Physical and Emotional Sibling Violence and Child Welfare: a

Portland State University Portland State University

PDXScholar PDXScholar

Dissertations and Theses Dissertations and Theses

Winter 3-14-2019

Physical and Emotional Sibling Violence and Child Physical and Emotional Sibling Violence and Child

Welfare: a Critical Realist Exploratory Study Welfare: a Critical Realist Exploratory Study

Katherine Elizabeth Winters

Portland State University

Follow this and additional works at: https://pdxscholar.library.pdx.edu/open_access_etds

Part of the Social Work Commons

Let us know how access to this document bene=ts you.

Recommended Citation Recommended Citation

Winters, Katherine Elizabeth, "Physical and Emotional Sibling Violence and Child Welfare: a Critical Realist

Exploratory Study" (2019).

Dissertations and Theses.

Paper 4808.

https://doi.org/10.15760/etd.6692

This Dissertation is brought to you for free and open access. It has been accepted for inclusion in Dissertations

and Theses by an authorized administrator of PDXScholar. Please contact us if we can make this document more

accessible: [email protected].

Physical and Emotional Sibling Violence and Child Welfare:

A Critical Realist Exploratory Study

Katherine Elizabeth Winters

A dissertation submitted in partial fulfillment of the

requirements for the degree of

Doctor of Philosophy

Social Work and Social Research

Dissertation Committee:

Stephanie Wahab, Chair

Joan Shireman

Jana Meinhold

Eric Mankowski

Portland State University

2019

SIBLING VIOLENCE AND CHILD WELFARE i

Abstract

Sibling violence is a pervasive, yet poorly understood and substantially

underreported phenomenon. Currently recognized as the most common form of intra-

familial abuse, various estimates suggest that 30 percent or more of children in the

general population experience severe acts of violence inflicted by a sibling each year.

Given that many young people in the child welfare system experience the family

conditions associated with abusive sibling violence, recent publications have implored

child welfare to embrace the notion that it is a form of child maltreatment. Practitioners

and policymakers have yet to reach agreement on what constitutes physical or emotional

abuse between siblings, and the perspectives of young people with lived experience of

abuse are largely absent from research and scholarship.

I designed the study, grounded in Critical Realism, to increase understanding of

how sibling violence manifests in child welfare, contribute to theory development, and

identify actions to protect children from harm. Based on in-depth interviews with eight

foster care alumni, I offer a refined definition of sibling violence and four family

conditions associated with sibling violence in child welfare. The findings also supported

a systems-based theory reflecting four stable family member roles. My recommendations

seek to leverage the infrastructure of the child welfare system while taking into

consideration the limitations imposed by neoliberal social and economic policy.

SIBLING VIOLENCE AND CHILD WELFARE ii

Dedication

This dissertation is dedicated to FosterClub for their partnership in this project,

the study participants who contributed so courageously to the interviews, and the foster

care alumni who served as advisers and whose guidance and recommendations

strengthened the study: Ashley Strange, Ke’Onda Johnson, TeAsia Hend, and Timothy

Dennis.

SIBLING VIOLENCE AND CHILD WELFARE iii

Acknowledgements

Utmost gratitude to my adviser, Stephanie Wahab - your encouragement,

responsiveness, and pointed questions kept me steadily moving forward, with confidence.

Joan Shireman, this project benefitted tremendously from your sage wisdom of child

welfare and unwavering attention to its practice and policy implications. Janna Meinhold,

I count myself lucky for having such an enthusiastic thought partner. Eric Mankowski,

your gentle reminder to attend to my own heart was well-timed and did not fall on deaf

ears. Beverly Parsons, your mentorship in systems thinking has beautifully altered the

trajectory of my life. Amber Fletcher, your expertise in Critical Realism was my

guidepost. Emily Lott and Lindsay Merritt, walking this path alongside you has been an

honor and a privilege. And finally, to my family, thank you for giving me everything I

needed to learn so much.

SIBLING VIOLENCE AND CHILD WELFARE iv

Table of Contents

Abstract ............................................................................................................................i

Dedication ...................................................................................................................... ii

Acknowledgements ....................................................................................................... iii

List of Tables............................................................................................................... viii

List of Figures ................................................................................................................ ix

Chapter 1: Introduction .................................................................................................... 1

Social Problem Description and Definitions ......................................................... 1

Sibling Violence Theory ...................................................................................... 3

Sibling Violence Theory and Intervention in the Context of Child Welfare .......... 4

Critical Realism for Sibling Violence Research in Child Welfare ......................... 9

Study Purpose, Research Questions, and Significance to Social Science

Scholarship ........................................................................................................ 10

Chapter 2: Literature Review ......................................................................................... 13

Defining Abusive Sibling Violence .................................................................... 13

Abusive Sibling Violence in the Context of Child Welfare ................................. 15

Abusive Sibling Violence Trauma ...................................................................... 20

A Call to Action ................................................................................................. 21

Family Conditions Associated with Sibling Maltreatment .................................. 22

Sibling Violence Theory .................................................................................... 23

Social learning theory. ............................................................................ 23

SIBLING VIOLENCE AND CHILD WELFARE v

Radical feminist theory. .......................................................................... 25

Conflict theory. ...................................................................................... 26

Coercive family process.......................................................................... 29

Social ecological model of violence. ....................................................... 31

Sibling Violence Interventions ........................................................................... 32

Micro- and meso-level interventions. ...................................................... 32

Macro-level influences and interventions. ............................................... 37

Holistic, coordinated violence prevention intervention. ........................... 39

Chapter 3: Methodology ................................................................................................ 42

Realist Ontology ................................................................................................ 44

Constructivist Epistemology .............................................................................. 47

Methodology and Methods ................................................................................. 48

Study Pilot ......................................................................................................... 49

Study participant recruitment and consent process. ................................. 54

Data collection........................................................................................ 55

Data analysis. ......................................................................................... 57

Establishing Trustworthiness .............................................................................. 61

Ethical Considerations ....................................................................................... 62

Researcher Positionality ..................................................................................... 64

Chapter 4: Results.......................................................................................................... 66

Sample Description ............................................................................................ 67

SIBLING VIOLENCE AND CHILD WELFARE vi

Research Question 1. How do foster care alumni describe their experiences with

physical and emotional sibling violence? ............................................................ 70

A nascent definition of abusive sibling violence in child welfare ............ 70

Family conditions associated with sibling violence ................................. 77

Resonance of linear and non-linear theories ............................................ 87

Research Question 2. From the perspective of foster care alumni, how do adults

who care for or work with young people involved in the child welfare system

(e.g., foster parents, case workers, mental health providers, kinship caregivers,

etc.) respond to sibling violence? ....................................................................... 99

Inadequacy of adult responses to sibling violence in child welfare .......... 99

Survivors’ recommendations to address the problem ............................ 103

Results of CR Retroduction .............................................................................. 104

Child maltreatment in political-economic context. ................................ 106

Chapter 5: Discussion .................................................................................................. 111

Raise Awareness and Provide Basic Training for Adults Charged with Ensuring

Family Safety ................................................................................................... 113

Refine and Adopt a Definition of Abusive Sibling Violence in Child Welfare .. 114

Systematically Assess and Track Abusive Sibling Violence Among Child

Welfare-involved Families ............................................................................... 115

Revisit Child Welfare Co-Placement Policy ..................................................... 116

Invest in Programs to Ensure Safe, Strong Sibling Relationships for Child

Welfare-Involved Youth .................................................................................. 116

SIBLING VIOLENCE AND CHILD WELFARE vii

Inform and Engage the Public to Catalyze Community-Based Solutions .......... 118

Organize for a More Just, Equitable Society ..................................................... 119

Chapter 6: Conclusions ................................................................................................ 121

Study Strengths ................................................................................................ 124

Youth Perspectives. .............................................................................. 124

Critical Realism. ................................................................................... 126

Systems Thinking. ................................................................................ 126

Implications for Future Research ...................................................................... 127

Implications for Social Work ........................................................................... 128

Study Limitations ............................................................................................. 129

All female sample ................................................................................. 129

Individual perspectives. ........................................................................ 130

Investigator limitations ......................................................................... 130

References ................................................................................................................... 132

Appendices .................................................................................................................. 151

Appendix A: Recruitment Flyer ....................................................................... 151

Appendix B: Participant Informed Consent Document ..................................... 153

Appendix C: Data Collection Instruments: Interview Protocol and Background

Form ................................................................................................................ 158

Appendix D: Final Coding Scheme .................................................................. 165

Appendix E: Family Map Example .................................................................. 172

Appendix F: Human Subjects Application to IRB ............................................ 181

SIBLING VIOLENCE AND CHILD WELFARE viii

List of Tables

Table 1. Characteristics of Sibling Rivalry and Sibling Violence .................................. 14

Table 2. NCANDS Child Maltreatment Perpetrator Categories and Percentages

in 2016 ......................................................................................................................... 16

Table 3. Sibling Study Adviser Conference Call Summary ........................................... 52

Table 4. Study Process Summary .................................................................................. 53

Table 5. Study Participants’ Reports of Physical and Emotional Sibling Violence ........ 68

Table 6. Family Conditions Associated with Sibling Violence ...................................... 78

Table 7. Final Set of Family Conditions Associated with Sibling

Violence in Child Welfare ............................................................................................ 80

Table 8. Family Conditions Associated with Sibling Violence Overall and by

Placement Type ............................................................................................................ 81

Table 9. Theory Code Descriptions ............................................................................... 88

Table 10. Final Set of Primary and Secondary Family Conditions for Sibling

Violence in Child Welfare with Associated Linear Theories ......................................... 92

SIBLING VIOLENCE AND CHILD WELFARE ix

List of Figures

Figure 1. Integrated theoretical model of sibling physical and psychological abuse

(Hoffman and Edwards, 2004, p. 191). ........................................................................ 27

Figure 2. Ecological model for understanding violence (adapted from the World

Health Organization, 2002, p. 12). ............................................................................... 32

Figure 3. Ecological model for understanding violence (adapted from the World

Health Organization, 2002, p. 12). ............................................................................... 38

Figure 4. Coordinated violence prevention model: Hypothetical common and

specific elements (Hamby and Grych, 2013, p. 85). ..................................................... 40

Figure 5. CR domains of reality presented as an iceberg. ............................................. 45

Figure 6. Instigator/retaliator escalating abusive sibling violence dynamic with

reinforcing feedback loops. .......................................................................................... 74

Figure 7. Two biological siblings and their father. ....................................................... 95

Figure 8. Kinship care setting with two biological siblings and their foster sister. ........ 97

Figure 9. Biological family with four siblings.............................................................. 98

Figure 10. Retroduction results summarized in terms of the three domains of

reality. ......................................................................................................................... 110

SIBLING VIOLENCE AND CHILD WELFARE 1

Chapter 1: Introduction

Social Problem Description and Definitions

Sibling violence is a pervasive, yet poorly understood and substantially

underreported phenomenon. Currently recognized as the most common form of intra-

familial abuse (Button & Gealt, 2010), various estimates suggest that 30 percent or more

of children in the general population experience severe acts of violence inflicted by a

sibling each year (Caffaro, 2014; Finkelhor, Tuner, & Ormrod, 2006; Straus, Gelles, &

Steinmetz, 2006; Tucker, Finkelhor, Shattuck, & Turner, 2013). Common discourse

continues to minimize the problem (e.g., sibling rivalry, rough housing), and the

phenomenon is often treated as an accepted rite-of-passage. Yet differences in size and

physique between siblings, the developmental immaturity of children, and everyday close

contact within the household are likely to increase the frequency, intensity, and duration

of violent sibling interactions (Finkelhor et al., 2006). These findings are cause for

concern given robust evidence of numerous harmful effects of sibling violence across the

lifespan (Button & Gealt, 2010; Caffaro, 2011; Caffaro, 2014; Feinberg, Solmeyer, &

McHale, 2012; Finkelhor et al., 2006; Graham-Bermann, Cutler, Litzenberger, &

Schwartz, 1994; Kramer & Bank, 2005; Straus et al., 2006; Tucker et al., 2013). In the

absence of a caring adult who intervenes, chronic sibling violence may cause “toxic

stress,” which has been linked to physical and mental illness later in life (Shonkoff,

Boyce, & McEwen, 2009; Finkelhor et al., 2006).

Sibling violence theorists, practitioners, and researchers have attempted to define

what constitutes abusive physical and emotional violence between siblings. Wiehe

SIBLING VIOLENCE AND CHILD WELFARE 2

(1997), one of the first and more widely cited authors on the topic of sibling violence,

aligned the phenomenon with parent-to-child maltreatment by using the word “abuse”

when defining violent interactions between siblings. Caffaro (2014), a clinical

psychologist and family therapist, effectively differentiated healthy sibling conflict (i.e.,

rivalry) and abusive sibling violence in terms of aims, frequency, power, change over

time, and the role of caregivers. According to the author, abusive sibling violence is

marked by stable victim and offender roles. Caffaro’s emphasis on repetition is notable,

given that even low frequency peer violence (i.e., occurring less than four times per year)

has been significantly associated with trauma symptoms in young children (Finkelhor et

al., 2006). Meyers (2017) built upon the work of Caffaro and others in a qualitative study

with adult survivors. She argued that a comprehensive definition of abusive physical and

emotional violence between siblings should include the victim/survivor’s lived

experience/perception of what took place.

Despite commendable progress made over the course of more than three decades

of sibling violence research and scholarship, a clearly-articulated and broadly accepted

definition of “sibling abuse” is yet to be developed. In a review of more than 100 journal

articles, books, chapters, and dissertations published between 1977 and 2008, Perkins

(2014) found 16 different labels used to discuss emotional and physical violence between

siblings: sibling abuse, sibling aggression, sibling agonistic behavior, sibling antagonism,

sibling assault, sibling conflict, sibling fighting, sibling hostility, sibling maltreatment,

sibling negativity, sibling psychological abuse, sibling psychological maltreatment,

sibling quarreling, sibling relational aggression, sibling rivalry, and sibling violence (pgs.

SIBLING VIOLENCE AND CHILD WELFARE 3

34-35). Definitional variation is problematic, limiting interpretation of individual sibling

violence studies and the capacity of the field to draw comparisons among them.

Sibling Violence Theory

Theorizing about sibling violence is similarly nascent, traversing levels of the

social ecology from the interpersonal to the global. Social learning theory positions

sibling violence as a learned behavior which is fortified through reinforcement (Bandura,

Ross, & Ross, 1963; Hoffman & Edwards, 2004). Radical feminist theory argues that the

patriarchal organization of society and the related, pervasive acceptance of the use of

power and control to achieve desired aims underlie the phenomenon (Button & Gealt,

2010; Graham-Berman et al., 1994; Hoffman & Edwards, 2004; Hoffman, Kiecolt, &

Edwards, 2005; Wiehe, 1997). With roots in Marxism, conflict theory assumes that

humans are innately self-interested, and when located within a context of scarcity, will

utilize any means available to obtain desired resources. Sibling violence is theorized to be

a response to a child’s perception that necessary resources (e.g., basic needs, parental

attention) are scarce or inadequate (Hoffman et al., 2005; Smith & Hamon, 2012). While

these theories identify potential causal mechanisms underlying sibling violence, they rely

on reductionist linear models and differentiations between violence and abuse are

unclear.

Moving beyond linear, mechanistic models, a small subset of the literature

discusses systems-based conceptions of sibling violence. Family systems theory

(Milevsky, 2011), coercion theory (Granic & Patterson, 2006; Patterson, 1982; Smith,

Dishion, Shaw, Wilson, Winter, & Patterson, 2014), and an ecological model

SIBLING VIOLENCE AND CHILD WELFARE 4

(Bronfenbrenner, 1979) employ key systems concepts including holism, nesting, and

non-linearity to examine interpersonal violence as a complex and dynamic process that

spans local to global levels of a structurally violent society (Gil, 1996). In a

comprehensive review of theories of sibling relationships across the lifespan, Whiteman,

McHale & Soli (2011) summarize the current state of sibling violence theorizing,

acknowledging that:

…the processes that affect sibling relationship dynamics operate at a variety of

levels, ranging from intrapsychic processes, such as attachment and social

comparison, to relational dynamics, such as social learning and more distal forces

beyond the family, such as sociocultural influences. (p. 135)

While a small number of systems-oriented scholars have explored these ideas,

substantial additional effort is needed to understand the extent to which these conceptions

approximate reality.

Sibling Violence Theory and Intervention in the Context of Child Welfare

The current definition of child maltreatment, per the most recent iteration of the

Child Abuse Prevention and Treatment Act (CAPTA) states:

…the term ‘child abuse and neglect’ means, at a minimum, any recent act or

failure to act on the part of a parent or caretaker, which results in death, serious

physical or emotional harm… or an act or failure to act which presents an

imminent risk of serious harm” (Administration for Children and Families,

Division of Health and Human Services, 2016, p. 7).

SIBLING VIOLENCE AND CHILD WELFARE 5

Though intentionally vague, the CAPTA definition of child maltreatment does not fit

well with analysis of abusive sibling violence in that it assumes the perpetrator is a

parent. However, the latter part of the definition, focused on “imminent risk of serious

harm” can be employed given strong evidence that sibling-perpetrated physical and

emotional violence produce negative outcomes across the lifespan (Button & Gealt, 2010;

Caffaro, 2011; Caffaro, 2014; Feinberg et al., 2012; Finkelhor et al., 2006; Graham-

Bermann et al., 1994; Kramer & Bank, 2005; Straus et al., 2006; Tucker et al., 2013).

Data on physical and emotional violence between siblings are not systematically

collected in any of the three federally-maintained (i.e., national) child welfare data

systems: National Child Abuse and Neglect Data System (NCANDS), Adoption and

Foster Care Analysis and Reporting System (AFCARS), and National Incidence Studies

of Child Abuse and Neglect (NIS) (Shireman, 2015). Moreover, many states exclude

sibling violence from definitions of child abuse (Caffaro, 2011), leaving direct service

practitioners ill-prepared to identify the problem among the children they serve

(Kominkiewicz, 2004). Increased emphasis on sibling co-placement in child welfare is

likely to further reduce attention to harmful sibling dynamics. This is notable given well-

intentioned and often beneficial federal policy mandating co-placement for siblings in

foster care.

In a review of the peer-reviewed empirical social work and psychology

literature published from 1988 through 2003, 17 studies addressing siblings in foster care

The Fostering Connections to Success and Increasing Adoptions Act of 2008 (P.L. 110–351) requires that

reasonable efforts be made to place siblings in the same foster home and with kin whenever possible, and

over half of state child welfare systems prioritize sibling co-placement as a means of maintaining sibling

bonds (Gustavsson & MacEachron, 2010). For an in-depth discussion of legal protections accorded to

siblings in foster care, see Shlonsky, Bellamy, Elkins, & Ashare, 2005.

SIBLING VIOLENCE AND CHILD WELFARE 6

or adoption collectively demonstrated that sibling co-placement can support placement

stability and reduce symptomology among child welfare-involved youth (Hegar, 2005). A

subsequent meta-analysis similarly concluded that “most children benefit significantly

from living with their siblings” (Washington, 2007, p. 431). Juxtaposed with reports

indicating that significant numbers of siblings were not co-placed (Staff & Fein, 1992;

Washington, 2007; Wulczyn & Zimmerman, 2005), these findings powered a shift in

federal and state legislation toward strong emphasis on co-placement and, when placed

separately, the importance of ensuring regular visitation between siblings removed from

biological caregivers, both in the United States and abroad (Hegar, 2005; Kothari et al.,

2014).

The Fostering Connections to Success and Increasing Adoptions Act of 2008 was

the first federal law to address the importance of sibling co-placement, requiring states to

make reasonable effort to maintain sibling connections when in foster care. Over half of

the states in the U.S. had established sibling placement or visitation policies prior to

passage of the Act, whereas others developed state-based policies specific to siblings in

foster care subsequent to its passage (Child Welfare Information Gateway, 2013). Yet

neither the Act, nor its accompanying Program Instruction for Title IV-E agencies

include the word “abuse” and the Act does not address cases where co-placement is not

appropriate. Moreover, no standard protocol is available for caseworkers to use in making

decisions about co-placement when sibling abuse is present. This “best practice” (Cohn,

2008) guides a nationalized approach to care in which siblings who should not be co-

placed, or who should be co-placed only with strong supports, may be overlooked. In an

SIBLING VIOLENCE AND CHILD WELFARE 7

effort to bring much needed nuance to co-placement policy, Linares, Li, Shrout, Brody, &

Pettit (2007) argued that separation of siblings during foster care placement may be

beneficial, particularly for children with elevated levels of behavior and conduct

problems.

The lack of attention to sibling violence in child welfare is concerning given that

rates are likely high among maltreated children. Family conditions among children

exposed to sibling violence include many which are common in child welfare: parental

physical or emotional unavailability, lack of supervision, differential treatment of

children in the same family such as scapegoating, inappropriate caretaking expectations

placed on older sibling(s), parental acceptance of sibling violence as a normal part of

family life, lack of parent intervention during acts of sibling violence, parental modeling

of physically or emotionally abusive behavior, drug or alcohol abuse in the home, chronic

mental or physical illness of parent(s), work or financial strain, and parental denial of the

problem (Caffaro & Con-Caffaro; 1998; Wiehe, 1997). Moreover, studies have

demonstrated that sibling violence often co-occurs with other types of family violence

(i.e., adult-perpetrated child maltreatment, interpersonal violence between adult

caregivers) (Henning, Leitenberg, Coffey, Bennet, & Jankowski, 1997; Spaccarelli,

Sandler, & Roosa, 1994; Wallace, 1999; Wiehe, 1997). A small number of studies

support the notion that maltreated children are prone to sibling/peer violence (Linares et

al., 2007; Linares et al., 2015; Shields & Cincchetti, 1998, 2001). Linares et al. (2015)

found that 82 percent of foster parents reported past-year physical aggression acts

SIBLING VIOLENCE AND CHILD WELFARE 8

between siblings in their care; 14 percent of foster children had been “beat-up” by a

sibling while in their care (p. 211).

While the causal mechanisms underlying sibling violence are tentative, various

interventions have been devised to train both parents and children to more productively

and peacefully navigate sibling disputes. As early as 1967, clinicians were testing

approaches to reduce sibling aggression and increase cooperative play among general

samples (Kramer, 2004). More recent interventions have recommended various

combinations of individual, sibling, family, and group therapy and training sessions on

topics such as mediation, to reduce sibling violence (Caffaro & Conn-Caffaro,

2005; Caspi, 2008; Feinberg, Sakuma, Hostetler & McHale, 2013; Kennedy & Kramer,

2008; Kiselica & Morrill-Richards, 2007; Reid & Donovan, 1990; Thomas & Roberts,

2009; Siddiqui & Ross, 2004; Smith & Ross, 2007). Two studies have tested

interventions devised specifically for child welfare-involved siblings (Kothari et al.,

2017; Linares et al., 2015). Both interventions employ skill-building to support siblings

to cultivate relational and self-regulatory skills to enhance relationship quality. While

initial testing generated statistically significant results, the study conducted by Linares et

al. (2015) was subject to a small sample size and Kothari et al. (2017) did not include a

measure of sibling violence. More work is needed to develop and test interventions to

ameliorate violence between siblings within the context of child welfare and beyond.

Theory and intervention development for child-welfare involved siblings is

promising, given a recent call for social workers and mental health practitioners to attend

to physical and emotional sibling violence in their work with children and families and

SIBLING VIOLENCE AND CHILD WELFARE 9

social work’s “Grand Challenge” to stop family violence

(Kulkami, Barth, & Messing

2016; Meyers, 2014; Perkins & O’Connor, 2015; Perkins & Stoll, 2016; Shadik, Perkins,

& Kovacs, 2013). Yet without holistic understanding of the causal mechanisms

underlying sibling violence, interventions for child welfare-involved siblings are likely to

exclude potentially efficacious components or strategies. Moreover, the perspectives of

young people with lived experience with sibling violence are absent from theory and

research on the phenomenon in the context of child welfare. This lack of youth-informed

research and practice is misaligned with a body of research which has demonstrated that

young people have much to offer when it comes to social problems that affect them.

When asked to describe their experiences, share their perspectives, and provide

recommendations, young people can meaningfully contribute to efforts to parse complex

interpersonal processes, develop accurate definitions and measures, capitalize on sources

of resilience in intervention designs, and inform direct practice and policy-making

(Horwath, Kalyva, & Spyru, 2012; Hyde & Kammerer, 2009; Riebschleger, Day, &

Damashek, 2015; Strolin-Goltzman, Kollar, & Trinkle, 2010).

Critical Realism for Sibling Violence Research in Child Welfare

Critical Realism (CR) emerged in the 1970s and 1980s as an alternative paradigm

to positivism and constructivism. CR opens space to investigate ontological questions

about causation by employing constructivist epistemologies and methods by separating

Led by the American Academy of Social Work & Social Welfare (AAASWSW), the Grand Challenges

for Social Work is an initiative to champion social progress through social science, collaboration, and

shared projects. For more information about the Grand Challenges visit http://aaswsw.org/grand-

challenges-initiative/about/. For information about the challenge to stop family violence, see Kulkarni,

Barth, and Messing’s (2016) policy recommendations.

SIBLING VIOLENCE AND CHILD WELFARE 10

reality into three domains: the empirical, the actual, and the causal. According to this CR

stratification, the empirical domain (where social science research takes place) consists of

a subset of all events produced by an underlying causal mechanism. CR research

embraces both intensive (i.e., in-depth interpretive data) and extensive (i.e., data on

widespread trends, typically gathered with quantitative data) methods to identify

empirically observable patterns of human behavior which are then subjected to theory-

laden researcher-driven thought exercises to clarify causal processes (Lennox & Jurdi-

Hage, 2017; Sayer, 2010). CR is a well-suited philosophy to guide the current project

because its goal is to inform emancipatory action to dismantle system structures that

cause human suffering. This project, grounded in the CR paradigm, aims to identify

causal mechanisms underlying abusive sibling violence as it is described by foster care

alumni. The long-term objective is to inform direct practice, intervention development,

and child welfare policies that effectively reduce its prevalence. A more detailed

discussion of CR as it relates to the methodology and methods for this project is included

in Chapter 3.

Study Purpose, Research Questions, and Significance to Social Science Scholarship

This dissertation was designed to begin to fill gaps in theory, direct practice, and

policy by asking foster care alumni to share their experiences with sibling violence and

offer recommendations to address the problem. The study examined abusive sibling

violence, employing a working definition based on the work of Caffaro (2011) and

Meyers (2017): (1) a repeated, escalating pattern of physically and/or emotionally violent

interactions; (2) with stable victim/offender roles (i.e., unidirectional, (Caspi, 2012)); (3)

SIBLING VIOLENCE AND CHILD WELFARE 11

in which the offender is motivated by a need for power and control, and; (4) the victim

experiences a chronic sense of terror and powerlessness. Specifically, the study

investigated the extent to which such a definition was represented in the descriptions of

physical and emotional violence provided by foster care alumni.

The study was designed to generate information to clarify (i.e., better define) what

constitutes abusive sibling violence and to assess the fit and utility of family conditions

and theories of sibling violence discussed in the extant literature by incorporating the

perspectives of foster care alumni who have experienced it in the context of child

welfare. The research questions guiding the study included:

1) How do foster care alumni describe their experiences with physical and emotional

sibling violence?

a) To what extent do their descriptions fit with Caffaro (2014) and Meyers’

(2015) definitions of abusive sibling violence?

b) To what extent are family conditions and theories of sibling violence

discussed in the extant literature represented in their descriptions?

c) What additional/refined causal mechanisms do their descriptions suggest?

The inquiry will intentionally exclude sexual abuse between siblings. While physical, emotional, and

sexual abuse are likely to co-occur (Caffaro & Conn-Caffaro, 1998; Wiehe, 1997), the latter is not subject

to the definitional problems that physical and emotional abuse are. Sibling incest is a criminal behavior

prohibited by law in every U.S. state (Myers, 1998 as cited in Perkins, 2014). The current study will focus

on physical and emotional abuse since these types of sibling abuse are definitionally ambiguous and largely

neglected in the child welfare literature.

SIBLING VIOLENCE AND CHILD WELFARE 12

2) From the perspective of foster care alumni, how do adults who care for or work with

young people involved in the child welfare system (e.g., foster parents, case workers,

mental health providers, kinship caregivers, etc.) respond to sibling violence?

a) To what extent are adult responses perceived as helpful?

b) What do young people recommend to address sibling violence?

The goals of the study for Social Science scholarship, research, and practice were

multifold:

1. increase understanding of physical and emotional sibling violence in the context

of child welfare;

2. contribute to sibling violence theory development; and,

3. identify innovative ways to protect children from sibling violence and/or help

foster children manage sibling violence.

The chapters that follow provide a comprehensive review of the sibling violence

literature, both within and beyond the context of child welfare, describe the methods for

the study including the ontology, epistemology and methodology guiding the project, the

results of the study, and conclude with a discussion of the findings with emphasis on

practical recommendations to address sibling violence in child welfare and beyond.

SIBLING VIOLENCE AND CHILD WELFARE 13

Chapter 2: Literature Review

Defining Abusive Sibling Violence

Wiehe (1997), one of the first and more widely cited authors on the topic of

sibling violence, aligned the phenomenon with adult-perpetrated child maltreatment by

using the word “abuse” when defining violent interactions between siblings. According to

the author, physical abuse includes “willful acts resulting in physical injury such as

slapping, hitting, biting, kicking, or more violent behavior that may include the use of an

instrument, such as a stick, bat, gun, or knife” (pp. 14-18). Emotional abuse was defined

as “verbal comments aimed at ridiculing, insulting, threatening, or belittling. Emotional

abuse is also inclusive of the destruction of personal property, such as a sibling who

deliberately destroys a prized possession or pet of another sibling” (Wiehe, 1997, pp. 33-

34). In addition to providing a variety of tangible examples of acts which constitute

physical and emotional abuse, both definitions highlight the intent of one sibling to harm

another. Wiehe’s definition of physical violence also specifies that injury must result

from an act of violence to qualify as abusive.

According to Caffaro (2011), sibling violence is defined as “a range of behaviors

including pushing, hitting, kicking, beating, and using weapons to inflict physical harm,”

whereas psychological maltreatment includes “exposing a sibling to violence by peers or

other siblings; comments aimed at ridiculing, insulting, threatening, terrorizing, and

belittling a sibling; rejecting, degrading, and exploiting a sibling, and; destroying a

sibling’s personal property” (pgs. 8-10). The author differentiated healthy sibling conflict

and abusive sibling violence in terms of aims, frequency, power, change over time, and

SIBLING VIOLENCE AND CHILD WELFARE 14

the role of caregivers (Table 1). Most notably, abusive sibling violence is marked by

stable victim and offender roles. A shortcoming of Caffaro’s list of characteristics is that

it largely neglects to describe the markers of psychological maltreatment between

siblings.

Table 1

Characteristics of Sibling Rivalry and Sibling Violence (Caffaro, 2011, p. 91)

Sibling Rivalry

• Conflict between siblings in which the reward is possession of something that

the other also wants

• Conflict between siblings that strengthens their relationship

• Fierce but balanced comparisons between siblings with regard to achievement,

attractiveness, and social relations with peers

Sibling Violence

• A repeated pattern of physical aggression directed toward a sibling with the

intent to inflict harm and motivated by an internal emotional need for power

and control

• Physical aggression directed toward a sibling that aims to leave the other feeling

humiliated, defeated, and/or unsafe

• An escalating pattern of sibling aggression and retaliation that parents seem

unwilling or unable to stop

• Role rigidity resulting in the solidification of victim and offender sibling roles

More recently, Meyers (2017) conducted a qualitative study employing

phenomenological and grounded theory methods to gather survivor accounts to develop a

working definition of sibling abuse. Based on in-depth interviews with 19 survivors aged

25 to 65, analysis identified the unpredictable nature of the abusive incidents as an

essential marker of sibling abuse. Survivors recounted violent interactions which

occurred consistently over the course of years. Despite the long-term and frequent nature

of the abuse, the perception that they could not anticipate or avoid the assaults resulted in

SIBLING VIOLENCE AND CHILD WELFARE 15

a chronic sense of terror coupled with a perpetual state of powerlessness to self-protect.

