems and Computer Networks

Question 5

This question tested candidates‟ knowledge of networking concepts such as network transmission media,

network configurations, connectivity devices, transmission standards, and the role of the open systems

interconnection (OSI) model in network communication. The least marks were scored in Part (d) wireless

and Part (f) the OSI model.

From all indications, this section of the syllabus was not thoroughly covered as evidenced by the quality

of candidates' answers.

Part (a) dealt with transmission media. Despite an attempt by most candidates to answer this question, it

was not answered correctly. Many candidates were unable to adequately describe the characteristic of

each for example, physical description, transmission capacity and mode of data transmission.

Part (b) tested candidates' knowledge of the ring topology. Most candidates who attempted this question

had some knowledge of this topology. However, they could not represent it diagrammatically, for

example, many drew a star topology instead of a ring when outlining the before and after situations, and

the solution in their response.

This question also tested the candidates' ability to troubleshoot this topology. Most candidates changed

the topology to a „star‟ instead of maintaining the ring topology with the addition of a device or simply

removing the malfunctioning computer to maintain the topology. Many candidates indicated the use of a

hub or switch but some indicated router or gateway which showed that they did not fully understand the

uses of these devices. There were those who converted the ring to a fibre distributed data interface

(FDDI) which was not an appropriate solution.

In Part (c), only a few candidates were able to describe the role of a hub. They showed some knowledge

of the device but lacked understanding of its purpose in a network. Diagrams were not well done, there

were many missing labels.

Part (d) tested candidates‟ knowledge of wireless networking. The question was poorly done and in many

cases was not attempted. Mainly, candidates did not know that IEEE802.11a is the established standard

for wireless networks. Many candidates had difficulty identifying the devices needed and the transmission

medium used in wireless networks. The diagrams were in many cases omitted or totally incorrect.

Part (e) which tested candidates' knowledge of firewalls was not well done, with most scoring one mark

out of the three marks. Candidates only outlined the role, without including the aspects of security,

hardware and/or software solutions, and blocking of IP addresses.

Part (f) assessed candidates' knowledge of the layers of the OSI and their corresponding functions.

Though it was well attempted, knowledge of the names of the layer, order of Layers and purpose of each

layer was lacking overall. The key to this question was to identify layers 1-5 only; Layer 1 being the

physical layer and 5 the session layer. A few candidates labelled the application layer as Layer 1, which

made the answer partially incorrect.

Question 6

This question tested candidates‟ understanding of various operating system concepts−deadlock, interrupt,

process states, process control block, and user interfaces.

For Part (a), candidates generally responded well, with most providing a „wait for‟ graph representing

deadlock along with a correct description of what causes deadlock. The outcome however, stating that

both processes must wait indefinitely, was often omitted.

In Part (b), most candidates provided at least a partially correct answer for how an interrupt is handled.

Most candidates who attempted Part (c) did not always define the process states by name, that is, ready,

running and blocked, but gave partial descriptions for the states.

Part (d), was poorly done, with many candidates unable to correctly identify the components of a process

control block. Some candidates stated components such as the program counter, accounting information

and process state information.

Part (e), was well done. Only a few candidates had difficulty identifying an advantage of a menu

interface over a command line interface.

RECOMMENDATIONS

Teachers should ensure that a thorough explanation of the printed circuit board (PCB) and its components

be provided to students as this was the area with the highest omissions and margins of error for Question

Teachers should also ensure that whilst they use visual aids to help students understand topics covered in

class, they should make students aware that only computer related diagrams are acceptable and they

should not provide those concept aids as responses to questions.

Paper 03 – School-Based Assessment (SBA)

UNIT 2

Students are reminded of the following:

 All programming language code should be printed from the C compiler (not from word

processor)

 Examiners can only mark programming code from printed output and, therefore, to show that the

program works, students need to print screen shots of the program in the testing phase and should

not send soft copies.

 Only one form of algorithm is required that is either

 Pseudocode OR

 Flow chart

 Students are not required to describe the results of the analysis techniques.

Marking Criteria

Definition of Problem

Students were required to state the shortcomings of the existing system, and generally this section was

well done by most of them. However, some students failed to identify a problem in the problem

statement and therefore full marks were not awarded.

Techniques of Analysis

Students were required to identify techniques of data collection and analysis. Most of them were able to

state the various techniques of data collection; however, they must be able to justify the reason for their

selection as well as show proof of analysis. For example questionnaires and/or interview transcripts

should be included in an appendix.

Context Level Diagram

In some instances, incorrect symbols were used for entities and processes. Some data flows were not

labelled, and data store/file were incorrectly included in the context level diagram.

Level 1 Diagram

Students were required to give a complete and accurate diagram with all relevant processes, data flows

and major data stores; however, some of them had links between files and entities which is incorrect.

ER Diagram

Students were required to give a complete and accurate diagram of all relevant entities and relationships.

This was generally well done by most students; however, some of the cardinalities and also the

description of the relationships were not done correctly.

Functional and Non-Functional Requirements

Students were required to give a complete and accurate description of all requirements. This was done

well by some of them; however, some students sometimes confused or misinterpreted functional and non-

functional requirements. Some students described project limitations for non-functional requirements.

Design Specification

Students were required to give a complete and accurate description of:

 System structuring

 User interface

 Report design

 Algorithm design

 Appropriate data structures

Some students did most of the design specifications, however:

 some of them went into too much detail in the systems structure by including modules or

branches that were not necessary

 some of them did not name or justify the type of user interface they used

 some of them had no evidence of report design but were given marks by teacher

 some of them did not produce any algorithm.

 some of them were not awarded any marks by the teacher for appropriate data structure even

though the program clearly showed the use of Struct, Array, Queues and Files declaration

and manipulations. This suggests that either the teachers did not know or needed

clarification on what to look for in this section.

Coding and Testing

Students were required to produce a complete and accurate program with:

 code that achieves functionalities such as documentations, outputs, usability and reporting

 code that corresponds to the design specification

 a test plan with exhaustive data set, (including unit testing, integrated testing and systems

testing).

Most students produced programs that achieved good functionality. However, some of them did not

produce proper test plans to cover unit testing, integrated testing and systems testing. Some students only

produced test results and screen shots but NO test plan.

RECOMMENDATION

Teachers should ensure that students are fully prepared for the examinations in both units. The poor

performance in some modules of the syllabus indicates that more time needs to be spent on some areas.

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2013

COMPUTER SCIENCE

St Michael Barbados

GENERAL COMMENTS

This was the fifth year in which the revised syllabus was examined. There were three examination papers

in each of Units 1 and 2, namely, Paper 01, Paper 02 and Paper 03. In each unit, Paper 01 and Paper 02

were examined externally by CXC while Paper 03, the School-Based Assessment, was examined by

teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple-choice questions that were designed to test candidates’

breadth of coverage of the syllabus. Paper 02 consisted of essay-type questions that were designed to test

candidates’ depth of understanding of the syllabus. Thus, candidates were expected to show deeper

insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50 per

cent, and 20 per cent, respectively.

Approximately 93 per cent of the candidates obtained Grades I─V in Unit 1 and approximately 93 per

cent of the candidates in Unit 2 obtained Grades I─V. Overall, there is still need for improvement in the

quality of responses for the programming questions in both units.

DETAILED COMMENTS

UNIT 1 – FUNDAMENTALS OF COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of approximately 60.51 out

of 90, standard deviation of 13.87 and scores ranging from 16 to 90.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

This question tested candidates’ knowledge and understanding of logic gates, circuits and binary

representation of negative integers and floating point numbers.

The question was generally well done with the majority of the candidates obtaining more than 50 per cent

of the marks available for this question.

Part (a) (i) was satisfactorily done. A few candidates either attempted a circuit diagram instead of a block

diagram or did not include selector lines.

Part (a) (ii) was fairly well done. Many candidates were able to partially explain how the multiplexer

operates but did not go on to disclose the values of selector lines used to obtain the signals in l

and l

respectively.

Part (a) (iii) was generally well done. The majority of candidates was able to state one property and one

use of flip flops.

Part (b) (i) was well done. A few candidates were unable to convert 5 to binary.

Part (b) (ii) was fairly well done. However, a number of candidates found the one’s complement of the

sign and magnitude representation from Part b (i).

Part (b) (iii) was also fairly well done because candidates went on to find the two’s complement using the

sign and magnitude value from Part (a) (i).

Part (c) was generally poorly done, with a significant number of candidates being unable to obtain any of

the three marks available. Many candidates simply converted the entire representation to decimal to

obtain a value of 165

10.

Part (d) was exceptionally well done with the majority of candidates being able to obtain full marks for

drawing the truth tables for the NOT, AND and OR logic gates respectively. Weaker candidates did not

correctly map each truth table to its corresponding logic gate.

Common Misconceptions

 Several candidates stated that a property of a flip flop is the ability to store one bit of data

when this is in fact a use of flip flops.

 Many candidates identified examples of bistable devices such as light switches, alarm

systems etc. instead of stating a use of a flip flop.

Question 2

This question mainly tested candidates’ knowledge of port connectivity, instruction sets and direct

addressing.

Part (a) was done relatively well. However, many candidates either did not explain the term port

connectivity as it related to the given situation or misunderstood the term and explained it in terms of the

ports malfunctioning due to rust, dust and general physical deterioration over time.

Candidates answered Part (b) well. Some, however, simply stated what the acronyms meant, as a way of

differentiating between the items given.

For Part (c), most candidates were able to arrange the given computers in order of size from smallest to

largest.

Part (d) was poorly answered overall. In Part (d) (i), most candidates were able to explain what is meant

by the instruction set of a computer.

In Part (d) (ii), many candidates were unable to accurately identify the three types of instructions and

often confused the examples given.

In Part (d) (iii), most candidates were unable to explain the term direct addressing and many explained

indirect addressing instead.

Part (e) was also poorly answered overall, with candidates having problems adequately explaining how

the fetch, decode and execute cycles work in a computer. The majority of candidates misused the terms

'data' and 'information' in their explanation. They also reused the terms fetch, decode and execute, to

explain the cycle on the computer.

Section B – Problem Solving with Computers

Question 3

This question tested candidates’ ability to

 identify and describe the stages in the problem-solving process

 generate algorithms based on a given scenario

 construct a flow chart based on a scenario

While the majority of the candidates was able to confidently attempt the question, there were a few who

demonstrated little knowledge of the subject content.

Part (a) was generally well done. A few candidates were only able to identify the problem-solving phase

but could not give an appropriate description.

Part (b) asked candidates to write an algorithm based on the given scenario. This question could have

been answered by using pseudocode, flow chart or narrative as examples of algorithms. Many candidates

wrote pseudocode and it was generally well done. A few candidates used a flow chart or a narrative. The

majority of candidates was able to write the algorithm using the constructs correctly and appropriately.

Only a few candidates wrote the algorithm using code-like statements and calculated the cost of the calls

incorrectly.

Part (c) required candidates to construct a flow chart based on the given scenario. It was generally well

done by the majority of the candidates. However, a few were not able to correctly and appropriately use

the flow chart symbols. For example, some candidates used the input/output symbol for processing and

vice versa.

Recommendations

Greater focus must be given to the following areas:

 Correct use of technical terms and giving appropriate descriptions of each

 Correct flow charting principles. For example,

− Correct and appropriate use of symbols

− Correct and appropriate representation of constructs such as loops (repetition)

Question 4

This question tested candidates’ ability to trace through an algorithm to determine its output. It also

examined their ability to write an algorithm for a particular case.

Part (a) was fairly well done. Most candidates attempted the question and were able to trace through the

algorithm. However, few candidates produced the correct pattern. Some candidates are still having

problems distinguishing between the write and writeln statements. The majority of candidates scored

between 5 and 6 out of a maximum of 11 marks for this part.

Part (b) was generally well done. Many candidates were able to produce an algorithm using either the

while or the for construct.

Part (c) was fairly well done. While most candidates were able to provide three properties of well-

designed algorithms, the difficulty came in accompanying them with examples in each case. This

suggests a lack of thorough understanding of properties of algorithms.

Section C – Programming

Question 5

This question tested candidates’ ability to unite a function swap and also write a C program based on

particular instructions.

Part (a) was generally well done. Candidates were required to explain two differences between compilers

and interpreters. Most candidates were able to properly explain at least one difference.

Part (b) was well done overall. Candidates showed good understanding of the need to introduce a

temporary variable in order to successfully execute the swap function.

Part (c) was generally well done. Most candidates were able to correctly write statements to compute and

output the integers in reverse order, along with the sum and product. However, in some cases, the actual

C code was not syntactically correct.

Part (d) was not well done. This question required candidates to write a recursive or non-recursive

program to output the factorial of an integer. Responses suggest that more practice is needed on questions

of this type.

Question 6

This question tested candidates’ knowledge of different programming paradigms and their ability to write

C code to manipulate records consisting of different data types.

Part (a) tested the basic knowledge of different programming paradigms. Most candidates did not

demonstrate understanding of the term paradigm, and hence could not answer the question satisfactorily.

Part (b) was attempted by most candidates. However, candidates gave differences between a mobile

device and a desktop instead of a reason for the difference in language implementation.

Part (c) was attempted by most candidates and was fairly well done. Most candidates correctly provided

the first four lines of the output. There seemed to be some difficulty coming to terms with the variable j as

a control variable in the outer for loop, and then being changed internally by the while construct. Many

candidates were able to score four out of six marks for this part.

Performance on Part (d) was satisfactory. For Part (d) (i), some candidates were unable to clearly define a

structure with fields. Many candidates were able to place values into the proper variables for Part (d) (ii)

but some did not use meaningful variable names based on the context of the question. The discount for

Part (d) (iii) was not properly applied generally because candidates were unable to access the structure

variables properly. Part (d) (iv) was fairly well done; some candidates did not use a temporary variable to

handle the swapping process.

Paper 03 – School- Based Assessment (SBA)

UNIT 1

Students must be commended for their creativity in choosing an SBA project. Although most topics were

recycled, students were able to be innovative in their delivery.

Definition of Problem

Many students were able to accurately describe the problem complete with the description of the current

system, giving examples of what takes place when the problems arise. However, some students’ projects

only included partial definitions of activities of the proposed system that they intended to create and did

not completely describe the current system and the problems which are to be solved using software.

Narrative and Flow Charts or Pseudocode

Narratives were fairly well written in most samples. However, some samples did not provide:

 A completely accurate description of the algorithm

 A description of the step by step procedure that will solve the problem as required by an

algorithm

Pseudocode algorithms were generally well done. Samples which included flow charts in some instances

incorrectly used some symbols, for example, system flow chart symbols and in others cases, diagrams

were poorly presented, had symbols with no labels (such as yes/no for the decision symbol), no connector

symbols to connect diagrams to the next page and therefore were difficult to follow. In the samples using

pseudocode algorithms, some were clearly modified copies of the programming source code, with a few

being an almost identical copy of the source code. Students should be encouraged to develop their

algorithms independent of the programming code.

Coding of Program

Students’ projects demonstrated that they were comfortable with procedural C programming language.