Meyers’ findings also confirmed intent, duration, and a power differential between

siblings as key attributes of sibling abuse. A notable insight which can be garnered from

the inquiry is the idea that a comprehensive definition of abusive sibling violence should

include the victim/survivor’s lived experience/perception of what took place. The current

project assessed the extent to which there was resonance between the definitions of

abusive sibling violence put forward by Caffaro and Meyers and respondents’

descriptions of sibling violence in the context of child welfare.

Abusive Sibling Violence in the Context of Child Welfare

While Caffaro and Meyers’ definitions position abusive sibling violence as a form

of child maltreatment, state and federal entities charged with ensuring the safety and

wellbeing of children have largely overlooked this form of family violence. In the United

States, the Department of Health and Human Services Administration for Children and

Families (ACF) is charged with promoting the economic and social well-being of

families, children, individuals, and communities. ACF programs and services for children

and youth include child abuse and neglect prevention and intervention and ensuring that

children who are victims receive treatment and care. ACF includes the Children’s Bureau

which collects case-level data on reports of child abuse and neglect via the National Child

Abuse and Neglect Data System (NCANDS). Annual analyses of maltreatment data are

summarized in publicly available reports that are also submitted to Congress (U.S.

Department of Health & Human Services, 2015).

SIBLING VIOLENCE AND CHILD WELFARE 16

In Child Maltreatment 2016, the 27

annual report on child abuse and neglect

generated by the Administration and the most recent report available, child maltreatment

was defined as “Any recent act or failure to act on the part of a parent or caretaker which

results in death, serious physical or emotional harm, sexual abuse or exploitation; or an

act or failure to act, which presents an imminent risk of serious harm” (U.S. Department

of Health & Human Services, 2016, p. viii). This definition is limited, given that many

people other than parents can be abusers. Regardless, for fiscal year 2016, “there were a

nationally estimated 676,000 victims of abuse and neglect…a rate of 9.1 victims per

1,000 children in the population” (U.S. Department of Health & Human Services, 2016,

p. x). The NCANDS does not capture sibling abuse rates, however, a significant

shortcoming of the system. Children under the age of 18 constitute under one percent of

perpetrators in the NCANDS (U.S. Department of Health & Human Services, 2016, p.

65). Moreover, the system does not disaggregate siblings among the perpetrator

categories captured, as shown in Table 2.

Table 2

NCANDS Child Maltreatment Perpetrator Categories and Percentages in 2016

Perpetrator’s Relationship to the Maltreated Child

Percentage

Parent

91.4%

Other relative

4.6%

Unknown

3.1%

Unmarried partner of parent

3.0%

Other

2.7%

More than one nonparental perpetrator

1.1%

Friend or neighbor

.8%

Child daycare provider or other professional

.7%

Foster parent

.2%

Legal guardian

.2%

SIBLING VIOLENCE AND CHILD WELFARE 17

According to the NCANDS, sibling perpetrators comprise a very small proportion

of substantiated child maltreatment cases; “other relatives”, who would include a variety

of family members in addition to siblings, account for slightly more than six percent of

perpetrators documented. Foster siblings are subsumed within the “other” category,

which also includes nonrelated adult, nonrelated child, foster sibling, babysitter,

household staff, clergy, and school personnel perpetrators (U.S. Department of Health &

Human Services, 2016, p. 23).

These data contrast sharply with numerous other

nationally-representative sources of family violence data. The National Family Violence

Survey, The National Family Violence Resurvey, the Developmental Victimization

Survey and, most recently, the National Survey of Children’s Exposure to Violence

reported rates of severe violence between siblings between 30 and 82 percent annually

(Caffaro, 2011; Finkelhor, Ormrod, Turner, & Hamby, 2005; Gelles & Steinmetz, 2006;

Tucker et al., 2013). While counting violent acts is an overly simplistic approach to

investigating physical abuse and overlooks emotional abuse altogether, it does provide

tangible evidence of the scope of the problem. These general population studies also

support the notion that sibling violence crosses economic boundaries and position

abusive sibling violence as a subset of family violence, comparable to interpersonal

violence between adults and adult-perpetrated child maltreatment with regard to its need

for both attention and intervention. The absence of sibling abuse cases in the NCANDS

Data on physical and emotional violence between siblings involved in child welfare are not systematically

collected in any of the three federally-maintained (i.e., national) data systems: National Child Abuse and

Neglect Data System (NCANDS), Adoption and Foster Care Analysis and Reporting System (AFCARS),

and National Incidence Studies of Child Abuse and Neglect (NIS) (Shireman, 2015).

SIBLING VIOLENCE AND CHILD WELFARE 18

exposes a lack of alignment between the findings of rigorous, nationally representative

sibling violence research and prevalence tracking conducted through child welfare.

The NCANDS is but one among numerous sources of information highlighting

the lack of attention to abusive sibling violence in child welfare. Many states exclude

sibling maltreatment from definitions of child abuse, or refer to it only indirectly,

influencing both reporting practices and perceptions of what constitutes abuse (Caffaro,

2011). In a state-wide study of child protection service caseworkers conducted in Indiana

in 2004, for example, almost one quarter of the child protection service caseworkers with

an undergraduate degree in social work did not identify physical, sexual, verbal, or

emotional violence when asked how they define sibling abuse (Kominkiewicz, 2008).

Without clearly articulated definitions of what constitutes abusive sibling violence, direct

service practitioners are ill-equipped to identify the problem among children they serve.

A focus on the parent as the primary point of intervention may further influence

caseworkers to discount the influence of sibling relationships.

Current child welfare policy recommends that siblings be placed together

whenever possible to support foster youth to maintain sibling ties (McBeath et al., 2014).

There is some evidence that co-placement mediates the relationship between adult-

perpetrated trauma and internalizing and externalizing symptoms among children in

foster care (Hegar & Rosenthal, 2011; Wojciak, McWey, & Helfrich, 2013). At

minimum, co-placement is generally assumed to do no harm (Hegar, 2005). While a

This study should be interpreted with caution, as the results were based on a small sample size. It also

appears that abuse was not defined in the survey that was administered to case workers.

SIBLING VIOLENCE AND CHILD WELFARE 19

growing body of research has quantitatively assessed the relationship between sibling co-

placement and placement stability, emotional adjustment, and foster children’s

perceptions of their placement situations (Hegar & Rosenthal, 2011), abusive sibling

violence is not discussed (Pinel-Jacquemin, Cheron, Favart, Dayan, & Scelles, 2013).

Oversimplified co-placement policy in the absence of information about harmful sibling

relationships may marginalize cases in which co-placement is not a good idea (Cohn,

2008). In a recent review of the literature on violence among siblings in joint placement,

the authors note a need to refine methodologies and tools used to analyze the

relationships between siblings across the continuum of care (Pinel-Jacquemin et al.,

2013). Linares et al. (2007) argue that separating a foster child from their sibling(s) may

be beneficial under certain conditions (i.e., behavioral or conduct problems).

The need to increase attention to abusive sibling violence in the context of child

welfare is amplified by studies demonstrating high rates of co-occurrence with other

forms of family violence. Children who witness domestic violence between their parents

are more likely to engage in violent behavior with their siblings and peers (Button and

Gealt, 2010; Spaccarelli et al., 1994; Wiehe, 1997). Noland, Liller, McDermott, Coulter,

& Seraphine (2004) found mother-to-father violence, father-to-mother violence, mother-

to-child violence, and father-to-child violence to be significant predictors of sibling

violence perpetration. With regard to the effect of child abuse perpetrated by adult

caregivers, Straus et al. (2006) documented a positive relationship between adult-

perpetrated violence toward a child and that child’s likelihood of severely attacking a

sibling. Children subjected to the most severe abuse from a parent were often intensely

SIBLING VIOLENCE AND CHILD WELFARE 20

violent with a brother or sister (National Family Violence Resurvey). Sibling abuse is

also more prevalent in families in which both spousal and child abuse are present

(Wallace, 1999; Wiehe, 1997). Given these findings, it is likely that sibling violence rates

are higher among youth served by the child welfare system than among general samples

(Hamby & Grych, 2013). In support of this notion, a recent study of foster youth found

that acts of extreme sibling violence were commonplace (Linares, 2008).

Abusive Sibling Violence Trauma

The lack of attention to abusive sibling violence in child welfare is concerning,

given that low frequency peer violence (i.e., occurring less than four times per year) has

been significantly associated with trauma symptoms in young children (Finkelhor et al.,

2006). The National Scientific Council on the Developing Child has defined “toxic”

stress as “strong, frequent, or prolonged activation of the body’s stress management

system. Stressful events that are chronic, uncontrollable, and/or experienced without

children having access to support from caring adults tend to provoke these types of toxic

stress responses” (2005/2014). As noted previously, Caffaro’s (2011) definition of

abusive sibling violence describes a repeated, “escalating pattern of sibling aggression

and retaliation that parents seem unwilling or unable to stop” (p. 91). This suggests that

sibling violence may cause toxic stress, which has been linked to physical and mental

illness later in life (Shonkoff et al., 2009). In a study cited by Caffaro (2011), “children

who were repeatedly attacked by a sibling were twice as likely as others their age to

demonstrate severe symptoms of trauma, anxiety, and depression, including

sleeplessness, suicidal ideation, and fear of the dark” (Finkelhor et al., 2006). Deleterious

SIBLING VIOLENCE AND CHILD WELFARE 21

outcomes also include mental health challenges, conduct disorders, neurotic traits,

suicidal ideation and attempts, and decreased self-esteem (Button and Gealt, 2010;

Caffaro, 2011; Feinberg et al., 2012; Finkelhor et al., 2006; Graham-Bermann & Cutler,

1992, as cited in Caffaro, 2011). As evidenced by these studies, the harmful effects of

abusive sibling violence manifest across the life course.

A Call to Action

Acknowledging that the phenomenon is often overlooked by social work

practitioners, a recent article in Social Work, a National Association of Social Workers

publication, advocated for the “necessary role for social work” in addressing physical and

emotional sibling violence, arguing that “violence against a child, regardless of the

individual perpetrator, is still violence against a child” (Perkins & O’Connor, 2015, p.

91). When siblings are engaged in the child welfare system, and are removed from their

original family, the state takes on the role of parent for those children and must conduct

adequate assessment to determine whether they are demonstrating healthy conflict or

abusive violence. Similarly, the state is responsible to intervene when abusive sibling

violence is taking place, and to ensure that siblings are not co-placed when risk is

eminent.

These findings point to a significant gap in the child welfare system, for the

published literature contains little or nothing about sibling abuse. This is a new idea for

child welfare. Apparently, little is currently being done, in either policy or practice, to

address abusive sibling violence. A recent publication advocated for inclusion of sibling

violence discussion in child abuse and neglect parent education curricula (Shadik at al.,

SIBLING VIOLENCE AND CHILD WELFARE 22

2013). This call for attention to abusive sibling violence in family-focused interventions

is aligned with Perkins and O’Connor’s recommendation that “social workers should

evaluate and develop interventions aimed at preventing or ameliorating physical and

emotional violence between siblings” (2016, p. 92). Yet without holistic understanding of

the causal mechanisms underlying sibling violence, interventions for child welfare-

involved siblings are likely to exclude potentially efficacious components or strategies.

The sections to follow present a synthesis of the literature on sibling violence theory and

intervention.

Family Conditions Associated with Sibling Maltreatment

Two seminal, qualitative studies carried out in the late 1990s identified numerous

individual characteristics and family conditions associated with sibling abuse including:

parental physical or emotional unavailability, lack of supervision, differential treatment

of children in the same family such as scapegoating, inappropriate caretaking

expectations placed on older sibling(s), parental acceptance of sibling rivalry as a normal

part of family life, lack of parent intervention during acts of inter-sibling violence,

parental modeling of physically or emotionally abusive behavior, drug or alcohol abuse

in the home, chronic mental or physical illness of parent(s), work or financial strain, and

parental denial of the problem (Caffaro & Con-Caffaro, 1998; Wiehe, 1997).

In addition to parent characteristics or behaviors, those embodied by perpetrators

and victims were also explored. According to Caffaro and Con-Caffaro (1998), offenders

are prone to thinking errors that minimize or distort their behavior, have suffered

victimization themselves, and/or demonstrate deficiencies of impulse control,

SIBLING VIOLENCE AND CHILD WELFARE 23

cognitive/developmental deficits, or an inability to empathize with others. Research

psychologists conducting lab-based investigations similarly identified stable personality

characteristics and cognitive processing deficits among aggressive children such as

impulsivity, callous-unemotional traits, narcissism, and a tendency to inaccurately

interpret social cues (Crick & Dodge, 1994; Frick & White, 2008; Kerig & Stellwagen,

2009; Loeber & Stouthhamer-Loeber, 1998). Large developmental, physical, or

intellectual differences between siblings were similarly associated with sibling abuse

(Caffaro & Con-Caffaro, 1998). While emphatic that victim characteristics should not be

used to place blame, Wiehe (1997) indicated that genetically determined physical and

behavioral qualities may make a child more prone to abuse.

Sibling Violence Theory

Various sources of nationally-representative data indicate that 30 percent or more

of children experience severe violence inflicted by a sibling (Caffaro, 2014; Finkelhor,

Tuner, & Ormrod, 2006; Straus, Gelles, & Steinmetz, 2006; Tucker, Finkelhor, Shattuck,

& Turner, 2013). While these studies reveal the extent of the problem, the research

community has yet to agree on the causal mechanisms underlying abusive sibling

violence. Linear and non-linear (i.e., systems-oriented) theories are discussed in the

literature.

Social learning theory. Numerous studies have documented the co-occurrence of

various forms of family violence, drawing linkages between a child’s exposure to

violence and their propensity to inflict violent acts upon others. Children who witness

interpersonal violence between their parents and/or experience adult-perpetrated abuse

SIBLING VIOLENCE AND CHILD WELFARE 24

are more likely to engage in violent behavior with their siblings and peers (Button &

Gealt, 2010; Noland et al., 2004; Spaccarelli et al, 1994; Straus et al., 2006; Wallace,

1999; Wiehe, 1997). Social learning theory, based on the process of modeling or

imitation in response to observation, has been offered as an explanatory theory for the

multigenerational transmission of violence in families (i.e., aggression as a learned

behavior). In a seminal study on the imitation process, Bandura et al. (1963)

demonstrated that children who observed adults behaving aggressively toward a plastic

Bobo doll later imitated this behavior in their own play, whereas a comparison group of

children who did not observe aggressive behavior modeled by adults did not engage in

aggressive play toward the doll. Parental demonstrations of aggression may

surreptitiously communicate to children that violence is a morally-just means to resolve

conflict and achieve desired ends in relationship with family members (Caffaro & Con-

Caffaro, 1998; Straus et al., 2006).

Social learning theory also draws upon reinforcement principles in which a

behavior results in receipt of desired rewards. Between siblings, reinforcement could

occur by gaining control of a desired object, receiving parental attention, or through the

pleasurable experience of power resulting from a sibling’s fearful response (Hoffman &

Edwards, 2004). If behaviors are repeated without redirection or punishment, and

continue to produce desirable effects, violence becomes a patterned response (Walker,

1986). In accordance with social learning theory, numerous researchers have proposed

parental modeling as a salient explanation for sibling abuse (Hoffman & Edwards, 2004;

Kiselica & Morrill-Richards, 2007; Straus et al., 2006).

SIBLING VIOLENCE AND CHILD WELFARE 25

Radical feminist theory. According to radical feminist theory, male domination

in all sectors of society reflects broadly accepted oppressive attitudes toward women

which have resulted in various forms of gender inequality including male-female

violence (Button & Gealt, 2010; Hoffman & Edwards, 2004; Hoffman et al., 2005;

Weihe, 1997). Research supports the notion that the patriarchal organization of society

contributes to abusive sibling violence. Studies have demonstrated that brothers have the

highest rates of sibling violence, boys tend to commit more serious acts than girls, and

male siblings are more likely to abuse female siblings (Hoffman et al., 2005; Wiehe,

1997). In Steinmetz and Straus’ seminal study of violence in the American family, those

with only male children had consistently more sibling violence than families composed of

only girls, a statistical relationship that increased markedly as children grew older (2006).

More recently, feminist theories have taken on greater breadth, theorized to

include all power differentials between oppressors and oppressed, such as those based on

race, ethnicity, socio-economic status, and age (Wiehe, 1998). This expansion of feminist

theory is particularly useful when examining potential antecedents for violence within

families in that it is more inclusive of the myriad exchanges that can take place among

members. When applied to the problem of sibling abuse, feminist theory can account for

oppressive dynamics occurring as the result of any source of power differential between

children. In such exchanges the perpetrator is positioned as holding greater power and/or

control such as via physical strength, intellectual/emotional maturity, or level of

responsibility (Button & Gealt, 2010; Hoffman & Edwards, 2004). Wiehe (1997)

described sibling violence as generally entailing an older/bigger/more powerful sibling

SIBLING VIOLENCE AND CHILD WELFARE 26

abusing a younger/smaller/less powerful one in order to compensate for a perceived lack

or loss of power. For example, an older sibling required to babysit his younger siblings

might use excessive force to enforce bedtime rules (Wiehe, 1997).

Intersectionalities can create tenuous dynamics where the location of power is not

clear cut. In cases where older female siblings advocate for themselves on the basis of

age hierarchy, younger brothers may successfully challenge their authority with greater

physical strength (Hoffman & Edwards, 2004). Regardless of the characteristics of the

individuals involved, the party holding more power within the conflicting dyad gains a

sense of control through the act of overpowering another. Research supports blending

these notions of feminist theory as an explanation for abusive sibling violence.

Quantitative analysis has demonstrated that sibling pairs comprised of an older (and

presumably larger/stronger) brother and younger (i.e., smaller/weaker) sister were at

greatest risk for serious conflict (Graham-Berman et al., 1994).

Conflict theory. With roots in Marxism, conflict theory assumes that humans are

innately self-interested and that societies are prone to scarcity of resources. In tandem,

these human and societal conditions are posited to create dynamics in which people use

violence to resolve conflicts that stem from competing interests (Smith & Hamon, 2012).

Jetse Sprey is the theorist credited with applying conflict theory to families, focusing on

the conditions under which stability and instability occur (Smith & Hamon, 2012).

Sibling jealousy, competition for parental attention, the expectation that siblings share

games, toys, or other valued items, and disagreements over assigned chores have all been

discussed through the lens of conflict theory (Hoffman et al., 2005). Siblings subjected to

SIBLING VIOLENCE AND CHILD WELFARE 27

differential treatment are likely to view each other as competitors for tangible resources

and parental attention (Hoffman et al., 2005). Caffaro and Con-Caffaro’s (1998)

determination that parental physical or emotional unavailability and differential treatment

of children (e.g., scapegoating) were family conditions associated with abusive sibling

violence is supported by conflict theory.

Hoffman and Edwards (2004) proposed a theoretical model of sibling violence

and abuse that integrated social learning theory, feminist theory, and conflict theory, in

addition to risk factors addressed in the extant literature (Hoffman and Edwards, 2004).

Based on numerous linear relationships, the model organizes unidirectional lines of

causation between parent and child characteristics, attitudes, and behaviors hypothesized

to produce physical violence and psychological abuse between siblings. Figure 1 depicts

the relationships between the components of the model.

Figure 1. Integrated theoretical model of sibling physical and psychological abuse

(Hoffman and Edwards, 2004, p. 191).

Characteristics of the

Parents’ Relationship

Characteristics of the

Parent-Child

Relationship

Characteristics of the

Sibling Relationship

Individual Attitudes and

Characteristics

Sibling Verbal Conflict

Physical Violence and

Psychological Abuse

SIBLING VIOLENCE AND CHILD WELFARE 28

The researchers tested the model with 651 university students via survey methods

with measures of sibling violence were drawn from the Conflict Tactics Scale and

additional variables operationalized by the researchers (Hoffman et al., 2005; Straus,

1979). At each stage of regression analysis gender was a significant factor. Males with

brothers committed more types of sibling violence and more injurious behaviors (e.g.,

choking, using a weapon) than any other type of sibling pair. Favoring assigning chores

to siblings according to traditional gender roles and approval of using physical force in

sibling conflict were positively associated with sibling violence. The authors argued that

the results were supportive of a feminist theoretical explanation for sibling violence.

Social learning theory also garnered consistent empirical support in that arguments and

interpersonal violence between caregivers and adult-perpetrated violence toward children

were independently associated with sibling violence. Finally, parental favoritism was

associated with heightened sibling violence, evidence that conflict theory was also

supported by the model (Hoffman and Edwards, 2005).

An independent study implemented with a statewide sample of public school

students tested the applicability of feminist and social learning theories to explain sibling

abuse (Button and Gealt, 2010). Among more than 8,000 8

and 11

grade students

surveyed in 2007, 42 percent reported some form of sibling violence. Females were

significantly more likely to report being victimized than males and youth who identified

their parents as abusive to them or to have witnessed adult violence in the home were

significantly more likely to report sibling violence. In addition to generating further

evidence in support of feminist theory and social learning theory, victimization was

SIBLING VIOLENCE AND CHILD WELFARE 29

significantly associated with reports of substance use and delinquency in middle and high

school.

The work of Hoffman and Edwards (2004) and Button and Gealt (2010) makes

useful progress, identifying variables in a linear model theorized to predict abusive

sibling violence. While an essential preliminary step, their approach is based on the

dominant methodological orientation of the social sciences, which is implicitly biased

toward a reductionist perception of reality. Analysis seeks to break a construct or process

down into its component parts. Understanding the relationships between them (which are

assumed to be unidirectional and stable) is the focus of the inquiry. By analyzing abusive

sibling violence in this way, the theories and models described in previous sections

oversimplify a complex, synergistic, and dynamic socio-behavioral process. Subsequent

sections discuss one systems-oriented theory and a systems-based model of sibling

violence.

Coercive family process. Applying general systems theory to family behavior,

family systems theory posits that individual members can be accurately understood only

within the context of the whole family, including past generations. Rather than targeting

individual members as the source of dysfunction, the locus of family problems is viewed

as a function of struggle among members. Numerous forces are seen as moving in many

directions simultaneously, with positive and negative feedback loops guiding behavior.

Family members take on defined roles, repeatedly demonstrating a narrow set of

behaviors across situations, resulting in a relative equilibrium of patterned rules of

interaction (Smith & Hamon, 2012). The family is also viewed as contained by a semi-

SIBLING VIOLENCE AND CHILD WELFARE 30

permeable boundary with the environment (Crosbie-Burnett & Klein, 2009). The family

system is perceived as the nexus between structure and function (Zwick, 2015). The

structure of the family system is the organization of members; function is the holistic

behavior of the family, including how it relates to its context. The family system is

hypothesized to engage in relatively stable patterns of behavior over time.

Coercion theory was developed by scientists at the Oregon Social Learning

Center and has been the focus of numerous studies of childhood aggressive behavior.

Based on principles of operant conditioning and negative reinforcement, coercion theory

posits that learning occurs through interpersonal exchanges. Members of a dysfunctional

dyad within the family mutually “train” each other in an ongoing process which

reinforces difficult child behavior including aggression (Granic & Patterson, 2006;

Patterson, 1982). Research on coercion theory has largely focused on parents and their

children, studying a process in which caregivers inadvertently reinforce difficult child

behaviors through repeated, cyclical reactions of emotional withdrawal and giving in

(Smith et al., 2014). Children with more frequent behavioral difficulties such as

aggression (and who are therefore more challenging to parent) amplify coercive parenting

practices which in-turn solidify aggression as a child develops (Granic & Patterson,

2006).

Coercion theory can be applied to the whole family system to describe a coercive

family process in which emotional, cognitive, and behavioral feedback loops manifest

within and among multiple family members to produce multi-directional and synergistic

coercive exchanges. Applying the theory in this way includes all dyadic subsystems; just

SIBLING VIOLENCE AND CHILD WELFARE 31

as marital conflict is associated with compromised sibling relationship quality, the

parental relationship, or relationships between caregivers and children, may become

strained in the presence of ongoing sibling conflict (Milevsky, 2011). In a coercive

family process, violence between dyads (or triads, etc.) becomes a stable, family-wide

pattern.

Social ecological model of violence. Human behavior is typically discussed as

generated through interactions among personal and contextual factors, such as by the

“person in environment” orientation of social work. One of the more commonly applied

systems-based models is the ecological framework originally developed by

Bronfenbrenner (1979), which organizes society into a set of nested systems that interact

in a synergistic fashion (Figure 2).

When applied to the problem of violence, the values

and actions of individuals, families, communities, and society are viewed as both

reflective and generative of each other. Just as nations employ war tactics to secure

global resources and gangs engage in acts of violence to control illegal drug sales within

impoverished communities, family members and siblings use force to resolve

disagreements. The “second wave” of interpersonal violence scholarship has begun to

Figure 2 presents a limited set of concepts from Bronfenbrenner’s bioecological theory of human

development. The “mature” form of bioecological theory focuses on proximal processes at the center of the

Process-Person-Context-Time model. According to the model, proximal processes of progressively

reciprocal interaction between an individual and the persons, objects, and symbols in their immediate

external environment are the “primary mechanisms” in development. The environment, or context, is

comprised of the microsystem (e.g., home, school, or peer group in which the person spends a good deal of

time engaging in activities and interactions), exosystem (i.e., contexts which have important indirect

influences on development), macrosystem (i.e., the broad set of cultural values that encompass the group

under study and influence the developing individual). In the mature model, time is also an essential

element, which is also comprised of micro-, meso-, and macro- subcomponents. For a complete synthesis,

see Tudge, Mokrova, Hatfield, and Karnik (2009).

SIBLING VIOLENCE AND CHILD WELFARE 32

examine how mechanisms of violence perpetration co-occur across individuals and

systems (Centers for Disease Control and Prevention, 2014; Hamby & Grych, 2013). A

recent study of preschoolers found that father-child physical aggression interacted with

community violence to predict aggression between siblings (Miller, Grabell, Thomas,

Bermann, & Graham-Bermann, 2012).

Figure 2. Ecological model for understanding violence (adapted from the World Health

Organization, 2002, p. 12).

Subsequent sections will discuss interventions designed to address sibling violence at the

micro-, meso-, and macro-levels of the social ecology.

Sibling Violence Interventions

Micro- and meso-level interventions. A variety of interventions have been

developed to train both parents and children to more productively and peacefully navigate

sibling disputes. As early as 1967, clinicians were testing micro-level approaches to

reduce sibling aggression and increase cooperative play (Kramer, 2004). Early

Relationship

Community

Societal

Individual

SIBLING VIOLENCE AND CHILD WELFARE 33

approaches were largely reactive rather than preventative, aimed at eliminating conflict

by augmenting parent behaviors. Examples included training parents to utilize time-outs,

overcorrection, verbal reprimands, logical consequences, token reinforcement systems,

and other re-directive techniques. While intervention developers reported encouraging

results in terms of reduced sibling aggression severity and frequency, the rigor of

research assessing the effectiveness of these interventions was compromised by small

sample sizes and single group designs (Kramer, 2004).

In response to evidence suggesting that sibling violence occurs as the result of

systemic family problems, subsequent clinical interventions recommended a combination

of individual, sibling, family, and group therapy sessions incorporating family rules for

nonviolent behavior, guidance on incentive and consequence systems, role plays,

identification of enjoyable sibling activities, perspective-taking, skill building, and/or

facilitation of grieving over past trauma (Caffaro & Conn-Caffaro, 2005; Kiselica &

Morrill-Richards, 2007). For example, the task-centered sibling aggression (TCSA)

treatment model emphasizes helping parents and siblings to co-create rules for

controlling conflict during problem solving sessions which engage siblings in direct

negotiation processes (Reid & Donovan, 1990; Caspi, 2008). In a single case design,

Caspi (2008) describes three cases in which sibling aggression rates decreased and

caregivers and children reported satisfaction with the intervention.

Based on the premise that constructive conflict interactions teach children to

negotiate and take another’s perspective, Canadian researchers identified mediation as a

promising strategy that balances caregiver intervention with child autonomy (Siddiqui &

SIBLING VIOLENCE AND CHILD WELFARE 34

Ross, 2004; Smith & Ross, 2007). Mediation-based interventions train parents to serve as

facilitators who guide children through the conflict resolution process while concurrently

allowing children to decide how the disagreement is resolved. Over the course of two

experimental studies with general samples of siblings under the age of 11 and their

female caregivers, mediation training resulted in increased discussion about emotions and

the negotiation process by caregivers and children, increased reasoning dialogue among

both caregivers and children, more accurate understanding of the other’s perspective

among siblings, increased rates of child-initiated resolution, and less authoritarian

parenting practices (Siddiqui and Ross, 2004; Smith and Ross, 2007). The researchers did

not measure types or rates of physical or emotional violence.

Based on the determination that healthy sibling relationships are defined by

ambivalence (i.e., social interactions which rapidly shift between positive and negative

behaviors), that moderate levels of conflict promote the acquisition of beneficial social

and emotional competencies, and evidence suggesting that conflict-focused interventions

often result in a shift toward disengaged sibling relationships, Kennedy and Kramer

(2008) developed and tested a preventative intervention for siblings and their caregivers.

More Fun with Sisters and Brothers (MFWSB) is a 5-week program for 4-8-year-old

siblings that targets: initiating play with a sibling, methods for accepting and

appropriately declining an invitation to play, perspective taking, identifying and

discriminating among emotions, regulating emotions, problem solving, and conflict

management. Caregivers observe trainer-led sessions for siblings; a subsequent home

training session is designed to reinforce and transfer skill to the home environment. A

SIBLING VIOLENCE AND CHILD WELFARE 35

randomized trial of MFWSB found a significant increase in demonstrations of warmth

between siblings and a significant decrease in down regulation by caregivers in the

treatment group. The researchers did not report data on violence rates.

A similar, though considerably more intensive preventative intervention was

subsequently tested in a randomized trial of 174 5

grade children and their younger (i.e.,

to 4

grade) siblings (Feinberg et al., 2013). Designed to enhance siblings’

interpersonal skills and increase parental involvement in the sibling relationship, Siblings

are Special (SIBS) includes 12 90-minute afterschool sibling training sessions delivered

by pairs of trained leaders and three 2.5-hour family nights where parents learn how to

transfer the skills conveyed during the youth training sessions to the home environment.

Significant, beneficial program effects for child adjustment and dimensions of sibling

relationship quality were detected, as were gains in parent adjustment and parenting

quality. Similar to the studies conducted by Siddiqui and Ross (2004), Smith and Ross

(2007), and Kennedy and Kramer (2008), Feinberg et al. (2013) did not report on

physical or emotional violence rates. While one could argue that a strength-based,

positive focus on improving relationship quality is a viable approach to understanding

sibling violence, their methodological choices also fortify the problematic notion that all

sibling violence as normative.

In addition to interventions designed for the general population, two preventive

interventions designed for siblings in foster care are discussed in the literature: Promoting

Sibling Bonds (PSB), an 8-week program for maltreated sibling pairs ages 5-11 who are

co-placed and, Supporting Siblings in Foster Care (SIBS-FC) for sibling dyads ages 7-15

SIBLING VIOLENCE AND CHILD WELFARE 36

regardless of placement situation. PSB includes components for siblings and foster

parents which are informed by MFWSB, Sibling Plus Parent Management, a social

learning-focused intervention (Bank, Snyder, & Prescott, 2002; Bank & Kothari, 2013).