This was quite evident in the various projects. Most programs were logically written and properly

decomposed. Nevertheless, some students’ sample

 did not comprise functions as independent units and were still awarded full marks by the teacher

 had few data structure in the program

 only had evidence of sequence and selection but no loops for the concept of structured

programming

 did not make use of the key data structures (struct, files and arrays)

Students are reminded to print their source code directly from the compiler as transferring to a word

processor changes the spacing of the code. This may lead to students losing marks for appropriate

programming style and documentation. Generally, most students did not include adequate comments at

key areas of their code and a few were not properly indented. However, this section was done well by

most students.

Evidence that Code Matches Algorithm

Most students were able to obtain full marks on this section.

Evidence of File Manipulation

Students were required to provide code that include file and show evidence of file manipulation (open

file, write to file, read from file, append file and close file). Most students demonstrated evidence of this.

However, some students were awarded marks by the teacher even though there was no evidence of file

manipulation and in some cases the use of file was non-existent.

Testing and Presentation

This year the majority of samples provided a suitable range of test data, but a few students:

 Did not have a test plan, but had screen shots only

 Did not include all four testing criteria (normal, extreme, erroneous and incomplete) in their test

plan

 Had test results but did not have a clear test plan

 Had test which results did not include actual screen shots of the working program and testing

must be done using the test data outlined in the test plan

Generally, this year’s SBA projects were well presented.

RECOMMENDATION

Teachers need to ensure that the students use the headings outlined in the Criteria for Marking Internal

Assessment Project in the syllabus, as well as follow the order in which these heading occurs. They

should also ensure that students check that the numbering of their table of contents corresponds to the

numbering in the document and that they provide information in a logical way using correct grammar and

appropriate jargon at all times in the presentation of their projects.

UNIT 2 – FURTHER TOPICS IN COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of 63.35 out of 90, standard

deviation of 11.78 and scores ranging from 25 to 88.

Paper 02 – Essay Questions

Section A 

Data Structures

Question 1

The question tested the stack ADT, linked lists and associated operations. It was attempted by most

candidates. Overall, candidates performed better in Part (a) compared with Part (b).

In Part (a) (i), candidates were expected to declare the stack variables using C syntax. Some syntactic

errors were found in responses. Some candidates declared the stack with an incorrect type. Overall, many

candidates generally gave proper declarations.

Parts (a) (ii) and (iii) required functions for push and pop respectively. Some candidates, although their

push and pop functions were not completely correct, were able to increment and decrement the top

variable properly.

In Part (b) (i), the linked list diagram needed to clearly illustrate the head/start and null positions, the

composition of a simple node and the direction of the link. Most candidates showed an understanding of

the linked list structure. In Part (b) (ii), candidates were expected to explain the actions involved when

inserting a node to the front and back of a linked list. Many candidates did not explain how the pointer

references would change during the insertion of nodes. Overall, more practice is needed with these types

of questions.

Question 2

This question tested candidates’ knowledge of sorting and searching concepts.

Part (a) tested candidates’ ability to interpret, explain and use a bubble sort function written in C. Most

candidates recognized that a swap was being performed in the inner for loop. They were also able to

determine the content of the array after one stage of the bubble sort. Some candidates had trouble

explaining the purpose of the for loops in the bubble sort.

Part (b) tested candidates’ ability to sort a given array using selection sort. Responses were satisfactory

but some candidates did not know what a pass was.

Part (c) tested candidates’ ability to describe the principle of the binary search. Most candidates knew the

basic concept of the binary search. The idea of a divide and conquer approach of binary search was

familiar to many candidates. Some confused binary search with linear search.

Part (d) tested candidates’ understanding of the operation of a linear and binary search. Most candidates

were able to determine the number of comparisons needed to determine if a key was not present using

both binary search and linear search. Although the basic concept of the binary search was well

understood, most candidates did not have a detailed understanding of its operations. Therefore, most were

not able to correctly trace a search.

Section B 

Software Engineering

Question 3

Part (a) (i) was fairly well done. Most candidates were able to link prototyping with building a system or

software testing and refining. However, some candidates failed to mention that the end-user was

responsible for testing and providing feedback. Also, a number of candidates incorrectly described beta

testing instead of prototyping.

Part (a) (ii) was generally well answered. A substantial number of candidates scored 3–4 marks out of a

maximum of 4. Candidates were able to describe the iterative process of building software, allowing users

to test and provide feedback and then refining the software, as beneficial to creating a system that meets

the needs of the user.

In Part (a) (iii), many candidates simply stated the weakness of prototyping as being time consuming and

expensive. While some candidates described how time consuming the iterative process of evolutionary

prototyping is and how expensive it is to constantly build prototypes, many candidates did not provide

any justification for their responses and lost marks as a result.

For Part (b), some candidates did very well to create an entity relationship diagram. Most candidates were

able to identify and draw the entities correctly. Candidates were also generally able to identify the

relationships between the entities, with a few candidates having duplicate entities and relationships. Some

of the errors candidates made included using improper notation for the entities and representation of

cardinalities. Also, many candidates failed to properly represent the attributes and the primary key for

Patient. Generally, candidates generally demonstrated an understanding of what entities are as well as the

cardinalities and relationships between these entities but the ability to represent this understanding in a

diagram using standard notation seemed to be an issue for some.

Many of the responses from candidates for Part (c) indicated that they did not understand what was being

asked. Consequently, many candidates scored zero because most of them wrote about different forms of

testing such as integration testing and abnormal testing. While some candidates were able to correctly

identify two appropriate tests by searching for a key that is known to be in the array and a key that is not

known to be in the array, many of these candidates failed to get the third test correct since they tested for

an inappropriate data type (for example, string or chart) rather than testing an empty or null array.

Question 4

This question mainly tested candidates’ knowledge of data flow diagrams (DFDs.) Part (a) was generally

well done. It tested the traditional methods of determining the requirements of a new system and the

advantages and disadvantages of each. Most candidates generally answered correctly by stating

questionnaire, interview, etc. However, some candidates misinterpreted the question and gave answers

related to a system development life cycle (for example, waterfall approach and prototyping).

Part (b) tested candidates’ knowledge of DFDs. Most candidates provided good responses. Many

candidates did not clearly state that an entity is a source of data external to the system. Many candidates

also stated the purpose of a data flow which was not specifically asked for. This part of the question

specifically asked for three separate diagrams for each of the symbols. It tested the drawing and the use of

the symbols. Some candidates used one diagram but labelled it correctly. However, some candidates drew

one diagram but did not label it so no marks could be awarded. In addition, some candidates drew

separate diagrams but did not label them. Some candidates also gave the correct symbol but no data flow.

Many candidates gave a context diagram when the question clearly asked for a level 0 DFD. Candidates

lost marks for the data flow in the diagram because they used verbs to describe the flow of data. Also, the

warehouse was placed as an entity in some DFDs when in fact the warehouse was not an external entity.

For Part (c), few candidates were able to correctly distinguish between functional and non-functional

requirements and give correct examples. Others gave correct definitions but the wrong examples and vice

versa.

Overall, most candidates attempted all parts of this question.

Section C 

Operating Systems and Computer Networks

Question 5

This question tested candidates’ knowledge of networking concepts such as network transmission media,

network configurations, connectivity devices, transmission standards, and the role of the OSI model in

network communication.

Part (a) dealt with a peer-to-to peer network. Despite an attempt by most candidates to answer this

question, it was generally not answered correctly. Many of them were unable to adequately mention that

the roles can interchange.

Part (b) (i) tested candidates' knowledge of network security. Most candidates understood the concept

very well and earned high marks.

Part (b) (ii) tested candidates’ ability to explain the network issue of expandability. Most candidates

understood that more devices or users would need to be added. However, some were not aware how to

modify to the network.

Part (b) (iii) tested candidates` knowledge of interconnectivity of the network. Most candidates wrote

about connecting within the network rather than with other networks. This part of the question was

generally poorly done by candidates.

In Part (c), only a few candidates were able to list the correct order of the OSI layers. Many candidates

listed layers starting from 7 to 3. Most candidates were unable to describe the role of three layers. They

did not understand the role of the different layers.

Part (d) tested knowledge of a network topology. Most candidates drew and labelled the diagram

correctly. However, some candidates had the server as the central device rather than the switch.

Question 6

This question tested candidates’ knowledge of process management, in particular their knowledge of

process states and process control blocks. It also examined their understanding of the different types of

operation systems and device management.

Parts (a) and (b) were poorly done. Most candidates recognized state of process and process identification

but could not identify the other components. Candidates could have stated that the process control black

(PCB) contains important information about specific processes. Some components of the PCB are:

 The current state of the process, for example, ready runway

 Priority of process which CPU can use for scheduling

 Register information where data can be stored

In Part (c), most candidates provided a partially correct answer for a device driver. The full response

expected was: divide drivers contain instructions to control the hardware components that are linked to

the operating system code used.

Part (d) was well done with candidates giving satisfactory answers for types of operating systems.

Paper 03 – School- Based Assessment (SBA)

UNIT 2

Students performed very well, with many following the marking guidelines.

Definition of Problem

Students were required to state the shortcomings of the existing system, and generally this section was

well done by most of them. However, some students failed to clearly describe the problem symptoms

evident in the existing system. There was too much description of the current system but not enough

explanations of the problems staff/clients face.

Techniques of Analysis

Students were required to identify techniques of data collection and analysis. Most of them were able to

state the various techniques of data collection. However, students must be able to justify the reason for

their selection as well as show proof of analysis. For example, questionnaires and/or interview transcripts

should be included in an appendix. The proper usage of the techniques was not described in the context in

which the problem was encountered.

Context Level Diagram

In some instances, incorrect symbols were used for entities and processes. Some data flows were not

labelled, and data store/file were incorrectly included in the context level diagram.

Level 1 Diagram

Students were required to give a complete and accurate diagram with all relevant processes, data flows

and major data stores; however, some of them had links between files and entities which were incorrect.

ER Diagram

Students were required to give a complete and accurate diagram of all relevant entities and relationships.

This was generally done well by most of them; however, some of the cardinalities and also the description

of the relationships were not done correctly.

Functional and Non-Functional Requirements

Students were required to give a complete and accurate description of all requirements. This was done

well by some of them. However, some students confused or misinterpreted the difference between

functional and non-functional requirements. Some of them described project limitations for non-

functional requirements.

Design Specification

Students were required to give a complete and accurate description of:

 System structuring

 User interface

 Report design

 Algorithm design

 Appropriate data structures

Some students did most of the design specifications; however, some of them

 went into too much detail in the systems structure by including modules or branches that

were not necessary

 did not name or justify the type of user interface they used

 had no evidence of report design but were given marks by teachers

 did not produce any algorithm

 were not awarded any marks by the teacher for appropriate data structure even though the

program clearly showed the use of struct, array, queues and files declarations and

manipulations. This suggests that either the teachers did not know or, needed clarification

on what to look for in this section.

Coding and Testing

Students were required to produce a complete and accurate program with

 code that achieves functionalities such as documentations, outputs, usability and reporting

 code that corresponds to the design specification

 a test plan with exhaustive data set (including unit testing, integrated testing and systems testing).

Most students produced programs that achieved good functionality. However, some of them did not

produce proper test plans to cover unit testing, integrated testing and systems testing.

RECOMMENDATIONS

 Teachers should ensure that students are fully prepared for the examinations in both units. The

poor performance on some modules of the syllabus indicates that more time needs to be spent on

these areas.

 All programming language code should be printed from the C compiler (not from word

processor)

 Examiners can only mark programming code from printed output and therefore, to show that the

program works, students need to print screen shots of the program in the testing phase and should

not send soft copies.

 Only one form of algorithm is required that is, either

− Pseudocode OR

− Flowchart

 Students are not required to describe the results of the analysis techniques.

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2014

COMPUTER SCIENCE

St Michael Barbados

GENERAL COMMENTS

This was the sixth year in which the revised syllabus was examined. There were three examination

papers in each of Units 1 and 2, namely, Paper 01, Paper 02 and Paper 03. In each unit, Paper 01 and

Paper 02 were examined externally by CXC while Paper 03, the School-Based Assessment (SBA),

was examined by teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple-choice questions which were designed to test candidates’

breadth of coverage of the syllabus. Paper 02 consisted of essay-type questions that were designed to

test candidates’ depth of understanding of the syllabus. Thus, candidates were expected to show

deeper insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50

per cent, and 20 per cent, respectively.

Approximately 93 per cent of candidates obtained Grades I─V in Unit 1 and approximately 92 per

cent of candidates in Unit 2 obtained Grades I─V. Overall, there is still need for improvement in the

quality of responses for the programming questions in both units.

DETAILED COMMENTS

UNIT 1 – FUNDAMENTALS OF COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of approximately 61.20

out of 90, standard deviation of 14.64 and scores ranging from 24 to 90.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

Part (a) tested candidates’ ability to draw a clearly labelled block diagram of a 2-to-1 line multiplexer.

Most candidates attempted this question gaining at least three of the five marks. Many candidates lost

marks mostly through failing to realize that a 2-to-1 line multiplexer has one select line. Candidates

also lost marks via omission of the select line and improper labelling.

Part (b) required candidates to define the term logic gate. Most candidates wrote that logic gates take

inputs and provide outputs according to their logical rules. Candidates did, however, lose marks by

failing to note that logic gates are elementary building blocks/components of digital circuits.

Part (c) provided a circuit diagram and asked candidates to provide the truth table associated with the

diagram. This part was fairly well done. Candidates are however encouraged to maintain order when

displaying the input values for the truth table (for example, 0 0, 0 1, 1 0, 1 1).

Part (d) required candidates to indicate which of decoder, flip-flop and multiplexor are associated with

a single input. Most candidates were able to identify at least one out of the two correct devices (flip

flop and multiplexor). Candidates were not required to provide definitions for the structures

identified.

For Part (e) (i), candidates were required to provide the number of digits that a 3-bit decoder would

display. Many candidates lost marks by answering 6 (2 x 3). A decoder can provide 2” outputs where

n is the amount of different inputs. Therefore the number of outputs would be 8.

For Part (e) (ii), candidates were required to examine the digital figure provided and list the segment

letters that must be switched on to display the number 5. This part was well done. Candidates

provided most or all of the segments needed. Some candidates provided the segments that should be

off and were awarded no marks for their response.

For Part (f), the responses provided by candidates were mostly correct although some candidates

included sign and magnitude alongside the complement process and added a sign bit before inverting

the value. Candidates must be mindful of the number of bits required for the question as those whose

final responses were not represented using 4 bits lost marks as a result.

Part (g) asked candidates to determine what decimal number was being represented by the given

binary string. Many candidates did not seem to have knowledge of this process and some candidates

simply treated the binary as an absolute value and converted it to its decimal equivalent. Candidates

were expected to have displayed the component parts of the string showing 0 for the sign, 011 = 3 for

the exponent and 10110 for the mantissa. Many candidates failed to recognize that the mantissa is a

fraction and should be written as 0.10110.