The component for foster parents provides positive child management training for sibling

aggression and mediation training for nonaggressive sibling conflict (Linares et al.,

2015). SIBS-FC is a 12-session sibling intervention curriculum designed to support

individual siblings’ socially skilled behavior and reduce sibling dyad-based conflict

(Kothari et al., 2014). The intervention engages siblings in 8 skill-building sessions,

weekly home activities (with foster parent collaboration), and four community activities

with project coaches. Other than supporting the home activities, the intervention does not

train foster parents. Both PSB and SIBS-FC were tested with randomized trials that

generated statistically significant results. Linares et al. (2015) measured sibling violence

with the Sibling Aggression Scale, a measure modeled after the Conflict Tactics Scale-2.

Foster parents in the intervention group reported lower sibling physical aggression from

the older toward the younger sibling than foster parents in the comparison group,

although the study was based on a small sample size. Kothari et al. (2017) did not include

a measure of sibling violence.

Whether reactive or preventative, multi-component interventions point to an

encouraging culmination of several decades of micro- and meso-level theory and

intervention research aimed at supporting healthy sibling interactions. By educating both

caregivers and youth in non-violent conflict resolution techniques these models have the

potential to address coercion and aggression throughout the family system.

SIBLING VIOLENCE AND CHILD WELFARE 37

Foster Parent Training (FPT) programs are another avenue for skill-building to

support safe sibling relationships in foster care. This would be an addition to most FPTs

in that they are largely focused on individual youth and/or their parents/families (Kothari

et al., 2017). While a subset of FPTs have demonstrated efficacy to improve foster parent

skill, many are not empirically supported, and they range widely in terms of their breadth

and depth (Solomon, Niec, & Schoonover, 2017). Improvement in FPTs, including

training to manage sibling violence, would likely be well-received. Foster parents have

expressed a need for skill building to address children’s behavioral problems (Spielfogel,

Leathers, Christian, & McMeel, 2011). At minimum, foster parents could be trained to

identify sibling abuse among foster children and provided with resources to access

additional support. Subsequent sections will turn to macro-level analysis of antecedents

and interventions to mitigate sibling violence.

Macro-level influences and interventions. Constraining consideration to the

micro- and meso-levels of the social ecology excludes the broader social context in which

children and families are nested. Feminist theory, initially conceived as a micro-level

dyadic phenomenon in which the masculine overpowers the feminine (e.g., intimate

partner violence), is now more broadly inclusive of all power differentials between

oppressors and oppressed (Wiehe, 1998). Applied to the macro-level of the social

ecology, feminist theory describes a process in which societal norms condoning the use

of power and oppression to achieve desired ends are transferred to, replicated within, and

maintained over time throughout all levels of the social ecology (Figure 3).

SIBLING VIOLENCE AND CHILD WELFARE 38

Figure 3. Ecological model for understanding violence (adapted from the World Health

Organization, 2002, p. 12).

When examined from a macro-level perspective, interventions to address inter-

personal violence, of which sibling abuse is a subset, are likely to be most effective if

designed to change social and behavioral norms that support the use of force to resolve

conflict at all levels of the social ecology. In 2009, the World Health Organization

published a briefing which advocated for mass media campaigns as a promising approach

for youth violence prevention. Mass media campaigns communicate information to broad

populations via print, radio, and television to correct misperceptions about norms, attach

a social stigma to unwanted behavior, or promote desirable behaviors through positive

appeals (World Health Organization, 2009).

In a recent and exhaustive review of mass media campaigns to change health risk

behaviors including tobacco use, sex-related behaviors, road safety, and child safety, the

authors determined that campaigns can produce desired change across large populations

(Wakefield, Loken, & Hornik, 2010). The authors specify, however, that mass media

Relationship

Community

Societal

Individual

Societal Patriarchal Values Orientation

SIBLING VIOLENCE AND CHILD WELFARE 39

campaigns are most effective when multiple marketing methods target non-habitual

behaviors with concurrent access to key resources and policy-level support and

enforcement. Additional recommendations to support mass media campaign effectiveness

include ensuring sufficient exposure to messaging, drawing on social marketing theory to

guide message development, and creating a supportive environment that allows the target

audience to make the desired change (Randolph & Viswanath, 2004). Creating an

environment supportive of non-violence messaging would be particularly difficult given

an overall culture in the United States which approves the use of power and sees power as

a masculine prerogative. There are signs that this is changing, including the current

#MeToo social media hashtag, removal of prominent men, particularly in Hollywood, for

widespread sexual misconduct allegations, and introduction of Senate and House bills to

combat sexual harassment on Capitol Hill.

Holistic, coordinated violence prevention intervention. The recommendations

accompanying evidence of the utility of mass media campaigns highlight the importance

of incorporating social-ecological (i.e., systems-oriented) principles to their design and

implementation. In alignment with this perspective, current violence prevention literature

advocates for interventions that systematically address factors contributing to

involvement in multiple forms of violence, account for developmental trajectories, and

address history of violence exposure (Hamby & Grych, 2013). As an example of a

possible intervention model, Hamby & Grych (2013) introduce a strategically

coordinated and holistic school-based prevention initiative which begins in elementary

school and is then “revisited, expanded, and tailored to different topics” as children

SIBLING VIOLENCE AND CHILD WELFARE 40

develop throughout middle and high school (p. 83). Reproduced in Figure 4, the design

combines numerous methods supported throughout the sibling abuse and interpersonal

violence literature and intervenes at the micro-, meso-, and macro-levels of the social

ecology. The center circle in the figure shows ongoing classroom curricula presented

across grade levels, media campaigns, and therapeutic adjunct and community

monitoring services. Peripheral circles in the diagram represent content-specific

intervention subcomponents implemented at developmentally appropriate stages to

address specific forms of interpersonal violence.

Figure 4. Coordinated violence prevention model: Hypothetical common and specific

elements (Hamby and Grych, 2013, p. 85).

Although the model remains to be piloted, a similar though less-intensive school-based

anti-violence intervention developed at Colorado State University demonstrated strong

evidence of impact over a 2-year implementation period (Swaim & Kelly, 2008). Resolve

Core classroom & social marketing curricula

content for all violence types:

Coping with prior victimization

Role of bystanders in violence prevention

Promoting self-regulation

Communication skills

Problem-solving

Promoting empathy, respect, & egalitarian values

Refusal skills

Integrating Therapeutic Adjunct Services with

Social Marketing or Classroom Curricula:

Coordinating family-centered services with school

& community services

Increasing security/adult presence in violence

hotspots

Specific

content for

dating

violence

Specific

content for

gang and

delinquency

prevention

Specific

content for

sexual

assault

Specific

content for

bullying &

peer

aggression

SIBLING VIOLENCE AND CHILD WELFARE 41

it, Solve it combines a mass media campaign developed by high school youth selected

and trained as non-violence trainers/role models combined with middle school

assemblies, classroom presentations, annual community events, and ongoing distribution

of promotional items. When tested in a randomized trial in rural communities in five

states, treatment site participants demonstrated statistically significant recall of media

campaign messaging, a faster rate of decline in physical violence, a significant decline in

verbal victimization, stable ratings of school safety (which declined for controls), and a

near-zero change in intentions for violence (while controls increased). Resolve it, Solve it

could generate stronger outcomes if expanded to reflect the recommendations made by

Hamby and Grych (2013).

The anti-violence literature reviewed here provides a wealth of instructive

guidance to ameliorate sibling abuse through both preventative and corrective

intervention. None of the studies mentioned inclusion of youth voice in their

development, however, a significant gap in efforts to address the phenomenon. Moreover,

just two studies have endeavored to intervene into relationship quality between siblings in

foster care, one of which focused specifically on sibling violence. Chapter 3 presents the

philosophy and methodology for the study, designed to build sibling violence theory in

the context of child welfare by incorporating the perspectives of foster care alumni.

SIBLING VIOLENCE AND CHILD WELFARE 42

Chapter 3: Methodology

The study was grounded in critical realism (CR), a philosophy of science

originating from the work of Bhaskar (1979, 1998) and then further developed by Sayer

(1992), Archer (1995), and others (Fletcher, 2017). The sections to follow present two

predecessors to CR, positivism and constructivism, with focus on their limitations in

causal analysis. After positioning their limitations as rooted in the “epistemic fallacy”

(Bhaskar, 1998, p. 27), CR is discussed as an alternative philosophy of science which

effectively delimits ontological and epistemological inquiry and opens the study of causal

mechanisms to qualitative methodology. The chapter concludes with the methodology for

the current study which employed in-depth interview methods and deductive analytic

techniques to investigate the causal mechanisms of abusive sibling violence and develop

recommendations for ameliorative action in the context of child welfare and beyond.

As a philosophy of science, positivism relies on a realist/objectivist ontology

which assumes the existence of an objective reality that is separate from our perceptions,

theories, and constructions of it (i.e., Nagel’s argument that there is, indeed, “a view from

nowhere in particular” (1989)). Based on Humean causal law, positivism employs

reductionist methods to parse a theoretically knowable reality into discrete, measurable

variables for statistical analysis (Fletcher, 2017; Lennox & Jurdi-Hague, 2017). The

researcher is positioned as objective and positivist study is presumed immune to

researcher bias by employing “rigorous” methodologies impervious to external influence.

Studies of this type are designed to describe associations or predict temporally bound,

linear relationships (i.e., if event x, then event y). Positivist research may predict that an

SIBLING VIOLENCE AND CHILD WELFARE 43

independent variable (or variables) predict dependent variables in a theoretical model, but

the causal mechanisms underlying the relationship(s) (i.e., how one variable influences

the other) remain elusive (Fletcher, 2017; Lennox & Jurdi-Hague, 2017).

Constructivism followed positivism,

a relativist/interpretivist ontology capable of

augmenting the limitations of positivism by gathering descriptions via verbal and

nonverbal symbol systems, primarily through qualitative research methodology. Based on

individuals’ lived experiences of social reality, constructivism proposes that human

experience is time- and context-dependent and can only be understood subjectively

(Morris, 2006); there are literally “different world versions” which are elicited during

dialectic exchanges (Schwandt, 1994, p. 126). Emphasizing social actors, constructivism

conflates individual experience with theoretical explanation, and as a result, largely

avoids the realm of causation. According to a constructivist philosophy of science, child

maltreatment “is more like pornography than whooping cough. It is a socially constructed

phenomenon which reflects values and opinions of a particular culture at a particular

time” (The British Department of Health, 1995 as cited by Houston, 2001, p. 848).

Considering this example, the limitations of constructivism for the social sciences are

obvious. The idea that many equally legitimate causes produce a given social

phenomenon in an infinite number of unique contexts makes determination of corrective

action extremely difficult.

This is an oversimplification. Numerous philosophies were generated in the time between positivism and

constructivism, and afterward. For more in-depth discussion of the philosophies of science, see Kuhn’s

(1962) The Structure of Scientific Revolutions, Morris’ (2006) Social Work Research Methods, and Brown

and Strega’s (2005) Research as Resistance: Critical, Indigenous and Anti-oppressive Approaches.

SIBLING VIOLENCE AND CHILD WELFARE 44

More recently, critical realism (CR) has gained traction in the social sciences as a

potentially generative alternative to positivism and constructivism. A key tenet

differentiating CR from its predecessors is the clear delineation of ontology and

epistemology. To the critical realist, reducing ontology to epistemology is a philosophical

error that unnecessarily constrains the research endeavor. CR separates the two,

employing a realist ontology and a relativist epistemology. Ontologically, CR holds that

there is a “real world” that our perceptions, theories, and constructions refer to, yet

epistemologically, our representations (which are also part of reality) are deemed fallible

given the constraints of human sensory capacity to fully grasp the complexity of social

systems. The notion of variant perspectives on reality (i.e., a constructivist epistemology)

is compatible with this realist ontology. Leaning heavily on explanatory theory, the

critical realist is free to employ constructivist, qualitative methodologies to investigate

ontological questions to more fully develop causal explanations for what occurs in the

“real” world. CR is emancipatory, with the goal of clarifying the causal mechanisms that

produce human suffering and identifying logical, ameliorative actions (Fletcher, 2017;

Maxwell, 2012).

Realist Ontology

According to CR, reality is stratified into three levels or domains, a “deep,”

differentiated ontology capable of overcoming the epistemic fallacy. The empirical

domain is comprised of observable events which are experienced with the five senses.

These events may be causal and are understood through the filter of human interpretation.

The actual domain is that in which all events occur, regardless of human sensory capacity

SIBLING VIOLENCE AND CHILD WELFARE 45

to detect them. These events occur regardless of whether they are observed (i.e., Nagel’s

“view from nowhere”). Finally, the real domain is comprised of the causal structures or

mechanisms underlying actual and empirically sensed events. These mechanisms are the

causal processes underlying a given phenomenon occurring in the actual and perceived in

the empirical domains. The three domains are not separate structures, but rather,

comprise a singular reality. They interact synergistically such that human interactions at

the empirical level can influence the causal mechanisms at the real level or vice versa

(Bhaskar, 1979; Danermark, Ekström, Jakobsen, & Karlsson, 2002). Fletcher (2017)

represents the three levels metaphorically as an iceberg. The real and actual levels are

below the waterline, unseen, whereas the empirical level is above the waterline,

perceptible by the observer with one or more of the five senses (Figure 5).

Figure 5. CR domains of reality presented as an iceberg.

In addition to the three domains of reality, a second key tenant of CR is the

interplay between structure and agency, a concept widely discussed in the systems

Empirical Domain: observable events

Actual Domain: All events,

regardless of observation

Real Domain: causal structures and

mechanisms underlying actual and

observable events

SIBLING VIOLENCE AND CHILD WELFARE 46

literature (Archer, 2010; Zwick, 2015). Bates (2006) represents this relationship as a

double-helix in which one arm symbolizes human agency and the other social structure.

The rods connecting the two arms of the helix are variant aspects of opportunity and

constraint, which change over time (i.e., the turning of the helix). Human agency is seen

as emergent and semi-autonomous from the underlying social structures that produce

empirically determinable patterns, yet human agency may also influence the conditions

that produce those patterns. Humans located within a given social system have choice in

their behaviors, and yet those behaviors are constrained by system structures which limit

their latitude of choice to some degree (Wright, 2011). Lennox and Jurdi-Hage (2017)

effectively concretize these metaphoric and theoretic representations in a study of street

harassment. Individual agents (i.e., women) are positioned as residing in a patriarchal

social system structure that condones physical and verbal assault in public spaces.

Returning to Bhaskar’s three-domain representation of reality, patriarchy is theorized as

the causal mechanism at the real level that produces gender socialization in the actual

level and which is perceivable at the empirical level (e.g., cat calls, physical

transgressions, etc.). Women’s agency to respond to harassment is constrained by fear of

retaliation but may also be amplified through new opportunities such as access to social

media platforms used for anti-harassment activism. Lennox and Jurdi-Hage’s example

effectively conveys how constraints and opportunities vary across time and contexts.

Once identified, they can inform emancipatory actions to dismantle oppressive system

structures.

SIBLING VIOLENCE AND CHILD WELFARE 47

Constructivist Epistemology

While constructivist ontology argues that reality is entirely socially constructed

through and within human knowledge or discourse (Fletcher, 2017), CR employs

constructivism as an epistemology to explore causal mechanisms in the real domain. The

critical realist assumes that empirical observations are influenced by the observer’s

perceptual filters (e.g., positionality, theoretical stance) while concurrently engaging a

realist ontology that assumes phenomena exist in the actual and real domains, beyond our

ability to experience them directly. The critical realist is unwilling to allow empirical

limitations to undermine the utility of the research endeavor to understand causal

mechanisms and identify emancipatory actions. Constructivist representations are

assumed to point to real things in a highly complex social world. Variation among human

representations of a given phenomenon are not problematic, but rather, are helpful to

identify patterns of action or behavior for a given object or structure (Danermark et al.,

2002; Fletcher, 2017; Wright, 2011). Critical realists actively engage with theory in a

dynamic analysis process which traverses the concrete and the metaphysical to develop

increasingly representative theories of reality. Their understanding of current scientific

knowledge on the topic is treated as both valuable and useful to explore multiple extant

theories, all of which are assumed to be both feasible and fallible (Bhaskar, 1979;

Fletcher, 2017). Empirical constructions can also challenge existing scientific knowledge

and theory (Redman-MacLaren & Mills, 2015 as cited in Fletcher, 2017). In the CR

practice of rational judgement, the researcher may elaborate on or deviate from

SIBLING VIOLENCE AND CHILD WELFARE 48

participants’ interpretations to ‘provide fuller or more adequate interpretations” (Parr,

2015, p. 10) of a given object or structure.

In summary, critical realists concern themselves with explaining what is occurring

in the real domain, utilizing the interplay between empirical data and extant explanatory

theory. By actively engaging with both the concrete (empiricism) and the abstract

(theory), critical realists take tenets from both the positivist and constructivist

philosophies of science without succumbing to their associated limitations (Lennox &

Jurdi-Hage, 2017). By describing not only what is taking place, but also how it manifests

(i.e., employing theory to more and more accurately approximate reality), CR supports

critique of the social conditions that produce empirical level patterns of human suffering

including social, political, and economic causes (Fletcher, 2017; Houston, 2001; Wright,

2011). This aspect of CR fits well with the grounding ethos of Social Work given its

potential to identify emancipatory solutions to social problems (Danermark et al., 2002;

Houston, 2001). The sections to follow describe the methodology for the study, grounded

in a CR philosophy of science.

Methodology and Methods

Critical Realism does not offer clear methodological guidance, but rather, tends to

advocate for a mix of quantitative and qualitative methods (Danermark et al., 2002).

Research questions should support empiricism that interrogates the utility of extant

theory, which may be denied, supported, or elaborated upon within a particular context

(Fletcher, 2017). The linear and non-linear theories examined in this project included:

social learning theory, radical feminist theory, conflict theory, and coercive family

SIBLING VIOLENCE AND CHILD WELFARE 49

process theory. A social ecological model was also incorporated into the analysis.

The

research questions guiding the study, replicated from Chapter 1, are presented below.

1) How do foster care alumni describe their experiences with physical and emotional

sibling violence?

a) To what extent do their descriptions fit with Caffaro (2014) and Meyers’

(2015) definitions of abusive sibling violence?

b) To what extent are family conditions and theories of sibling violence

discussed in the extant literature represented in their descriptions?

c) What additional/refined causal mechanisms do their descriptions suggest?

2) From the perspective of foster care alumni, how do adults who care for or work with

young people involved in the child welfare system (e.g., foster parents, case workers,

mental health providers, kinship caregivers, etc.) respond to sibling violence?

a) To what extent are adult responses perceived as helpful?

b) What do young people recommend to address sibling violence?

Study Pilot

In preparation for the study I engaged with a youth-driven advocacy and support

program for youth in care with the goal of piloting the study and gaining assistance

recruiting participants for the current project.

The Executive Director was open to these

The social ecological model locates an individual within a set of nested systems that interact in a

synergistic fashion. The environment, or context, is comprised of the microsystem, exosystem, and

macrosystem. In the mature model, time is also an essential element, which is also comprised of micro-,

meso-, and macro- subcomponents (Tudge et al., 2009). Although not a theory, the model is helpful to

interpret interactions within and across levels of the social ecology that are hypothesized to produce

interpersonal violence. The model is also well aligned with the “person in environment” orientation of

Social Work.

I have concealed the name of the agency to maintain study participant confidentiality.

SIBLING VIOLENCE AND CHILD WELFARE 50

ideas and after negotiating a contract with clearly articulated tasks, I worked with the

Program Manager to plan a one-day site visit to carry out the pilot with young people

participating in a summer internship program focused on leadership development. I

planned to use the pilot study to improve upon the design by testing the consenting

process and data collection instruments and facilitating a group conversation to gather

feedback about the research questions and data collection process.

The week prior to the scheduled pilot study date the Program Manager informed

me that the interns had opted not to participate. While she alluded to events occurring

within the program that were creating disruptive dynamics among the interns, she and the

Executive Director also recommended that I enlist a group of study advisers from their

national network of “young leaders” to improve the project (i.e., youth who had aged out

of care and were achieving academic, professional, and social success in their lives). I

embraced their recommendation and over the course of the subsequent year I advertised

for the opportunity through the program’s various social networks and other

communication modalities, reviewed the resumes and interest statements of more than 20

candidates, conducted interviews with a subset of candidates whose background included

physical and emotional violence with siblings, and ultimately enlisted four advisers to

provide consultation on the study. In total the advisers participated in four conference

calls, the content of which is summarized in Table 3.

As shown in the right-most column of the table, the advisers affirmed the utility

of several aspects of the planned design (i.e., research questions, systems concepts, and

the questions in the data collection instruments). However, I revised two key aspects of

SIBLING VIOLENCE AND CHILD WELFARE 51

the study because of our conversations. First, it became clear that the advisers, who

included both self-identified victims and perpetrators of sibling violence, were eager to

talk about their experiences. The advisers demonstrated through their own sharing during

the calls and their study design recommendations that I should ask directly about what

had occurred during violent exchanges and allow ample time for study participants to

share about their experiences in detail. This was particularly meaningful given that the

Executive Director and Program Manager had communicated a strong concern about the

study re-traumatizing study participants, recommending that I avoid asking young people

to disclose the details of violent episodes.

Second, the advisers were very inquisitive about my interest in the subject,

inviting me to discuss my experiences with sibling violence during the second call. I

followed their lead, sharing about the types of violence I experienced and the ways they

had impacted my life into adulthood. I did not dwell in the details of individual acts or

events, nor share anything I was not ready to disclose, maintaining an interpersonal

boundary that balanced courage and self-care. After I finished providing this information

about my background, the advisers offered supportive condolences and appreciation for

my choice to study the topic.

This early exchange with the advisers made it clear that creating a safe space to

talk about sibling violence should begin with my own honest sharing about my

background and interest in the topic, a conclusion that the advisers solidified during the

final call by recommending that I build more information about myself into the

introduction section of the interview protocol. While the young people did not state this

SIBLING VIOLENCE AND CHILD WELFARE 52

explicitly, my takeaway was that personal disclosure could serve to disrupt the

researcher-participant power imbalance (Ross, 2017), while concurrently allowing me to

model a level of vulnerability appropriate to the research endeavor.

Because of these learnings, I determined that individual interviews, rather than

focus groups, would be a more appropriate method of data collection for the project. This

determination is in alignment with the peer-reviewed literature which offers numerous

examples of in-depth interviewing to gather data on sensitive topics including bullying,

high-stakes decision-making, and physical, emotional, and sexual sibling abuse (Allen,

2013; Cranley, Doran, Tourangeau, Kushniruk, & Nagle, 2012; Keddell, 2011; Meyers,

2014; Rowntree, 2007).

Table 3

Sibling Study Adviser Conference Call Summary

Call

Agenda

Study Implications

Call 1

Welcome

Introductions

Roles and responsibilities

N/A

Call 2

Establish an agreement for our work together

Discuss our interest in the topic

Young people want to share their stories

about sibling violence; retraumatization

is not a concern. Ask directly about

violent episodes.

Call 3

Discuss the research questions for the study

Discuss systems concepts

Research questions are on-target

Systems concepts resonate as applicable

Call 4

Discuss the interview protocol

Discuss data collection logistics

Importance of researcher self-disclosure

Individual interviews (not focus groups)

It is noteworthy that the timeline from my first contact with the agency through

completion of the pilot spanned 18 months. This lengthy phase of engagement generated

numerous benefits: (1) I was able to demonstrate my commitment to the project, gaining

credibility as a student researcher with Executive Director, Program Manager, and foster

SIBLING VIOLENCE AND CHILD WELFARE 53

care alumni; (2) I gained experience with youth-informed research; (3) I learned about

the agency, including the Executive Director and Program Manager’s interests and

perceptions about research; (3) I brainstormed with the Executive Director and Program

Manager about new ideas for the organization to support young people to gain leadership

experience as research advisers and research-informed policy advocates.

After concluding engagement with the study advisers, I conducted a pilot study

with four young people recruited through the agency’s summer internship program. I

obtained IRB approval for the pilot study and the data were included in the analysis. The

sections to follow describe the methods for the study, which were implemented

successfully during the Summer 2017 pilot and applied in the same manner for the

remaining interviews, completed the summer of 2018. Table 4 summarizes the study

process, including the study event, date, the individuals involved, and the count. Eight

interviews were included in the analysis, conducted during the “successful pilot” and

“final data collection” events.

Table 4

Study Process Summary

Event

Date

Individuals

Involved

Count

Planned pilot Summer 2016 N/A 0

Study adviser engagement

Winter 2017-present

Study Advisers

Successful pilot: Data

included in final analysis

Summer 2017 Interview Participants 4

Final data collection Summer 2018 Interview Participants 4

SIBLING VIOLENCE AND CHILD WELFARE 54

Study participant recruitment and consent process. Study participants were

recruited through the youth-driven advocacy and support program for foster youth

described in the previous section, by disseminating a recruitment flyer through its e-

communication system and during two summer internship program sessions. To further

increase the likelihood of a robust response, I included the contact information for three

sibling study advisers on the recruitment flyer who agreed to serve as references for me.

This gave interested respondents the opportunity to speak with a peer about their

experience working with me prior to contacting me directly. A copy of the recruitment

flyer is contained in the Appendix.

Study participants were recruited to meet the following criteria: (1) foster care

alumni ages 18-24 years old; (2) with at least one (self-defined) sibling; (3) and for whom

the study participant describes their sibling relationship as having been physically or

emotionally violent in the period leading up to or during their involvement in foster care.

I recruited a convenience sample of participants who reported emotional and/or physical

violence within biological and/or foster family/kinship care settings. To accommodate the

familial diversity represented among those served by child welfare, and to avoid

erroneously privileging heteronormative, white, western family structures, self-defined

siblings included biological full- and half-siblings, adoptive and step-siblings, as well as

fictive kin and social siblings (Walker, Allen & Connidis, 2005). The inquiry focused

primarily on physical and emotional violence because sexual violence is already a widely

acknowledged (and more clearly defined) form of abuse between children and youth.

Alumni who reported sexual abuse in addition to physical and emotional violence were

SIBLING VIOLENCE AND CHILD WELFARE 55

not be excluded, however, since the study pilot and previous research (Hamby & Grich,

2013) have demonstrated that physical and emotional violence may co-occur with sexual

violence.

The flyer initially directed alumni to contact me by telephone or email to express

their interest and learn more about the study. One participant contacted me following

dissemination through the agency’s e-communication network. I scheduled an initial

telephone call with the potential participant to share more about the project and field her

questions. After agreeing to participate, we identified a time and location to meet for the

face-to-face interview. I recruited the remaining study participants during two sessions of

the 2018 summer internship program at the youth-driven advocacy and support program

for foster youth. I completed the final consent process for all the study participants

immediately prior to conducting the interview. A copy of the consent form is contained in

the Appendix.

Data collection. According to CR, data collection occurs in the empirical domain

of reality. Intensive data (i.e., in-depth interpretive data) are gathered from individual

people to learn how particular mechanisms manifest within the specific contexts of their

lives (Bhaskar, 1998). Intensive data for this study consisted of semi-structured

interviews with foster care alumni to explore theories of sibling violence in the context of

child welfare. CR also advocates for complementary collection of extensive data (i.e.,

data on widespread trends, typically gathered with quantitative data) to look for

systematic differences across a population (Lennox & Jurdi-Hage, 2017; Sayer, 2010).

While collection of extensive data would be an ideal accompaniment to the primary in-

SIBLING VIOLENCE AND CHILD WELFARE 56

depth interview methods described here, no such data are currently collected from child

welfare-involved youth and to include such an approach in the current study was beyond

the scope of a feasible dissertation. Ideally, the intensive data gathered in this project will

inform a subsequent extensive approach.

Intensive data collection included three components: (1) creation of a family map;

(2) a semi-structured interview, and; (3) completion of a background information form.

The interviews began with the participant drawing a series of family maps, graphic

representations of the family members present over the time period when the most

pronounced sibling violence occurred (see Appendix). The purpose of the map was

multifold. First, it helped me to understand the structure of the young person’s family

when they were experiencing the most significant sibling violence. For young people

involved in the child welfare system, family organizations can be highly complex,

including a variety of parents, step-parents, other adult caregivers such as grandparents,

aunts and uncles, as well as full-, half-, fictive kin and social siblings. Participants drew

multiple iterations of the map to represent various structures of biological and foster

families over time. These visual representations helped me to quickly learn participants’

family formations. The maps also allowed me to easily refer to individuals by name or by

pointing to their location on a map, and to probe about dynamics between/among dyads,

triads, or other more complex combinations of family members with a focus on their

influence upon violent exchanges. Once the participant had mapped the family

structure(s), I carried out a semi-structured in-depth interview with questions designed to

gather information on the following topics:

SIBLING VIOLENCE AND CHILD WELFARE 57

1. Relationship(s) with sibling(s), including physical and emotional violence

2. Relationships with other family members, and their relationships with each other

3. The community/communities where the respondent lived when they experienced

sibling violence

4. The things that adults did (or did not) do to address sibling violence

5. Ideas to help young people who experience sibling violence

A copy of the interview protocol is contained in the Appendix. With participant

permission, the interviews were recorded to accurately capture the conversations. I

transcribed the recordings to generate rich field texts (Butler-Kisber, 2010) for

subsequent review and analysis.

After completing each interview, the participant was invited to complete a

background form to gather basic demographic information (i.e., age, gender, sexual

orientation, ability status, race/ethnicity) and determine their willingness to be contacted

for future research (see Appendix). This information was collected for descriptive

purposes only; establishing representativeness of the sample was an inappropriate goal

given the constructivist epistemology grounding the study.

Data analysis. The data analysis process was based on the three key steps

prescribed in CR: identification of demi-regularities, abduction, and retroduction

(Fletcher, 2017). The process was non-linear, moving from the concrete, to the abstract,

and then returning to the concrete (Sayer, 1992). Each step is described in more detail in

the following sections.

SIBLING VIOLENCE AND CHILD WELFARE 58

Identification of demi-regularities occurs through qualitative data coding (i.e.,

analytical resolution) (Danermark et al., 2002). The data are used to identify rough trends

or broken patterns of common human behavior. Given the theory-laden CR grounding of

the study, I employed Mayring’s (2000) deductive category application process, which

works with theoretically derived codes to organize textual data. The codes for the initial

coding scheme were derived from the sibling violence definitions, family conditions, and

linear and non-linear theories discussed in the literature. Additional codes captured

examples of physical and emotional sibling violence in biological and foster family

contexts, the point at which sibling violence ended, actual and recommended

interventions, and sources of resilience to navigate sibling violence. I conducted and

coded interviews until I determined that the data corpus was adequate to generate

information to extend and advance knowledge in relationship to the literature-derived

definitions, family conditions, and theories in the context of child welfare.

Applying Mayring’s coding agenda, each code included a clearly articulated

definition, coding rules articulating the circumstances under which a given code was to

Data saturation continues to be a contentious topic among qualitative researchers. An article by Morse

(1995) positioned saturation as the hallmark of a valid qualitative study. To provide concrete guidance as to

the number of interviews needed to achieve data saturation, Guest, Bunce, and Johnson (2006) documented

the progression of theme identification after each set of six interviews in a study of 60 sex workers in

Africa. The researchers concluded that 6-12 interviews would likely be adequate to identify meaningful

themes and useful interpretations in studies with homogenous samples, high quality data, and a clearly

defined domain of inquiry. O’Reilly and Parker (2012) argue that thematic saturation is not always an

appropriate marker of quality assurance, but rather, that quality is inextricably linked to the epistemology

undergirding the project. According to O’Reilly and Parker, quality should be determined with

consideration for the nature of the topic and the resources available to conduct the study. In cases where

saturation is not achieved, the authors cite the position of Morse (1995): this simply means that the

construct has not been fully explored rather than that the findings are invalid. If a study generates rich

information and extends or advances knowledge, it is deemed successful despite falling short of the

saturation ideal. In such cases, transparency about any limitations to achieving saturation should be

included with the findings.