Question 2

Part (a) tested candidates’ ability to describe in the correct order, the three main activities in an

instruction cycle. Candidates were able to correctly identify- in the correct order- the three main

activities that take place in an instruction cycle as fetch, decode and execute. Many candidates

attempted to discuss the activities outlined but gave vague explanations of each.

In Part (b) (i), candidates were required to give one similarity and one difference between RAM and

hard disks. This part was well done.

Part (b) (ii) tested candidates’ knowledge of registers used in the CPU of a computer. Few candidates

were able to provide the correct responses to this part. Most neglected to thoroughly explain the

reasons for including registers in the CPU.

For Part (c), almost all candidates attempted to describe situations where a Supercomputer, PDA and

Mainframe can be used. However, many candidates gave definitions or explanations of these

categories of computers rather than their uses.

Part (d) was very well done.

For Part (e), whilst the majority of candidates correctly identified port connectivity, speed and

software incompatibility as reasons why current computers may not be able to work efficiently with

new storage devices, very few adequately discussed these points resulting in their inability to obtain

full marks.

Section B – Problem Solving with Computers

Question 3

This question tested candidates’ ability to:

 Identify and give examples of the main control constructs used in programming.

 Generate an algorithm to select the favourite subjects of 100 students.

 Generate an algorithm to calculate the sum of the multiples of six and seven between 100 and

250.

For Part (a), the majority of candidates described the control constructs and gave an example of each.

This part was fairly well done.

Part (b) was done well by the majority of candidates. A few candidates did not initialize the variables

needed to perform the counting operation. Some candidates also used the selection construct

improperly.

Part (c) proved challenging for some candidates. Those who did not complete the question had

difficulty testing correctly for the multiples of the given numbers.

More attention should be focused on the following:

 correctly using the sequence, selection and iteration control constructs

 using functions such as modulus to test numbers for specific properties

 correctly selecting the limits of loops based on given scenarios

Question 4

Part (a) tested candidates’ ability to write an algorithm using the iterative control construct to solve a

given problem; Part (b) required candidates to represent the algorithm with the use of a flowchart and

Part (c) tested candidates’ ability to trace an algorithm to determine its output.

More than 90 per cent of the candidates attempted this question and gave satisfactory responses.

Part (a) was fairly well done. The weaker candidates confused the data in the example given as a

solution to their algorithm. Some candidates used the iterative control construct improperly, using the

“$” in the IF statement as well as failing to initialize the variable for the running total. Some

candidates also answered the question using the C programming language rather than the construction

of the algorithm.

Part (b) was attempted by most candidates and was generally well done. Some of the weaker

candidates used the flowchart symbols incorrectly and some excluded the iterative statement in the

flowchart.

Part (c) was also attempted by the majority of candidates and they performed very well. In most

cases, they scored full marks. Some candidates did not change the value of the variable ‘x’ for the

trace table which resulted in incorrect values.

Section C – Programming

Question 5

This question tested candidates’ knowledge of the stages in the translation process of a program, and

the purpose of the watch window in the C compiler. Candidates were also required to write a C

function based on a given scenario and apply knowledge of FILE processing.

Part (a) was poorly done by most candidates. Candidates attempting this part seemed to be familiar

with the lexical analysis stage only.

Part (b) was poorly done. Most candidates simply did not know the purpose of the watch window in

the C compiler.

Part (c) was poorly done. Candidates seemed not to know the difference between a main function and

a programmer defined function, hence many wrote their solution using main. Those candidates who

did well were able to write a proper function header, sum the elements of an array and return its total.

Candidates lost marks if they did not declare and initialize their variables.

Part (d) was poorly answered. Most candidates who attempted this part of the question knew how to

create a FILE pointer. However, candidates had difficulty writing the appropriate structure of fscanf,

fprintf, and fopen statements. Candidates seemed not to know when to use the different modes of

opening a file, for example, (r, w, a, r+, w+, a+).

Question 6

This question was designed to test candidates’ ability to:

 Explain the importance of good programming style.

 Interpret source code and utilize their knowledge to formulate the output of given code.

 Write a C program that tests their knowledge of program structure, variable declarations,

input and output of variables, and their ability to formulate an equation to solve the addition

of taxes on an existing price.

 Write C code that tests their ability to read and print numerical values after performing an

arithmetic calculation to solve the average of those numbers.

Part (a) was generally well done. Some candidates failed to explain the importance of indentation

clearly and lost marks. A few candidates confused the purpose of indentation and associated it with a

benefit towards the computer or compiler.

Part (b) was well answered; only a few candidates did not elaborate their answers properly.

Part (c) was not clearly understood by many candidates; the majority failed to provide the actual

output that was being requested. The majority of candidates took the approach of providing a trace

table and some simply attempted to show some sort of calculation to arrive at their answers.

Most candidates failed to analyse the loops properly. Upon entering the while loop, candidates

performed well in obtaining the result of the first stage but failed to re-enter the for loop with the

appropriate data hence, obtaining wrong data thereafter.

Part (d) was generally well understood by most candidates but some failed to write correct syntax

either in the form of program structure: missing out # or braces; variable declarations: declaring as int;

reading of inputs: missing out the &; and in their output: not formatting their result as %6.2f. Some

candidates failed to analyse the mathematical problem as most failed to divide the tax by 100.

Part (e) seemed to be unclear to most candidates as it pertains to the required result. Most candidates

read variables properly and solved the result but failed to provide their answer as a valid average of

type float. Some candidates failed to apply proper precedence of operations by omitting brackets and

some even ended up dividing the sum of all numbers read by values such as 2 and 3 when the

question clearly asked for 4 numbers. The majority of candidates failed to provide their result as a

float data type.

Candidates are encouraged to pay close attention to the use of proper coding syntax when asked to

write C programming code. Candidates are also encouraged to carefully read and analyse the

questions to fully understand what is being requested of them.

Paper 03 – School-Based Assessment (SBA)

General Requirements

Students were expected to choose a problem for which a software solution was appropriate; develop

and present the solution in a logical way using correct grammar and appropriate jargon at all times by

doing the following:

 Create an algorithm for the solution using modules, sequencing, selection, assignments, and

iteration (bounded and unbounded).

 Represent their algorithms using narrative format and also as either a flow chart or pseudo-

code.

 Implement the algorithm in C using various data structures such as arrays, struct, strings and

files, with no less than five functions as independent units.

 Ensure that the source code produced, matches the algorithm.

 Create a test plan with exhaustive data set, test and produce test results and appropriate error

messages.

The following specific aspects of the project were assessed:

 Definition of problem;

 Narrative and flow charts or pseudocode;

 Coding of program;

 Testing and presentation;

 Communication of information.

For each component, the aim was to find the level of achievement reached by the student. It is

recommended that teachers make the assessment criteria available to students at all times.

Definition of problem

Many students were able to accurately describe the problem complete with the description of the

current system and giving background to the problem, the issue or problem, recommended solution

and examples of what takes place when the problem arises. However, some students’ projects only

included partial definition of activities of the proposed system that they intended to create without

completely describing the current system and the problems which are to be solved using software. In

addition, some problems that were identified could not be solved by using a computer program. For

example, filling out of a form is a time-consuming and error-prone process for both paper-based and

computer-based systems. However, there are advantages of using computer-based forms that were

not identified.

Narrative and flow charts or pseudocode

Narratives were fairly well written in most samples. However some samples:

 Provided only a partially correct description of the algorithm as a solution.

 Provided descriptions of an intended module or unit instead of describing the step by step

procedure that would solve the problem as required by an algorithm.

Pseudocode algorithms were generally well done. Samples which included flowcharts in some

instances incorrectly used some symbols. For example, system flowchart symbols were used. In

others cases, diagrams were poorly presented, had symbols with no labels (such as yes/no for the

decision symbol), had no connector symbols to connect diagrams to the next page. Charts were

therefore difficult to follow. Most students produced samples using pseudo-code algorithms. Some

samples were clearly modified copies of the programming source code, and a few almost identical

copies of the source code. It is therefore recommended that in the future, students be encouraged to

develop their algorithms independent of the programming code.

Coding of program

Students’ projects demonstrated that they were comfortable with procedural C programming

language. This was quite evident by the responses of the various projects. Most programs were

logically written and properly decomposed. Nevertheless, some students:

 Did not use functions as independent units in their programs and yet some teachers still

awarded full marks.

 Used too few data structures in their programs.

 Did not demonstrate appropriate use of the concept of structured programming.

Some students did not print their programs from the C compiler, instead choosing to make print

screen copies of their projects. Students are therefore reminded to print their source code directly

from the C compiler as transferring the code to a word processor often changes the code. For

example, transferring to a word processor may cause adjustments to the spacing and cause strange

symbols to occur in the code, making it difficult to read. This may also lead to students losing marks

for inappropriate programming style and documentation. In addition, marks were lost because some

students did not include adequate comments at key areas in their source code, and they used a poor

indentation style that did not make the code easier to read. However, this section was generally done

well by most students.

Evidence that code matches algorithm

Most students were able to obtain full marks on this section as this was well done.

Evidence of file manipulation

Students were required to provide code that included a file and show evidence of file manipulation

that is, open file; write to file; read from file; append file and close file.

This section was well done. However, some students were awarded marks by the teacher even though

there was no evidence of file manipulation and in some cases the use of files was non-existent or was

declared but not used, and the teacher still awarded full marks. Projects that showed screenshots of

the file data being added, edited and deleted were awarded full marks.

Testing and presentation

Students were required to prepare a test plan with an exhaustive data set (test data) to test their

programs. Students were also expected to use the test data set to produce test results with normal

input giving correct results, extreme input giving correct results and erroneous (abnormal) input

giving appropriate error messages. The majority of samples moderated this year did not have a

suitable range of test data. However, a few students:

 Did have a test plan and screen shots while others provided screen shots without any test

plan.

 Did not include all testing criteria (normal, extreme and erroneous) in their test plan.

 Had test results but did not have a clear test plan.

 Had no test results but were awarded marks by teacher.

 Showed test results but did not include actual screen shots of the working program. Testing

was not done using the test data outlined in the test plan.

 Tested their menus with the appropriate test data, but all input must be tested in the same

way.

Generally, this year’s SBA projects were well presented.

Recommendation

Teachers need to ensure that students use the headings as outlined in the syllabus. In addition,

students should follow the order in which these heading occur in the Form CSCI 1-3. They should

also ensure that students check that the numbering in the table of contents corresponds to the

numbering in the document and that students provide information in a logical way using correct

grammar and appropriate jargon at all times in the presentation of their projects.

UNIT 2 – FURTHER TOPICS IN COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of approximately 62.19

out of 90, standard deviation of 13.09 and scores ranging from 17 to 86.

Paper 02 – Essay Questions

Section A  Data Structures

Question 1

Part (a) tested candidates’ knowledge of the creation and deletion of a STACK (ADT). The majority

of candidates was able to correctly identify that a stack only needed to exist in order to execute the

Destroy Stack function. The modal mark was two in this section.

For Part (b) (i), candidates were expected to perform the basic PUSH and POP operations on a stack;

(Module 1: Specific Objective 3). This question tested analysis and interpretation of the given

operations. Generally, candidates were able to acquire total marks when they clearly illustrated the

contents of the stack at various stages as the operations progressed. In a few cases candidates focused

on the computation and neglected to show the contents of the stack. Marks were not awarded in those

cases.

The responses to Part (b) (ii) were generally accurate. Most candidates were able to identify that there

should be at least two elements in the stack in order for the MULT and ADD operations to work

effectively. It should be noted that there were cases where candidates gave a vague response, for

example, ‘Stack should not be empty.’ Marks were not awarded in these cases.

For Part (b) (iii), candidates were expected to analyse the above scenario and deduce that the contents

of the stack would be reduced to one element and eventually the ADD operation will attempt to POP

from an empty stack.

For Part (c) (i), the majority of candidates obtained four out of five marks because they were able to

accurately identify that the ENQUEUE operation adds an element to the rear/tail of a queue and the

DEQUEUE operation removes/retrieves an element from the front of the queue.

Candidates performed poorly on Part (c) (ii). The majority of their algorithms were incomplete and

incoherent. The DEQUEUE function was not implemented properly and the iterative construct and its

condition were in most cases incorrect or omitted. Generally, candidates were expected to:

 Initialize a ‘count’ variable

 Implement an iterative construct which will terminate when the queue is empty (‘Dequeue’

the queue correctly and Increment count)

 Return/Display count

Question 2

Part (a) tested candidates’ ability to explain how two numbers, 5 and 15, would be inserted into a

singly linked list and then to represent the final list in the form of a diagram. Most candidates were

able to give some level of explanation, however many were somewhat vague. In most cases they

demonstrated the concept of a link between nodes but often were not clear on the sequence of steps

involved in inserting items into the linked list. Candidates needed to clearly indicate in their

explanation that a head was present and linked to either a node or null. Additionally, candidates

needed to indicate that a new node must be created first then the number is stored in it and not merely

“insert the number into the list”.

About 80 per cent of the candidates were able to produce reasonable diagrams to represent the nodes,

however many did not correctly represent key components. For example, some candidates omitted

arrow heads, while others had the head pointer arrow in the wrong direction. In some cases, the head

pointer was drawn to look like a node; candidates needed to demonstrate some distinction between

them. In other cases the notation used for the node was incorrect in that it did not clearly show the

two parts, that is, the data and the address fields.

Part (b) tested candidates’ ability to declare an integer array, read values into it and then search for the

presence of a particular value.

The majority of candidates were able to correctly declare an array. Candidates almost always

indicated the correct data type as well as the array size. However, some candidates used the wrong

name for the array although a specific one was given in the question.

In terms of storing items in the array which was the second requirement of this part of the question,

most candidates demonstrated the need for a loop in order to perform this task efficiently. They were

also able to produce the correct or near correct code to accept values and assign them to the array

locations. However, many candidates omitted one or two key components such as the ampersand or

the index. In some cases they also used the wrong format specifier instead of %d.

In the final section of Part (b), most candidates were able to produce C code to accept the search key.

Some of them were able to write reasonable code to conduct the search, however quite a few used the

wrong upper limit for the loop. They could have used < 100 since the C programming language starts

to index an array from 0 causing index value 99 to represent the 100

location.

For the comparison many used incorrect notation for ‘equal to’. Candidates should note that two equal

signs ‘==’ are used for this purpose while a single ‘=’ is used for assignment.

For the last statement which should state ‘Key Not Found’, many candidates placed it within the loop

so that the error message would be printed repeatedly. Other candidates did not have a statement to

allow the program to break out of the loop after the key was found. This meant that the loop would

continue processing unnecessarily after the key was found.

Section B  Software Engineering

Question 3

In general, almost all candidates attempted this question.

For Part (a), many candidates incorrectly wrote that a deliverable was the end product of the Systems

Development Life Cycle (SDLC) to be ‘delivered’ to the customers. The majority of candidates did

however identify a deliverable as some form of documentation or tangible/intangible result of each

phase of the SDLC. Candidates lost marks for not indicating that a deliverable needed to be signed

off and used as input for subsequent phases.

In Part (b), candidates responded well to the reasons why an information system may need to be

changed. Candidates easily identified issues that would cause a system to change including change in

requirements, inefficiency and outdated systems. Yet candidates could have done better by

suggesting how a new system with modern updated technology would alleviate these issues.

In Part (c), while some candidates were able to correctly identify the factors: operational, technical,

economic and legal, in some cases candidates’ responses lacked proper explanation of the terms and

often candidates would confuse operational feasibility with technical feasibility.

For Part (d), the ERD was done quite well by most candidates. It should be noted that many

candidates who should have received 12 marks lost marks for simple errors including:

 Plural entities (ending with ‘s’)

 Not indicating the primary key by underlining the attribute

 Incorrect cardinalities including the notation

Question 4

This question generally tested candidates’ knowledge of the design of input forms and of data flow

diagrams (DFDs). Part (a) (i) was generally well done. However, some candidates provided one

example only of data in the required formats.

Part (a) (ii) asked candidates to justify their selection of one of four input options. However, many

candidates did not match the input option to a specific data item. Others misinterpreted the question

and recommended different types of user interfaces for instance command-line or menu based.

Part (a) (iii) was poorly done. Most candidates incorrectly described range verification checks and

did not relate this part of the question to the previous parts.

Part (b) tested candidates’ ability to create a data flow diagram based on a given scenario. Most

candidates provided satisfactory responses, however many candidates incorrectly labelled the

processes as departments; drew context diagrams which included each department listed as a

source/external entity; and included departments along with their processes as separate symbols.

Candidates also lost marks for the data flow in the diagram because they incorrectly used verbs to

describe the flow of data. Additionally, marks were not awarded when candidates used incorrect

symbols for the four data flow diagramming objects (source/sink, data store, data flow, process).

Section C  Operating Systems and Computer Networks

Question 5

This question tested candidates’ knowledge of network devices, including: router, switch and modem

along with their roles; hybrid network topologies, fibre optic cables, the IEEE802.11b network, VOIP,

and GPRS.

Part (a) (i) dealt with setting up a small network using the following network devices: modem, router,

and switch. Most candidates attempted this question, however, even though the devices were shown,

some of them did not connect the devices correctly and critical ports were not shown.

Part (a) (ii) tested candidates’ knowledge of the role of the modem, switch and router. Most

candidates demonstrated an understanding of the role of the modem being analog to digital conversion

and vice versa. However, many candidates failed to adequately state the role of the router and switch.

Part (b) tested candidates’ knowledge of the hybrid network topology. This question was well done.

Almost all candidates were able to provide a correct definition of a hybrid network topology.

Part (c) tested candidates’ knowledge of the advantages and disadvantages of fibre optic cables. This

part was poorly done since candidates failed to adequately describe the advantages and disadvantages.

Part (d) was poorly done since many candidates failed to clearly explain how data is communicated in

an IEEE802.11b network. Many of them did not demonstrate understanding of the use of wireless

access points in this kind of network.

Part (e) required candidates to explain why the quality of voice over IP (VOIP) communication might

be different from telephone communication service as offered by a telephone company. Most

candidates demonstrated understanding that VOIP uses the Internet but failed to discuss the

deterioration of the quality of the call where there are Internet problems or data packet delays.

Part (f) tested candidates’ knowledge of the main purpose of GPRS. This question was fairly well

done. Many of the candidates were able to link the term GPRS with the transmission of internet

packets through a cellular network.

Question 6

This question tested candidates’ knowledge of operating systems, security of data and device

management. In addition it tested their knowledge of

 process management and in particular, process states

 interrupts and their application to scenarios

 scheduling processes with emphasis on pre-emptive scheduling.

For Part (a), most candidates were able to correctly identify that batch systems compiled jobs in

groups. However, they were unable to specify that these jobs would be processed at a later time. The

majority of candidates was also able to say that multi-user systems allow the use of many users at the

same time but failed to identify their processing power.

For Part (b), most candidates recognized that a process may be blocked but were unable to define

a blocked process as a process that is paused, pending an event. Candidates also failed to specify that

a running process is one where instructions were being executed by the CPU. This part of the

question was poorly answered by most candidates. Many candidates understood that the game would

be halted to process the interrupt but most candidates did not identify the type of interrupt.

Candidates also failed to recognize that a context switch occurs, that is, saving the state of the running

process in registers for later resumption, and that the game process resumes after the processing of the

interrupt.

Part (c) was poorly answered. While some candidates recognized that pre-emptive scheduling meant

that the scheduler had the power to pre-empt the tasks, they did not specify that the processor would

resume the previously running tasks.

In Part (d), the majority of candidates was able to correctly identify three ways of protecting data.

However, some candidates were not able to sufficiently discuss their methods in order to gain full

marks.

In Part (d), candidates were able to identify the type of software being used as device drivers.

However, most candidates did not give the definition of a device driver as the set of instructions used

to communicate with the device. They also neglected to mention that each device driver uses the same

protocol to communicate with the OS.

Paper 03 – School-Based Assessment (SBA)

Requirements

Each student was expected to choose a problem for which a software solution exists and then develop

the software using software engineering techniques. In particular, the student was expected to

demonstrate appropriate choice of the tools and techniques used in the analysis of the software to be

developed. They were then expected to design, code, and test their software using appropriate

techniques.

General Comments

 Students should follow the order laid out in the criteria for marking when arranging the

sections of the SBAs.

 Teachers should avoid using red ink pens to correct SBAs.

 Teachers should ensure that each SBA is clearly labelled with the student’s name and

centre number.

Marking Criteria

Specification of Requirements

Definition of Problem

Students were required to give a complete and accurate description of the problem. Most students

handled this section fairly well but there were a few students still unclear as to what to include in their

definition of the problem. A brief description of the context in which the problem occurred is

required but details about actual problems staff/clients face and proposed steps to correct such

problems must be emphasized.

Techniques of Analysis

Students were required to identify techniques of data collection and analyse the data collected.

 Most students were able to state the various techniques of data collection; however, some

students were not able to justify their selections. Several students misinterpreted justification

to mean definition. This is incorrect. Students should clearly explain why the chosen

technique was used as it relates to the business/company/institution and not regurgitate

advantages of the technique.

 Proof of analysis should also be given. All proof must be included directly after the analysis

not in the appendix. For example, sample questions from the questionnaires and/or interviews

should be included. It is understood that it may be difficult to include proof of observation;

hence, if this technique is used, students should clearly describe what was observed.

Use of Data Flow Diagrams and E-R Diagrams

This section of the SBA was poorly done by most students.

Context Level Diagram

Students were required to give a complete and accurate diagram of all relevant entities and data flows.

 In some instances, incorrect symbols were used for entities and processes.

 Some data flows were not labelled.

 Data stores/files were incorrectly included in the Context Level Diagram.

Level 1 Diagram

Students were required to give a complete and accurate diagram with all relevant processes, data

flows and major data stores.

 Most students are unaware that the Level 1 diagram is an expansion of the context level

diagram. Hence, new external entities were created for the Level 1 diagram, as well as new

data flows were created for previously used entities.

 Students had links between

o data stores and entities

o data stores and data stores

o entities and entities

ER Diagram

Students were required to give a complete and accurate diagram of all relevant entities and

relationships.

 Most students did not use the correct symbol for a relationship (a diamond).

 Some students did not include attributes for the entities.

Functional and Non-Functional Requirements

Students were required to give a complete and accurate description of all requirements.

 For the functional requirements, students did not clearly state what the system is expected to

do; instead, they stated what the user will be doing. An example of a good functional

requirement is the system will be able to delete a patient’s record.

 For the non-functional requirements, students did not state the limitations of the system. An

example of a good non-functional requirement is the system can only store 1000 patient

records.

Design Specification

Students were required to give a complete and accurate system structuring diagram containing all

processes and a description of the user interface, report design, algorithm design and appropriate data

structures. Most students handled this section fairly well. However, the following should be noted:

 Students should not just include screens shots of interface and report design but also a

justification in order to gain maximum marks.

 Narratives will not be accepted as an algorithm. Students are reminded to submit

pseudocodes.

Coding and Testing

Students were required to produce a complete and accurate C program solution for the problem stated

in the Definition of Problem section. Most students produced programs that achieved good

functionality. Students should be aware that:

 All code must be written in procedural C. No other programming language will be accepted.

Code must also be printed from the compiler, NOT transferred to a word processor before

printing.

 Soft copies will not be marked.

 Most students did not include enough screen shots (and in some cases none at all) to support

their functioning of the code.

 Some codes presented did not match the screen shots given. In such cases students were not

awarded any marks.

 Test plans should be written in a tabular format and it should include normal, extreme,

erroneous and incompatible data.

 All input data must be tested.

 Test results without related test plans will not be awarded any marks.

Communication and Presentation

This section of the SBA is often overlooked by students. Teachers and students are urged to pay close

attention to their use of grammar and the overall presentation of the SBA.

Recommendation

Teachers should ensure that students are fully prepared for the examinations in both units. Poor

performance in some modules of the syllabus indicates that more time needs to be allocated to these

areas.

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2015

COMPUTER SCIENCE

St Michael Barbados

GENERAL COMMENTS

This was the seventh year in which the revised syllabus was examined. There were three examination

papers in each of Units 1 and 2, namely, Paper 01, Paper 02 and Paper 03. In each unit, Paper 01 and

Paper 02 were examined externally by CXC while Paper 03, the School-Based Assessment, was

examined by teachers and moderated by CXC.

In each unit, Paper 01 consisted of multiple-choice questions that were designed to test candidates’

breadth of coverage of the syllabus. Paper 02 consisted of essay-type questions that were designed to

test candidates’ depth of understanding of the syllabus. Thus, candidates were expected to show deeper

insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50

per cent, and 20 per cent, respectively.

Approximately 94 per cent of the candidates obtained Grades I─V in Unit 1 and approximately 94 per

cent of the candidates in Unit 2 obtained Grades I─V. Overall, there is still need for improvement in

the quality of responses for the programming questions in both units.

DETAILED COMMENTS

UNIT 1 – FUNDAMENTALS OF COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of approximately 64.3 out

of 90, standard deviation of 13.9 and scores ranging from 24 to 90.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

Part (a) (i) required candidates to write down the truth table for a given security alarm system scenario.

Most candidates attempted this part but many failed to use three inputs, using only two instead. Most

candidates also failed to indicate clearly what labels represented, that is, switch, door or window. While

most candidates produced a table with bit patterns, a few used phrases in their table, which were

unacceptable.

Part (a) (ii) required candidates to use primary logic gates to design and draw circuit associated with (a)

(i). Most candidates attempted this part but complicated their diagrams by using several gates, including

NOT gates. Only two gates an OR and an AND gate were required. Some accurately produced diagrams

were not labelled resulting in the loss of marks. In some instances, candidates’ representation of the OR

gate was unclear, closely resembling the NOT gate.

Part (b) required candidates to explain the term ‘demultiplexer’. Most candidates attempted this part

but some chose to simply say that it was the opposite of a multiplexor. Others gave long definitions of

multiplexors and did not focus on the definition of the demultiplexer. In many cases, candidates

incorrectly stated that it accepts one output signal to produce several input signals.

Part (c) required candidates to draw a clearly labelled block diagram of a 3 to 8 line decoder. Most

candidates attempted this part successfully with some losing marks only for inadequate labelling. It was

expected that all input and output lines, as well as the diagram itself be properly labelled.

Part (d) required candidates to determine the decimal equivalent of the hexadecimal number 1A53.

Most candidates attempted this part, but the quality of the responses was generally unsatisfactory. While

most of the candidates were able to identify the base as 16, many could not identify the value of A as

10 and some did not apply the weighting of the first digit correctly.

Part (e) required candidates to find the 8 bit two’s complement of -21. Most candidates attempted this

part, but the quality of responses was generally unsatisfactory. While most candidates were able to

correctly convert the number to binary, many were unable to go on to find the correct two’s complement.

Part (f) required candidates to calculate the decimal equivalent of a given floating point representation.

Most candidates attempted this part. While most candidates were able to identify the exponent correctly,

some were unable to determine the correct position of the decimal point by moving the point three

spaces to the right, causing them to get an incorrect answer. Only a few candidates failed to recognize

the number as being positive. Also, many candidates did not clearly separate and label the sign,

exponent and mantissa components of the given binary representation.

Question 2

This question was attempted by the majority of candidates however the majority of them obtained

between 11 and 15 marks with only about 5 per cent earning between 20 and 25 marks.

Part (a) tested candidates’ ability to define the CPU and describe TWO of its main functions. Almost

all candidates omitted to clearly define the CPU as hardware though they noted that it carries out the

instructions of a computer programme. Candidates were able to identify the two major components of

the CPU as the ALU and CU but the descriptions of the functions of each were often incomplete.

Part (b) (i) required candidates to state the purpose of the fetch operation while Part (b) (ii) required

candidates to state the purpose of decode and execute operations. Candidates were generally correct in

explaining the purpose of each operation, clearly stating that:

 The fetch operation is used to get data and instructions from main memory

 The decode operation determines what the given instruction means,

 The execute operation performs the instructions.

Part (b) (iii) required candidates to define the term instruction set. Part (b) (iii) was poorly answered as

candidates failed to define the instruction set as a specific set of commands which the CPU is designed

to understand.

Part (c) (i) required candidates to differentiate between ROM and RAM. The majority of candidates

was able to identify two differences between ROM and RAM but some candidates confused the two

and did not identify RAM as volatile and ROM as non-volatile.

Part (c) (ii) tested candidates’ knowledge of different types of computer memory. Many candidates were

unable to distinguish between cache and RAM and, hence, inadequately discussed the advantages and

disadvantages of each option. Whilst many candidates indicated as an advantage, that for Option 1

having more cache would increase the speed of the computer as an advantage they did not identify being

expensive as a disadvantage. For Option 2, many candidates did not recognize multitasking or cheaper

costs as advantages of RAM but stated slower processing speeds as a disadvantage of having less cache.

Part (d) tested candidates’ knowledge of the purpose and functions of registers found in the CPU. The

majority of candidates correctly stated that registers hold all the data currently being handled by the

CPU. However, about 70% of candidates were unable to correctly identify and describe registers such

as the instruction register, program counter, memory address register and memory data registers.

Section B – Problem Solving with Computers

Question 3

Part (a) required candidates to explain what happens during the implementation and review phase of

the problem-solving process. This question was poorly done by many candidates as they failed to

correctly describe each phase. The, implementation phase was generally incorrectly described as

identifying and selecting possible solutions. Reviewing was generally incorrectly described as the phase

which involves the testing of the solutions. Much greater focus is needed in understanding and

explaining the main activities in each stage of the problem-solving process.