SIBLING VIOLENCE AND CHILD WELFARE 59

be applied to a text unit, and example text units. Through constant comparison analysis I

identified areas of overlap/duplication among the family condition codes, reducing the set

from seven to four. I then identified areas of overlap/duplication between the family

condition codes and the linear theories, recoding text units representing aspects of the

linear theories not subsumed within the family conditions. Finally, I worked back and

forth between the family condition and linear theory codes and the non-linear theory and

model codes, determining that the non-linear theory and model offered unique insight

into sibling violence in the context of child welfare. As such, the non-linear theory and

model codes were retained as separate codes. A copy of the final coding scheme (codes,

definitions, and coding rules) is included in the Appendix. Example text units for the

codes are presented in Chapter 4 (Results).

After identifying the demi-regularities I engaged in abduction and retroduction,

two CR thought exercises designed to elucidate meaningful causal theory that

incorporates not only underlying structures, but also the human actions that fortify or

dismantle them. Abduction is a process of “inference or thought operation, implying that

a particular phenomenon or event is interpreted from a set of general ideas or concepts”

(Danermark et al., 2002, p. 205). In this study, abduction consisted of a holistic analysis

process to determine how the demi-regularities (i.e., final codes) were represented within

participants’ family systems with the goal of identifying patterns of interaction both

within and across the families represented in the data. To do this, I recreated the family

maps for each participant, identifying the gender of each member and their familial

relationship to the study participant (mother, brother, nephew, etc.). I labeled three traits

SIBLING VIOLENCE AND CHILD WELFARE 60

associated with the family conditions and theories (i.e., drug use, alcohol use, and mental

health challenges) for individual members, and identified the presence and direction of

physical and emotional violence between members. Beneath each family map, I

summarized the key interview content associated with the time-frame (i.e., participant

age range) represented in the map. These summaries included family member

characteristics, examples of physical and emotional violence, patterns of interaction, and

critical events, such as those that triggered reorganization of the family structure. I also

drafted brief explanations of how the definitions and theories were represented in data,

along with the text unit(s) associated with them. I concluded abduction by determining

that a non-linear theory focused on the family level of the social ecology was the best

representation of the phenomenon in the data. An example of a de-identified family map

created during this phase of analysis is included in the Appendix.

The last phase of analysis was the CR process of retroduction, where I explored

the “real” causal mechanisms necessary for abusive sibling violence to manifest at the

empirical level. Retroduction focuses on theorizing about the social, cultural, and/or

historical conditions in which a causal mechanism takes effect (Fletcher, 2017; Lennox &

Jurdi-Hage, 2017). Retroduction includes people’s knowledge, and their reasons or

motivations for doing things, which are treated as real causal mechanisms that can

dismantle or fortify the objects or structures influencing their lives. I incorporated the key

CR concepts of structure, agency, and open systems in the retroduction process, working

to identify the causal mechanisms underlying abusive sibling violence (i.e., domain of the

“real”) in the context of child welfare. I applied the questions developed by Danermark et

SIBLING VIOLENCE AND CHILD WELFARE 61

al. (2002) in this phase of analysis: (1) What is fundamentally constitutive for the

structures and relations of abusive sibling violence? (2) How is abusive sibling violence

possible? (3) What properties must exist for abusive sibling violence to be what it is? (4)

What causal mechanisms are related to abusive sibling violence? I applied this thought

exercise to the child welfare system, with consideration for the broader system conditions

(i.e., social, political, and economic) that bear upon its structure and function.

Establishing Trustworthiness

In this study I served as the “instrument” in that I devised and refined the research

questions, conducted the interviews with the study participants, and completed the data

analysis and synthesis. According to Finlay (2002), the practice of reflexivity, or

“thoughtful, conscious self-awareness… encompass[ing] continual evaluation of

subjective responses, intersubjective dynamics, and the research process itself” (p. 532) is

essential to ensuring the integrity and trustworthiness of a study. To ensure

trustworthiness, I engaged in a variety of activities in partnership with my academic

advisers, colleagues, and peers.

After transcribing the data, I coded the interviews, preserving copies of data

synthesis documents (i.e., participant-specific and aggregated text unit summary tables)

to make it possible to follow the process of data analysis and synthesis back to the

original transcripts. I engaged members of my committee as auditors who reviewed the

coding scheme guiding the initial phase of analysis (i.e., identification of demi-

regularities). At appropriate junctures over the course of data collection and analysis, I

again engaged members of my committee to review the intermediate synthesis documents

SIBLING VIOLENCE AND CHILD WELFARE 62

(i.e., family map reproductions with text units). I drafted reflexive memos to explore how

my own conceptions were influencing my interpretation of the data. I also debriefed with

my committee members, student peers, and professional colleagues to discuss the

preliminary findings and the issues that arose in the memos.

Ethical Considerations

By nature of being in the child welfare system, the foster care alumni who

contributed to the study they have experienced significant trauma, including removal

from their family of origin into out-of-home care. They represent a “vulnerable”

population in that they are at increased risk for adverse social outcomes including

homelessness, un-/under-employment, substance use/abuse, and physical and mental

health challenges (Courtney, Piliavin, Grogan-Kaylor, & Nesmith, 2001; Lenz-Rashid,

2004; Lopez & Allen, 2007; Nicoletti, 2001; Reilly, 2003). Moreover, they may have

experienced negative interactions with a variety of well-intentioned helping professionals

who may have dismissed their allegations (McDonald & Martinez, 2016), made

seemingly illogical decisions about their care (Hyde & Kammerer, 2009), unexpectedly

severed interpersonal ties with them (Strolin-Goltzman et al., 2010), or subjected them to

additional trauma (Riebschleger et al., 2015). Consideration for their sense of safety was

paramount to ensuring that they were left no worse off for having taken part in the project

and to increase the likelihood that the study would generate meaningful data. In addition

to the oversight provided by the Portland State University Institutional Review Board

(see Appendix for IRB Application) I addressed these concerns in ways which are

informed by the literature and the guidance of the sibling study advisors.

SIBLING VIOLENCE AND CHILD WELFARE 63

First, I built information about myself into the data collection instrument and I

used this rapport-building process to foster a two-way relationship that modeled a safe

level of self-disclosure with the hope that it would serve to disrupt the research-

participant power imbalance (Ross, 2017; Oakley, 1981). Second, I explicitly conveyed

that the interviewee held the power to guide the process in that they could decide where

we would meet, which questions to answer, and that they could stop participating at any

time. Third, throughout my engagement with each participant I emphasized an ethic of

care that privileged empathy and mutual respect over capturing data. In one case I

conducted an interview with a participant whose experience did not fit with the inclusion

criteria. This became clear shortly after beginning the interview, but the participant was

eager to talk about her experience losing touch with a foster sister whose relationship she

valued. I chose to carry the conversation through to completion as a demonstration that

her needs were important, valid, and equal to my own (Hoffman, 2007; Parr, 2015).

Despite my best efforts, some of the participants experienced psychological

distress during the interviews. Three participants became tearful and two others took long

pauses to regain their emotional composure following disclosure of particularly painful

events. To avoid interrupting their experience or sending the signal that I was

uncomfortable with their displays of emotion, I remained quiet and still until they

regained their composure. Once calm, I expressed that I was sorry to hear about the

painful experiences they had been through and asked if they would like to take a break or

discontinue. In each case the participant communicated that they wanted to continue and

appeared calm, so I proceeded with the remaining questions.

SIBLING VIOLENCE AND CHILD WELFARE 64

At the conclusion of the interviews I observed each participant to be in a good

emotional space; several expressed gratitude for having the opportunity to share their

stories and thanked me for carrying out the project. Seven of the eight interviews took

place at the residential internship program sight where participants had numerous peers

and providers available for additional support. The consent form also included contact

information for resources for individuals experiencing interpersonal violence, sexual

abuse, or mental health challenges. Collectively, these cues and safeguards gave me

confidence that I could conclude the interviews without additional assurances of safety or

follow-up.

Researcher Positionality

I am a survivor of abusive sibling violence. In preparation to conduct this study, I

completed a review of the literature and had many conversations with other survivors. I

noticed a tendency to compare the literature to my own experience as I was preparing the

proposal for this project, assessing the extent to which they fit or did not, and the

associated feelings and judgements that arose in relation to those comparisons. Given the

depth of my personal experience and reactions to the literature, I anticipated that some

definitions, family conditions, or theories would resonate more deeply than others, and

that I might focus on those similarities while missing other key findings in the data. I

employed various methods (i.e., reflexive memos, adviser and peer consultation) to

monitor the ways that my lived experience influenced the project, to strengthen the

quality and usefulness of the research, and to assist me to manage the emotional labor

required to sustain engagement with this sensitive topic. While these tasks were helpful,

SIBLING VIOLENCE AND CHILD WELFARE 65

bi-monthly counseling sessions with a trauma-informed therapist were integral to

maintaining my own wellbeing while completing the study.

SIBLING VIOLENCE AND CHILD WELFARE 66

Chapter 4: Results

After obtaining committee and IRB approval, I worked with the youth-driven

advocacy and support program to recruit participants for the interviews. The final sample

included eight participants; I interviewed four individuals during the study pilot in 2017

and four additional participants in the spring and summer of 2018. As mentioned

previously, the study pilot was conducted with IRB approval and the data were included

in the final analysis. As each interview was completed, I transcribed the conversation

verbatim from the recording, and entered the participants’ demographic information from

the background information form into a spreadsheet.

I then applied the coding scheme, assigning text units to transcript passages that

resonated with or were counter to the definitions, family conditions, and theories

discussed in the sibling violence literature. Additional codes captured examples of

physical and emotional sibling violence, the “break-point” at which sibling violence

ended (if at all), and sources of resilience that supported participants to survive the

violence they experienced. After coding the interviews, I electronically reproduced the

family maps, adding more detail to the graphic representations and developing brief

descriptions that identified linkages between the maps and the definitions, family

conditions, and theories in the literature.

Working back and forth between the coded interviews and the family maps in a

process of constant comparison, I identified demi-regularities (i.e., rough trends or

broken patterns in the data (Danermark et al., 2002, p. 70)). The demi-regularities were

comprised of the components of a nascent definition of sibling violence in the context of

SIBLING VIOLENCE AND CHILD WELFARE 67

child welfare, a condensed set of key family conditions associated with the phenomenon,

aspects of the linear and non-linear theories of sibling violence that resonated with the

data, key perspectives about sibling violence among affected youth, and responses among

the adults in their lives.

Finally, I engaged in abduction and retroduction, two CR thought exercises to

elucidate meaningful causal theory. Abduction focused on causal mechanisms rooted in

individual choice discernable from the interview data (i.e., at the empirical level).

Retroduction expanded the analysis to include the broader social, political, and economic

conditions under which the causal mechanisms of sibling violence take effect, with focus

on the child welfare system. The results of the analysis are presented in the sections to

follow.

Sample Description

The final sample was comprised of eight women ages 19-24 who self-selected to

participate.

Five of the participants self-identified as white; three others self-identified

as Black/African American, Hispanic/Latinx, and multi-racial (Black/African American,

Hispanic/Latinx, and American Indian). Seven of eight participants were recruited

through the youth-driven advocacy and support program’s summer internship program.

While the interviews were all conducted face-to-face in Oregon, the participants had

traveled to the program site from seven U.S. states: California, Michigan, Nevada,

Oklahoma, South Carolina, Texas, and Wisconsin. The eighth participant was recruited

through the program’s e-network and was from Oregon. One participant self-reported

No males opted into the study. This is discussed in Chapter 5.

SIBLING VIOLENCE AND CHILD WELFARE 68

multiple disabilities; others selected “none/prefer not to answer.” Seven of eight

participants self-identified as heterosexual and one as bisexual.

Table 5 summarizes physical and emotional sibling violence data for the eight

study participants. All were removed from their family of origin to out-of-home care. The

out-of-home care settings where participants’ experienced the most pronounced physical

and/or emotional sibling violence are listed. Four participants were placed in kinship care

settings and three were placed in non-relative foster/adoptive care settings. One

participant did not experience physical or emotional violence while placed in out-of-

home care. Rather, this participant was involved in the child welfare system for many

years, during which time she transitioned between her original family and foster care.

Each time she returned home, physically and emotionally abusive sibling violence

resumed; she was under the care of the child welfare caseworker during this time. As

shown, most participants experienced physical and emotional violence when living in

their original home and after removal by child welfare.

Table 5

Study Participants’ Reports of Physical and Emotional Sibling Violence

Sibling Violence Out of Home Care Type

Count

Kinship care

Foster/adoptive care

No sibling violence in out-of-home care

Sibling Violence Type

Physical sibling violence - family of origin

Physical sibling violence - in care

Emotional sibling violence - family of origin

Emotional sibling violence - in care

One study participant was fostered and then adopted by her foster parents.

SIBLING VIOLENCE AND CHILD WELFARE 69

The study was guided by two research questions to learn how the foster care

alumni experienced sibling violence and gather their perspectives on intervention into the

phenomenon. The two research questions, and related sub-questions, are shown below.

1) How do foster care alumni describe their experiences with physical and emotional

sibling violence?

a) To what extent do their descriptions fit with Caffaro (2014) and Meyers’

(2015) definitions of abusive sibling violence?

b) To what extent are family conditions and theories of sibling violence

discussed in the extant literature represented in their descriptions?

c) What additional/refined causal mechanisms do their descriptions suggest?

2) From the perspective of foster care alumni, how do adults who care for or work with

young people involved in the child welfare system (e.g., foster parents, case workers,

mental health providers, kinship caregivers, etc.) respond to sibling violence?

a) To what extent are adult responses perceived as helpful?

b) What do young people recommend to address sibling violence?

I conducted the analysis with the overarching goals of:

1. increasing understanding of physical and emotional sibling violence in the context

of child welfare;

2. contributing to sibling violence theory development; and,

3. identifying innovative ways to protect children from sibling violence and/or help

foster children manage sibling violence.

SIBLING VIOLENCE AND CHILD WELFARE 70

In the following sections I present the results of the thematic analysis and CR abduction

for Research Question 1 and Research Question 2. Subheadings identify demi-regularities

developed through constant comparison. The narrative describes each demi-regularity

with additional synthesis generated through CR abduction.

Research Question 1. How do foster care alumni describe their experiences with

physical and emotional sibling violence?

A nascent definition of abusive sibling violence in child welfare. The initial

coding scheme identified text units that supported and challenged the definitions put

forward by Caffaro (2014) and Meyers (2015) which were developed in studies with the

general population of sibling violence survivors: a repeated, escalating pattern of violent

interactions; stable victim/offender roles; perpetrator motivated by a need for power and

control, and; victim experiences a sense of powerlessness/terror and perceives the

behavior as abusive. The first demi-regularity, a refined, nascent definition of abusive

sibling violence in the context of child welfare, supported the notion that abusive sibling

violence is a repeated escalating pattern of violent interactions between two or more

siblings that results in physical or emotional harm to a child. Stable roles were evident in

the data, but the roles were that of instigators and retaliators, rather than victims and

offenders. Instigator motivations were more nuanced than the previous definition

suggests, and while some retaliators feared instigators, many did not and none of the

participants described their interactions as abusive. Subsequent sections discuss these

findings in greater detail.

SIBLING VIOLENCE AND CHILD WELFARE 71

A repeated escalating pattern. When asked to describe their experiences with

sibling violence, participants were able to easily recall examples due to their frequency

and stable patterns of interaction. Participants used language such as “every week,”

“daily,” “a lot,” “over and over,” and “common” when recalling the frequency of their

violent interactions with siblings, which spanned five to 15 years. All but two participants

experienced abusive sibling violence for six or more years. Many examples began with

verbal arguing, “nitpicking,” or horseplay between siblings, but without adequate adult

supervision or intervention, escalated until one or more children was physically and/or

emotionally harmed. In the quote below, a participant describes the persistent, violent

pattern of interactions with her older brother which began in early childhood and

continued until age 12 when she was placed in foster care:

So it would start out with something like him flicking his boogers on me. I’d be

like, “Stop,” even though I’d know he’d be joking. So I’d be like, “Stop, or I’m

gonna hit you.” And he’d keep flicking his boogers at me and it would be

frustrating, so I’d hit him like that [slaps with back of flat hand] because he’d be

sitting next to me. And he’d take that seriously, and that was when we would

actually get physical, and he’d shove me, and I’d get mad…. it always started out

simple, but then it’d get to the point where my grandparents would have to

intervene. They’d grab him and hold him back and my grandmother would have

to pull me back and, “Okay, you guys need to calm down.” But we’d want to

attack each other. I’d want to kick him. That was the thing I always wanted to do,

is kick him. But he’d punch me. So it got pretty bad. We definitely would give

SIBLING VIOLENCE AND CHILD WELFARE 72

bruises to each other. It would never get to the point where he’d punch me in the

face, but he’d punch me in the arm, in my gut. I’d get bruises, or I’d be winded

where I couldn’t breathe. And kicking him, I don’t know if I left bruises on him

but I probably did. Kicking leaves bruises. So it got pretty physical at times. It

never got to the point where we had to go to the hospital or anything. We

definitely did get to the point where my brother had to go to juvi a couple of

times. He would get a little bit too serious and he would go into the garage and

he’d grab the chainsaw, I think that’s what it’s called, the chainsaw and he’d bring

it in. My grandpa freaked out that one time, and he called the cops. And me, I ran

to my room so scared. So it got to serious moments and he’d go away for a little

while and then he’d come back and he’d be okay for a while but he was so easy to

get elevated. Easier than me. I only got mad when he got mad at me.

Despite often beginning with relatively harmless interactions, participants described

sustained patterns of violence that regularly escalated into abuse in that they experienced

physical and/or emotional harm from the interaction. This was one of three examples in

the data, each from a different participant, where the police were summoned to manage

violence between siblings due to the extremity of the violence. The use of a weapon was

also common – three participants reported that a sibling had threatened them with a knife.

Participants’ descriptions of their experiences with sibling violence strongly resonated

with Caffaro’s defining abusive sibling violence as a repeated, escalating pattern of

interaction.

SIBLING VIOLENCE AND CHILD WELFARE 73

Sibling violence instigators and retaliators. Caffaro’s definition also includes

stable victim and offender roles, a linear conceptualization of interpersonal violence

between two siblings in which the victim does not retaliate. This type of interaction was

represented in one of the interviews, but more common in the data were interactions

where one child would instigate and another (or others) would retaliate. One pattern was

that of an instigator employing emotional violence to engage a sibling. Examples spanned

minor irritations like name-calling to emotionally abusive behavior such as degrading a

sibling’s sense of self-worth or destroying their possessions. The retaliator would respond

with emotional or physical violence, which would then escalate until one or more of the

children was physically and/or emotionally harmed. Describing the violence with her

older sister, one participant recalled:

She would instigate it. She knew that was what I would turn to, physical violence.

She was more like, verbal. Like she would abuse you verbally until you do

something about it… she would run... that's what she does. She runs and then she

does this thing where she falls on her back. She would like, kick, so you can't hit

her. This one time she was doing that, and she kicked me hard, and I flew all the

way down the stairs.

Figure 6 displays a nonlinear instigator/retaliator dynamic with stable roles. The arrows

with plus signs represent reinforcing feedback loops (i.e., a repeated, escalating pattern).

In cases of sibling violence between two siblings, participants’ descriptions resonated

more strongly with the nonlinear representation displayed in Figure 6 than Caffaro’s

unidirectional linear conceptualization.

SIBLING VIOLENCE AND CHILD WELFARE 74

Figure 6. Instigator/retaliator escalating abusive sibling violence dynamic with

reinforcing feedback loops.

In families with more than two children in the home, the stable instigator role was also

represented in the data. For example, in a complex family system with eight children in

the home, the participant identified her nephew, a self-defined sibling in her kinship care

setting, as the sibling violence instigator:

He would coordinate fights with all of us... He would get us together and say

“Let's play body shots,” so what that means is “Fight!” Fight pair to pair. So he

would coordinate like you and you go fight… So it was mainly more so him, than

anyone else.

Another participant who was the eldest of five siblings identified one of her younger

brothers as the primary instigator:

[Younger half-brother 1] was extremely violent with pretty much everyone.

Meaning he would just like, grab [younger half-brother 2] and choke him, pretty

much the same stuff my stepdad was doing, like pinch him… or hit my brother or

hit my sister… [Later he] started getting more violent by like grabbing knives and

trying to stab people while they were sleeping and I would have to tackle him and

Instigator

Retaliator

SIBLING VIOLENCE AND CHILD WELFARE 75

take the knife away from him… my brother was probably the more violent

person.

Seven of eight participants identified a primary instigator in the family system, regardless

of the number of children in the family. In households with more than two siblings,

however, additional abusive sibling dynamics co-occurred along with the primary

instigator/retaliator(s) dynamic. In the situation above, two younger children were also

instigators. The result was a complex dynamic in which five siblings shifted between

instigator and retaliator roles as they engaged in multiple dyadic and triadic patterns of

violence.

Finally, Caffaro defines abusive sibling violence perpetrators as motivated by a

need for power and control, a characteristic which was supported in the data, along with

others. Siblings employed violence to enforce rules and to establish their dominance as

decision-makers. One participant used violence to “teach” her sister not to emotionally

abuse their neglectful mother. Another overpowered her brother to prevent him from

abusing a younger sibling. An older brother beat up his sister to keep her from reporting

their lack of parental supervision to adult authorities. In addition to a need for power or

control, other instigation drivers were discussed. In some cases, siblings living in

impoverished families engaged in violence to obtain scarce resources such as food.

Others suggested that instigators’ violent behaviors were due to behavioral,

developmental, or mental health challenges. As evidenced by the data, children’s

motivations for engaging in sibling violence were more nuanced than Caffaro suggests.

SIBLING VIOLENCE AND CHILD WELFARE 76

Victim/retaliator perceptions. Extending Caffaro’s definition, Meyers (2017)

argued that the phenomenon is also defined by the victim experiencing a sense of terror

and powerlessness, along with a perception that the perpetrator’s behavior is abusive. The

data offered support for this notion, provided by two of eight participants whose older

brothers had abused them. One reenacted her 911 call: “I’m scared for my life. I don’t

have a parent to support me. My brother is doing this to me every single day.” Another

noted that she “didn’t need to be scared all the time” after she was removed to foster care,

placed separately from her older brother.

More commonly, the foster care alumni contrasted violent interactions between

siblings with adult-perpetrated violence in the home. Participants were exposed to such

frequent and extreme adult-perpetrated violence, they were prone to viewing sibling

violence as normal. This perception was fortified through social isolation – participants

actively hid their home circumstances from outsiders, changed residences frequently due

to poverty/eviction, or both. A lack of adult intervention affirmed this perception. In

some cases, participants were hesitant to identify the behavior of their instigating siblings

as abusive because they had retaliated against the behavior. As one participant recalled,

“I did it too. I followed the footsteps. In the back of my mind I knew it was wrong,

because it didn't feel good.” The data gathered in the current study suggest that Meyers’

extension of Caffaro’s definition is less resonant in the context of child welfare.

My analysis suggests that a refined definition of abusive sibling violence is

needed in the context of child welfare that considers the circumstances of child welfare-

involved children’s lives. Specifically, foster children’s exposure to adult-perpetrated

SIBLING VIOLENCE AND CHILD WELFARE 77

violence and the lack of adult supervision they receive appears to bear upon children’s

motivations for engaging in sibling violence and their perspectives about it. While

Caffaro argued that a perpetrating sibling is motivated by a need for power and control,

participants in the current study offered alternate motivations including a need to secure

basic resources and un-/under-addressed behavioral, developmental, and mental health

challenges among instigators. Meyers’ additions to the definition were also less resonant

with the data in that participants were more apt to perceive sibling violence as normal and

to retaliate rather than identify as victims. A refined, nascent definition of abusive sibling

violence in the context of child welfare is as follows: a repeated, escalating pattern of

violent interactions between/among two or more siblings that results in physical or

emotional harm to at least one child, regardless of that child’s perception of the violence

as it occurs.

Family conditions associated with sibling violence. The extant literature

identifies a plethora of family conditions associated with sibling violence in the general

population (Caffaro & Con-Caffaro, 1998; Crick & Dodge, 1994; Frick & White, 2008;

Kerig & Stellwagen, 2009; Loeber & Stouthhamer-Loeber, 1998; Wiehe, 1997). I

distilled the conditions into seven codes which were included in the original coding

scheme. Table 6 summarizes the data for the sample, displaying the code, definition, and

the number of participants who discussed the family condition during their interview. As

shown, five or more participants reported that they experienced or were exposed to each

one. Individual participants experienced three to seven of the family conditions at the

time that sibling violence was happening, with an average of six (data not tabled).

SIBLING VIOLENCE AND CHILD WELFARE 78

Table 6

Family Conditions Associated with Sibling Violence

Code

Definition

Number of

Participants

(n=8)

Acceptance/

Modeling of

Violence

Parent(s) communicate(s) that violence is an appropriate

solution to interpersonal problems within the family system by:

• Physical/emotional violence (IPV) between adult

caregivers

•

Adult-perpetrated physical/emotional child maltreatment

Parental

Unavailability

Parents are physically and/or emotionally unavailable as a

result of:

• Inability to express love, empathy, caring for the

child(ren)

• Regular/severe alcohol/drug use

• Chronic/severe mental health challenges

• Chronic/severe physical health challenges

Lack of

Supervision

Parent(s) do(es) not provide expectations for or enforce healthy

child behaviors

Siblings Differ

Sibling qualities that prime perpetrators to enact

physical/emotional violence or predispose victims to receive

abuse:

• Perpetrator Qualities - thinking errors,

cognitive/developmental deficits, lack of impulse

control, lack of empathy, callout-unemotional traits,

narcissism, tendency to inaccurately interpret social cues

• Victim Qualities - genetically determined physical,

emotional, mental characteristics (disabled,

developmentally delayed, exhibit behaviors others

experience as irritating

• Large developmental, physical, intellectual, or

social/emotional differences between siblings in a dyad

in which the perpetrator is superior and the victim is

inferior

Differential

Treatment

Parent(s) favor or scapegoat one sibling as compared to

other(s)

Work/

Financial

Strain

Parent(s) described as experiencing work/financial strain such

that:

• Parent(s) is/are unable to secure stable employment

• Parent(s) work is illegal (e.g., drug dealer)

• Parent(s) work(s) more than full-time to meet basic

needs

• Parent(s) unable to meet family's basic needs due to

un/under-employment

Parentified

Child

One sibling is expected to parent others; caretaking

expectations are developmentally inappropriate given the

child's age/abilities

SIBLING VIOLENCE AND CHILD WELFARE 79

Areas of overlap and alignment among the conditions arose through constant

comparison, resulting in a refined set of two primary and two secondary conditions that

were directly associated with sibling violence in the data. Primary conditions were

strongly resonant in the data, with most participants reporting exposure and directly

linking their exposure to sibling violence. Secondary conditions were also prevalent in

the data, but linkages to sibling violence were less pronounced. Table 7 presents the final

set of family conditions and their definitions along with the initial code(s) each was

drawn from. Table 8 displays the number of respondents who reported each family

condition, both in their family of origin and in out-of-home care. For those respondents

who reported the condition in care, results are also split between participants living in

kinship care and those in foster/adoptive care. After the tables, each of the family

conditions is discussed in greater detail with example text units drawn from the

interviews.

SIBLING VIOLENCE AND CHILD WELFARE 80

Table 7

Final Set of Family Conditions Associated with Sibling Violence in Child Welfare

Initial Code

Primary/

Secondary

Final Code

Final Definition

• Acceptance/

Modeling of

Violence

• Lack of

Supervision

• Differential

Treatment

Primary

Modeling/

Acceptance of

Violence

Caregivers communicate that violence is an

appropriate solution to interpersonal

problems within the family system:

• Adult-perpetrated physical/emotional

child maltreatment

• Interpersonal violence (IPV) between

adult caregivers

• Caregivers condone sibling/peer

violence

• Parent(s) favor one child (the sibling

violence instigator)

• Parental

Unavailability

• Parentified

Child

Primary

Drugs,

Alcohol, and

Mental Health

Challenges

Caregivers are mentally, emotionally,

and/or physically absent, failing to provide

adequate supervision

• Work/Financial

Strain

Secondary

Poverty

Caregivers fail to meet children’s basic

needs

• Siblings Differ

Secondary

Instigator

Characteristics

Instigators experience behavioral,

developmental, and mental health

challenges

SIBLING VIOLENCE AND CHILD WELFARE 81

Table 8

Family Conditions Associated with Sibling Violence Overall and by Placement Type

TOTAL

Foster/Adoptive

SUBTOTAL

Kinship

SUBTOTAL

Modeling/Acceptance of Violence

Family of origin 8

Out-of-home care 4 2 2

Drugs, Alcohol, Mental Health Challenges

Family of origin 6 3 3

Out-of-home care 2 1 1

Poverty

Family of origin 6 2 4

Out-of-home care 3 1 2

Instigator Characteristics

Family of origin 4 1 3

Out-of-home care

Primary family condition 1. Modeling/acceptance of violence. Caregivers

directly and indirectly communicated to participants and their siblings that violence is an

appropriate solution to interpersonal problems in the family system. Every participant in

the sample experienced some form of adult-perpetrated physical or emotional violence,

either directed toward another adult in the home or toward the child(ren). Seven of eight

participants were physically abused by an adult caregiver in their family of origin. Four

were physically abused or witnessed a sibling being physically abused after removal to

out of home care. Among the four, two were in kinship care and two were in

foster/adoptive care. Participants and their siblings were beaten with fists, belts, and

utensils, choked, thrown into walls, tied up, and force-fed, often for minor transgressions

such as talking back, making a mess, or failing to do chores. One participant’s father

SIBLING VIOLENCE AND CHILD WELFARE 82

“would rather hit me than ground me, because hitting me is just for the day but

grounding, then it has to be for weeks.” Another described a kinship caregiver shoving

her fingers down her throat to make her eat dinner. A third participant’s kinship

caregivers:

…hit us a lot with a belt… it was over the littlest, the dumbest things. Like

somebody would eat something and we would get in trouble. Or somebody would

mess something up and we would get in trouble… the only time they would hit us

in the face was when we insulted them.

Emotional abuse was also widespread. Participants and their siblings were shamed and

screamed at for “leaving a cup out” or arguing. A particularly cruel example was one

mother’s repeated accusation that the participant was to blame for her father abandoning

the family.

Participants drew direct connections between the adult-perpetrated child

maltreatment they experienced in their families of origin and in care, and the physical and

emotional violence between siblings. An interview participant who regularly fought

physically with her sister stated, “I would hit her and stuff. Probably because my dad was

hitting me before, and then I was like, this is how you deal with problems. This is what

you do.” In a family where the stepfather was abusive to both the children and their

mother, the participant said that her stepdad “would talk down to my sister and call her

ugly and fat and all these other things, so my brother followed in his footsteps and said

the same things.” When describing a pattern of physically violent interactions with her

brother, another participant recalled her mother’s alcohol-induced abuse: “It definitely

SIBLING VIOLENCE AND CHILD WELFARE 83

put a strain on me and my brother. Because it happened to us the most, so we kind of did

the same to each other.” As evidenced by these comments, the foster care alumni viewed

sibling violence as a replication of violent behavior modeled by their caregivers.

In addition to pervasive modeling of violent behavior, caregivers also condoned

violence between/among children by allowing, ignoring, or encouraging sibling/peer

violence. In an example from early childhood, a participant recalled her father verbally

encouraging sibling violence between she and her sister. As a teenager the same

participant’s mother did not take issue when she told her that she planned to fight a peer

after school. The participant connected these exchanges with sibling violence: “Nobody

made it not normal… my mom wasn’t like ‘Don’t hit your sister. Don’t go beat this girl

up.’”

Passive acquiescence was also commonly discussed in the interviews. Six

participants described caregivers who had allowed biological or foster siblings to abuse

each other in their presence or failed to respond when children reported sibling violence.

As stated by one participant, “My mother would just sit back and let it happen.” Another

recalled:

I can remember when I was younger my brother just would like, tease me all the

time. Call me names and treat me like trash and he would throw things at me and

hit me. He was horrible to me. And that was in front of my mom.