Part (b) presented candidates with two algorithms and required them to identify which illustrated

bounded and unbounded iteration. Candidates were also required to explain how the loops in the two

algorithms would terminate. The majority of candidates was able to correctly identify the algorithm

which were bounded and unbounded. However, many candidates were unable to correctly explain how

the loops were terminated. For example, candidates incorrectly explained that the ‘for’ loop terminated

on the 20th repetition. The ‘for’ loop terminates after the limit has been reached and on the final

repetition when the limit is checked and is found to be outside of the range. The majority of the

candidates correctly described how the ‘While’ loop was terminated. More focus in general must be

given to understanding how loops are implemented.

Part (c) required that candidates identify and correct errors in a given algorithm. This question was

generally well done by most candidates. However, greater focus is needed in writing the correct

statements when errors are identified.

In Part (d), the candidates were required to write an algorithm which used iteration to find the sum of

all multiples of 8 and all multiples of 11 between two positive integers. Candidates were required to

sum the multiples of 8 and 11 within a given range. Many candidates were able to write statements to

control the lower and upper limits of the range, as well as to determine the multiples of 8 and 11.

However, more focus must be given to understanding the proper use of ‘DIV’ and division ‘/ ‘. These

statements were used incorrectly on many occasions in an effort to determine if a given value is a

multiple of 8 or 11.

Question 4

Part (a) tested candidates’ ability to trace through the execution of a given algorithm and draw the output

exactly as it would be generated by the algorithm. In particular, it tested candidates' ability to

differentiate between functions which continue output on the current line and those which begin output

on a new line. A key skill tested was the ability to traverse a loop accurately and determine the exit

values of loop variables.

 This question was attempted by most candidates, however, many candidates failed to attain

more than 50 per cent of the marks.

 Candidates failed to interpret the function of the output functions accurately even though they

were clearly defined in the question paper.

 Many candidates iterated the loops too many times producing too many lines of output for

each loop.

 Many candidates misinterpreted the end of the ‘for’ statement and included the next output

statement as a loop statement even though indentation was used.

- Inaccuracies produced were clearly linked to candidates’ inability to calculate loop

variables correctly.

Part (b) required candidates to construct a flowchart to represent a given algorithm. It tested candidates'

ability to represent different control structures using flowcharts. Many candidates answered this

question well gaining full marks. However, some candidates failed to demonstrate credible flowcharting

skills. For example they:

 failed to accurately and consistently use the basic flowchart symbols

 included the use of the preparation symbol for initialization of variables and the display

symbol for output

 used a circle as a start/stop or termination symbol

 neglected to use connection symbols completely

 often used the decision symbol without either the two flow lines proceeding from the symbol

or did not label the flow lines y/n or t/f

 used flow lines without no arrows.

 incorrectly placed flow lines to show looping which also did not flow back to the decision in

the structure.

Section C – Programming

Question 5

Part (a) tested candidates’ knowledge of the lexical analysis process. Part (a)(i) tested knowledge of the

purpose of the lexical analysis stage and Part (a)(ii) asked candidates to comment on the existence of

error detection during the lexical analysis phase.

Part (a) most candidates did fairly well. Candidates were required to describe the process the lexical

analysis goes through to convert a program from sources code to machine language. Most candidates

failed to mention that during the lexical analysis process it converts a sequence of characters into a

sequence of tokens which is read from left to right.

Part (b) was done well by most candidates. Those candidates who scored full marks were able to state

what modulariation is and one of its benefits.

Part (c) was attempted by most candidates and was generally well done. Most candidates knew how to

declare a file pointer, open a file in read mode and read from a file. Candidates who did well on this

question were able to evaluate the grades using the appropriate selection control structure while

accumulating each category.

Most candidates lost marks when they did not use an accumulator to track the number of grades for

each category. Candidates also lost marks when they utilized a sentinel value instead of an end of file

character in their looping condition.

Part (d) was poorly done. Most candidates who attempted the question used a string function to

determine whether the string entered was a palindrome or not. Candidates lost marks when they made

reference a single character instead of a ‘string’ inside their string function.

Candidates lost mark when they requested the length of the string entered instead of determining the

length. Also, some candidates found it difficult to traverse the string to determine whether it is a

palindrome.

Question 6

In general the question was poorly done with approximately 80% of the candidates scoring in the lowest

range, that is, between 0 and 5 marks.

It was noted that while many candidates had a general sense of the logic, they were challenged with

regard to the syntax of the language.

Part (a) required candidates to describe three types of programming languages. While most candidates

attempted this section, most appear to be unclear in their understanding particularly of the scripting

language. Most definitions were only partly correct.

Part (b) required candidates to trace through a program and record the output that would be produced.

A large number of candidates were able to produce some part or all of the output. However, a significant

number either could not do so or recorded the data in a trace table but did not display the results in the

order they would appear on the screen.

Part (c) required candidates to write segments of code for an application that would allow a vendor to

store data about some fruits that he sells. Each fruit was given an ID, cost price, selling price and

quantity in stock. In writing the code candidates were expected to declare and use a C struct for the

storage of data.

In Part (c) (i), while about 60 per cent of the candidates were able to declare the struct correctly many

included items in the declaration that were not necessary or incorrect.

In Part (c) (ii) candidates had a fair understanding of how to declare the struct variables but often did

not note the difference in syntax that was necessary for this section if they had used “typedef” in Part

For Part (c) (iii), candidates were required to assign values to the previously declared variable. Some

did it through assignment while others prompted the user to enter from the keyboard as the question did

not explicitly say how it should be done. Again, candidates had the logic in place but were challenged

by the syntax of the language.

In Part (c) (iv), candidate were required to write code to exchange values that were previously loaded

into the struct “mango” and “pine”. It was very clear that candidates could do the exchange if given an

algorithm but many had difficulty doing so in the format that governs the C structure. This difficulty

indicates a lack in their understanding of the language.

Finally in Part (c) (v), candidates were required to calculate and print the total profit to be made from

the sales of the two fruits with preloaded data. Although candidates were quite creative in their

solutions, many lost marks because their solution did not conform to C or were not logically sound.

In conclusion, candidates need to become more familiar with the language and in particular the use of

structs for storing data.

Paper 03 – School-Based Assessment (SBA)

UNIT 1

General Requirements

Students were expected to choose a problem for which a software solution was appropriate; develop

and present the solution in a logical way using correct grammar and appropriate jargon at all times by

doing the following:

 Creating an algorithm for the solution using modules, sequencing, selection, assignments, and

iteration (bounded and unbounded).

 Representing their algorithms using narrative format and also as either a flowchart or pseudo-

code.

 Implementing the algorithm in C using various data structures such as arrays, struct, strings and

files, with no less than five functions as independent units.

 Ensuring that the source code produced matched the algorithm.

 Createinga test plan with exhaustive data set, testing and producing test results with normal

inputs giving correct results, extreme input giving correct results and erroneous input giving

appropriate error message.

The following specific aspects of the project were assessed:

(a) Definition of problem

(b) Narrative and flowcharts or pseudocode

(d) Testing and presentation

(e) Communication of information

For each component, the aim was to find the level of achievement reached by the student. It is

recommended that the assessment criteria be available to student at all times.

DETAILED COMMENTS

Definition of Problem

Many students were able to accurately describe the problem complete with the description of the current

system and giving background to the problem, the issue or problem, recommend solutions and examples

of what takes place when the problems arise. However some students projects only included partial

definitions of activities of the proposed system that they intended to create without complete

descriptions of the current system and the problems which are to be solved using software. In addition,

some problems that were identified would not be solved by using a computer program. For example,

filling out of a form is a time consuming and error prone process for both a paper and computer-based

form, however, there are advantage of using computer-based forms that were not identified.

Narrative and Flowcharts or Pseudocode

Narratives were fairly well written in most samples. However some samples:

 Provide only a partially correct description of the algorithm as a solution

 Provide description of an intended module or unit instead of describing the step-by-step

procedure that would solve the problem as required by an algorithm

Pseudocode algorithms were generally well done. Samples which included flowcharts in some

instances incorrectly used some symbols. For example, system flowchart symbols were used and in

others cases, diagrams were poorly presented, had symbols with no labels such as (yes/no for the

decision symbol), no connector symbols to connect diagrams to the next page and, therefore, the charts

were difficult to follow. Most students produced samples using pseudocode algorithms, some samples

were clearly modified copies of the programming source code, with a few almost identical copied of

the source code. It is therefore recommended that in the future, students be encouraged to develop their

algorithms independent of the programming code.

Coding of Program

Students’ projects demonstrate that they were comfortable with procedural C programming language.

This was quite evident by the response of the various projects. Most programs were logically written

and properly decomposed. Nevertheless, some candidates sample did not

 comprise functions as independent units and some teachers still awarded full marks

 provide sufficient data structure in the program

 demonstrate appropriate use of the concept of structured programming using evidence of

sequence and selection but no loops

 use appropriate data structure (struct, files and arrays). Only a few samples made use of all or

most of the key data structures.

Some students did not print their project from the C compiler, instead choosing to make print screen

copies of their projects. Students are therefore reminded to print their source code directly from the C

compiler as transferring the code to a word processor often changes the code. For example, it can adjust

the spacing and strange symbols to occur in the code, making it difficult to read. This may also cause

students losing marks for appropriate programming style and documentation. In addition, marks were

lost because some students did not include adequate comments at key areas of their source code and

used a poor indentation style that did not make the code easier to read. However this section was done

well by most students.

Evidence that Code Matches Algorithm:

Most students were able to obtained full marks on this section for this was well done.

Evidence of File Manipulation:

Students were required to provide code that included a file and show evidence of file manipulation, that

is, open file, write to file, read from file, append file and close file. Most students demonstrated evidence

of this, however some students were awarded marks by teachers even though there was no evidence of

file manipulation and in some cases the use of file was none existent or was declared but not used,

non-existent but teachers awarded full marks. Projects that showed screenshots of the file data being

added, edited and deleted were awarded full marks.

Testing and Presentation:

Students were required to prepare a test plan with an exhaustive data set (test data) to test their program.

They were also expected to use the test data set to produce test results with normal input giving correct

results, extreme input giving correct results and erroneous (abnormal) input giving appropriate error

messages. The majority of samples provided this year did not have a suitable range of test data, but a

few candidates:

 Had a test plan and screenshots while others provide screenshots without any test plan

 Did not include all testing criteria (normal, extreme and erroneous) in their test plan

 Had test results but did not have a clear test plan

 Had no test results but candidate were awarded marks by the teacher

 Had test results which did not include actual screenshots of the working program and testing

was not done using the test data outlined in the test plan

 Were awarded marks for a test plan, when they had no plan

 Therefore teachers may have not understood the marking criteria or did not understand what a

test plan should be

 Only tested their menus with the appropriate test data, but all input must be tested in the same

way

Generally, this year’s IA projects were well presented. Twenty-five per cent of the candidates gained

marks between 51 to 60, twenty eight per cent had marks between 41 to 50, twenty two per cent had

marks between 31 to 40 and only twenty five made less than 31.That is about 75 per cent of the

candidates made a mark of over 30, this demonstrates that most IA were satisfactorily done.

Recommendation

Teachers need to ensure that the students use the headings outlined in the criteria for marking the SBA

Project in the syllabus, as well as follow the order in which these headings occur on Form CSCI 1−3.

They should also ensure that students, check that the numbering in their table of contents numbering

correspond to the numbering in the document and candidates provide information in a logical way using

correct grammar and appropriate jargon at all times in the presentation of their projects.

UNIT 2 – FURTHER TOPICS IN COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of 64.4 out of 90, standard

deviation of 12.3 and scores ranging from 21 to 88.

Paper 02 – Essay Questions

Section A Data Structures

Question 1

Part (a) required candidates to describe the concept of ADTs (Abstract Data Types). Most candidates

interpreted the ‘D’ to mean dynamic and spoke about changes in the size of the stack or queue during

program execution. Some candidates were able to give basic explanations but could not expand further.

Part (b) required candidates to describe the differences between stacks and queues and give real world

examples. This question was very well done and most candidates were able to identify and describe the

differences; however, a few candidates struggled to give correct real world examples of stacks. Most

candidates got full marks for this question.

Part (c) (i) required candidates to write C code for a stack’s PUSH function. The logic suggest that the

function must first check that the stock is full then increment the top position before storing the value.

These steps were often confused. Candidates struggled to write correct function prototypes (arguments

and return type) and write valid C code. Some candidates assumed the existence of a separate function

(isFull) when the question specified not to use predefined functions.

Part (c) (ii) required candidates to write C code for a queue’s ENQUEUE function. Most candidates did

not consider a circular queue. The logic suggest that the function must first check that the queue is full

then increment the tail position before storing the value. These steps were often confused. Candidates

struggled to write correct function prototypes (arguments and return type) and valid C code. Some

candidates assumed the existence of a separate function (isFull) which the question specified not to do.

Some candidates gave the same function as Part (c) (i) which added to the top and not the rear/tail.

Part (d) required candidates to demonstrate how a stack’s POP operation works on a stack containing

four letters. The majority of candidates was able to answer this question completely. Some candidates

began popping (removing letters) the stack from the incorrect end. Some candidates did not show the

final contents of the stack after all the POP operations were completed.

Question 2

Part (a) tested candidates’ ability to sort a given array using the selection sort algorithm, but only

displaying the first three passes. Many candidates did not perform the swap correctly by displaying two

integers exchanging places, instead many showed one move of an integer which did not earn them the

full two marks. Some candidates did not realize that on the 3 pass, the position of the integers remained

in place and no swapping took place. Some candidates misunderstood what was required from the

question and chose instead to draw 3 instances of the array on the 4, 5 and 6 pass, thus earning

themselves no marks. Some candidates did not comprehend the selection sort algorithm at all, gave

erroneous responses and so scored no marks. About 6 per cent of the candidates earned full marks on

this part.

Part (b) (i) tested candidates’ ability to draw a linked list containing three nodes. In Part (b) (i), the

majority of responses earned the candidates one or more marks, but many did not correctly indicate

nodes as having two parts: data and pointer − and that between two nodes an arrow (not just a line)

pointed to the other node. Some candidates’ indicated two arrows flowing in opposite directions

between the nodes (a doubly linked list), which was acceptable and earned them the mark. The head is

a pointer (and not a node) and was not indicated as such in many candidates’ responses. Some

candidates incorrectly indicated the end node (node containing integer 5) as pointing to a null node.

Acceptable responses indicating the end node were the word NULL, the earth or ground symbol, an X,

0 or Ø.

In Part (b) (ii) candidates were asked to explain how the first integer from the linked list could be

deleted. This was not well done; in that about 90 per cent of candidates’ responses did not earn them

full marks. Candidates needed to search for the node (node 3), set a temporary pointer to point to the

head node (node 3), set the original head to point to 4, then free the memory allocated to 3.

Part (c) had two parts and required that candidates write code to obtain a value (target) entered by the

user, and to perform a search, given a sorted array (numbers). Most candidates scored full marks for

writing their code, correctly using scanf, and &target. In Part (c) (ii), however, about 10% of the

responses earned full marks, many candidates did not understanding the binary search concept.

Teachers should be reminded that C is the required language for CAPE Computer Science, and that

other languages including pseudocode are unacceptable. Candidates should also note that where the

question posed refers to a specific variable, the use of these variables is mandatory in the required

solution.

Section B  Software Engineering

Question 3

For Part (a), candidates were required to explain the rapid prototyping approach (RAD to systems

development. The majority of candidates was able to describe RAD as a model/simulation of a system.

Candidates lost marks for not specifying that it was a quick fabrication of what the real system might

be like.

In Part (b), candidates were required to describe the main challenge which RAD seeks to overcome.

While some candidates were able to identify that the RAD approach helped clarify customer

requirements, other candidates responded that the process was time consuming and a waste of resources.

For Part (c), candidates were required to give FOUR reasons why it is believed that the SDLC may be

initiated and completed during the maintenance. Some candidates were able to correctly identify the

four major points that is, analysis, design, coding and testing which are conducted during the

maintenance phase of SDLC. However candidates’ responses lacked proper explanation of how each

point could be integrated in the maintenance cycle. For example, when maintenance is conducted the

analysis phase is initiated as new requirements must be gathered; feasibility tests and scheduling and so

on must be carried out.

For Part (d), candidates were required to draw a Level−0 DFD to describe the flow of data in a given

scenario. The Level−0 diagram was very poorly drawn as the rules for constructing these diagrams

were often ignored. Many candidates:

 Confused Level−0 with Context diagrams

 Used inconsistent and incorrect symbols to represent sources, data stores and processes

 Used verbs for data flows

 Connected sources directly to each other

 Drew processes with data in-flows only

Question 4

Part (a) tested candidates’ knowledge of the significance of data flow diagrams (DFDs) in the Analysis

and design phases of the SDLC. Most candidates were unable to distinguish between the purposes of

the DFDs in these phases, incorrectly stating that DFDs in the analysis phase depict how the system

‘should’ operate. However, some candidates were able to explain that DFDs can be used in comparative

analysis which can in turn lead to business process re-engineering.

Part (b) required that candidates state three benefits of the object−oriented approach to software design

such as reusability, maintainability and scalability. Many candidates did not answer this question well;

most candidates provided answers that were too vague, for example, objects are easier to use and code.

In Part (c) candidates were asked to describe two (2) of the major deliverables from the design phase of

the SDLC. Many candidates either omitted this part of the question or incorrectly identified various

components, tools and techniques (including HIPO charts, DFDs, algorithms, interface design and

prototypes) as major deliverables. Many responses included the feasibility study or report which was

incorrect, as this is done in the analysis phase. Few candidates were able to correctly identify and

describe the system architecture and design specification as major deliverables resulting from the design

phase.

Part (d) required that candidates draw an ERD based on a given scenario. Candidates were able to gain

most of their marks for this subpart as opposed to Parts (a), (b) or (c). Though most candidates showed

major improvements when compared with last year’s ERD diagram question, additional emphasis

should be placed on the consistency of the symbology. Common errors included the following:

 Pluralization of individual entity names: EMPLOYEE not EMPLOYEES.

 Vague relationship descriptions: has, is

Section C  Operating Systems and Computer Networks

Question 5

This question tested candidates’ knowledge of types of networks, network architectures, the OSI model,

routers and GPRS.

Part (a) required that candidates describe two advantages of a client/server network over a peer to peer

network. This question was poorly answered as many candidates were vague in their responses.

Candidates confused the client/server network and the peer-to-peer network with network topologies.

They also failed to indicate that the centralized nature of the client/server network allowed for easier

management. While many candidates mentioned security as an advantage, few included that special

software would be needed for better security control.

In Part (b) (i) candidates were asked to define the term firewall. Most candidates acknowledged that it

was a hardware/software security measure but were unable to specify that it protects against external

network threats.

In Part (b) (ii) candidates were asked to explain why firewall software needs to be updated frequently.

This question was well answered with the majority of candidates correctly identifying that since new

malware, viruses and threats are created daily via the Internet, firewall software must be updated.

Part (c) required that candidates state two differences between FDDI and Ethernet network

architectures. Many opted not to answer this question and those who did failed to describe the

architecture of FDDI and Ethernet, instead they described the speeds and physical cabling. Few

candidates gave the correct answer in their response, that is, FDDI has a dual ring layout with the second

ring acting as a back-up while Ethernet does not have a back-up configuration like the secondary ring

of FDDI.

Part (d) dealt with the role of the network and transport layer of the OSI model. Some candidates failed

to differentiate between the layers and did not include the role of the network layer in path determination

and IP addressing. They also failed to discuss the transport layer’s role in breaking up a file into parts

and the reliability of packages in transport. Few candidates specified the particular protocol associated

with each layer.

Part (e) was well answered by most candidates. This question required that candidates outline three

benefits of wireless routers over wired routers. Few candidates confused the mobility of the devices on

the network with the mobility of the router. However, most were able to correctly identify the portability

of the devices, the inconvenience of installation, the compatibility issues with newer devices and the

susceptibility of cables to damage as valid answers.

Part (f) (i) dealt with the purpose of GPRS. Most candidates were able to identify that GPRS was used

in the transmission of data but failed to elaborate that it is a wireless technology used for data

transmission through a cellular network. A few candidates mistook GPRS for GPS.

In Part (f) (ii) required candidates to name and describe two applications of GPRS. Most candidates

were able to gain marks by correctly identifying applications such as point-of-sale terminals, MMS,

tracking and navigation devices, web browsers and wireless electricity meters as solutions. However,

they failed to describe fully how they operate in order to gain full marks. For example, POS devices are

portable devices which allow customers to make payments without the use of a phone line. This device

uses a spectrum that one of the carriers has licensed and a fee has to be paid.

Question 6

This question tested candidates’ knowledge of device management, file security and user interfaces. In

addition it tested their knowledge of:

 Process management and in particular components in a process control block (PCB)

 Scheduling algorithms with emphasis on round-robin scheduling algorithms

Part (a) required that candidates explain the role of a device driver in an operating system. Most

candidates recognized that a driver was software that allowed the operating system to communicate

with a device but failed to note that drivers provided the operating system with instructions on how to

control the device.

Part (b) tested candidates’ knowledge of spooling. This question was generally well done as many

candidates noted this was where documents where loaded into temporary storage location pending an

I/O device availability.

Part (c) was also generally well done as many candidates were able to outline TWO advantages of a

menu interface over a command interface. However, candidates frequently made mention of ‘lines of

code’ or a ‘machine language’ instead of commands when referencing typing instructions in a command

interface.

Part (d) asked candidates to describe TWO components of a process control block (PCB) in an operating

system. Many of them supplied suitable components that is, process ID, process state, and registers.

However, many candidates correctly identified the program counter by itself as a component.

In Part (e) candidates correctly mentioned encryption, access control lists and firewalls as suitable ways

in which files can be protected from unauthorized access. However, many failed to explain how a

firewall prevents access to files.

Candidates also made mentioned of activity logs and file compression, methods which by themselves

would not protect files from unauthorised access to a file.

Part (f) was generally well done, with the majority of candidates recognizing that each process is given

a certain amount of CPU time (time slice or quantum). They were also able to provide adequate details

of how the scheduling algorithm functioned.

In Part (g), candidates were asked to explain how an interrupt is handled by the process in the computer.

Many candidates provided generally suitable answers. However, quite a few of them instead described

what an interrupt is or identified instances that would cause an interrupt in the system as opposed to

focusing on what happens in the processor when an interrupt occurs. Namely, a context switch occurs

or the running process is saved in registers for later resumption, after the processing of the interrupt.

Paper 03 – School-Based Assessment (SBA)

UNIT 2

Each student was expected to choose a problem for which a software solution exists and then develop

the software using software engineering techniques. In particular, students were expected to

demonstrate appropriate choice of the tools and techniques used in the analysis of the software to be

developed. They were then expected to design, code, and test their software using appropriate

techniques.

Definition of Problem:

Students were required to give a complete and accurate description of the problem. Most students

handled this section fairly well but there are a few students who are still unclear of what to include in

their definition of the problem. A brief description of the context in which the problem occurred is

required but details about actual problems staff/clients face and proposed steps to correct such problems

should be emphasized.

Techniques of Analysis:

Students were required to identify techniques of data collection and analyse the data collected.

Most students were able to state the various techniques of data collection; however, some students were

not able to justify the reasoning for their selection. Several students misinterpreted justification to mean

definition. This is incorrect. Students should clearly explain why the chosen technique was used as it

relates to the business/company/institution and not regurgitate advantages of the technique.

Proof of analysis should also be given. All proof must be included directly after the analysis and not in

the appendix. For example sample questions from the questionnaires and/or interviews should be

included. It is understood that it may be difficult to include proof of observation; hence, if this technique

is used, students should clearly describe what was observed.

Use of Data Flow Diagrams and ER Diagrams

This section of the SBA has improved significantly. Students were able to properly draw relevant

diagrams that correctly mapped to their problem statement.

Context Level Diagram

Students were required to give a complete and accurate diagram of all relevant entities and data flows.

The weaker students generally failed to label their data flows.

Level-1 Diagram

Students were required to give a complete and accurate diagram with all relevant processes, data flows

and major data stores.

Most students are unaware that the level 1 diagram is an expansion of the context level diagram. Hence,

new external entities were created for the level 1 diagram, as well as new data flows were created for

previously used entities.

Students had links between (1) data stores and entities, (2) data stores and data stores, (3) entities and

entities, which are all incorrect.

Some students’ diagrams were too small, which made reading and marking difficult.

ER Diagram

Students were required to give a complete and accurate diagram of all relevant entities and relationships.

Most students did not use the correct symbol for a relationship (a diamond). Some students did not

include attributes for the entities.

Functional and Non-Functional Requirements:

Students were required to give a complete and accurate description of all requirements. For the

functional requirements, students did not clearly state what the system is supposed to do but instead

stated what the user will be doing. An example of a good functional requirement is the system will be

able to delete a patient’s record. For the non-functional requirements students did not state the

limitations of the system. An example of a good non-functional requirement is the system can only store

1000 patient records.

Design Specification

Students were required to give a complete and accurate system structuring diagram containing all

processes and a description of the user interface, report design, algorithm design and appropriate data

structures. Most students handled this section fairly well.

Most students were unaware of how to correctly draw a system structuring diagram. Students should

note that this diagram is similar to the HIPO chart as it shows the breakdown of the system into its

submodules and the submodules into its further components.

Students should not just include screenshots of interface and report design but also a justification in

order to gain maximum marks.

Narratives will not be accepted as an algorithm. Students are reminded to submit pseudocodes.

Coding and Testing

Students were required to produce a complete and accurate C program solution for the problem stated

in the Definition of Problem stage. Most students produced programs that achieved good functionality.

Only procedural C is accepted as the programming code. No other programming language will be

accepted. Code must also be printed from the compiler, not transferred to a word processor before

printing. Students should not include soft copies of their SBA. Soft copies will not be marked.

Most students did not include enough screenshots (and in some cases no screenshots) to support their

functioning of the code. Some codes presented did not match the screenshots given. In such cases

students were not awarded any marks.

Test plans should be written in a tabular format. It should include normal, extreme, erroneous and

incompatible data. All input data must be tested. Test results without related test plans will not be

awarded any marks.

Communication and Presentation

This section of the SBA is often over looked by students. Lectures/teachers and students are urged to

pay close attention to the use of grammar and the overall presentation of the SBA.

Further Comments

i. Students should follow the order and the headings laid out in the criteria for marking

when compiling the SBA.

ii. Students should avoid using watermarks when printing the SBA as it makes it difficult

for the examiner to read and assess their work.

iii. Lecturers/teachers should avoid using red ink pens to correct the SBA.

iv. Lectures/teachers should ensure that each SBA is clearly labelled with the candidate’s

name and centre number.

Recommendations

Teachers should ensure that students are fully prepared for the examinations in both units. The poor

performance in some modules of the syllabus indicates that more time needs to be these areas.

C A R I B B E A N E X A M I N A T I O N S C O U N C I L

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

MAY/JUNE 2016

COMPUTER SCIENCE

St Michael, Barbados

GENERAL COMMENTS

This was the eighth year in which the revised syllabus was examined. There were three examination

papers each in Units 1 and 2; namely, Paper 01, Paper 02 and Paper 03. In each unit, Paper 01 and Paper

02 were examined externally by the Caribbean Examinations Council (CXC) while Paper 03, the School-

Based Assessment, was examined by teachers and moderated by CXC.

In each unit, Paper 01 consisted of 45 multiple-choice questions that were designed to test candidates’

breadth of coverage of the syllabus. Paper 02 consisted of six compulsory essay-type questions that were

designed to test candidates’ depth of understanding of the syllabus. Thus, candidates were expected to

show deeper insight and understanding of the topics examined in Paper 02.

The individual contributions of Paper 01, Paper 02, and Paper 03 to the final grade are 30 per cent, 50 per

cent, and 20 per cent, respectively.

Approximately 91 per cent of the candidates obtained Grades I─V in Unit 1 and approximately 95 per cent

of the candidates in Unit 2 obtained Grades I─V. Overall, there is still need for improvement in the quality

of responses for the programming questions in both units.

DETAILED COMMENTS

UNIT 1 – FUNDAMENTALS OF COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of approximately 64.26 out

of 90, standard deviation of 13.57 and scores ranging from 18 to 90.

Paper 02 – Essay Questions

Section A – Computer Architecture and Organization

Question 1

Part (a) required candidates to write down the truth table for a given combination of switches. Most

candidates attempted this part of the question but many did not interpret the switches correctly and

often noted the wrong combination of bit patterns. In many cases the columns representing the inputs

were not correctly labelled.

Part (b) required candidates to write down the truth table for a given proposition. Most candidates

attempted this part of the question but it was clear that many confused the operand ‘AND’ with ‘OR’ or

did not understand the proposition at all.

Part (c) required candidates to define the term ‘program counter’. Most candidates attempted this part

of the question but were not always clear in stating that a program counter is a register that stores the

address of the next instruction in memory. Some candidates confused this term with a loop counter and

in these cases marks were not awarded.

Part (d) required candidates to add two 8-bit binary numbers stored in Registers 1 and 2 respectively.

Most candidates were correct in adding the two numbers together but marks were lost in explaining what

would be stored in Register 3. While the calculation generated a 9-bit binary number, Register 3 would

store 8-bits only. As a consequence, an overflow error would occur and the overflow bit on the left would

be truncated. Alternatively, candidates could indicate that nothing would be stored because of the

overflow error.

Part (e) required candidates to determine the binary equivalent of the decimal number 13.25. Most

candidates attempted this part of the question but lost marks for lack of working despite this being clearly

requested in the question posed. Many candidates successfully converted the whole number part of the

given decimal number but failed to correctly convert the fractional part. In some cases, the same

conversion technique used for the whole number component was used to convert the fractional

component thus yielding the wrong answer.