In addition to adult-perpetrated child maltreatment, all participants disclosed emotional or physical

violence (IPV) between adult caregivers in their family of origin or while living in care. There were no

direct connections drawn between witnessing IPV and sibling violence, however.

SIBLING VIOLENCE AND CHILD WELFARE 84

Passive acquiescence was often linked to descriptions of preferential treatment. Five

participants described parental favoritism for their male siblings who received more

emotional nurturance, were fed first at meals, and received better clothing, toys, and

technology than their female siblings. These privileged children could behave violently

without consequence. One participant recounted her father standing by as her brother

attempted to drown her in a lake. All five participants whose brothers received

preferential treatment identified them as the primary sibling violence instigator in the

family.

In summary, every participant in the sample received the explicit or implicit

message that violence is an appropriate way to address interpersonal problems in the

family system. Study participants lived in families where emotional and physical violence

were frequently modeled by adult caregivers. When siblings behaved violently,

caregivers did not intervene, and in one case verbally encouraged the behavior. Exposure

to and acceptance of interpersonal violence in the family system surfaced as a primary

family condition for sibling violence in child welfare.

Primary family condition 2. Drugs, alcohol, and mental health challenges. Two

of the initial codes (i.e., parental unavailability, parentified child) were combined into a

single code describing family systems where parents were mentally, emotionally, and/or

physically absent, and as a result failed to provide adequate supervision for children.

Parental unavailability was largely due mental health challenges (e.g., PTSD, bipolar

disorder, schizophrenia) and/or addiction issues (e.g., alcohol, methamphetamine, crack

cocaine, heroin, and prescription drugs). Half reported having a parent with co-occurring

SIBLING VIOLENCE AND CHILD WELFARE 85

mental health and addiction challenges. In most cases, caregivers’ addiction issues and

mental health challenges caused them to leave participants and their siblings

unsupervised for lengthy periods of time. When caregivers were physically present, they

were described as emotionally or mentally unavailable.

Children in these family situations commonly took on developmentally

inappropriate caregiving responsibilities for themselves and their siblings. One

participant identified as a “mom” to her siblings and regularly resorted to violence with

her brother (i.e., slapping him, pushing him into corners, standing on him) to keep him

from harming their younger brother. Another participant “felt like [she] had to be the

adult when [her] mom left.” When she attempted to dissuade her older brother from

selling drugs, he responded by screaming at her and beating her up. In both instances,

interactions of this type were described as a repeated pattern as opposed to one-time

incidents and were explicitly linked with children being left alone. When parents who

were minimally engaged assigned older children to babysit younger ones, violent

dynamics similarly followed. In an adoptive family where an older brother was charged

with babysitting four younger siblings, they ganged up on him and beat him up. The

younger children also ganged up on their older sister when she was put in charge. Un-

/under-treated caregiver addiction and mental health challenges resulting in inadequate

supervision of children surfaced as another primary family condition for sibling violence

in child welfare.

Secondary family condition 1. Instigator characteristics. When describing

sibling violence instigators, participants recalled behaviors associated with various

SIBLING VIOLENCE AND CHILD WELFARE 86

mental health diagnoses, disabilities, and developmental delays. Examples included

motor tics, significantly delayed toilet training, showing no sign of remorse after an

aggressive episode, and manipulative, oppositional, and violent behavior directed toward

people and animals in a variety of settings. Participants reported that sibling violence

instigators had received mental health diagnoses including attention deficit hyperactivity

disorder, oppositional defiance disorder, and bipolar disorder. In most cases, participants

reported that these challenges were untreated, even after removal to foster care. The data

suggest that child welfare-involved children with un-/under-addressed challenges such as

these may be prone to physical or emotional sibling violence instigation.

Secondary family condition 2. Poverty. All participants reported living in low-

income families and many lacked adequate access to basic needs such as food, clothing,

and stable housing. One participant explicitly connected living in extreme poverty with

sibling violence:

Since we were always all over the place, mom could never afford to feed us all, so

we were stuck in that survival mode. There were a lot of times when we got in fist

fights over things like who was gonna have the last can of green beans or who

could have the piece of bread that we had in the cupboard. We've gone to school

in the morning with bloody noses, cuts, bruises all over us because we were

fighting over food. There's been times when we fought to get the last bit of water,

because she couldn't pay for the water bill or electricity a lot of times. If we were

running out of water we had to fight over who was gonna get to take the last

shower, get a glass of water. It was pretty brutal.

SIBLING VIOLENCE AND CHILD WELFARE 87

Although most participants did not link poverty with sibling violence, it is likely that the

stress this placed on children was a contributing factor. One participant described

arguments with her brother who was selling drugs, a choice that was likely driven by a

need for income. Others recalled arguments over shared clothing that escalated into

abuse.

Resonance of linear and non-linear theories. The extant sibling violence

literature offers both linear and non-linear (i.e., systems-oriented) theories to describe the

causal mechanisms that foment family violence. The coding scheme included a code to

identify text units associated with components of three linear and one non-linear theory to

determine the extent to which they were represented in participants’ descriptions.

Summarized in Table 9, the codes identified key components of each theory. Codes were

assigned to text units describing any aspect of the theory.

SIBLING VIOLENCE AND CHILD WELFARE 88

Table 9

Theory Code Descriptions

Theory/Model

Description

Social

Learning

Theory

Physical/emotional sibling violence is:

• Learned through observation

• Results in receipt of desired rewards

• Not redirected/does not result in consequences

Radical

Feminist

Theory

Physical/emotional sibling violence is described as:

• Justified means for the more powerful person in a dyad/situation to get what they

want

• Due to a power differential (i.e., physical strength, intellectual/emotional maturity,

level of responsibility) between siblings

• Giving a sibling a sense of control by overpowering another

Conflict

Theory

Sibling violence is described as occurring in a setting in which siblings experience:

• Scarcity - lack of access to emotional or tangible resources

• Competition - for parental attention

• Parental favoritism - one sibling gets better tangible/emotional resources

Coercive

Family

Process

Sibling violence is described as occurring in a family system with:

• Multiple concurrent dynamics among 3+ members

• Family members embody stable roles

• Mutual training - repeated, cyclical interpersonal reactions reinforce violence

• Feedback loops - emotional, behavioral, or cognitive responses perpetuate violence

Linear theories employ reductionist methods to parse a theoretically knowable

reality into discrete, measurable variables. Based on Humean causal law, linear theories

describe chains of temporally bound relationships (i.e., if event x, then event y). The three

linear theories included in the inquiry were social learning theory, radical feminist theory,

and conflict theory. After discussing the resonance of the three linear theories with

participants’ descriptions of sibling violence in the context of child welfare, I will present

the study findings related to a non-linear (i.e., systems-oriented) theory of sibling

violence, coercive family process.

Social learning theory explains interpersonal violence as a behavior learned

through observation that results in receipt of desired rewards and which does not produce

SIBLING VIOLENCE AND CHILD WELFARE 89

negative consequences. As discussed in the previous sections on family conditions

associated with sibling violence, participants and their siblings experienced considerable

adult-perpetrated violence. In addition to learning through experience and observation,

children gained a variety of rewards by engaging in sibling violence. Participants

instigated violence to gain control of a television remote or an item of clothing, to get to

eat the remaining food in the house, to take the last shower before the water was cut off,

and to receive a gaming system (a reward one child’s grandparents offered to him to

encourage him to stop being violent with his sister). Another violent episode was

triggered when a foster sister stole the participants’ tablet computer. In other examples,

children gained a sense of power or control in situations lacking parenting guidance or

oversight. Rarely redirected, children were able to behave violently without being held

accountable by adult caregivers.

The causal mechanism defined by social learning theory was represented in the

data, but given the violent contexts that many child welfare-involved children endure, its

primary use is to hone in on children’s motivations for engaging in violence. Moreover,

the theory does little to inform understanding of sibling violence as a dynamic process

between two or more children that is replicated over time within context. Radical feminist

theory and conflict theory, discussed next, are subject to similar limitations.

Radical feminist theory identifies patriarchy and the related, pervasive acceptance

of the use of power and control to achieve desired aims as the causal mechanisms

underlying interpersonal violence. Applied to sibling violence, radical feminism argues

that any power differential between siblings such as in physical strength, emotional

SIBLING VIOLENCE AND CHILD WELFARE 90

maturity, or assigned level of responsibility could foment violence. Whatever the

perspective of the older, bigger, or more powerful sibling, the younger, smaller, less

powerful sibling(s) must follow suit or suffer violent consequences.

Among seven of the eight participants interviewed, the primary instigator was

older, bigger, or stronger than the sibling(s) they abused. In a situation where a

participant’s brother was abusive to three younger siblings, the participant related:

“…because I was significantly taller than him… he didn’t try [to be violent with me]. But

he was also significantly taller than all my other younger siblings… So having that

towering effect was beneficial to him in that control over something or some people at

least.” Employing her greater size and strength, the participant managed her brother’s

violent behavior by overpowering him physically. A power differential in age, size, and

strength was particularly evident in three interviews where the primary instigator was an

older brother who instigated violence with their younger sister. Each sister retaliated,

whether verbally or physically, but was ultimately overpowered when their brother

employed more extreme acts of violence or used a weapon. One brother destroyed his

sister’s prized possessions and pulled out clumps of her hair. Another emotionally abused

and physically beat up his sister, while the third punched his sister and threatened her

with a chainsaw. Across the interviews, the children who instigated or ultimately “won”

physically violent exchanges tended to be taller, older, and/or physically stronger than the

siblings they engaged in violence with.

As evidenced by the data, within a family configuration radical feminist theory is

useful to understand why a child feels empowered to instigate violence, their choice of a

SIBLING VIOLENCE AND CHILD WELFARE 91

sibling to target (i.e., one who is smaller/less powerful), and why a child succeeds in

overpowering their sibling in an exchange. The theory is less helpful to identify the

causal mechanisms underlying a child’s determination to engage in violence, and similar

to social learning theory, fails to inform understanding of abusive sibling violence as a

dynamic process, especially in multi-sibling families or when the sibling configuration is

comprised of children who do not differ markedly (i.e., siblings who are close in age,

size, etc.).

Conflict theory positions violence in the family as a reaction to perceived

scarcity, assuming that humans are innately self-interested and will utilize any means

necessary to obtain desired resources. Conflict theory is discussed in the literature as the

causal mechanism underlying sibling jealousy, competition for parental attention, and the

expectation that children share valued items. In the current study, the conflict theory code

was applied to text units where participants described inadequate access to tangible

resources and scarcity of emotional nurturance, including parental favoritism. All text

units initially coded for conflict theory were subsumed within the drugs, alcohol, and

mental health challenges code and the poverty code – by nature of their involvement in

child welfare, every participant in the study was subjected to emotional and/or material

scarcity. While conflict theory could prove useful in a general population study with

more nuanced experiences of scarcity, it does little to extend understanding of the

phenomenon among child welfare-involved youth.

Table 8 maps the linear theories onto the family conditions associated with sibling

violence that were represented in the data. The utility of the linear theories to explain

SIBLING VIOLENCE AND CHILD WELFARE 92

sibling violence beyond the family conditions would likely depend on the role of the

individual or group applying the theory. While a therapist, intervention designer, or

researcher could find social learning theory or radical feminist theory useful, a caregiver,

case worker, or policy maker might find the primary and secondary family conditions

adequate to inform their understanding or guide decision-making in practice.

Table 10

Final Set of Primary and Secondary Family Conditions for Sibling Violence in Child

Welfare with Associated Linear Theories

Primary/

Secondary

Final Family

Condition

Final Definition

Linear Theories

Primary

Modeling/

Acceptance of

Violence

Caregivers communicate that violence is an

appropriate solution to interpersonal problems

within the family system:

• Adult-perpetrated physical/emotional child

maltreatment

• Interpersonal violence (IPV) between adult

caregivers

• Caregivers condone sibling/peer violence

• Parent(s) favor one sibling (the instigator)

Social learning

theory

Radical feminist

theory

Primary

Drugs,

Alcohol, and

Mental Health

Challenges

Caregivers are mentally, emotionally, and/or

physically absent, failing to provide adequate

supervision

Conflict theory

Secondary

Poverty

Caregivers fail to meet children’s basic needs Conflict theory

Secondary

Instigator

Characteristics

Instigators experience behavioral,

developmental, and mental health challenges

N/A

After reviewing the linear theories of sibling violence, I examined coercive

family process in relation to the data. The analysis focused on the extent to which this

non-linear theory of sibling violence, or elements of the theory, could extend

SIBLING VIOLENCE AND CHILD WELFARE 93

understanding of the causal mechanisms underlying the phenomenon in the context of

child welfare beyond the family conditions and linear theories.

Coercive family process. Combining family systems theory with principles of

operant conditioning and negative reinforcement, coercive family process theory employs

the key systems concepts of holism, nesting, and non-linearity to examine interpersonal

violence in relation to the whole family, including its context. Rather than targeting an

individual member as the source of family discord, the theory conceives interpersonal

violence to be the product of interactions among every member in the family. Individuals

are described as maintaining stable roles within the family system, repeatedly engaging in

a narrow set of behaviors. This role stability creates a state of relative equilibrium,

producing a family-wide pattern of interaction. The pattern is comprised of multiple

coercive processes that produce physical and emotional violence between/among various

dyads and triads. Sibling violence results from the collective influence of all members,

who are both influenced and constrained by their context.

Holistic analysis via constant comparison between the interviews and the family

maps demonstrated substantial resonance between coercive family process theory and the

data. All eight participants described multiple, concurrent, stable behavioral interactions

among three or more family members that created a repeated pattern of violence among

caregivers and children. Four stable roles were represented in the data. Parents and

caregivers were modelers and fortifiers, often due to un-/under-treated mental health and

addiction challenges. Modelers demonstrated how to behave violently through adult-

perpetrated child maltreatment, engaging in violence with their adult partner(s), or both.

SIBLING VIOLENCE AND CHILD WELFARE 94

Fortifiers condoned sibling violence, either by rewarding children through verbal

encouragement or passively allowing the behavior to occur without consequence. Many

caregivers embodied both roles.

Children also demonstrated two stable roles. Instigators were the children who

initiated violent interactions with their siblings. They were prone to behaviors associated

with various mental health diagnoses and disabilities and were typically bigger and more

physically powerful than the siblings they initiated violence with. When instigators had

multiple siblings, they sometimes served as secondary modelers to younger siblings who

then instigated physical or emotional violence with other children in the family system.

Retaliators were the children who responded to instigators, typically as an act of self-

preservation. Instigators and retaliators employed violence to rescue other family

members from instigator-perpetrated violence.

With regard to families’ community contexts, most lived in low-income

communities and some experienced community violence due to high rates of drug use or

gang activity. Caregivers who were employed worked in low-wage jobs; many were

unemployed and some sold or bartered drugs to supplement their income. Children

experienced bullying in their neighborhoods and schools. In the sections that follow,

three synthesized typologies of family systems drawn from the most salient findings in

the data demonstrate how the four family roles can manifest in child welfare-involved

families to produce patterns of sibling violence.

In the family system displayed in Figure 7, two sisters live with their biological

father who regularly disciplined them with beatings until he hurt his back at his

SIBLING VIOLENCE AND CHILD WELFARE 95

construction job. He is addicted to pain medication and has been out of work for more

than a year. The family has been evicted several times and their food stamps do not

adequately meet their needs. Between apartments they live out of their car and the sisters

are bullied at school for poor hygiene. Weekly, and sometimes more often, Sister 1

becomes overwhelmed by their circumstances, emotionally shaming her father for failing

to provide for the family. Sister 2 wants to preserve her father’s dignity, so she instigates

violence by hitting Sister 1. This triggers a physical fight that escalates until one or both

sisters are physically harmed. Their father, whose physical mobility and mental clarity

are compromised by the pain medication, does not intervene. The pattern persists until

the sisters are removed to foster care due to neglect and placed separately. Both sisters

engage in sibling violence in foster care.

Figure 7. Two biological siblings and their father.

Figure 8 presents a kinship care setting where a foster child has been placed with

her aunt, uncle, and their two biological children. Her parents lost custody due to their

Sister 1 - Retaliator

9 years old

Sister 2 – Instigator

(Rescuer)

8 years old

Bio Father

Passive Fortifier

Emotional violence

Physical violence

feedback loop

SIBLING VIOLENCE AND CHILD WELFARE 96

methamphetamine addiction. Before being removed she observed numerous fights

between her biological parents and was physically and emotionally maltreated. The foster

family lives in an apartment in a low-income community and only has two bedrooms.

Brother already slept on the sofa so Sister could have her own room. Sister now shares

her room with Foster Sister and takes it out on her by destroying her belongings and

shaming her about her parents. Foster Sister responds to Sister’s behavior with physical

violence. Brother takes Sister’s side, retaliating with physical violence. He’s been getting

bullied at school and it feels good to get his anger out. Bio Dad/Uncle works long hours

at a factory to support the family. He can hear the noise in the other room but doesn’t

have the energy to do anything about it. The physical violence between Brother and

Foster Sister escalates until Bio Mom/Aunt intervenes with a bribe (e.g., the children can

all stay up an hour later if they stop fighting). The pattern of violence persists until Foster

Sister’s aunt and uncle decide they can’t handle caretaking a foster child – she is removed

to a new foster home with five other foster siblings. It is not a good situation but she is

about to age out of care and her social worker seems to have other priorities.

SIBLING VIOLENCE AND CHILD WELFARE 97

Figure 8. Kinship care setting with two biological siblings and their foster sister.

Figure 9, the final example, presents a biological family system prior to the

children being removed to foster care. Dad is in the military and has completed several

tours. When he is home he has been increasingly violent with mom, giving her a black

eye in their most recent fight. Mom started drinking to manage her stress and she often

leaves the children unsupervised. Both mom and dad have used beatings with belts and

other objects as a form of discipline over the years. Brother 2 has been violent since he

was a toddler, hitting, pinching, and kicking his younger siblings. Mom used to intervene

(and so did dad, when he was home), but for the past several years that has been Brother

1’s responsibility. Usually he grabs Brother 2 and hits him several times before locking

him in the closet – sometimes it takes a very long time for him to calm down and be let

out. Brother 3 has tried to retaliate against Brother 2, but he’s much smaller and Brother 2

just laughs at him. Brother 3 takes his anger out on Sister with emotional violence. Most

Emotional

violence

Physical

violence

feedback loop

Sister - Instigator

16 years old

Foster Sister –

Retaliator

(Self-Preservation)

17 years old

Foster/Bio Mom

Active Fortifier

Brother – Retaliator

(Rescuer)

15 years old

Foster/Bio Dad

Passive Fortifier

SIBLING VIOLENCE AND CHILD WELFARE 98

recently he killed her pet hamster and threw it at her. When she told her mom what had

happened, mom told her to heed her brothers – they are the men of the house while dad is

away.

Figure 9. Biological family with four siblings.

As evidenced by these hypothetical family system examples, coercive family

process theory is helpful to understand sibling violence as a family-wide pattern of

interaction that is influenced by the community context. The theory supports a more

holistic analysis of the problem, expanding beyond the family conditions and linear

theories to assess how all members in the family system, including key contextual

Mom

Modeler/Active

Fortifier

Dad

Modeler/Passive

Fortifier

Brother 2

Instigator/Secondary

Modeler

9 years old

Brother 3

Retaliator/Secondary

Instigator

7 years old

Brother 1

Retaliator (Rescuer)

11 years old

Sister

Retaliator (Self-

Preservation)

5 years old

Emotional

violence

Physical violence

feedback loop

SIBLING VIOLENCE AND CHILD WELFARE 99

factors, contribute to sibling violence. Family system maps that display member

characteristics, interpersonal violence, and reinforcing feedback loops are a useful tool to

examine how the theory applies, both within and across families.

Research Question 1 sought to elucidate how the foster care alumni experienced

sibling violence, and to determine the extent to which their experiences resonated with

the family conditions and theories discussed in the general population sibling violence

literature. The analysis produced a nascent, refined definition and a reduced set of four

key family conditions for sibling violence in the context of child welfare. Moreover, the

results underscore the utility of a non-linear theory to understand the phenomenon as a

complex, synergistic, and dynamic socio-behavioral process. Coercive family process

theory effectively transcends the family conditions and linear theories to holistically

incorporate the roles and behaviors of all family members with consideration for the

family context. In sum, this non-linear theory offered the most resonant reflection of the

causal mechanisms of sibling violence among child welfare-involved families as

represented in the data. Subsequent sections present the results for Research Question 2.

Research Question 2. From the perspective of foster care alumni, how do adults who

care for or work with young people involved in the child welfare system (e.g., foster

parents, case workers, mental health providers, kinship caregivers, etc.) respond to

sibling violence?

Inadequacy of adult responses to sibling violence in child welfare. Study

participants were asked to describe adult responses to sibling violence and their

perceptions of the helpfulness of the responses. Just one participant able to recall a

SIBLING VIOLENCE AND CHILD WELFARE 100

caregiver who effectively addressed sibling violence. The intervention served to interrupt

physical violence between the participant and her sister, but they continued to engage in

emotional violence. Participants were more apt to report that the adults in their lives,

including mandatory reporters and kinship/foster caregivers, did not intervene into sibling

violence when it occurred. Four participants did not recall any adult intervention into

their abusive sibling relationships. When caregivers, caseworkers, and law enforcement

did respond to sibling violence, their responses were inadequate, failing to ensure the

safety of the children. Subsequent sections present study results highlighting the

inadequacy of adult responses to sibling violence in the context of child welfare.

Caregiver responses. Among participants who experienced adult intervention into

sibling violence, two described ineffectual verbal redirections from caregivers that did

not interrupt individual exchanges or the pattern of violence over time. For example, one

participant’s mother said, “If you’re going to fight go outside.” Another participant was

involved in family meetings guided by her adoptive parents that deteriorated into

“everybody yelling” and the children being sent to their rooms. A participant who was

physically and emotionally abused by her foster siblings reported it to her foster mother

who then sided with her siblings.

Just one participant discussed a helpful caregiver response after she and her sister

were placed in kinship care:

My aunt, she came home from work that day that we got in a huge fight and said,

“You guys may have used to hit each other and that's how you dealt with your

problems but that's not how we're going to deal with them here.” She just kind of

SIBLING VIOLENCE AND CHILD WELFARE 101

talked to us about how hurting each other isn't the way to do it… And since we

had so many rules and all this stuff. Our caseworker was always around. I just felt

like the rules were way more important… I was just like, “I don't want to mess

anything up.” …I didn't want her to have to call the caseworker because we were

hitting each other... she had like, a SWAT behind her. It was more impactful.

From the perspective of this participant, clear expectations from her aunt, coupled with

an engaged caseworker who was perceived as a strong backup, effectively interrupted the

sibling violence.

Caseworker responses. Four participants discussed responses to sibling violence

after child welfare reports were made. Two participants personally informed their

caseworkers that they were being abused by their siblings. After being returned to her

biological mother’s care one participant reported her biological brother but her “social

workers would literally blame it on [her mom] or blame it on [her].” Another reported her

foster siblings and had recordings and pictures as evidence to back up her claim, but

“[caseworkers] see no point [in addressing the problem] because I was gonna age out and

they… don’t want to go through the paperwork and stuff like that.” Caseworker

responses to sibling violence reports were limited to two reports that adults filed, and

which resulted in sibling separation. One participant was placed in a new foster home

while her brother, who had enacted more frequent and extreme acts of violence, remained

with their kinship caregivers. Another was glad to have been removed because she was

afraid of her brother, but was then forbidden from having any contact with him:

SIBLING VIOLENCE AND CHILD WELFARE 102

That was worse than taking me away from my mom. The one person who has

been consistent in my life, I can’t even talk to on the phone? I was like, “I

understand where you’re coming from because I don’t want to be abused

anymore, but he can’t jump through the phone. If he starts saying stuff all I have

to do is hang up.” But they were like, “We’re not letting you talk to him.”

The only respondent who discussed a moderately helpful child welfare

intervention attended a family session as a supplement to her sister’s individual therapy.

She recalled, “I remember liking it. I remember thinking that it was good, and that we

should express to each other more,” but never had another opportunity as family therapy

was not repeated. Two participants disclosed that they had intentionally avoided

disclosure to their caseworkers because they were afraid of being separated from their

siblings.

Law enforcement. Three participants discussed police intervention. Two

participants, one in kinship care and another in foster care, had an adult caregiver call law

enforcement because a sibling was wielding a weapon. In both cases, the response kept

siblings safe in the moment, but the children did not receive any further intervention or

follow-up. In both cases the violence between siblings continued until the siblings were

separated. A third participant who was living alone with her brother called the police

during a particularly violent incident. Shortly after arriving the officers determined that

there was not an adult in the home and left her alone with him:

I was like, “What is this world? How can you just sit there? That’s a danger to my

safety.” I was literally crying, 9-1-1. I called 9-1-1. I was like, “My brother is

SIBLING VIOLENCE AND CHILD WELFARE 103

beating me up!” It was really bad. I was really upset when they didn’t do anything

about it... calling the police should have been a huge game-changer. Before that, I

can understand that maybe it wasn’t as clear. Maybe the social workers or parents

or people involved, outside of us, maybe they didn’t see it happening because it

was behind closed doors and nobody would know, but once the police have been

called it should be intervened… [The officers] literally stayed there for one

minute and left.

In this situation the officers did not perceive sibling violence to be a true threat to the

child. It is notable that they opted to leave the child in the home with her abuser after she

so clearly communicated that she feared for her safety. The officers notified child

protective services of the child’s living circumstances and she was placed in foster care

shortly afterward.

Survivors’ recommendations to address the problem. Given the limited

response of caregivers, child welfare practitioners, and law enforcement, it is not

surprising that participants’ recommendations were simple and practical. Two

participants called out caregivers for failing to address the problem, recommending that

they intervene when siblings are being violent by “actually taking hands into the matter,

actually trying to do something” or by “teach[ing] people in general what discipline and

consequences are.” Another offered that “Somebody should have said… maybe let’s

figure out what’s going on… what are the factors that are playing into this issue? And

who can we get [sibling] connected to, that could offer support?” Four participants

SIBLING VIOLENCE AND CHILD WELFARE 104

suggested that improved communication among siblings, facilitated by their caregivers,

would be helpful. Another wanted to talk things through with her foster sibling one-on-

one “to find a solution together.” One participant noted that none of her siblings had

access to a CASA (Court Appointed Special Advocate) while they were in foster or

adoptive care, one of two who recommended youth mentorship as a solution. Mentors

were identified as a potential resource outside of the family system to talk with about

sibling violence, identify workable solutions, and to advocate for appropriate action.

Three participants suggested education about sibling violence for adults who work with

and care for children.

The study participants reported little if any adult intervention into sibling

violence. Moreover, that which was offered was not effective, especially in the long-term.

The only examples that successfully ensured sibling safety included sibling separation,

but participants lamented lost contact with their siblings and the lack of follow-up to heal

their relationships once separated. Given the inadequacy of adult responses, participants

advocated for straightforward solutions: increased attention to the problem, direct action

to intervene, and support to maintain and strengthen their sibling relationships. These

findings reflect a significant gap in awareness that is negatively impacting children

involved in child welfare.

Results of CR Retroduction

The first two phases of analysis, thematic analysis and abduction, produced three

key findings based on the resonance between participants’ descriptions of sibling

violence and the definitions, family conditions, and theories in the literature: (1) a more

SIBLING VIOLENCE AND CHILD WELFARE 105

nuanced definition of abusive sibling violence in the context of child welfare; (2) a

refined set of four key family conditions (two primary and two secondary) associated

with sibling violence, and; (3) a non-linear theory to explain how the phenomenon

manifests within families. These analytic activities also generated information about

affected youths’ experiences with adult intervention into sibling violence and their

recommendations to more effectively address the problem. Collectively, the findings

offer insight into sibling violence in relation to existing theoretical frameworks and

concepts, with emphasis on the CR “actual” domain of reality that attends to people’s

motivations for doing things.

To stop analysis here would be an epistemic fallacy, failing to consider the deeper

causal structures and mechanisms underlying events observed empirically through the in-

depth interviews. In the CR process of retroduction, the researcher applies rational

judgment to investigate the social conditions under which the causal mechanism(s) of a

given phenomenon take(s) effect. The primary goal of retroduction is to modify, support,

or reject existing theories to provide the most accurate explanation of reality. Danermark

et al. (2002) offer four guiding questions for retroduction, which I revised to apply to the

current study: (1) What is fundamentally constitutive for the structures and relations of

abusive sibling violence? (2) How is abusive sibling violence possible? (3) What

properties must exist for abusive sibling violence to be what it is? (4) What causal

mechanisms are related to abusive sibling violence? I applied the questions to assess the

extent to which the child welfare system achieves its multi-fold purpose of ensuring the

safety, permanency, and wellbeing of children within the broader U.S. political-economic

SIBLING VIOLENCE AND CHILD WELFARE 106

context. I built the retroductive analysis from the two primary family conditions

associated with sibling violence in the data (i.e., adult modeling/acceptance of violence

and un-/under-treated drug, alcohol, and mental health challenges among caregivers). I

selected these family conditions because they offered the most accessible in-roads for

emancipatory action among child welfare-involved families, an approach well-aligned

with the emphasis on critique in CR.

Child maltreatment in political-economic context. For much of modern history

child maltreatment lived in the shadows, a phenomenon largely ignored. Then, in the

1960s “battered child syndrome” (i.e., a pattern of multiple fractures in the long bones of

very small children) was discovered by pediatric radiologists (Shireman, 2015). This

discovery horrified the public and new laws were established to protect children from

physical abuse. Passage of the Federal Child Abuse Prevention and Treatment Act of

1974 identified “mandatory reporters” (i.e., working professionals required to notify child

protective services of suspected maltreatment) and allocated funding to states to develop

child protection programs. This legislation, along with newly developed state laws and

media publicity, were highly effective in bringing the issue to light. Child maltreatment

reports increased from 416,000 in 1976 to 3.4 million in 2011, an 800 percent increase

(Kudashin and Martin, 2009 as cited in Shireman, 2015).

During the same time period, the U.S. experienced two financial crises with

substantial political-economic effects. In response to the first crisis in the mid-1970’s

neoliberal ideology took hold in American politics, fueling economic and social policies

SIBLING VIOLENCE AND CHILD WELFARE 107

in support of free market capitalism and rendering substantial cuts to social welfare

funding that have persisted to present day. As described by Abramovitz (2014, as cited in

Abramovitz and Zelnick, 2015):

…in the 1970s the elite blamed “big government,” and especially the welfare

state, for their financial woes. Seeking to restore growth and profits by

redistributing income upwards and downsizing the state, they sought to undo the

New Deal and Great Society programs that they regarded as unwieldy, socially

ineffective and too expensive. The now familiar retrenchment tactics included (1)

tax cuts, (2) retrenchment of social programs, (3) devolution (shifting social

welfare responsibility from the federal government to the states), (4) privatization

(shifting social welfare responsibility from the public to the private sector), (5)

support for traditional “family values” and a color blind social order and (6)

reducing the influence of social movements best positioned to resist this austerity

program.

According to neoliberal ideology, social problems are the responsibility of the individuals

who experience them and infinite economic growth is paramount to the needs of children

and families. Rather than adequately resource the social service system through robust

policies that benefit the citizenry, family problems are treated as individual failures

deserving the bare minimum of state-funded support. The broader system is structured to

benefit the wealthy through policies that privilege capitalist interests over the public

good. As evidence of this ideological shift, from 1980 to 2013, federal discretionary

SIBLING VIOLENCE AND CHILD WELFARE 108

spending for human services dropped from 5.1 percent to 3.5 percent of GDP

(Congressional Budget Office, 2014 as cited in Abramovitz and Zelnick, 2015). Less

obvious are the many national and international economic policies that funnel resources

away from individuals and communities in service to a free market economy.