Part (f) required candidates to define the term ‘multiplexor’ and draw a block diagram to support the

answer. Most candidates attempted this part of the question by defining a multiplexor as a device which

selects one of several input signals and forwards the selected input into a single line. Though the

supporting diagram was accurately presented and many candidates correctly illustrated 1 output channel,

several candidates failed to correctly represent 2

input channels.

Part (g) required candidates to explain why multiplexing was important. The question challenged

candidates to apply their knowledge of multiplexing by providing a real-world example. In general this

question was poorly answered. Central to the solution was the understanding that multiplexing makes

better use of available bandwidth across a communication network. Specific scenarios included the

provisioning of cable TV and telephones as well as CCTV cameras. A further explanation to illustrate the

use of multiple inputs versus a single output was required to support the given scenario.

Question 2

Candidates were presented with a diagram in Part (a) and asked to provide examples of non-volatile and

secondary memory. This question was generally well answered by all candidates, with various versions of

ROM presented as examples of non-volatile memory, and hard disks, floppy disks, USB drives, CDs, DVDs

and SSDs (Solid State Disks) presented as examples of secondary memory.

Part (b) described a scenario in a fictitious community where electricity was available but inconsistent.

The details of a cell phone were provided and candidates were asked to explain the purpose and benefit

of the using each of three devices in the scenario. It was therefore important to link the use of the devices

to the given scenario. Most candidates attempted this question but many were not clear in stating the

benefit of the device in the context of the scenario.

In Part b (i), candidates were expected to note that a surge protector would limit any unwanted voltage

above a safe threshold, thus protecting the cell phone in the case of a power spike. In Part b (ii), candidates

were expected to note that a voltage regulator automatically maintains a constant voltage and is a

proactive device that provides clean electricity regardless of changes to its input voltage. In Part b (iii),

candidates were expected to note that an uninterruptible power supply is a device that allows the cell

phone to keep running for a short time when the primary power source is lost. The device could therefore

be charged by the UPS for a limited time.

Part (c) required candidates to define the term ‘clock speed’ as it relates to the processor and give

examples of CPU clock speeds using different units of measure. In general, most candidates attempted

this question and stated that the clock speed is the speed at which the processor executes each

instruction. Candidates were expected to give two distinct examples, such as 2.5 MHz and 1.8 GHz to

demonstrate their knowledge of the different units. Candidates lost 1 mark where distinct units of

measure were illustrated but no quantifying speed was given.

Part (d) required that candidates apply their knowledge of input/output devices by providing examples of

devices that would be appropriate for a person who is visually impaired. Candidates were expected to

identify the device and describe how it would be useful in this case. Most candidates attempted this

question, with many scoring full marks. The notion of voice to text and text to voice were important to

note as these points indicated a clear knowledge of the requirements of a visually impaired person.

In Part (e), candidates were asked to explain, with the use of examples, the main difference between a

supercomputer and a mainframe computer. Most candidates attempted this question and gave

reasonable examples of the use of these systems though many failed to highlight the main difference

clearly. Candidates were expected to note that the main difference between the two systems was the

type of problems each tries to solve. Candidates were able to achieve only partial credit when the cost

difference between the two types of systems was emphasized. In general, candidates were expected to

note that supercomputers are ideal for complex calculations such as weather forecasting/predictions

while mainframe computers focus on problems which are limited by input/output and demand reliability.

Part (f) asked that candidates use the processor’s instruction set to write a set of instructions to perform

a mathematical calculation. This question was poorly answered as many candidates did not demonstrate

a knowledge of appropriate instructions that could be used. Candidates were expected to use an

appropriate mnemonic to represent instructions which included specific operands (registers) as is

expected for calculations. For example,

LOAD R1, R2

MULTIPLY R1, R2

STORE R1

Here LOAD, MULTIPLY and STORE are the instructions used while R1 and R2 refer to registers.

Section B – Problem Solving with Computers

Question 3

The scores on this particular question were not as evenly distributed as those recorded for previous

questions. That is, though a reasonable number of candidates scored between 15 and 25 inclusive, almost

10 per cent of the total number of candidates scored 0, while many did not attempt this question at all.

Part (a) required candidates to write an algorithm to read a given set of data related to library fines. The

data comprised a customer ID, number of books outstanding and a category corresponding to the rate at

which each fine was charged. Results required included the total outstanding fine for each customer, total

amount of fines paid by category, total amount of fines, number of customers and customer id for the

highest fine paid. Many candidates recognized the need for an unbounded looping structure in this case

but at times confused the loop condition that should be used (customer ID < > 9999). Many candidates

did not include the input statement before the start of the looping structure and often omitted the input

at the end of the loop. IF statements within the looping structure were generally well used, as were

cumulative variables. Many candidates did not recognize, however, that outputting the ID of the customer

with the highest fine would require both the customer ID and customer fine to be stored in order for the

correct comparison to be made.

Part (b) asked that candidates distinguish between ‘bounded’ and ‘unbounded iteration’. Many

candidates noted examples only and this was not sufficient to answer the question. In distinguishing

between the two terms, it was important for candidates to note that though both terms referred to a

looping structure with bounded iteration, the number of passes through a loop is known beforehand while

with unbounded iteration the number of passes through the loop is not known beforehand.

Part (c) required that candidates write an algorithm to find the sum of all multiples of 7 between 21 and

210 inclusive. Many candidates recognized the need for a looping structure in this case and, though not

required, demonstrated their knowledge of the modulus function and this was commendable. Several

candidates did not initialise the counter variables used and therefore failed to earn the maximum marks

on this part of the question. There were a few candidates who did not understand the caveat to include

21 and 210 in the summation and thus provided an incorrect loop condition.

Question 4

Part (a) presented an algorithm and required candidates to construct a flowchart to represent it.

Candidates were assessed on appropriate signals, as well as the ability to accurately represent the logic.

A few candidates satisfied all the assessment requirements and achieved the maximum score available

for this part. Often, however, candidates made a few common mistakes, such as:

 Failing to recognize that two separate conditions are represented in the algorithm

 Failing to place correct labelling on decision outputs.

In addition, a few candidates had no arrows on flow lines and many candidates did not use arrows to

properly represent the loop structure.

Part (b) required candidates to write an algorithm to perform a calculation that finds the cost based on

an input variable and a condition that specifies whether one formula or another will be applied. Some of

the errors observed were:

 Many candidates achieved full marks on this question.

 A high number of candidates represented one or more formulae inaccurately.

 A few candidates did not print the output.

 A few candidates did not recognize that a condition structure was to be applied.

Part (c) required that candidates describe four steps in the software development process. Many

candidates were able to state four steps, but only a few followed through properly with a description of

all the mentioned steps. Some candidates stated the steps, but did not provide a description.

Section C – Programming

Question 5

Part (a) described the requirements for a system that generates a random number and then allows a user

to repeatedly guess the number until the guess is correct.

A few candidates scored fairly well. A high number of candidates, however, provided answers that

contained minor syntax errors. Many candidates described syntactically incorrect loop conditions. Some

of the candidates did not provide appropriate output and omitted the loop structure. A small number of

candidates generated the random number in the wrong part of the program (for example, in the loop).

Part (b) (i) asked candidates to explain what a procedural language is. A high number of candidates

correctly indicated step-by-step operation of a computer program; however, some candidates did not give

a sufficient explanation.

Part (b) (ii) required candidates to describe object oriented programming. Some candidates scored full

marks, but an inordinately high number of candidates did not give a satisfactory answer.

Part (c) required that candidates write a program that opened a formatted file, read the contents, and

printed some summary information. Few candidates scored full marks on this question. Some candidates

did not properly open the file. Some candidates did not initialize variables properly. Many candidates used

an improper loop condition (for example, using the equality test to compare the contents of a string).

Many candidates did not read the data properly from the file. Many candidates did not properly test for

the exit condition for the program. A few candidates did not properly aggregate the summary data, and a

few candidates did not print the output properly.

Question 6

Part (a) of the question required candidates to state three differences between applications for mobile

devices and those developed for desktop computers. A large number of candidates were able to properly

describe two differences, the most common ones being the design for smaller screens, the design with

limited memory, and the differences in input devices. A few candidates misinterpreted the question and

provided solutions that described differences in the devices, rather than applications for the devices.

Part (b) required candidates to write down the output of a program that was presented. Few candidates

provided a complete answer to this question. Most candidates, however, were able to produce some

correct values generated on the first line.

Part (c) required candidates to write code that defined and manipulated a structure that stored

information on books.

Part (c) (i) required candidates to declare the structure, and a large number of candidates declared the

structure properly. Some candidates used the typedef command correctly, which affects where the

structure name “BookRec” is expected, and also affects manipulation in other sections. Some candidates

who did not use the typedef command did not name the structure before specifying contents.

Part (c) (ii) required candidates to populate the structure. A large number of candidates completed this

task properly, but some candidates did not reference the data in the structures correctly.

Part (c) (iii) required candidates to increase a numeric data item by a percentage. Many candidates

answered this question properly. Some candidates did not reference the data item properly. Some

candidates also did not use the correct formula when increasing the percentage.

Part (c) (iv) expected candidates to write a code that would swap the values in two records. Candidates

who completed the task properly would have taken on two primary approaches, namely, using an auxiliary

record and swapping the entire record, or using auxiliary data items to swap individual items within the

record. Some candidates presented incorrect swap logic. Some candidates did not reference data

correctly. Some candidates did not declare auxiliary data structures or variables properly.

Paper 03 – School-Based Assessment (SBA)

General Requirements

Candidates were expected to choose a problem for which a software solution was appropriate and

develop and present the solution in a logical way, using correct grammar and appropriate jargon at all

times. In accomplishing this they were required to do the following:

 Create an algorithm for the solution, using modules, sequencing, selection, assignments, and

iteration (bounded and unbounded).

 Represent their algorithms, using narrative format and also as either a flow chart or pseudo-code.

 Implement the algorithm in C, using various data structures such as arrays, struct, strings and files,

with no less than five functions as independent units.

 Ensure that the source code produced matched the algorithm.

 Create a test plan with exhaustive data set; test and produce test results with normal inputs giving

correct results, extreme input giving correct results, and erroneous input giving appropriate error

message.

The following were specific aspects of the project that were assessed:

(a) Definition of problem

(b) Narrative and flow charts or pseudocode

(d) Testing and presentation

(e) Communication of information.

For each component, the aim was to find the level of achievement reached by the candidates. It was

recommended that the assessment criteria be available to candidates at all times.

DETAILED COMMENTS

Definition of Problem

Many candidates were able to accurately describe the problem complete with the description of the

current system, and give examples of what takes place when the problems arise. However, some

candidates’ projects only included partial definition of activities of the proposed system that they

intended to create without completely describing the current system and the problems or issues which

are to be solved using software.

Narrative and Flow Charts or Pseudocode

Narratives were fairly well written in most samples. However, some samples did not provide a completely

accurate description of the algorithm. Proper algorithm structure was not used. For example, many

algorithms submitted had no name. Some only provided descriptions of each intended module or unit

instead of describing the step-by-step procedure that would solve the problem as required by an

algorithm.

Pseudocode algorithms were generally well done. Samples which included flowcharts in some instances

incorrectly used some symbols, for example, system flowchart symbols, and in other cases diagrams were

poorly presented, had symbols with no labels (such as yes/no for the decision symbol) or no connector

symbols to connect diagrams to the next page and therefore were difficult to follow.

In the many samples using pseudo-code algorithms, some were clearly modified copies of the

programming source code, with a few being almost identical copies of the source code. Candidates should

be encouraged to develop their algorithms independent of the programming code, using proper algorithm

structure and design.

Coding of Program

Candidates’ projects demonstrated that they were comfortable with procedural C programming language.

Most programs were logically written and properly decomposed. Nevertheless, some candidates’

samples did not comprise functions as independent units and, despite this, were still inappropriately

awarded full marks by the teacher, while some projects had few data structures in the program. Further,

some projects showed evidence of sequence and selection only, but no loops for the concept of structured

programming, while a few samples did not make use of the key data structures (struct, files and arrays).

Candidates are reminded to print their source code directly from the compiler as transferring to a word

processor changes the spacing of the code. This may lead to candidates losing marks for appropriate

programming style and documentation.

Generally, most candidates did not include adequate comments at key areas of their code and a few were

not properly indented. However, this section was done well by most candidates.

Evidence that Code Matches Algorithm

This section for this was well done as most candidates were able to obtain full marks.

Evidence of FILE Manipulation

Candidates were required to provide a code that included FILE and showed evidence of FILE manipulation

(open file, write to file, read from file, append file and close file).

Most candidates demonstrated evidence of this. However, some candidates were awarded marks by the

teacher even though there was no evidence of FILE manipulation, and in some cases the use of FILE was

non-existent yet marks were awarded.

Testing and Presentation

This year the majority of samples provided a suitable range of test data, but a few candidates

 did not have a test plan, but had screen shots only

 did not include all four testing criteria (normal, extreme, erroneous and incomplete) in their test

plan

 had test results but did not have a clear test plan

 were awarded marks but had no test results

 did not include actual screen shots of the working program and testing was not done using the

test data outlined in the test plan.

In one example a candidate was awarded marks for test results, when the candidate had not produced

a test plan or any results. Therefore, the teacher might not have understood the marking criteria or

did not understand what a test plan should be like.

Generally, the SBA projects this year were well presented. Most candidates gained marks between 41 and

50, followed by marks between 51 and 60, then marks between 31 and 40 marks and few candidates

gained under 20 marks .This demonstrated that most SBAs were satisfactorily done.

Recommendation

Teachers need to ensure that the candidates use the headings outlined in the criteria for marking the

School-Based Assessment Project in the syllabus, and that they also follow the order in which these

headings occur. They should also ensure that candidates check that the numbering in their Table of

Contents corresponds to the numbering in the body of the document and that candidates provide

information in a logical way, using correct grammar and appropriate jargon at all times in the presentation

of their projects. Teachers also need to use the correct forms recommended by CXC to mark and/or submit

candidates’ marks.

UNIT 2 – FURTHER TOPICS IN COMPUTER SCIENCE

Paper 01 – Multiple Choice

Performance on the 45 multiple-choice items on this paper produced a mean of 64.26 out of 90, a

standard deviation of 13.57 and scores ranging from 22 to 90.

Paper 02 – Essay Questions

Section A  Data Structures

Question 1

Part (a) required candidates to suggest possible data structures that would fit applications that were

described. Most candidates were able to score full marks on this part of the question.

Part (b) expected candidates to insert a string value into a linked list node and insert the node at the

beginning of a linked list. Very few candidates scored full marks on this part, due to failure to place the

string value in a linked list node, and failure to execute the pointer manipulations required to place the

node at the beginning of the list.

Part (c) required candidates to diagrammatically represent a queue as data is inserted into it and deleted

from it. Most candidates answered this part properly. Some candidates, however, confused the operation

of a stack and a queue.

Part (d) expected candidates to describe three operations of a stack. Many candidates scored full marks

on this part of the question; however, an inordinately high number described an operation not in the set

of operations for a stack, such as a display operation that shows all the contents of the stack.