Neoliberalism has allowed for a systematic dismantling of the social welfare

system (Tonry, 1995; Wacquant, 2009; Western, 2006 as cited in Kim, 2013) with

substantial implications for child welfare. States have responded to funding shortfalls by

narrowing their definitions of what constitutes child maltreatment. Focusing attention on

the most serious cases, most financial and human resources are allocated to reactive

responses to imminent harm despite research demonstrating that prevention-based

interventions are most effective (Shireman, 2015). Many children and families with

substantiated maltreatment cases receive no services at all. Among those that do receive

services, they are often inadequate to address complex needs (i.e., short-term, focused on

a single issue). Training and retaining skilled child protective service caseworkers is also

a challenge. Large caseloads and significant emotional labor place considerable stress on

caseworkers who are perennially expected to do more with less. As the social service

system increasingly rewards efficiency, performance, and accountability, practitioners

must dedicate ever more time to tracking outcome data, further increasing workload

while detracting from direct service (Abramovitz and Zelnick, 2015). Turnover

commonly exceeds 30 percent, further increasing demands and undermining good

practice among those who remain (Nissly, MorBaarak, and Levin, 2005 as cited in

Shireman, 2015). In addition to undermining family wellbeing, members of the human

SIBLING VIOLENCE AND CHILD WELFARE 109

service workforce who are disproportionately comprised of women and people of color

suffer low wages, high stress, and burnout (Abramovitz and Zelnick, 2015).

These systemic shortcomings have spanned an era of increasing challenge for

American families. Four decades of neoliberal economic policy have produced vast

increases in income inequality, stymieing working families’ best efforts to move out of

poverty (Rogowski, 2012). A series of national drug epidemics (i.e., crack cocaine,

methamphetamine, pharmaceutical/opioid) coupled with a poorly funded behavioral

health system have further undermined family wellbeing. Employing CR retroduction to

engage critically with the phenomenon, one can logically surmise why child welfare has

struggled to address all forms of child maltreatment including abusive sibling violence.

By placing the onus on an under-funded and reactive social safety net to solve social

problems that are, from a more radical point of view, structural in nature (Mehrotra,

Kimball, and Wahab, 2016), neoliberalism operates in an oppressive function while

maintaining the false narrative that the citizenry will benefit from upward income

redistribution. Figure 10 summarizes the results of the CR retroduction phase of analysis,

organizing the three domains of reality with the iceberg metaphor. As shown, the deep

causal structures that make violent sibling interactions possible constitute the neoliberal

political-economic mechanisms that have gutted the social safety net, including the child

welfare system.

SIBLING VIOLENCE AND CHILD WELFARE 110

Figure 10. Retroduction results summarized in terms of the three domains of reality.

In summary, key findings included a refined, nascent definition of abusive sibling

violence, four family conditions associated with sibling violence, and a family systems-

based theory of sibling violence with four stable roles in the context of child welfare.

From the perspectives of the foster care alumni interviewed, very little was done to

address sibling violence by the caregivers and providers involved in their lives. The final

phase of CR analysis, retroduction, identified neoliberal political-economic mechanisms

as the deep causal structures that make violent sibling interactions possible. Chapter 5

presents recommendations informed by all three phases of analysis to address sibling

violence in the context of child welfare and beyond.

Empirical domain: Interview participants’

descriptions of sibling violence.

Actual domain: Economic disparities and an

under-resourced social safety net limit adults’

ability to intervene.

Real domain: Neoliberal political-economic

mechanisms drive social and economic policies

that privilege capitalist interests over the public

good.

SIBLING VIOLENCE AND CHILD WELFARE 111

Chapter 5: Discussion

As presented through the experiences of the eight foster care alumni who

participated in the interviews, abusive sibling violence is a problem worthy of increased

attention in the context of child welfare. The participants described living in

impoverished families, regularly evicted and without adequate access to basic needs like

food and stable housing. Caregivers struggled with serious un-/under-treated drug,

alcohol, and mental health challenges, subjecting children to horrific physical and

emotional abuse. Without adequate supervision, siblings engaged in patterns of abusive

physical and emotional violence within family-wide coercive processes that spanned

changing family configurations and years of their young lives. Once removed to out-of-

home care, siblings who were co-placed continued to engage in abusive sibling violence;

children who were placed with foster siblings experienced physical and emotional sibling

violence in those relationships as well.

Participants’ descriptions of sibling violence expose a source of “serious harm”

that falls within the federal definition of child maltreatment.

Given that many child

welfare-involved youth experience the family conditions associated with abusive sibling

violence in the data (i.e., violence modeled by adult caregivers, inadequate supervision

due to caregivers’ un-/under-treated behavioral health challenges), sibling violence is

likely to be common among foster children. The study results support the notion that

Child maltreatment is defined as “Any recent act or failure to act on the part of a parent or caretaker

which results in death, serious physical or emotional harm, sexual abuse or exploitation; or an act or failure

to act, which presents an imminent risk of serious harm” (U.S. Department of Health & Human Services,

2016, p. viii).

SIBLING VIOLENCE AND CHILD WELFARE 112

there is a necessary role for social work in addressing physical and emotional sibling

violence, yet based on the current sample, child welfare can do much more to protect

children from sibling-inflicted abuse. Among the young people interviewed, only extreme

acts of physical violence were addressed, and only in some cases. The interventions

applied, whether emergency response by law enforcement or therapeutic supports, were

inadequate to stop the behavior long-term. The only intervention that effectively

interrupted physical sibling violence was separation to different home settings. This was

applied without support to maintain or heal the sibling relationship(s), and in some cases

emotional sibling violence persisted through long-distance contact.

The lack of adult attention to sibling violence discussed in the interviews is not

surprising, given that this form of interfamilial abuse has failed to penetrate discourse on

family safety and wellbeing. Despite more than three decades of compelling scholarship

demonstrating a variety of deleterious effects among siblings in the general population,

“social norms around the acceptability of sibling aggression have likely discouraged

interest in [its] prevalence and severity” (Tucker et al., 2013, p. 214). Most states lack

specific laws protecting children from their siblings, providing mandated reporters and

child welfare caseworkers with little guidance as to how to interpret child abuse statutes

in cases of abusive sibling violence (Meyers, 2014). A focus on the parent as the primary

point of intervention may further influence caseworkers to overlook harmful sibling

relationships.

In cases where abusive sibling violence comes to the attention of a child welfare

worker, their ability to provide needed services is hampered by numerous structural

SIBLING VIOLENCE AND CHILD WELFARE 113

factors. As a result of the neoliberal economic-political project, child welfare already

struggles to respond in a timely or effective manner to adult-perpetrated maltreatment

cases. Therapeutic supports are rationed to children with extreme needs and are typically

provided to an individual child rather than a sibling group or the whole family.

Interventions focused on sibling relationship quality among foster youth are in the early

phases of development, leaving workers with few options to support siblings to change

behavior or heal from past trauma. Despite these barriers, in recent years, attention to

abusive sibling violence has increased in child welfare. Shadik et al. (2013) recommend

including discussion of sibling violence in child abuse and neglect parent education

curricula, while Perkins and O’Connor (2016) advocate for attention to sibling violence

in family-focused interventions. These ideas offer a useful starting point but given the

potential for sibling violence to exact harm, additional effort is warranted. The following

sections offer recommendations for child welfare.

Raise Awareness and Provide Basic Training for Adults Charged with Ensuring

Family Safety

The notion that sibling violence is a form of abuse is a new idea for child welfare.

Without understanding what constitutes physical and emotional abuse between siblings,

those who are charged with ensuring the safety and wellbeing of children and families

(i.e., caregivers, mandatory reporters, caseworkers, crisis service providers, and clinical

interventionists) will continue to overlook the problem. The results of the current study

offer a starting place for child welfare to begin raising awareness of abusive sibling

violence and providing basic tools to intervene. Training on the nascent definition,

SIBLING VIOLENCE AND CHILD WELFARE 114

traumatic effects, and primary and secondary family conditions associated with sibling

violence could significantly increase the likelihood that caregivers and providers will

identify abusive sibling violence when it occurs. Ready contexts for such training include

undergraduate and graduate degree programs, new employee onboarding and in-service

trainings for direct service practitioners, licensure processes for clinicians, and Foster

Parent Training (FPT) programs.

Refine and Adopt a Definition of Abusive Sibling Violence in Child Welfare

Without a clearly articulated and nationally-adopted definition of what

constitutes abusive sibling violence, child welfare workers will have difficulty identifying

the problem in the families they work with. It will not be possible to systematically track

prevalence data, and sibling violence scholars and researchers will have difficulty

bringing the results of individual studies into conversation with one another. The current

study offers a nascent definition of abusive sibling violence in the context of child

welfare: a repeated, escalating pattern of violent interactions between/among two or more

siblings that results in physical or emotional harm to at least one child, regardless of the

child’s perception of the violence as it occurs. The definition builds from the work of

more than three decades of sibling violence research and scholarship in the general

population, while adding refinements that take into consideration the contexts of foster

children’s lives. A sibling violence working group comprised of experienced caregivers,

practitioners, scholars, policy makers, and young people with lived experience could

collaboratively refine the definition and facilitate efforts to ensure that it is nationally

adopted and universally applied.

SIBLING VIOLENCE AND CHILD WELFARE 115

Systematically Assess and Track Abusive Sibling Violence Among Child Welfare-

involved Families

Sibling violence is currently recognized as the most common form of intra-

familial abuse in the general population, with various estimates suggesting that 30

percent or more of children experience severe acts of sibling violence each year (Caffaro,

2014; Finkelhor et al., 2006; Straus et al., 2006; Tucker et al., 2013). Studies have

demonstrated that sibling violence commonly co-occurs with adult-perpetrated child

maltreatment and interpersonal violence between adult caregivers (Henning et al., 1997;

Spaccarelli et al., 1994; Wallace, 1999; Wiehe, 1997). A small number of studies support

the notion that maltreated children are likely to engage in sibling/peer violence (Linares

et al., 2007; Linares et al., 2015; Shields & Cincchetti, 1998, 2001), yet the National

Child Abuse and Neglect Data System (NCANDS) does not capture sibling abuse rates.

To accurately identify cases, determine the scope of the problem, and appropriately

allocate resources to prevention and intervention, a sibling violence assessment tool

specific to child welfare-involved families should be developed, validated, and

systematically incorporated into maltreatment investigations and treatment planning.

Informed by the results of the current study, the measure should account for maltreated

children’s desensitization to interpersonal violence. Substantiated cases can then be

incorporated into state data systems, aggregated nationally, and used to devise national

and state policies that support child welfare-involved siblings in developmentally- and

culturally-appropriate ways.

SIBLING VIOLENCE AND CHILD WELFARE 116

Revisit Child Welfare Co-Placement Policy

Research has demonstrated that sibling co-placement offers numerous benefits to

child welfare-involved youth and current federal child welfare policy recommends that

siblings be placed together whenever possible. In the absence of adequate consideration

for abusive sibling violence, child welfare is applying an approach to care that appears to

be failing to protect children from sibling-inflicted trauma. To ensure that the system

provides adequate protections to children in foster care, co-placement policies should be

revisited such that cases where co-placement is not warranted are clearly defined and

ensure the agency of children to self-determine the level of contact with their siblings.

Mandated supports for children who have experienced sibling violence should be

incorporated into child welfare policies, undergo pilot testing, and include rigorous,

youth-informed evaluation to ensure effectiveness and identify unanticipated outcomes.

While sibling violence intervention research is limited, examples of potentially

efficacious offerings include individual, sibling, and family therapy, as well as the

programs discussed below.

Invest in Programs to Ensure Safe, Strong Sibling Relationships for Child Welfare-

Involved Youth

Sibling relationships are developmentally significant in childhood and across the

lifespan. Young children learn social understanding through sibling exchanges and the

quality of those exchanges influence adjustment. In early childhood, social and emotional

development, identity, and attachment are significantly affected by the sibling

relationship (Bank, Burraston, & Snyder, 2004; Caffaro, 2014; Feinberg et al.,

SIBLING VIOLENCE AND CHILD WELFARE 117

2013). Robust evidence of linkages between sibling relationship quality and peer

acceptance and social competence in childhood exists, as well as school achievement and

romantic relationships in adolescence (Feinberg et al., 2012; Feinberg et al., 2013; Pike,

Coldwell, & Dunn, 2005; Richmond, Stocker, & Rienks, 2005). Supportive sibling

relationships can serve a protective function, buffering against stressful life events such

as accidents, illnesses, disasters, and parental conflict (Cummings & Davis, 2002; Gass,

Jenkins, & Dunn, 2005). Participating in moderate sibling conflict and seeing it through

to resolution through parent-facilitated mediation can teach children to negotiate and take

another’s perspective (Siddiqui & Ross, 2004; Smith & Ross, 2007).

Sibling relationships, if nurtured, can be the most long-term and consistent

interpersonal connections in foster youths’ lives. The many instructive and protective

functions of healthy sibling relationships make a compelling case for investment in

universal sibling supports for children involved in child welfare. Clinical interventions

based on coercive family process theory hold promise to support families to remain

intact. If out-of-home placement is necessary, sibling-focused interventions may reduce

the trauma of separation from caregivers and increase the likelihood that children will

maintain meaningful, supportive sibling relationships that last into adulthood. The study

participants recommended that child welfare intervene to maintain and strengthen their

sibling relationships, with a self-determined level of sibling contact. If separation is

required to ensure safety, children should be given agency to choose whether and how to

stay in touch with their siblings and, if desired, provided with developmentally- and

culturally-appropriate supports to heal and strengthen their sibling relationship(s).

SIBLING VIOLENCE AND CHILD WELFARE 118

Interventions should be youth-informed, undergo pilot testing, and include evaluation to

build effectiveness and identify unanticipated side-effects among demographic

subpopulations.

Inform and Engage the Public to Catalyze Community-Based Solutions

The recommendations in previous sections focus on the child welfare system (i.e.,

practitioners, policy makers, and young people who have experienced maltreatment).

While it is reasonable to begin addressing sibling violence by leveraging the extensive

social welfare infrastructure already in place, the above recommendations will require a

substantial, sustained investment from an already strained system. Moreover, it is likely

that most sibling violence cases will never be reported to child welfare. American society

values family privacy and social norms largely condone sibling violence (Tucker et al.,

2013; Wiehe, 1997). Further compounding the issue, many parents and are not

knowledgeable of the traumatic effects of physical and emotional sibling violence,

dismissing the behavior as a normal demonstration if sibling rivalry that will pass as

children mature (Meyers, 2014; Caffaro & Con-Caffaro, 2005; Wiehe, 1997). Even in

cases where a parent believes one child is inflicting abusive acts upon another, to bring

the occurrence to the attention of the authorities they must file charges against one child

on behalf of the other, a rare occurrence for obvious reasons (Eriksen & Jensen, 2006).

Given these barriers, a broad-based social movement toward nonviolent sibling

interactions could be a more impactful lever for social change. Informing the public with

simple messaging about the harms that result from sibling violence and ways to

communicate to children that it is not acceptable could reduce sibling violence over time.

SIBLING VIOLENCE AND CHILD WELFARE 119

Aligning sibling violence mitigation efforts with anti-bullying campaigns and programs

could expedite efforts to increase public awareness and drive collective action.

Incorporating sibling violence into community-driven solutions to interpersonal violence

such as those arising from social justice movements could also prove fruitful. Just as

interpersonal violence between adults and adult-perpetrated child maltreatment have

penetrated public awareness as social problems worthy of attention and intervention,

systematic information dissemination coupled with a sustained call for collective action

could create a future in which sibling violence is no longer tolerated.

Organize for a More Just, Equitable Society

Much of sibling violence theorizing in the general population is based on an

uncritical acceptance of current economic and social arrangements, focusing on

individual, dyadic, and family factors. Yet the CR retroduction phase of analysis for this

project and the critical anti-violence literature identify structurally violent social and

economic policies as the real causal mechanism underlying actual and observable violent

events within families. As demonstrated through CR retroduction, the deep causal

mechanism at the root of sibling violence is neoliberal political-economic ideology,

blocking individual agency to meet basic economic and behavioral health needs and

limiting the capacity of the child welfare system to respond when family violence occurs.

In the context of child welfare, efforts to mitigate family violence are also largely based

on neoliberal ideology. The child welfare caseworker’s intervention toolkit is

predominantly focused on the family, devised to interrupt abusive behavior assumed to

be due to individual characteristics and factors in the immediate environment.

SIBLING VIOLENCE AND CHILD WELFARE 120

Researchers, theorists, and activists are increasingly advocating for systemic

solutions to social problems in hopes of creating a more just, equitable society. Recent

theoretical frameworks discuss a “crisis of modernity” in which the governing logics of

neoliberal democracy have caused rampant human suffering such as the family

conditions for sibling violence presented in the results of the current study (Cherry,

2018).

A transdisciplinary community of innovators has proposed a variety of direct

actions to confront and dismantle economic and political neoliberalism. In the field of

social work, an American New Social Work Left advocates for eliminating engagement

with neoliberalism’s insidious form of “cruel optimism” (i.e., perpetuation of the fantasy

of an achievable American Dream), militantly confronts neoliberalism’s economic and

political forms, and encourages creative thinking to imagine an entirely different world

(Cherry, 2018). The strength of these emergent strategies is in their acknowledgement of

the entanglement of the complex problems we face. Increasing attention to the deep

causal structures that produce inequity could ameliorate many of today’s most wicked

social problems, including sibling violence.

As a current example of the impact of neoliberalism on American life, the United Nations recently

determined that the living conditions of the homeless population in San Francisco, one of the most affluent

cities in the United States, is a violation of human rights (United Nations General Assembly).

SIBLING VIOLENCE AND CHILD WELFARE 121

Chapter 6: Conclusions

Adult-perpetrated child abuse is generally viewed as a well-established social

problem and public health concern, the reduction of which is “a moral imperative with

clear benefits to children and society” (Lundahl, Nimer, and Parsons, 2006, p. 251).

Since coming to the attention of the public in the 1970s, a decades-long process of

incremental transformation has shifted perceptions about child maltreatment and

significant infrastructure is now in place to protect children from adult-perpetrated harm.

While this progress is commendable, physical and emotional sibling violence persist in

the shadows, largely ignored. Positioned as the most common form on interfamilial

violence, a compelling line of research has demonstrated that children exposed to sibling

violence experience trauma symptoms and suffer deleterious effects across the lifespan.

Building on a decades-long line of research and scholarship focused on the

general population, a small number of publications have recently implored social work,

child welfare, and children’s mental health to embrace the notion that sibling violence is

a form of child maltreatment and take action to protect children from harm (Meyers,

2014; Perkins & O’Connor, 2015; Perkins, 2016; Shadik, 2013). These entities have yet

to reach agreement on what constitutes physical or emotional abuse between siblings and

theorizing about its causes is similarly nascent. To begin to fill gaps in theory, practice

and policy in the context of child welfare, I conducted in-depth interviews with eight

foster care alumni who self-identified as having experienced physical or emotional

sibling violence in their family of origin, or after transitioning into out of home care. The

SIBLING VIOLENCE AND CHILD WELFARE 122

study was grounded in critical realism (CR) with the goal of identifying actions to protect

children from sibling violence.

The study results offer much guidance to those committed to ensuring the safety

and wellbeing of children. As represented in the data, sibling violence manifests

differently in the lives of child welfare-involved youth as compared to the general

population. The study participants were unique in that they had all experienced extreme

physical and/or emotional violence exacted by adult caregivers in their families of origin;

many also reported maltreatment in out-of-home care. Participants also lived in

circumstances scarce of both instrumental and emotional supports. Building from the

work of Caffaro (2011) and Meyers (2017), the interviews informed a refined, nascent

definition of abusive sibling violence that reflects the social isolation, instrumental and

emotional scarcity, and adult-perpetrated violence so common among child welfare-

involved families: A repeated, escalating pattern of violent interactions between/among

two or more siblings that results in physical or emotional harm to at least one child,

regardless of that child’s perception of the violence as it occurs. The definition places the

onus on the adults who care for and work with child welfare-involved youth to effectively

inquire into sibling relationship quality and use what they learn to identify supports and

guide placement decisions.

The study also examined the extent to which various individual and family

conditions associated with sibling violence in the general population were represented in

the interviews. Through constant comparison analysis, I identified four family conditions

associated with sibling violence in child welfare: (1) modeling/acceptance of violence by

SIBLING VIOLENCE AND CHILD WELFARE 123

adult caregivers; (2) un-/under-treated drug, alcohol, and mental health challenges among

caregivers; (3) un-/under-treated instigator characteristics associated with various mental

health diagnoses, disabilities, and developmental delays, and; (4) poverty. Given the

prevalence of these conditions among child welfare-involved families, sibling violence is

likely to be common, an idea with preliminary support in the literature (Linares, 2015).

Yet data systems are not currently structured to accurately track abusive sibling violence

and the rates documented by child welfare are much lower than other nationally

representative sources. The study findings highlight a need for data system reform so that

child welfare can accurately assess the scope of the problem and allocate resources to

support children exposed to this form of abuse.

In addition to offering a definition of abusive sibling violence and identifying four

family conditions associated with the phenomenon, the study results present sibling

violence as a complex, synergistic, and dynamic socio-behavioral process in the context

of child welfare. While the extant literature defines sibling violence in terms of a dyadic,

unidirectional exchange with stable victim and offender roles, the study findings

resonated most strongly with coercive family process theory. All eight participants

described multiple, concurrent, stable behavioral interactions among three or more family

members that created a repeated pattern of violence. Within the coercive family processes

discussed, there was evidence of four family member roles: (1) adult and child modelers

demonstrate how to behave violently; (2) adult fortifiers explicitly or implicitly condone

sibling violence; (3) child instigators initiate violent interactions with their siblings, and;

(4) child retaliators respond to instigators with reactive violence. Many adult caregivers

SIBLING VIOLENCE AND CHILD WELFARE 124

were both modelers and fortifiers. In some cases, children were both instigators and

retaliators, taking on different roles depending on the family configuration and contextual

factors present at a given point in time. These findings suggest that holistic interventions

that engage all family members are apt to be most effective, a notion well-aligned with

participants’ recommendations to address the problem. Study participants advocated for

caregivers (and the adults who work with child welfare-involved youth) to engage in

direct action to interrupt physical and emotional violence and support children to

maintain and strengthen their sibling relationships.

Finally, CR retroduction exposed neoliberal ideology, and the associated practices

and policies that have systematically dismantled the social safety net, as the deep causal

structure that produces abusive sibling violence in the empirical domain. By constraining

caregiver agency to meet basic needs and address behavioral health challenges and

undermining the effectiveness of the child welfare system, the study participants and their

siblings were exposed to a variety of harms and challenges and the entities charged with

ensuring their safety and wellbeing lacked the fiscal and human resources necessary to

meet their needs.

Study Strengths

The study was strengthened by engaging the perspectives of young people and

blending CR with systems thinking.

Youth Perspectives. There is significant evidence that youth engaged in child

welfare do not fare well, despite substantial efforts to ensure their safety, permanency,

and wellbeing. Most research, scholarship, policy-making, and direct practice is carried

SIBLING VIOLENCE AND CHILD WELFARE 125

out in the absence of consultation from the young people whose lives are so significantly

impacted by the system. The current study attempted a youth-engaged and youth-guided

approach, inviting foster care alumni with lived experience with physical and emotional

sibling violence to contribute in a number of ways.

During conceptualization, a group of sibling study advisers from across the

country participated in a series of conference calls to offer their recommendations for the

research questions, methods, and data collection processes. In response to their guidance,

I revised the initial study design to conduct individual interviews rather than focus groups

and offered much more background about my own experiences with sibling violence than

I had originally intended. I am confident that the data were much richer because of their

guidance. I am also hopeful that their recommendations increased the likelihood that the

young people who volunteered to be interviewed for the study will be open to

contributing to research in the future.

The in-depth interview methods further strengthened the study in that the

participants were able to describe their experiences in detail while also providing

important contextual information about family and community circumstances they

perceived as influencing sibling violence. This approach revealed a dynamic process

replicated across family configurations and contexts. Their courageous offerings exposed

a significant gap in child welfare that warrants multi-faceted, youth-informed attention

and intervention. Their descriptions clearly demonstrate that much more can be done to

address abusive sibling violence in child welfare, offering many inroads to facilitate safe,

stable sibling relationships.

SIBLING VIOLENCE AND CHILD WELFARE 126

Critical Realism. Grounded in a Critical Realist paradigm, the study brought the

study participants’ experiences and perspectives into conversation with the linear and

non-linear theories discussed in the extant literature. While the more common

explanations of sibling violence rely on reductionist linear models (i.e., social learning

theory, radical feminism, and conflict theory), the current project suggests that a theory

inclusive of intrapsychic, relational, and sociocultural influences (Whitman et al., 2011)

offers a closer approximation to the realities of abusive sibling violence in the context of

child welfare. Among the families represented in the interviews, coercive family process

theory resonated most strongly, a holistic accounting of the contribution of each family

member, with consideration for immediate and more distal contextual conditions that

influence sibling violence.

Subsequent to abduction, during the retroduction phase of CR I engaged critically

with the data with consideration for the historic and political-economic context of the

phenomenon. Through this final thought exercise, I was able to identify neoliberalism as

the deep causal structure that makes sibling violence possible. From study conception to

completion, employing CR enhanced the trustworthiness of the findings by demanding

sustained engagement with the data through iterative thought exercises that facilitate

critical analysis. This approach also supported me to identify actions for caregivers,

practitioners, and policy makers to protect children from abusive sibling violence that

span the dyadic to the societal levels of the social ecology.

Systems Thinking. The study was also strengthened by incorporating systems

thinking, an emerging paradigm that includes a multitude of models, theories, and

SIBLING VIOLENCE AND CHILD WELFARE 127

practices. As a systems-oriented researcher, I developed the study with attention to the

key systems concepts of holism, nesting, and non-linearity, and executed data analysis

and synthesis with heavy reliance on visual representations. After coding the interviews, I

created a series of family system maps that displayed member characteristics,

interpersonal violence, and reinforcing feedback loops. During abduction and

retroduction, I sketched numerous diagrams representing interactions among social-

ecological structures discussed in the family violence literature, working back and forth

between the theories and data. By thinking systemically, I was able to describe the

movement, dynamics, and patterns of sibling violence as represented in the interviews

and informed by deep causal analysis. Combining youth engagement with CR and

systems thinking, the study offers practical insights and guidance for those invested in the

safety and wellbeing of children.

Implications for Future Research

While there were many strengths in the study design, a key weakness was the lack

of “extensive” data (i.e., data on widespread trends, typically gathered with quantitative

data) to look for systematic differences across a population (Lennox & Jurdi-Hage, 2017;

Sayer, 2010). It was beyond the scope of a feasible dissertation to include such an

approach in the current study, but given the findings of this project, a subsequent phase of

extensive data collection is certainly warranted. Informed by the results of the current

project, an extensive approach should account for the likelihood that young people may

avoid disclosure because they: (1) are afraid of being separated from their siblings; (2)

view sibling violence as normal due to their exposure to adult-perpetrated violence, social

SIBLING VIOLENCE AND CHILD WELFARE 128

isolation, or failure of adult caregivers to intervene in sibling violence, and; (3)

participated in the violence by retaliating against an instigating sibling. In addition to

gathering extensive data, more “intensive” research is needed to build on the results of

the current study. The study results suggest that there may be a relationship between the

extreme adult-perpetrated violence the participants experienced and their exposure to

abusive sibling violence. This is a relationship that would benefit from additional inquiry,

both in the context of child welfare and in the general population. Moreover, the current

study did not explore the relationship between demographic characteristics such as race,

ethnicity, or culture and sibling violence, another important area for future research.

Implications for Social Work

The current study has demonstrated that abusive sibling violence takes place in

the context of child welfare, and those charged with ensuring the safety of children are

not responding adequately. Given social work’s “Grand Challenge” to stop family

violence (Kulkami, Barth, & Messing 2016), efforts should be made to build attention to

sibling relationship quality into scholarship, practice, and policy-making in social work,

child welfare, and children’s mental health. Social workers are already a touchstone in

the lives of child welfare-involved youth. Though often overextended and under-

resourced due to a national political-economic context that privileges profit over family

safety and wellbeing, even simple actions such as naming the behavior and stating that it

is unacceptable could be highly impactful.

Social work would also benefit from a more radical stance on violence, embracing

the notion that all forms of violence are harmful and conducting practice and policy with

SIBLING VIOLENCE AND CHILD WELFARE 129

the goal of achieving a national culture of nonviolence. In cases of adult-perpetrated

maltreatment, an essential component of an investigation is consideration for whether an

adult has committed a crime and must be prosecuted in a court of law. The child welfare

system has thus dedicated considerable effort to identifying specific violent acts that

constitute adult-perpetrated maltreatment, necessitated by its relationship with the legal

system. I caution against taking a similar approach to sibling violence, for it is not the

extremity of an individual act of physical or emotional violence that impacts a child’s

brain architecture, but rather, continued exposure to violence in the absence of a

supportive adult (Shonkoff et al., 2009). Too much focus on definitional exactitude may

limit necessary attention to less extreme forms of violence and could diminish the

likelihood of universal access to services for child welfare-involved siblings. Moreover,

such an approach could undermine the level of agency the study participants requested, in

determining if/how to interact with their siblings when the relationship includes a violent

dynamic.

Study Limitations

All female sample. The study was exploratory and the participants self-selected

through convenience sampling methods. While discussion of representativeness is not

relevant given the epistemological underpinnings of the inquiry, it is notable that only

female participants agreed to participate. There are a variety of reasons why this may

have been the case, such as gendered differences in willingness to contribute emotional

labor, availability of support for victims of violence, the discourse of violence in policy,

practice, and theory, and norms for vulnerability and help-seeking. While it was not

SIBLING VIOLENCE AND CHILD WELFARE 130

within the scope of the study to examine gender differences in relation to willingness to

talk about sibling violence, the all-female sample has implications for future research.

Violence is a gendered phenomenon and the inclusion of male, transgender, and non-

binary voices would likely have influenced the findings. As an example, masculine-

identified individuals may be more apt to view sibling violence as normative, or see it as

a beneficial skill-building interaction that supports survival with the family or

community. Going forward, it will be important to identify ways to add male,

transgender, and additional non-binary voices to research on sibling violence in the

context of child welfare.

Individual perspectives. I conducted the interviews with foster care alumni who

identified as having experienced sibling violence before and during foster care. In the

exploratory spirit of the study, the only constraints I placed on the sampling criteria were

to ensure that participants were over the age of 18 and spoke English at the time of the

interview. The study findings are limited to the perspectives of the individuals

interviewed, discussing what appears to be a family-wide coercive process of multiple

dynamics involving children and caregivers. Gathering the perspectives of family

members involved in a violent sibling dynamic would surely enrich understanding of the

phenomenon, another area for future research.

Investigator limitations. There were many aspects of the study that I was

attempting for the first time. While I began the project with experience conducting

interviews, these were on a far more sensitive topic and covered significantly more

SIBLING VIOLENCE AND CHILD WELFARE 131

breadth and depth than any I had done previously. As I transcribed the interviews and

reviewed the transcripts, I noted missed opportunities to probe for additional clarification,

and in some cases failed cover every question in the protocol. Every study has trade-offs;

I compromised detail in order to cover more ground. Despite these shortcomings, the

protocol successfully covered all the study questions. More importantly, participants

appeared to be glad to have the opportunity to share their stories within a minimal

structure.

This was also my first attempt executing a study grounded in critical realism. The

CR literature was helpful to develop the methodology for the enquiry, but I found little

guidance about how to carry out the retroduction phase of analysis and few articles where

CR researchers discussed their process in enough detail for me to replicate it.

Retroduction blends analysis and interpretation, a tricky endeavor for any researcher. I

felt least confident in this phase of analysis, consistently wondering if my own values

were exacting too much influence on the results. I was fortunate to have the opportunity

to dialogue (by email) about retroduction with a CR researcher in Canada who is

currently publishing on the “nuts and bolts how to” of CR. Based on my exchanges with

her, as well as the recommendations from my committee chair, I believe that I was

ultimately able to complete the process in alignment with the spirit of the approach.