Part (e) required candidates to reverse a queue. Few candidates scored high marks on this part of the

question, and those with better scores used an auxiliary structure or a swap to achieve this objective.

Question 2

Part (a) tested candidates’ ability to perform linked list manipulations that delete data. This question was

generally poorly answered. A large number of candidates did not appear to comprehend the importance

of preserving a pointer location, and then executing pointer manipulations to achieve an objective.

Part (b) (i) tested the ability to declare an array of a specified size, and was fairly well answered. Most

candidates scored full marks on this question, but there was a minority who specified an incorrect array

size or had syntax errors.

Part (b) (ii) required that candidates populate the array with user-entered data. A high number of

candidates scored full marks on this section. Some candidates, however, presented solutions with errors

in the loop condition, or syntax errors while reading data.

Part (b) (iii) required candidates to write a program that uses a sequential search to look within the array

for a user-entered value. Most candidates provided a fairly good answer to this question; however,

common problems included errors in the loop condition, errors in reading the user value, or errors in

output. A small number of candidates provided a solution that was not related to the sequential search.

Section B  Software Engineering

Question 3

For Part (a) candidates were expected to state two factors for risk assessment in software development,

apart from time and costs. Most candidates answered this part of the question correctly, but a few

provided only one appropriate answer.

Part (b) required candidates to describe CASE tools. Few candidates scored full marks on this part. Some

did not mention that the tools were computer based, or that the tools were designed to aid the software

development process, or explain the acronym.

Part (c) asked candidates to describe four weaknesses of the waterfall model. Few candidates scored full

marks on this part. Common mistakes included repeating points in the response, and simply stating a

property of the waterfall model without highlighting why that point could be a weakness.

Part (d) asked candidates to explain two weaknesses of rapid prototyping. Candidates who scored full

marks stated two valid weaknesses with supporting information. Other candidates omitted supporting

information, stated a property of rapid prototyping without stating why it was a weakness, or repeated a

previously stated point.

Part (e) assessed candidates on the ability to draw a Level-0 DFD to describe information flow in a school

registration system. A large number of candidates provided satisfactory responses; however, common

errors included incorrect symbols to represent sources, data stores and processes, use of verbs for data

flows, and nouns for processes, and neglecting the flows between processes as described in the

registration system.

Question 4

This question tested candidates’ knowledge of software development and database design. Part (a)

required that candidates define the term ‘cardinality’ and provide one example. Whereas many

candidates were able to give an appropriate example of cardinalities, that is, one to one, one to many,

and many to many, the large majority of candidates failed to give an appropriate definition. Cardinality is

the number of instances of an entity that may exist between two related entities.

Part (b) required that candidates list three characteristics of a well-engineered software product. This

question was generally well answered with candidates noting that a well-engineered software product is

maintainable, dependable, efficient, usable, portable, scalable and appropriately documented, among

other qualities.

In Part (c), candidates were asked to differentiate between functional and non-functional requirements

and to give one example of each term. The responses to this question were fair; many candidates were

correct in distinguishing that functional requirements describe what the system should do but many

candidates were not clear in stating that non-functional requirements are requirements that judge the

operations of a system or constraints on how the system should work. In many cases, correct examples

were not given to support the given definition. A good example of a functional requirement is that the

system should reject character entries in the data field. A good example of a non-functional requirement

is the response times the system must guarantee.

Part (d) asked that candidates explain two types of tests which could be completed during the testing

phase, other than white box and black box testing. Most candidates attempted this question and correctly

noted various types of tests, namely unit testing, which is used to validate that individual units of code or

modules operate as intended, and integration testing, which combines software modules as a group to

ensure that the program operates as intended. Other types of tests include stress/performance testing,

recovery, usability, alpha testing and beta testing.

In Part (e), candidates were required to draw an entity relationship diagram for a given scenario related

to a university called the Science College. Whereas weaker candidates did not score high marks in Parts

(a) through (d), many were able to gain reasonable marks in this part of the question. In general, it appears

that candidate performance on this topic is showing improvement based on the comments in the report

for 2015. Once again common errors included the following:

 Pluralisation of individual entity names; LECTURER not LECTURERS, COURSE not COURSES

 Inconsistent ERD notation denoting cardinalities; Crow’s Foot or UML accepted.

Section C  Operating Systems and Computer Networks

Question 5

This question tested the candidates’ knowledge of types of networks, network architectures, the OSI

model, routers and GPRS. Part (a) required candidates to describe the characteristics of coaxial cable and

fibre-optic cable transmission media. Physical characteristics were the main focus here. For coaxial cable,

there is a single centre copper wire symmetrically surrounded by a braided or foil conductor while fibre-

optic cables consist of a bundle of extremely thin tubes of glass transmitting optical signals in the form of

light. Though several candidates attempted this question, many opted to compare the given terms as

opposed to what was required.

In Part (b), candidates were asked to explain what an IEEE 802.16 network is. In general, this part of the

question was poorly answered. Many candidates appeared not to understand that this type of network

caters for long-range wireless, broadband networks. Though several candidates noted that this type of

network allowed for wireless communication, this was not sufficient as the notion of long-range wireless,

WAN or MAN was important to differentiate the 802.16 standard from other wireless network standards.

Part (c) required candidates to explain the role of a switch in a local area network and accompany this

explanation with a diagram. Many candidates attempted this question and were able to draw the

supporting diagram correctly. Candidates lost marks for failing to note that the switch acts as a

connectivity device on the network and sends signals from node to node efficiently without broadcasting

messages to every node on the network.

Part (d) required candidates to describe how data is communicated in an IEEE 802.11b network, using a

diagram to support the response. Candidates were required to note that the medium for transmission is

wireless. Nodes must first associate with the access point, then for a node to transmit to another node

the communication must pass through the access point. Though in many cases wireless transmission was

noted, the details regarding the access point were omitted. As far as the diagram was concerned,

candidates were expected to clearly depict the access point along with the wireless node while using the

appropriate symbol to represent wireless communication.

In Part (e), candidates were asked to briefly explain specific factors that should be considered when

building a network. These factors included cost, security, management, expandability and

interconnectivity. Though many candidates attempted this question, several candidates did not clearly

expand on the given factors. For example, where security is concerned, issues of encryption, protection

of data and resources, firewalls, and virus protection are some of the detailed considerations, while in

terms of management, it is important to oversee the network and fix the problem while routinely

maintaining the network. Many candidates did not provide correct responses where expandability and

interconnectivity were concerned. Expandability is concerned with ensuring that the addition of new

nodes is easy while interconnectivity ensures that connecting to other networks and the Internet is

seamless.

Finally, Part (f) dealt with the OSI model for network communication and asked that candidates explain

the role of any two of the first five layers of this model. Correct responses would include the physical, data

link, network, transport and session layer. Though this question was generally well answered, some

candidates only noted the actual layer without a supporting explanation. This indicated that the purpose

of the layer was not clearly understood.

Question 6

This question tested the candidates’ knowledge of operating systems, namely, memory management,

device management, interrupts, types of operating systems, file security and the process control block.

Part (a) required candidates to explain the terms ‘virtual memory’, ‘paging’ and ‘thrashing’. This question

was poorly answered. Candidates were expected to note that virtual memory is disk space used to

maximize the amount of RAM available to programs, while paging is a memory management scheme used

by operating systems where data is copied from secondary storage in blocks for use in main memory. On

the other hand, thrashing occurs when a computer’s virtual memory subsystem is in a constant state of

paging, rapidly exchanging data in memory for data on disk.

Part (b) required candidates to explain how excessive paging affects a computer. Many candidates

attempted this question and were able to state that excessive paging causes degradation in the

computer’s performance. However, many candidates did not note that this degradation could lead to the

collapse of the system.

Part (c) tested candidates’ knowledge of device drivers in an operating system. Though this question was

widely attempted, it was not always clear that a device driver contains a special code which the operating

system uses to connect to the device.

Candidates were provided with a short scenario in Part (d) and asked to explain how the operating system

uses interrupts in the given scenario. Few candidates identified the type of interrupt that occurred as an

i/o interrupt though many were able to describe the subsequent context switch, reading from the file

followed by another interrupt in order to continue the previous process.

In Part (e) candidates were asked to distinguish between a multi-user system and a batch processing

system. This question was generally well answered, with candidates stating that a multi-user system is a

computer with an operating system that supports multiple users at once or at different times. In contrast,

jobs can be stored/accumulated and later executed in batches in a batch processing system.

In Part (f), candidates were asked to outline one way, besides passwords, in which files can be protected.

Several candidates correctly noted encryption or the encoding of data in their solution while other

candidates noted access control lists which specify permissions for users. Though lock words were also

noted, these are recognized as a type of password.

Finally, in Part (g), candidates were asked to identify three components of a process control block. Many

candidates answered this question successfully by noting the process state, process ID, program counter,

CPU registers, memory management information and accounting information as components of a process

control block.

Recommendations

 This subject is now marked using the on-screen marking system and as such when scripts are

scanned, care is taken to scan only those areas provided for writing. Teachers should emphasize

that candidates should write their solutions only in the boxes/lines provided. Where this space

appears insufficient, the extra pages provided should be used. Candidates should avoid writing in

the white spaces surrounding the boxes/lines provided, as there is a risk of truncation during the

scanning exercise.

 Candidates’ ability to express themselves clearly in writing is at times weak. This makes the

process of assessment difficult. Candidates should be mindful that clear and succinct answers are

best, ensuring that the solution matches the question posed.

 Teachers should ensure that candidates are fully prepared for the examinations in both units. The

poor performance in some modules of the syllabus indicates that more time is needed or that

different pedagogical strategies should be utilised when teaching these subject areas.

Paper 03 – School-Based Assessment (SBA)

Each candidate was expected to choose a problem for which a software solution exists and then use

software engineering techniques to develop the software. In particular, the candidate was expected to

demonstrate appropriate choice of the tools and techniques used in the analysis of the software to be

developed. They were then expected to design, code and test their software, using appropriate

techniques.

General Comments

 Candidates should follow the order and the headings laid out in the criteria for marking when

compiling the SBA report.

 Candidates should avoid using water marks when printing the SBA as it makes it difficult for the

examiner to read and assess their work.

 Teachers should avoid using red ink pens to correct SBA.

 Teachers should ensure that each SBA is clearly labelled with the candidate’s name and centre

number.

Marking Criteria

1. Specification of Requirements

- Definition of Problem:

Candidates were required to give a complete and accurate description of the problem. Most candidates

handled this section fairly well but there were a few candidates who were still unclear of what to include

in their definition of the problem. A brief description of the context in which the problem occurred is

required but details about actual problems staff and/or clients face and proposed steps to correct such

problems should be emphasised. Most candidates focused on the proposed system instead of adequately

describing the problem to be solved, while some omitted a proposal altogether.

- Techniques of Analysis:

Candidates were required to identify techniques of data collection and to analyse the data collected.

i. Some candidates were able to state the various techniques of data collection; however, most

candidates did not provide a justification for their selection. Those candidates misinterpreted

justification to mean definition. Marks are not awarded for definitions. Candidates should clearly

explain why the chosen technique was used as it relates to the organisation and should not

regurgitate advantages of the technique in order to gain marks.

ii. Proof of analysis should also be given. All proof must be included directly after the analysis and

not in the appendix. For example, sample questions from the questionnaires and/or interviews

should be included. It is understood that it may be difficult to include proof of observation; hence,

if this technique is used, candidates should clearly describe what was observed.

- Use of Data Flow Diagrams and E-R Diagrams

This section of the School-Based Assessment has improved significantly. Candidates were able to

properly draw relevant diagrams that correctly mapped to their problem statement.

Context Level Diagram

Candidates were required to give a complete and accurate diagram of all relevant entities and data flows.

The weaker candidates generally failed to label their data flows.

Level 1 Diagram

Candidates were required to give a complete and accurate diagram with all relevant processes, data flows

and major data stores.

In this regard, the following points are to be noted:

i. Only one level 1 diagram is required.

ii. Most candidates are unaware that the level 1 diagram is an expansion of the context level

diagram. Hence, new external entities were created for the level 1 diagram, and new data flows

were created for previously used entities.

iii. Candidates had links between data stores and entities; data stores and data stores; entities and

entities, which are all incorrect.

iv. Some candidates’ diagrams were too small, which made reading and marking difficult.

ER Diagram

Candidates were required to give a complete and accurate diagram of all relevant entities and

relationships.

i. Most candidates did use the correct symbol for a relationship (a diamond). Nevertheless, there

were still candidates who did not use a symbol but wrote on the line instead. This is not

acceptable.

ii. Some candidates did not include attributes for the entities.

Functional and Non-Functional Requirements:

Candidates were required to give a complete and accurate description of all requirements.

i. For the functional requirements, candidates did not clearly state what the system is supposed to

do. Instead they stated what the user would be doing. An example of a good functional

requirement is the system will be able to delete a patient’s record.

ii. For the non-functional requirements candidates did not state the limitations of the system. An

example of a good non-functional requirement is the system can only store 1000 patient records.

iii. Candidates are encouraged to use bullets or number their functional and non-functional

requirements.

2. Design Specification

Candidates were required to give a complete and accurate system structuring diagram containing all

processes and a description of the user interface, report design, algorithm design and appropriate data

structures. Most candidates handled this section fairly well.

i. Most candidates were unaware of how to correctly draw a system structuring diagram.

Candidates should note that this diagram is similar to the HIPO chart as it shows the breakdown

of the system into its submodules and the submodules into its further components.

ii. Candidates should not just include screens shots of interface and report design but also a

justification in order to gain maximum marks.

iii. Narratives will not be accepted as an algorithm. Candidates are reminded to submit either

pseudocodes or flow charts. They must not submit both.

3. Coding and Testing

Candidates were required to produce a complete and accurate C program solution for the problem stated

in the Definition of Problem stage. Most candidates produced programs that achieved good functionality.

i. Only procedural C is accepted as the programming code. No other programming language will be

accepted. Code must also be printed from the compiler, not transferred to a word processor

before printing.

ii. Candidates should not include soft copies of their School-Based Assessment report. Soft copies

will NOT be marked.

iii. Most candidates did not include enough screen shots and in some cases no screen shots were

included. Since a soft copy is not required, the screen shots are very important in allocating marks

for the functionality of the code.

iv. Some codes presented did not match the screen shots given. In such cases candidates were not

awarded any marks.

v. Test plans should be written in a tabular format. They should include normal, extreme and

erroneous.

vi. All input data must be tested using the three testing criteria mentioned above.

vii. Test results submitted without related test plans will not be awarded any marks.

4. Communication and Presentation

This section of the School-Based Assessment is often overlooked by candidates. Teachers and candidates

are urged to pay close attention to their use of grammar and the overall presentation of the SBA report.

Candidates often lose marks due to poor sentence structure and failure to use appropriate technical

jargon in their reports

Recommendations

Teachers should ensure that candidates are fully prepared for the examinations in both units. The poor

performance in some modules of the syllabus may indicate that more time needs to be allocated to these

areas. Encouraging students to review this subject report can be very helpful in improving performance.