SIBLING VIOLENCE AND CHILD WELFARE 132

References

Abramovitz, M. & Zelnick, J. (2015). Privatization in the human services: Implications

for direct practice. Clinical Social Work, 43, 283-293.

Administration for Children and Families, Division of Health and Human Services.

(2016). The child abuse prevention and treatment act.

https://www.acf.hhs.gov/sites/default/files/cb/capta2016.pdf

Allen, K.A. (2013). Understanding bullying in an affluent, academically rigorous U.S.

high school: A grounded theory analysis. Journal of human Behavior in the Social

Environment, 23, 413-436.

Archer, M.S. (1995). Realist social theory: The morphogenetic approach. Cambridge:

Cambridge University Press.

Archer, M.S. (2010). Morphogenesis versus structuration: On combining structure and

action. The British Journal of Sociology, 225-252.

Bandura, A., Ross, D. & Ross, S. (1963). Vicarious reinforcement and imitative learning.

Journal of Abnormal and Social Psychology, 63, 575-582.

Bank, L., Burraston, B., & Snyder, J. (2004). Sibling conflict and ineffective parenting as

predictors of adolescent boys’ antisocial behavior and peer difficulties: Additive

and interactional effects. Journal of Research on Adolescence, 14(1), 99-125.

Bank, L., Snyder, J., & Prescott, A. (2002). Sibling relationship intervention in the

prevention and treatment of antisocial behavior: Rationale, description, and

advantages. Unpublished manuscript, Oregon Social Learning Center, Eugene,

OR.

SIBLING VIOLENCE AND CHILD WELFARE 133

Bank, L., & Kothari, J. (2013). Sibling intervention for conduct-problem children:

Benefits for both older and younger siblings. Manuscript submitted for

publication.

Bates, S.R. (2006). Making time for change: On temporal conceptualizations within

(critical realist) approaches to the relationship between structure and agency,

Sociology, 20(1), 143-161.

Bhaskar, R. (1979). The possibility of naturalism: A philosophical critique of the

contemporary human sciences. Atlantic Highlands, NJ: Humanities Press.

Bhaskar, R. (1998). Philosophy and scientific realism. In M. Archer, R. Bhaskar, A.

Collier, T. Lawson, & A. Norrie (Eds.), Critical realism: Essential readings (pp.

16–47). London: Routledge.

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature

and design. Cambridge, MA: Harvard University Press.

Brown, L. & Strega, S. (2005). Research as resistance: Critical, indigenous, and anti-

oppressive approaches. Toronto: Canadian Scholars’ Press/Women’s Press.

Butler-Kisber, L. (2010). Qualitative inquiry: Thematic, narrative and arts-informed

perspectives. Thousand Oaks, CA: Sage Publications Inc.

Button, D.M., & Gealt, R. (2010). High risk behaviors among victims of sibling violence.

Journal of Family Violence, 25, 131-140.

Caffaro, J.V. (2011). Sibling violence and systems-oriented therapy. In J. Caspi (Ed.),

Sibling development: Implications for mental health practitioners (pp. 245-272).

New York, NY: Springer Publishing Company.

SIBLING VIOLENCE AND CHILD WELFARE 134

Caffaro, J.V. (2014). Sibling abuse trauma: Assessment and intervention strategies for

children, families, and adults, Second Edition. New York, New York: Routledge.

Caffaro, J.V., & Conn-Caffaro, A. (1998). Sibling abuse trauma: Assessment and

intervention strategies for children, families, and adults. Binghamton, NY: The

Haworth Maltreatment and Trauma Press.

Caffaro, J.V., & Conn-Caffaro, A. (2005). Treating sibling abuse in families. Aggression

and Violent Behavior, 10, 604-623.

Caspi, J. (2008). Building a sibling aggression treatment model: Design and development

research in action. Research on Social Work Practice, 18, 575-585.

Caspi, J. (2012). Sibling aggression: Assessment and treatment. New York: Springer.

Centers for Disease Control and Prevention, Division of Violence Prevention, National

Center for Injury Prevention and Control (2014). Connecting the dots: An

overview of the links among multiple forms of violence. Retrieved from

http://www.cdc.gov/violenceprevention/pdf/connecting_the_dots-a.pdf

Cherry, K. (2018). Critical thoughts on American social work and the crisis of modernity:

Lessons from theory and current events. Journal of Progressive Human Services,

29(1), 40-60.

Child Welfare Information Gateway (2013). Sibling issues in foster care and adoption.

Washington DC: U.S. Department of Health and Human Services, Children’s

Bureau.

SIBLING VIOLENCE AND CHILD WELFARE 135

Cohn, M. (2008). Sibling placement: The importance of the sibling relationship for

children in foster care. National Resource Center for Permanency and

Family Connections Information Packet, New York, New York.

Courtney, M.E., Piliavin, I., Grogan-Kaylor, A., & Nesmith, A. (2001). Foster youth

transitions to adulthood: A longitudinal view of youth leaving care. Child Welfare

Journal, 80, 685-717.

Cranley, L.A., Doran, D.M., Tourangeau, A.E., Kushniruk, A., & Nagle, L. (2012).

Recognizing and responding to uncertainty: A grounded theory of nurses’

uncertainty. Worldviews on Evidence-Based Nursing, Third Quarter, 149-158.

Crick, N.R., & Dodge, K. (1994). A review and reformulation of social information-

processing mechanisms in children’s social adjustment. Psychological Bulletin,

115(1), 74-101.

Crosbie-Burnett, M., & Klein, D.M. (2009). The fascinating story of family theories. In

Bray, J. H. and Stanton, M. (Eds.), The Wiley-Blackwell handbook of family

psychology (37-52). Malden, MA: Wiley-Blackwell.

Cummings. E.M. & Davis, P.T. (2002). Effects of martial conflict on children: recent

advances and emerging themes in process-oriented research. Journal of Child

Psychology and Psychiatry, 43(1), 31-63.

Danermark, B., Ekström, M., Jakobsen, L., & Karlsson, J.C. (2002). Explaining society:

An introduction to critical realism in the social sciences. London: Routledge.

SIBLING VIOLENCE AND CHILD WELFARE 136

Eriksen, S., and Jensen, V. (2006). All in the family? Family environmental factors in

sibling violence. In H. Williams (Ed.), Child abuse (pp. 54-64). Farmington Hills,

MI: Greenhaven Press.

Feinberg, M.E., Solmeyer, A.R., & McHale, S.M. (2012). The third rail of family

systems: Sibling relationships, mental and behavioral health, and preventative

intervention in childhood and adolescence. Clinical Child and Family Psychology

Review, 15, 43-57.

Feinberg, M.E., Sakuma, K., Hostetler, M., & McHale, S.M. (2013). Enhancing sibling

relationships to prevent adolescent problem behaviors: Theory, design and

feasibility of Siblings are Special. Evaluation and Program Planning, 26, 97-106.

Finkelhor, D., Ormrod, R., Turner, H., & Hamby, S.L. (2005). The victimization of

children and youth: A comprehensive, national survey. Child Maltreatment,

10(1), 5-25.

Finkelhor, D., Turner, H., & Ormrod, R. (2006). Kid’s stuff: The nature and impact of

peer and sibling violence on younger and older children. Child Abuse & Neglect,

30, 1401-1421.

Finlay, L. (2002). “Outing” the researcher: The provenance, process, and practice of

reflexivity. Qualitative Health Research, 12(4), 531-545.

Fletcher, A.J. (2017). Applying critical realism in qualitative research: methodology

meets method. International Journal of Social Research Methodology, 20(2), 181-

194.

SIBLING VIOLENCE AND CHILD WELFARE 137

Frick, P., & White, S.F. (2008). The importance of callous-unemotional traits for

developmental models of aggressive and antisocial behavior. Journal of Child

Psychology and Psychiatry, 29(4), 359-375.

Gass, K., Jenkins, J., and Dunn, J. (2007). Are sibling relationships protective? A

longitudinal study. Journal of Child Psychology and Psychiatry, 48(2), 167-175.

Gelles R.J. & Steinmetz, S.K. (2006). Behind closed doors: Violence in the American

family. New Brunswick, N.J.: Transaction Publishers.

Gil, D.G. (1996). Preventing violence in a structurally violent society: Mission

impossible. American Journal of Orthopsychiatry, 66(1), 77-84.

Graham-Bermann, S.A., Cutler, S.E., Litzenberger, B.W., & Schwartz, W.E. (1994).

Perceived conflict and violence in childhood sibling relationships and later

emotional adjustment. Journal of Family Psychology, 8(1), 85-97.

Grannic, I. & Patterson, G.R. (2006). Toward a comprehensive model of antisocial

development: A dynamic systems approach. Psychological Review, 113(1), 101-

131.

Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An

experiment with data saturation and variability. Field Methods, 18(1), 59-82.

Gustavsson, N.S., & MacEachron, A.E. (2010). Sibling connections and reasonable

efforts in public child welfare. Families in Society, 91, 39–44

Hamby, S., & Grych, J. (2013). The web of violence: Exploring connections among

different forms of interpersonal violence and abuse. New York, London:

Springer.

SIBLING VIOLENCE AND CHILD WELFARE 138

Hegar, R.L. (2005). Sibling placement in foster care and adoption: An overview of

international research. Children and Youth Services Review, 27(7), 717-739.

Hegar, R.L., & Rosenthal, J.A. (2011). Foster children placed with or separated from

siblings: Outcomes based on a national sample. Children and Youth Services

Review, 33(7), 1245-1253.

Henning, K., Leitenberg, H., Coffey, P., Bennet, R., & Jankowski, M. (1997). Long-term

psychological adjustment to witnessing interparental physical conflict during

childhood. Child Abuse & Neglect, 21, 501-515.

Hoffman, K.L., & Edwards, J.N. (2004). An integrated theoretical model of sibling

violence and abuse. Journal of Family Violence, 19(3), 185-200.

Hoffman, E. (2007). Open-ended interviews, power, and emotional labor. Journal of

Contemporary Ethnography, 36, 318-346.

Hoffman, K.L., Kiecolt, K.J., & Edwards, J.N. (2005). Physical violence between

siblings: A theoretical and empirical analysis. Journal of Family Issues, 26(8),

1103-1130.

Horwath, J., Kalyva, E., & Spyru, S. (2012). “I want my experiences to make a

difference” promoting participation in policy-making and service development by

young people who have experienced violence. Children and Youth Services

Review, 34, 155-162.

Houston, S. (2001). Beyond social constructionism: Critical realism and social work.

British Journal of Social Work, 31, 845-861.

SIBLING VIOLENCE AND CHILD WELFARE 139

Hyde, J. & Kammerer, N. (2009). Adolescents’ perspectives on placement moves and

congregate settings: Complex and cumulative instabilities in out-of-home care.

Children and Youth Services Review, 31, 265-273.

Keddell, E. (2011). Reasoning processes in child protection decision making: Negotiating

moral minefields and risky relationships. British Journal of Social Work, 41,

1251-1270.

Kennedy, D.E. & Kramer, L. (2008). Improving emotion regulation and sibling

relationship quality: The More Fun with Sisters and Brothers Program. Family

Relations, 57, 567-578.

Kerig, P.K. & Stellwagen, K.K. (2009). Roles of callous-unemotional traits, narcissism,

and Machiavellianism in childhood aggression. Journal of Psychopathology and

Behavioral Assessment, 32, 343-352.

Kim, M.E. (2013). Challenging the pursuit of criminalization in an era of mass

incarceration: The limitations of social work responses to domestic violence in the

USA. British Journal of Social Work, 43,1276-1293.

Kiselica, M.S., & Morrill-Richards, M. (2007). Sibling maltreatment: The forgotten

abuse. Journal of Counseling and Development, 85, 148-160.

Kominkiewicz, F. B. (2004). The relationship of child protection service caseworker

discipline-specific education and definition of sibling abuse: An institutional

hiring impact study. Journal of Human Behavior and the Social Environment,

9(1-2), 69-82.

SIBLING VIOLENCE AND CHILD WELFARE 140

Kothari, B.H., McBeath, B., Lamson-Siu, E., Webb, S.J., Sorenson, P., Bowen, H. … &

Bank, L. (2014). Development and feasibility of a sibling intervention for youth

in foster care. Evaluation and Program Planning, 47, 91-99.

Kothari, B.H., McBeath, B., Sorenson, P., Bank, L., Waid, J. Webb, S.J., & Steele, J.

(2017). An intervention to improve sibling relationship quality among youth in

foster care: Results of a randomized clinical trial. Child Abuse & Neglect, 63, 19-

29.

Kramer, L. (2004). Experimental interventions in sibling relations. In R.D. Conger, F.O.

Lorez, and K.A.S. Wickrama (Eds.), Continuity and change in family relations:

Theory, methods and empirical findings (pp. 345-380). Mahway, NJ: Erlbaum.

Kramer, L. & Bank, L. (2005). Sibling relationship contributions to individual and family

well-being: Introduction to the special issue. Journal of Family Psychology,

19(4), 483-485.

Kuhn, T. (1962). The structure of scientific revolutions. Chicago: University of Chicago

Press.

Kulkarni, S.J., Barth, R.P., & Messing, J.T. (2016, September). Policy recommendations

for meeting the Grand Challenge to Stop Family Violence (Grand Challenges for

Social Work Initiative Policy Brief No. 3). Cleveland, OH: American Academy of

Social Work & Social Welfare.

Lennox, R. & Jurdi-Hage, R. (2017). Beyond the empirical and the discursive: The

methodological implications of critical realism for street harassment research.

Women's Studies International Forum, 60, 28-38.

SIBLING VIOLENCE AND CHILD WELFARE 141

Lenz-Rashid, S. (2004). Employment experiences of homeless young adults: Are they

different for youth with a history of foster care? Children and Youth Services

Review, 28, 235-259.

Linares, L.O. (2008). The sibling bonds project. Unpublished manuscript. NYU Child

Study Center, New York University, New York, NY.

Linares, L.O., Jimenez, J., Nesci, C., Pearson, E., Beller, S., Edwards, N., & Levin-

Rector, A. (2015). Reducing sibling conflict in maltreated children placed in

foster homes. Prevention Science, 16, 211-221.

Linares, L.O., Li, M.M., Shrout, P.E., Brody, G.H., & Pettit, G.S. (2007). Placement

shift, sibling relationship quality and child outcomes in foster care: A controlled

study. Journal of Family Psychology, 21, 736-743.

Loeber, R. & Stouthamer-Loeber, M. (1998). Development of juvenile aggression and

violence. American Psychologist, 53(2), 242-259.

Lopez, P., & Allen, P.J. (2007). Addressing the health needs of adolescents transitioning

out of foster care. Pediatric Nursing, 33, 345-356.

Maxwell, J.A. (2012). A realist approach to qualitative research. Los Angeles: Sage

Publications, Inc.

Mayring, P. (2000). Qualitative content analysis. Forum: Qualitative Social Research,

1(2).

McBeath, B., Kothari, B.H., Blakeslee, J., Lamson-Siu, E., Bank, L., Linares, L.O.... &

Shlonsky, A. (2014). Intervening to improve outcomes for siblings in foster care:

SIBLING VIOLENCE AND CHILD WELFARE 142

Conceptual, substantive, and methodological dimensions of a prevention science

framework. Children and Youth Services Review, 39, 1-10.

McDonald, C. & Martinez, K. (2016). Parental and others’ responses to physical sibling

violence: A descriptive analysis of victims’ retrospective accounts. Journal of

Family Violence, 31, 401-410.

Mehrotra, G. R., Kimball, E., and Wahab, S. (2016). The braid that binds us: The impact

of neoliberalism, criminalization, and professionalization on domestic violence

work. Journal of Women and Social Work, 31(2), 153-163.

Meyers, A. (2014). A call to child welfare: Protect children from sibling abuse.

Qualitative Social Work, 13(5), 654-670.

Meyers, A. (2017). Lifting the veil: The lived experience of sibling abuse. Qualitative

Social Work, 16(3), 333-350.

Milevsky, A. (2011). Sibling relationships in childhood and adolescence. New York:

Columbia University Press.

Miller, L.E., Grabell, A., Thomas, A., Bermann, E., & Graham-Bermann, S.A. (2012,

February 13). The associations between community violence, television violence,

intimate partner violence, parent-child aggression, and aggression in sibling

relationships of a sample of preschoolers. Psychology of Violence. Advance

online publication. doi: 10.1037/a0027254

Morris, T. (2006). Social work research methods. Thousand Oaks, CA: Sage

Publications.

SIBLING VIOLENCE AND CHILD WELFARE 143

Morse, J.M. (1995). The significance of saturation. Qualitative Health Research, 5(2),

147-149.

Nagel, T. (1989). The view from nowhere. Oxford: Oxford University Press.

National Scientific Council on the Developing Child (2005/2014). Excessive stress

disrupts the architecture of the developing brain: Working paper No. 3. Updated

edition. Retrieved from www.developingchild.harvard.edu.

Nicoletti, A. (2007). Perspectives on pediatric and adolescent gynecology from the allied

health professional. Journal of Pediatric Adolescent Gynecology, 20, 205-206.

Noland, V., Liller, K., McDermott, R., Coulter, M., & Seraphine, A. (2004). Is adolescent

sibling violence a precursor to college dating violence? American Journal of

Health Behavior, 28, S13-S23.

Oakley, A. (1981). Interviewing women: A contradiction in terms. In H. Roberts (Ed.),

Doing feminist research (pp. 30-61). London: Routledge and Kegan Paul.

O’Rielly, M. & Parker, N. (2012). ‘Unsatisfactory saturation’: A critical exploration of

the notion of saturated sample sizes in qualitative research. Qualitative Research,

13(2), 190-197.

Parr. S. (2015). Integrating critical realist and feminist methodologies: Ethical and

analytical dilemmas. International Journal of Social Research Methodology,

18(2), 193-207.

Patterson, G.R. (1982). Coercive family process. Eugene, OR: Castalia.

SIBLING VIOLENCE AND CHILD WELFARE 144

Perkins, N.H. (2014). Perceptions and experiences of physical and emotional violence

between siblings: A mixed-methods, comparative case study (Doctoral

dissertation). Virginia Commonwealth University, Richmond, Virginia.

Perkins, N.H., & O’Connor, M.K. (2016). Physical and emotional sibling violence: A

necessary role for social work. Social Work, 61(1), 91–93.

Perkins, N.H., & Stoll, A.J. (2016). Practical considerations of physical and emotional

sibling violence in foster families: A summary of the current literature. Journal of

Family Social Work, 19(5), 407-419.

Pike, A., Coldwell, J., & Dunn, J.F. (2005). Sibling relationships in early/middle

childhood: Links with individual adjustment. Journal of Family Psychology,

19(4), 523-532.

Pinel-Jacquemin, S., Cheron, J., Favart, E., Dayan, C., & Scelles, R. (2013). Violence

among siblings and joint placement: A review of the literature. Early Child

Development and Care, 183(7), 963-980.

Randolph, W., & Viswanath, K. (2004). Lessons learned from public health mass media

campaigns: Marketing health in a crowded media world. Annual Review of Public

Health, 25, 419-437.

Reid, W.J., & Donovan, T. (1990). Treating sibling violence. Family Therapy, 17(1), 49-

59.

Reilly, T. (2003). Transition from care: Status and outcomes of youth who age out of

foster care. Child Welfare, 82, 727-746.

SIBLING VIOLENCE AND CHILD WELFARE 145

Richmond, M.K., Stocker, C.M., & Rienks, S.L. (2005). Longitudinal associations

between sibling relationship quality, parental differential treatment, and children’s

adjustment. Journal of Family Psychology, 19(4), 550-559.

Riebschleger, J., Day, A., & Damashek, A. (2015). Foster care youth share stories of

trauma before, during, and after placement: Youth voices for building trauma-

informed systems of care. Journal of Aggression, Maltreatment, & Trauma, 24,

339-360.

Rogowski, S. (2012). Social work with children and families: Challenges and possibilities

in the neo-liberal world. British Journal of Social Work, 42, 921-940.

Rowntree, M. (2007). Responses to sibling sexual abuse: Are they as harmful as the

abuse? Australian Social Work, 60(3):347-361.

Ross, K. (2017). Making empowering choices: How methodology matters for

empowering research participants. Forum: Qualitative Social Research, 18(3),

Art. 12.

Sayer, A. (1992). Methods in social science: A realist approach. London: Routledge.

Sayer, A. (2010). Method in social science: A realist approach, revised (2

ed.). New

York: Routledge.

Schonkoff, J. P., Boyce, W. T., & McEwen, B. S. (2009). Neuroscience, molecular

biology, and the childhood roots of health disparities: building a new framework

for health promotion and disease prevention. JAMA, 301(21), 2252-2259.

Schwandt, T.A. (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage.

SIBLING VIOLENCE AND CHILD WELFARE 146

Shadik, J.A., Perkins, N.H., & Kovacs, P.J. (2013). Incorporating discussion of sibling

violence in the curriculum of parent intervention programs for child abuse and

neglect. Health and Social Work, 38(1), 53-57.

Shireman, J.F. (2015). Critical issues in child welfare, second edition. New York, New

York: Columbia University Press.

Shields, A., & Cicchetti, D. (1998). Reactive aggression among maltreated children: The

contributions of attention and emotion dysregulation. Journal of Clinical Child

Psychology, 27(4). 381-395.

Shields, A., & Cicchetti, D. (2001). Parental maltreatment and emotion dysregulation as

risk factors for bullying and victimization in middle childhood. Journal of

Clinical Child & Adolescent Psychology, 30(3). 349-363.

Siddiqui, A., & Ross, H. (2004). Mediation as a method of parent intervention in

children’s disputes. Journal of Family Psychology, 18(1), 147-159.

Smith, S.R. & Hamon, R.R. (2012). Exploring family theories (3

Ed.). New York:

Oxford University Press.

Smith, J. & Ross, H. (2007). Training parents to mediate sibling disputes affects

children’s negotiation and conflict understanding. Child Development, 78(3), 790-

805.

Smith, J.D., Dishion, T.J., Shaw, D.S., Wilson, M.N., Winter, C.C., & Patterson, G.R.

(2014). Coercive family process and early-onset conduct problems from age 2 to

school entry. Development and Psychopathology, 26, 917-932.

SIBLING VIOLENCE AND CHILD WELFARE 147

Solomon, D.T., Niec, L.N., & Schoonover, C.E. (2017). The impact of Foster Parent

Training on parenting skills and child disruptive behavior: A meta-analysis. Child

Maltreatment, 22(1), 3-13.

Spaccarelli, S., Sandler, I., and Roosa, M. (1994). History of spouse violence against

mother: Correlated risks and unique effects in child mental health. Journal of

Family Violence, 9, 79-95.

Spielfogel, J.E., Leathers, S.J., Christian, E., & McMeel, L.S. (2011). Parent management

training, relationships with agency staff, and child mental health: Urban foster

parents’ perspectives. Children and Youth Services Review, 33, 2336-2374.

Staff, I. & Fein, E. (1992). Together or separate: A study of siblings in foster care. Child

Welfare, 71(3), 257-270.

Straus, M. A. (1979). Measuring intrafamily conflict and violence: The conflict tactics

(CT) scales. Journal of Marriage and Family, 41, 75-88.

Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (2006). Behind closed doors: Violence in

the American family. New York, NY: Anchor Press/Doubleday.

Strolin-Goltzman, J., Kollar, S., & Trinkel, J. (2010). Listening to the voices of children

in foster care: Youths speak out about child welfare workforce turnover and

selection. Social Work, 55(1), 47-53.

Swaim, R.C. & Kelly, I. (2008). Efficacy of a randomized trial of a community and

school-based anti-violence media intervention among small-town middle school

youth. Prevention Science, 9, 202-214.

SIBLING VIOLENCE AND CHILD WELFARE 148

Thomas, B.W. & Roberts, M.W. (2009). Sibling conflict resolution skills: Assessment

and training. Journal of Child and Family Studies, 18, 447-453.

Tucker, C.J., Finkelhor, D., Shattuck, A.M., & Turner, H. (2013). Prevalence and

correlates of sibling victimization types. Child Abuse & Neglect, 37, 213-223.

Tudge, J.R.H., Mokrova, I., Hatfield, B.E., & Karnik, R.B. (2009). Uses and misuses of

Bronfenbrenner’s bioecological theory of human development. Journal of Family

Theory & Review, 1, 198-201.

U.S. Department of Health & Human Services (2015). Children's Bureau focus areas.

[Web content]. Retrieved from http://www.acf.hhs.gov/programs/cb/focus-

areas/child-abuse-neglect

U.S. Department of Health & Human Services. (2016). Child maltreatment 2016.

Washington DC: Administration of Children and Families. Retrieved from

https://www.acf.hhs.gov/sites/default/files/cb/cm2015.pdf

United Nations General Assembly (2018). Report of the special rapporteur on adequate

housing as a component of the right to an adequate standard of living, and on th

eirhg to non-discrimination in this context. Retrieved from

http://www.undocs.org/A/73/310/rev.1

Wakefied, M. A., Loken, B., & Hornik, R. C. (2010). Use of mass media campaigns to

change health behavior. The Lancet, 376, 1261-1271.

Walker, L.E. (1986). Psychological causes of family violence. In Lystad, M. (ed.),

Violence in the Home: Interdisciplinary Perspectives (pp. 71-98), New York, NY:

Brunner/Mazel.

SIBLING VIOLENCE AND CHILD WELFARE 149

Walker, A.J., Allen, K.R., & Connidis, I.A. (2005). Theorizing and studying sibling ties

in adulthood. In Bengtson, V. L., Acock, A. C., Allen, K. R., Dilworth-Anderson,

P., and Klein, D. M. (Ed.). Sourcebook of family theory & research (pp. 167-190).

Thousand Oaks, CA: Sage.

Wallace, H. (1999). Family violence: Legal, medical, and social perspectives. Needham

Heights, MA: Allyn & Bacon.

Washington, K. (2007). Research review: Sibling placement in foster care: a review of

the evidence, Child and Family Social Work, 12, 426-433.

Whiteman, S. D., McHale, S. M., & Soli, A. (2011). Theoretical perspectives on sibling

relationships. Journal of Family Theory and Review, 3(2), 124–139.

Wiehe, V. R. (1997). Sibling abuse: Hidden physical, emotional, and sexual trauma.

Thousand Oaks, CA: Sage Publications, Inc.

Wiehe, V. R. (1998). Understanding family violence: Treating and preventing partner,

child, sibling, and elder abuse. Thousand Oaks, CA: Sage Publications, Inc.

Wright, K. (2011). Reality without scare quotes: Developing the case for critical realism

in journalism research. Journalism Studies, 12(2), 156-171.

World Health Organization (2002). World report on violence and health. Retrieved from

http://whqlibdoc.who.int/publications/2002/9241545615_eng.pdf?ua=1

World Health Organization (2009). Changing cultural and social norms that support

violence. Retrieved from

http://www.who.int/violence_injury_prevention/violence/norms.pdf

SIBLING VIOLENCE AND CHILD WELFARE 150

Wojciak, A., McWey, L. M., & Helfrich, C. M. (2013). Sibling relationships and

internalizing symptoms of youth in foster care. Children and Youth Services

Review, 35(7), 1071-1077.

Wulczyn, F. & Zimmerman, E. (2005). Sibling placements in longitudinal perspective.

Children and Youth Services Review, 27, 741-763.

Zwick, M. (2015). Elements and relations: Aspects of a scientific metaphysics.

Manuscript in preparation.

SIBLING VIOLENCE AND CHILD WELFARE 151

Appendices

Appendix A: Recruitment Flyer

SIBLING VIOLENCE AND CHILD WELFARE 152

A research study is seeking foster care alumni to learn more about physical and

emotional sibling violence.

Your participation may help improve child welfare services for others.

You will get $20 for your participation.

You are eligible if:

• You are 18-24 years old

• You have a history of involvement in foster care

• You have at least one sibling

• You were a perpetrator or victim of physical or emotional sibling violence before

or during your involvement in child welfare

Physical violence includes hitting, punching, kicking, biting, or using a weapon like a

knife or a gun against a sibling. Emotional violence can include humiliating a sibling,

using words or threats to make a sibling feel unsafe, destroying a prized possession or

harming a pet. These are just examples; there are lots of other ways siblings are violent

with each other.

Your participation in the study will include:

• A face-to-face interview with the researcher – Katie will work with you to

arrange a meeting location that is convenient for you

• Filling out a background information form

If you are interested in participating, please contact the researcher:

Katie Winters, M.A.

Ph.D. Student, Portland State University

k.winters@pdx.edu

Cell phone: (503) 956-6432

Foster Care alumni who have worked with Katie are also available if you would like to

ask them about their experiences:

Tim Dennis (Knoxville, TN): tburns58@gmail.com (865) 256-1391

Ashley Strange (Washington, D.C.): ashleystrange@outlook.com (202) 468-8041

Ke’Onda Johnson (West Palm Beach, FL): keondaj14@gmail.com (561) 572-7141

Do you have experience with foster care?

Did you experience physical or emotional violence

with a sibling?

SIBLING VIOLENCE AND CHILD WELFARE 153

Appendix B: Participant Informed Consent Document

SIBLING VIOLENCE AND CHILD WELFARE 154

Sibling Violence Study - Consent to Participate

Please sign and return one copy of this consent form. Keep the other for your records.

What is this research study about?

A PhD student at Portland State University named Katie Winters is interested in learning

about sibling violence among child welfare-involved youth. You are being asked to be in

this study because you have past experience with the child welfare system and sibling

violence.

You are eligible if:

• You are 18-24 years old

• You have a history of involvement in the child welfare system

• You have at least one sibling

• You were a perpetrator or victim of physical or emotional sibling violence before

or during your involvement in child welfare

What will I be asked to do?

If you decide to be in this study, you will be asked to do the following:

1. Participate in a 90-minute interview with the researcher

2. Fill out a background information form

The interview will be used to talk about your experiences with sibling violence. Physical

violence includes hitting, punching, kicking, biting, or using a weapon like a knife or a

gun against a sibling. Emotional violence can include humiliating a sibling, using words

or threats to make a sibling feel unsafe, destroying a prized possession or harming a pet.

These are just examples; there are lots of other ways siblings are violent with each

other. You will be asked to talk about:

• Your relationship(s) with your sibling(s)

• Your relationships with your other family members, and their relationships with

each other

• The community where you were living when you experienced sibling violence

• The things that adults such as your case worker, counselor, or foster parent(s)

did (or didn’t do) to address the violence between you and your sibling(s)

• Your ideas to help siblings who are experiencing violent interactions

These discussions will be audio-recorded, with your permission.

What will I get in return?

 You will receive a $20 to thank you for your participation

SIBLING VIOLENCE AND CHILD WELFARE 155

 Satisfaction of knowing your participation may help improve child welfare

services for other youth who experience sibling violence

What are the risks and benefits?

Talking about your experiences with sibling violence can be very personal. Some of the

questions may be sensitive for you and you may feel uncomfortable, angry, sad, guilty,

or embarrassed. You do not have to answer any questions that you’re not comfortable

with and can stop whenever you’d like. If you are upset after the interview and need to

talk with someone, I encourage you to use the resources below.

• 211 info – Bi-lingual: By dialing 211 on your phone or going to 211info.org, you

will receive confidential assistance to access roughly 3,000 agencies and over

50,000 low cost and free programs to help people in Oregon.

• Oregon Coalition Against Domestic Violence: 503-230-1951

• National Alliance of Mental Illness Oregon Chapter: 503-230-8009

• National Suicide Prevention LifeLine: 1-800-273-8255

There is also a small risk that your information will be seen by somebody other than

Katie and her research supervisor. However, she will follow strict rules to protect your

confidentiality. Steps she will take to protect you are listed in the next section. There are

no direct benefits for participating in this study.

Mandatory Reporting

It is the investigator’s legal obligation to report child abuse, child neglect, elder abuse,

harm to self or others or any life-threatening situation to the appropriate authorities,

and; therefore, your confidentiality will not be maintained.

What are you doing to protect me?

Your confidentiality is very important. Katie will do many things to protect you:

• She won’t tell anyone if you take part in this study or not.

• What you share will be kept confidential. Your name and any information about

you will be kept confidential. By “kept confidential” I mean that the names of

people in the study will not be given to anyone else and, when I write or talk

about what I learned in this study, I will do my best to make it so no one will be

able to tell who I am talking about.

• I will store all information you give me in a locked briefcase, a locked cabinet, or

on a password-protected computer and/or secure server.

Participation is Voluntary

It is up to you if you want to be in this study. If you choose to be in the study, you do not

have to answer any questions you don’t want to and you can stop at any time. If you

SIBLING VIOLENCE AND CHILD WELFARE 156

decide not to be in the study, or to drop out, you may do so without it affecting any

services you receive.

Alternative to Being in the Study

The alternative to being in this study is to choose not be in the study. Your participation

is voluntary. Your choice to participate will NOT have an impact on ANY of the services

you receive now or in the future.

Who to Call with Questions

If you have any concerns about the research study, you may contact Stephanie Wahab

at (503) 725-5083. She is the person who directs the study. If you have any concerns

about your rights in the study, you may contact the PSU Office of Research Integrity,

1600 SW 4th Ave., Market Center Building, Ste. 620, Portland, OR 97201; phone (503)

725-2227 or 1 (877) 480-4400.

If I sign the form, what does it mean?

This is a consent form. Your signature below means that:

1. You have read and you understand what this form says.

2. You are willing to be in the study.

3. You know that you do not have to be in this study and you can change your mind

and stop at any time.

4. You know that being in this study has nothing to do with any of the services you

receive.

5. You keep a copy of this form.

I have read this form and agree to be in this study.

First Name Last Name

Signature Date

Audio Recording

In order to ensure that I can accurately document responses to the questions, I am

asking for your permission to audio-record the interview. The audio-recording will be

kept confidential and stored in a password-protected file. The recordings will be

destroyed no earlier than three years after the completion of the study. You have the

right to stop the recording at any time during the interview or your optional follow-up

interview without penalty. You also have the right to agree to participate in the study

without being audio-recorded. Your signature below means that:

• You have read and you understand what this form says.

SIBLING VIOLENCE AND CHILD WELFARE 157

• You are giving permission to be audio-recorded.

Signature Date

SIBLING VIOLENCE AND CHILD WELFARE 158

Appendix C: Data Collection Instruments: Interview Protocol and Background

Form

SIBLING VIOLENCE AND CHILD WELFARE 159

Interview Protocol

Foster Care Alumni Perspectives on Abusive Sibling Violence

Introduction

Welcome! Before we get into any of this, I first want to thank you for your willingness to

talk with me today. This is an important topic, and it can also be a tender topic, so I

want to acknowledge your courage first thing.

I think it’s also important for me to let you know about my experience with sibling

violence, to give you a bit of understanding about where I’m coming from. I have a

brother who is two years older than me, and who was very violent with me when we

were growing up [provide examples]. Those interactions affected different parts of my

life in powerful ways, and so I’ve chosen to focus on learning about siblings in my

graduate school program.

There have been some big studies about violent sibling relationships, and I would say

that there is strong evidence that it’s a pretty common problem. None of the studies I

have come across have included the voices of foster care alumni, so I would like to use

my research project to hear young people’s perspectives on sibling violence. It’s my

hope that this research will help case workers, therapists, foster parents… really

anybody who works with young people involved in child welfare to have information

about sibling relationships and what to do if siblings are having trouble.

So today we’ll be using the time to hear your perspectives on the topic of sibling

violence. By violence I mean physical violence and emotional violence. Physical violence

includes hitting, punching, kicking or biting, or using a weapon like a knife or a gun

against a sibling. Emotional violence can include humiliating a sibling, using words or

threats to make a sibling feel unsafe, destroying a prized possession or harming a pet.

These are just examples, and there are lots of other ways that siblings are violent with

each other. I’d like to keep this conversation focused on physical and emotional sibling

violence, so if you have experiences with sexual violence, please try to keep your sharing

focused on physical and/or emotional violence.

This conversation will be confidential. What I mean by confidential is that I won’t share

your comments with anyone else.

Do you have any questions for me before we begin? [Respond to any questions posed

by the interview participant]

Would it be okay for me to audio-record our conversation? I won’t share the recording

with anyone- it’s just to ensure that I accurately capture the information you share

SIBLING VIOLENCE AND CHILD WELFARE 160

during the interview. I can take notes instead of recording if that is your preference. [If

yes, begin recording. If no, use paper/pencil to take notes during the interview.]

OK, that’s enough from me.

Interview Questions

1. I’d like to begin by having you tell me about your biological family (your family of

origin). While we talk I’d like to work together to draw a map of the people in

your family system (here’s an example). This will help me to understand the

family members who were involved in your life when you were experiencing

sibling violence (show example).

a. What was/were your relationship(s) like with each of the people in your

family system before you became involved with child welfare?

b. What were their relationships like with each other? Was there physical or

emotional violence between other people in your family?

i. Were there sources of stress in your family that affected the

relationships you’ve described?

1. Did your family have financial struggles?

2. Did members of your family have problems with drugs or

alcohol?

3. Did family members have mental health challenges?

2. I’d also like to hear about the community where you lived, when you lived with

your biological family.

a. What was your neighborhood like? Did it feel like a safe place to live?

b. What kinds of families lived in your neighborhood? What were your

neighbors like? Did you know your neighbors? How was your relationship

with them?

c. What was your school like? How did you feel when you were at school?

Were there problems at your school with drugs or violence?

3. Now I’d like to have you tell me about the physical and or emotional violence

between you and your sibling(s).

a. What did your interactions look like when you weren’t getting along?

i. Could you please describe a specific example of a physically and

or emotionally violent exchange?

ii. From your perspective, what usually lead up to (or caused)

physical or emotional violence with your sibling?

iii. Was one person usually the perpetrator and another person

usually the victim?

iv. What would happen afterward?

SIBLING VIOLENCE AND CHILD WELFARE 161

v. How often did these kinds of exchanges happen?

vi. When this was happening, did you see it as abusive?

vii. How did your parents/caregivers respond when you and your

sibling(s) were being physically or emotionally violent with each

other?

1. How helpful were the things they did to support your

sibling relationship?

2. Can you think of other things they could have done that

might have worked better?

viii. Were there other adults in your life that tried to help you and

your sibling(s) to have a better relationship at that time?

1. How helpful were the things they did?

2. Can you think of other things they could have done that

might have worked better?

ix. Thinking about the ways that adults intervened (or didn’t), how

has that shaped your perspective about adults or authority?

4. Now I’d like to talk about your relationship with your sibling(s) after you went

into care.

a. Did you live together or apart?

b. What was/were your relationship(s) like with your siblings during the

time you were in foster care? How would you compare your relationship

at that time, to when you were living with your biological family?

5. During that time, when you were in foster care, what kinds of things did adults in

your life, such as your caseworker, foster parents, counselors, or others do that

were focused on your relationship(s) with your sibling(s)?

a. What kinds of things did they do to help you and your sibling(s) have a

good relationship?

b. How helpful were the things they did to support your sibling relationship?

c. Did they do anything that made your relationship with your sibling(s)

worse?

d. Can you think of other things they could have done that might have

worked better?

e. What do you recommend to address sibling violence?

6. What helped you to get through those tough times with your sibling(s)?

a. Were there things that you did to keep yourself safe?

b. Can you think of qualities about yourself that helped you through those

experiences?

SIBLING VIOLENCE AND CHILD WELFARE 162

c. Were there any other sources of support or resilience that you can think

of?

7. What is/are your relationship(s) with your sibling(s) like now?

a. Did you rekindle or heal the relationship? How? What supports were

helpful?

b. What are your wishes for the future of your relationship(s)?

Thank the study participant. Complete the Background Information Form.

SIBLING VIOLENCE AND CHILD WELFARE 163

Participant Background Information

1. Age in years:

2. Gender:

□ Female

□ Genderqueer/Androgynous

□ Intersex

□ Male

□ Transgender

□ Transsexual

□ Cross-dresser

□ FTM (female-to-male)

□ MTF (male-to-female)

□ Other (please specify):

□ Prefer not to answer

3. Do you consider yourself to be:

□ Heterosexual or straight

□ Gay or lesbian

□ Bisexual

□ Other

□ Prefer not to answer

4. What is your ability status? (please check all that apply):

□ Deaf or have serious difficulty hearing

□ Blind or have serious difficulty seeing even when wearing glasses

□ Because of a physical, mental, or emotional condition, I have serious

difficulty concentrating, remembering, or making decisions

□ Serious difficulty walking or climbing stairs

□ Have difficulty dressing or bathing

SIBLING VIOLENCE AND CHILD WELFARE 164

□ Because of a physical, mental, or emotional condition, I have difficulty doing

errands alone such as visiting a doctor’s office or shopping

□ Other:

□ None/Prefer not to answer

5. What is your race or ethnicity? (Please check all that apply)

□ Alaska Native

□ American Indian

□ Asian

□ Black/African American

□ Hispanic, Latino/-a, or of Spanish origin

□ Middle Eastern

□ Native Hawaiian/Other Pacific Islander

□ White

□ Other:

□ Unknown/Prefer not to answer

6. Would you be willing to have another conversation with Katie about sibling

violence?

□ Yes □ No

If yes, please provide your contact information below

Name:

Telephone Number:

Email Address:

SIBLING VIOLENCE AND CHILD WELFARE 165

Appendix D: Final Coding Scheme

SIBLING VIOLENCE AND CHILD WELFARE 166

Code

Definition

Coding Rule(s)

D1. Caffaro &

Meyers' definitions

supported

Supports any aspect of the definition of

abusive sibling violence put forward by

Caffaro and/ or Meyers:

A repeated pattern of violent

interactions

An escalating pattern of violent

interactions

Stable victim/ offender roles

Perpetrator motivated by a need for

power and control

Victim experiences a sense of terror

Victim experiences a sense of

powerlessness

Victim perceives perpetrator's behavior

as abusive

Any single or multiple

aspects of the

definition is/ are

supported

D2. Caffaro &

Meyers' definitions

challenged

Any single or multiple aspects of the

definition is/are challenged (i.e.,

opposite description is provided):

One-time events

Stable pattern that does not escalate

Victim/ offender roles are not clear

and/ or shift over time

Perpetrator motivation is something

other than a need for power or control

Victim is not afraid of the perpetrator

Victim feels powerful in the face of

perpetrator's aggressions/ responds in-

kind

Victim minimizes perpetrator's

behavior or perceives it as normal/

typical/ acceptable

Code if any aspect of

the definition is

challenged

FC1. Family

condition -

parental

unavailability

Parents are physically and/ or

emotionally unavailable as a result of:

Inability to express love, empathy,

caring for the child(ren)

Regular/severe alcohol/ drug use

Chronic/ severe mental health

challenges

Chronic/ severe physical health

challenges

Any single or multiple

aspects of the

definition is/ are

supported

SIBLING VIOLENCE AND CHILD WELFARE 167

FC2. Family

condition - lack of

supervision

Parent(s) do(es) not provide

expectations for or enforce healthy

child behaviors

Must co-occur in the

data along with/be

described as resulting

from parental

unavailability (FS1)

FC3. Family

condition -

parentified child

One sibling is expected to parent

others; caretaking expectations are

developmentally inappropriate given

the child's age/ abilities

Must co-occur in the

data along with/ be

described as resulting

from parental

unavailability (FS1)

FC4. Family

condition -

differential

treatment

Parent(s) favor or scapegoat one sibling

as compared to other(s)

Favoritism/

scapegoating must be

a stable, long-term

pattern of interaction;

not a one-time event

FC5. Family

condition -

work/financial

strain

Parent(s) described as experiencing

work/ financial strain such that:

Parent(s) is/are unable to secure stable

employment

Parent(s) work is illegal (e.g., drug

dealer)

Parent(s) work(s) more than full-time

to meet basic needs

Parent(s) unable to meet family's basic

needs due to un-/ under-employment

Any single or multiple

aspects of the

definition occur

immediately prior to

and/ or at the same

time as sibling

violence

FC6. Family

condition -

acceptance/

modeling of

violence

Parent(s) communicate(s) that violence

is an appropriate solution to

interpersonal problems within the

family system by:

Encouraging/ allowing/ ignoring

sibling

physical/ emotional violence (SEE

FS2)

Physical/ emotional violence (IPV)

between adult caregivers

Adult-perpetrated physical/ emotional

child maltreatment

Physical abuse

includes: hitting,

kicking, biting,

beating up, use of a

weapon

Emotional abuse

includes: ridiculing,

insulting, threatening,

belittling, destruction

of personal property,

harming a pet

Also code if

participant describes

behavior as "abuse"/

"abusive"

SIBLING VIOLENCE AND CHILD WELFARE 168

FC7. Family

condition - siblings

differ

Sibling qualities that prime perpetrators

to enact physical/ emotional violence or

predispose victims to receive abuse:

Perpetrator Qualities: thinking errors,

cognitive/ developmental deficits, lack

of impulse control, lack of empathy,

callout-unemotional traits, narcissism,

tendency to inaccurately interpret

social cues

Victim Qualities: genetically

determined physical, emotional, mental

characteristics (disabled,

developmentally delayed, exhibit

behaviors others experience as

irritating

Large developmental, physical,

intellectual, or social/ emotional

differences between siblings in a dyad

in which the perpetrator is superior and

the victim is inferior

Description of a

sibling(s) experiencing

physical/ emotional

abuse embody any of

the traits listed

SLT1. Social

Learning Theory

Sibling violence described as resulting

from social learning. Physical/

emotional violence is...

Learned through observation

Results in receipt of desired rewards

(e.g., gaining control of an object,

parental attention, the pleasurable

experience of power from a sibling’s

fearful response)

Not redirected/ does not result in

consequences (SEE FS2)

Any single or multiple

aspects of the theory

is/ are supported

RFT1. Radical

Feminist Theory

Sibling violence described as…

Justified means for the more powerful

person in a dyad/ situation to get what

they want (SEE FS2)

Due to a power differential (i.e.,

physical strength, intellectual/

emotional maturity, level of

responsibility) between perpetrator and

victim (SEE FS7)

Giving the perpetrator a sense of

control by overpowering another

Either or both aspects

of the theory is/ are

supported

SIBLING VIOLENCE AND CHILD WELFARE 169

CT1. Conflict

Theory

Sibling violence is described as

occurring in a setting in which siblings

experience…

Scarcity - Lack of access to emotional

or tangible resources perceived as

scarce (SEE FS5, FS1)

Competition - for parental attention

Parental favoritism - One sibling gets

better "stuff", whatever the stuff is

(SEE FS4)

Any single or multiple

aspects of the theory

is/ are supported

CFP1. Coercive

family process

Sibling violence is described as

occurring in a family system with…

Multiple concurrent dynamics between

3+ members

Family members embody stable roles

Mutual training - Repeated, cyclical

interpersonal reactions reinforce

violence

Feedback loops - Responses support

perpetuation of violence

Emotional, behavioral, or cognitive

responses perpetuate violence

Any single or multiple

aspects of the theory

is/ are supported

SEM1. Social

Ecological Model

Interactions across community and

family contexts produce/ support

violence (i.e., "open systems"

conception of reality). Focus is on

community (school, neighborhood,

etc.) factors that interact with the

family level of the ecology to produce/

support interpersonal violence.

Community factors

include poverty,

crime, and community

violence

I1. Intervention -

Real Time -

Helpful

Description of helpful adult

intervention into sibling violence:

Verbal, physical, or other efforts to

intervene into violent exchanges in real

time

Code any real-time

intervention effort

described as helpful to

halt/ reduce sibling

violence

I2. Intervention -

Real Time - NOT

Helpful

Description of unhelpful adult

intervention into sibling violence:

Verbal, physical, or other efforts to

intervene into violent exchanges in real

time

Code any real-time

intervention effort

described as unhelpful

to halt/ reduce sibling

violence

SIBLING VIOLENCE AND CHILD WELFARE 170

I3. Intervention –

Formal - Helpful

Description of helpful adult

intervention into sibling violence:

Preventative or reactive formal

interventions - therapy, programs

Code any formal

intervention effort

described as helpful to

halt/ reduce sibling

violence

I4. Intervention –

Formal - NOT

Helpful

Description of unhelpful adult

intervention into sibling violence:

Preventative or reactive formal

interventions - therapy, programs

Code any formal

intervention effort

described as unhelpful

to halt/ reduce sibling

violence

I5. Intervention -

Recommendations

Participant-generated recommendations

to intervene into sibling violence

Code any participant-

generated

recommendations to

intervene

PSV Origin.

Physical Sibling

Violence in Family

of Origin

Hitting, kicking, biting, beating up, use

of a weapon

Any single or multiple

physical sibling

violence types are

reported in family of

origin

PSV Foster.

Physical Sibling

Violence in Foster

Care

Hitting, kicking, biting, beating up, use

of a weapon

Any single or multiple

physical sibling

violence types are

reported in foster care

ESV Origin.

Emotional Sibling

Violence in Family

of Origin

Ridiculing, insulting, threatening,

belittling, destruction of personal

property, harming a pet

Any single or multiple

emotional sibling

violence types are

reported in family of

origin

ESV Foster.

Emotional Sibling

Violence in Foster

Care

Ridiculing, insulting, threatening,

belittling, destruction of personal

property, harming a pet

Any single or multiple

emotional sibling

violence types are

reported foster care

Break1. Stop-point

of Sibling

Violence

The point at which sibling violence

ends

Code the final stop-

point- may include

physical separation or

end to violence while

cohabitating

SIBLING VIOLENCE AND CHILD WELFARE 171

R1. Resilience

Sources of resilience that support

subjects to navigate sibling violence

Code reply to

interview question:

What helped you get

through it?

SIBLING VIOLENCE AND CHILD WELFARE 172

Appendix E: Family Map Example

SIBLING VIOLENCE AND CHILD WELFARE 173

Physical violence

Emotional violence

Drug abuse

Alcohol abuse

Mental health challenges

Map 1. Participant 0-5 years old

• Dad was a drug dealer and user. Bipolar – diagnosed and

medicated.

• Dad was verbally and physically violent with mom (e.g.,

controlling, choked her, threw her up against the wall). “You could

tell my mom was afraid of my dad.”

• Mom abused alcohol and described as “paranoid.”

• Mom was a teen parent to participant and older sister.

• Dad was “controlling” of participant and sister

• Participant and older sister fought over toys; Dad verbally

reinforced that violence was an appropriate way for siblings to

resolve disagreements.

• Dad was incarcerated for attempt to do bodily harm  Map 2

SIBLING VIOLENCE AND CHILD WELFARE 174

• Mom “was a mess” (heavy user of alcohol and meth) and “not

around much.”

• Sister was away a lot - started having sex “because no one was

around… all the time trying to be gone,” … “we didn’t have a

relationship almost.”

• Mom went to jail for meth  Map 3

♀

Participant

Older Sister

3 yrs. older

Bio Mom

Map 2. Participant 5-9 years old

• Brief family formation.

• Dad “liked to have control” and sister didn’t like it. Sister

opted to move in with mom  Map 4

Map 3. Participant 9 years old

♀

Participant

Older Sister

3 yrs. older

Bio Dad

SIBLING VIOLENCE AND CHILD WELFARE 175

• Sister lived separately, with mom.

• Dad was physically abusive. Hitting that left marks. Also

“scaring you” and “trying to be in control” by “having a

screaming feud.”

• Noteworthy that dad’s physical violence followed his time in

prison (likely a highly violent environment).

• Dad went to prison again for selling drugs  Map 5

Map 4. Participant 9-12 years old

♀

Participant

Bio Dad

SIBLING VIOLENCE AND CHILD WELFARE 176

• Most pronounced sibling violence (i.e., weekly)

• Mom was depressed (meth use side-effect)

• Sibling Violence Patterns:

o Sister would be “really mean” to mom (response to differential

treatment?)  Participant would then hit sister to “set her

straight”

o Sister “would instigate it… she will abuse you verbally until you

do something about it”  Participant would retaliate against

verbal abuse with physical violence.

 Sister would physically fight back (lay on her back

and kick).

• Mom and boyfriend have an intense drug episode  Map 6

♀

Participant

Older Sister

3 yrs. older

Bio Mom

Map 5. Participant 12-13 years old

SIBLING VIOLENCE AND CHILD WELFARE 177

• Adoptive home

• Sibling physical and emotional violence until aunt

intervened. Emotional violence continued until older sister

moved out.

• Sister went to “juvi”, residential home, and participant

never lived with her again.

♀

Participant

Older Sister

3 yrs. older

Aunt

(mom’s sister)

Map 6. Participant 13-14 years old

SIBLING VIOLENCE AND CHILD WELFARE 178

Theory Synthesis

Caffaro & Meyers – Definitions of Abusive Sibling Violence

SUPPORTED – repeated escalating pattern, stable victim/offender roles (sister initiates

conflict with emotional violence, participant responds with physical violence – stable

instigator/retaliator roles)

[Participant] She would know. I would literally just look at her and she would know

and she would run... that's what she does. She runs and then she does this thing

where she falls on her back. She would like, kick, so you can't hit her. This one time

she was doing that and she kicked me hard, and I flew all the way down the stairs.

[Interviewer] How often was that?

[Participant] Probably every week or something... Not always her kicking me down

the stairs or me punching her five times in the head, but just like, we would fight.

She would instigate it. She knew that was what I would turn to, physical violence.

She was more like, verbal. Like she would abuse you verbally until you do something

about it.

If my sister said something really mean to my mom, on multiple occasions it

happened, I would hit her. I would run after her, wherever she would go, and just hit

her wherever I could. And so that was typical... I would usually hit her on places like

her head so you could never see it. She'd just be like "My head hurts, I have a big

knot now." “Well, you shouldn't have said that. You shouldn't have done that." That's

what I would say.

NOT SUPPORTED – no evidence of need for power/control, perception of perpetrator’s

behavior as abusive, victim sense of terror, or victim sense of powerlessness

Social Learning Theory

SUPPORTED – participant learned and then replicated physical violence from bio dad.

This occurred in the absence of redirection. The “reward” was a sense of

accomplishment for protecting mom

We really started fighting, like being physical toward each other, after I lived with

my dad. I think if I was never really like, introduced to that, I wouldn't be hitting her.

I would hit her and stuff. Probably because my dad was hitting me before, and then I

came and I was like, "This is how you deal with problems. This is what you do."

Because that's what [dad] does. He hits people, and so, that's the way to do things.

Radical Feminist Theory

SIBLING VIOLENCE AND CHILD WELFARE 179

SUPPORTED – General acceptance of violence to meet desired ends is supported by bio

dad (modeling) and bio mom (failure to intervene)

She used to take my toys from me all the time, and then she would hit me (giggles).

And I would hit her with my toys. Like I specifically remember a time she took this

phone away from me. We both had our own phones, but she took mine. I took it back

from her and I hit her on the head. And I will never forget it. And my parents were

like ‘Don’t take her stuff’, and I felt pretty cool. I was like, you learned a lesson,

right?

My mom was fine with it. I said, "Mom, I'm gonna fight this girl tomorrow, so be

prepared to pick me up" and she was like "Okay." She was like, "I'll be waiting at

work for you to give me a call." I was like, "Okay, cool."... you see, nobody made it

not normal. It happened to me, so it was normal, from my dad. And then my mom

wasn't like "Don't hit your sister. Don't go beat this girl up."

Conflict Theory

SUPPORTED – Poor family, parents are drug dealers with addiction and mental health

problems, siblings have differential access to fashionable clothes, good food, clear

demonstrations of parental favoritism

I was eating all these cool things [from our food stamps] at the table and then my

sister came to get something to eat… and my mom comes out there like "What do

you think you're doing?" and my sister was like, "Getting some food," and my mom

was like "No. You can eat bread and butter for all I care." ...I always felt bad when

my mom would be really mean to my sister, because she was so nice to me, and that

obviously hurt my sister... I can't be out here eating a platter of all these different

great foods we just bought and you be like "You can eat bread and butter for all I

care" to my sister, while she's watching me eat whatever I want.

Coercive Family Process

SUPPORTED – Triad of mom, sister, and participant creates a patterned dynamic in

which sister initiates through verbal abuse, to either mother or sister. Participant then

retaliates with physical violence (replicated from abuse received from father), and sister

responds with physical violence

[Participant] She would like, be really mean to my mom, like make my mom cry, and

I was like "I have to straighten her out," that's what I thought.

[Interviewer] Straighten her out?

[Participant] Yeah, like I would hit her and stuff.

SIBLING VIOLENCE AND CHILD WELFARE 180

I feel like my mom put us in that situation. So like, what could she have done?

Nothing. She was putting me in the situation, so if she was like "Don't hit her," I

would have been like, "I'm hitting her for you."

Social Ecological Model

SUPPORTED – community violence, “drug” communities, school bullying

We always lived in the same type of town… drug communities. Where I'm from meth

is huge. That's like, normal, not a big deal, honestly. People being on drugs, like

pills, and that kind of stuff… a lot of kids in foster care.

Fourth grade was the first time I experienced being bullied... I feel like rules were

not implemented. It was not a good school. I was bullied on the bus, I was bullied in

school... then in 5th grade, 6th and 7th grade I lived with my dad... I also

experienced bullying there... I was learning how to be really mean in school.

Because I had to be... I just feel like everyone did it... I got in a fist fight [in middle

school] and got suspended. Middle school I was bullying people, people were

bullying me.

SIBLING VIOLENCE AND CHILD WELFARE 181

Appendix F: Human Subjects Application to IRB

SIBLING VIOLENCE AND CHILD WELFARE 182

IRB APPLICATION for EXPEDITED / FULL REVIEW

IMPORTANT: PSU faculty and students must submit any research plan involving Human

Subjects to the IRB for review. Use this application to request Expedited or Full review human

subjects’ research approval. If you believe the activities are Exempt, you may use the IRB

Exempt Application. If you believe the activities do not meet the definition of “human

subjects research” complete the Review Not Required Form and submit to hsrrc@pdx.edu.

See Instructions page of this application for more details.

Important: (1) Hard-copy submissions will not be accepted; submit electronically. (2) All

questions must be answered, please enter N/A for questions that do not apply. (3) If the

research is funded, a copy of the research proposal must be submitted to complete the

application.

Section I: Investigator’s Assurance

This is a new protocol submission

This is a revised initial review protocol submission with requested modifications

This is an amendment submission

Indicate which Sections are revised: (Check each applicable section and include all

protocol revisions in red text or use track changes – see Instructions on Pg. 3)

Section I: Section II (indicate which parts: A-T): Section III (indicate

changed attachments/addendums):

Principal Investigator (or faculty advisor for students): Stephanie Wahab, Ph.D. (faculty advisor)

E-Mail: wahabs@pdx.edu

Co-Principal Investigator: E-Mail:

Other Personnel (GA, Project Mgr., etc.): E-Mail:

Department: School of Social Work Campus Mail Code: SSW Preferred Phone #: (503) 475-

3442

Title of Protocol: Study Examining Foster Care Alumni Perspectives on Abusive Sibling Violence

Mailing Address: P.O. Box 751-SSW, Portland, OR 97207-0751

Proposed Duration of Project (months/years): 1 year Anticipated Start Date: February, 2018

Is this project funded? Yes Not yet (Application has been submitted) No

Type of Funding: Federal State Foundation Other Funding Agency:

PIAF #:

NOTE: If this is a funded project, a copy of the research proposal must be submitted.

SIBLING VIOLENCE AND CHILD WELFARE 183

STUDENTS ONLY:

Master’s Thesis PhD/EdD Dissertation (Approval Date:) Other:

Under advisement from the above faculty member, I verify that I will conduct this

research in accordance with PSU’s Human Subjects Research Review Policy.

Student Name: Katie Winters, M.A. (type in your name and email electronic copy to your PSU mentor)

PSU Student ID #: 925034915 Email: k.winters@pdx.edu Date: January 12, 2018

Investigator’s Responsibilities and Assurances:

(Mark each box with an  when understood/agreed/certified)

I understand PSU’s policies concerning research involving human subjects and:

1. I understand that I have ultimate responsibility for the protections of the rights and

welfare of human participants, the conduct of this study, and the ethical performance of

this research.

2. I will maintain IRB related documents (including signed consent forms, as applicable)

for a minimum of three years after the completion of the study.

3. I understand that it is my responsibility to ensure that all study personnel receive the

mandatory human subjects’ research protection education (either CITI or NIH) and to

maintain a training documentation file.

I agree to:

4. Comply with all PSU/IRB policies, decisions, conditions and requirements.

5. Obtain prior approval from the IRB before amending or altering the research

protocol or changing the approved consent/assent form.

6. Notify the Office of Research Integrity of the development of any financial interest

not already disclosed.

7. Notify the Office of Research Integrity for all adverse events and unanticipated

problems as soon as possible. In case of DHHS supported activities, I will also report

these problems to the Department of Health and Human Services (through the

respective granting office).

I certify that:

8. The time and resources are available to complete this project.

9. The equipment, facilities, and procedures to be used in this research meet

recognized standards for safety.

10. New information that may affect the risk-benefit assessment for this research will be

reported to the Office of Research Integrity.

11. I agree to ensure adequate supervision of all research study personnel and to meet

with the investigator(s), if different then myself, on a regular basis to monitor progress.

SIBLING VIOLENCE AND CHILD WELFARE 184

12. The information provided in this application and all attachments is complete and

correct.

Signature of Principal Investigator or Faculty Advisor: Stephanie Wahab

Date:

1/12/18

(Type in name and submit by email to hsrrc@pdx.edu )

Instructions-IRB Application for Expedited/Full Review

Application Requirements:

• The IRB application for Expedited/Full Review has three parts:

1. Investigator’s Assurance cover sheet (Section I)

2. Project Narrative (Section II)

3. Appendices (Section III)

• All questions must be answered. Please enter N/A for questions that do not

apply.

• Consent documents must be written in at least 12 pt. font.

• Applications must be page numbered, including Appendices.

• Submit complete applications by email, to hsrrc@pdx.edu

. Only Microsoft Word

of Adobe PDF files will be accepted.

• IRB amendment submissions: Amendments to protocols initially submitted on or

after May 1, 2014 must use this form. Include new information in red text or

Track Changes.

• The Investigators Assurance serves as the researcher’s contact information page

and signature of assurance. This form must be filled out completely and

accompanied by proper signatures. Electronic signatures will be accepted from

@pdx.edu email accounts.

• Information for student research only:

1. Graduate/undergraduate students cannot function as Principal Investigators

(PIs).

2. Application must be signed and submitted by the advisor/PI (i.e., the faculty

advisor must complete and sign the Investigator’s Assurance as PI).

3. The student must sign the “Students Only” box on the bottom portion of the

Investigator Agreement and provide their PSU ID number.

4. Graduate Studies requires PhD students to have committee approval of their

dissertation prior to IRB submission (contact GSE for more details).

5. Student investigators may not include themselves as a human participant in

their research. Also, recruitment of human participants from their immediate

family, friends and associates should be avoided.

‘Human Subjects’ Definition and Resources:

SIBLING VIOLENCE AND CHILD WELFARE 185

‘Human subjects’ is a legally defined term. In this document the terms participant,

respondent and human subject or subject are all referring to the legally defined term

“Human subject.”

Information on activities meeting the definition of “human subjects’ research” can be

found on the “Review Not Required Form.” Additionally, there are two areas

researchers often ask about:

The following MEET the definition of human subjects’ research:

1. NSF human subjects’ activities, including performance data shared with NSF.

(NSF retains all shared data in a data repository for future research purposes.)

2. Data collection about live humans that involves maintaining this data in a

repository for future research purposes.

How to “Unprotect” Application to Insert Red Text or Use Track Changes:

In Microsoft Word 2010 or 2013, click the “Restrict Editing” icon on the Developer Tab.

In the sidebar, click “Stop Protecting.” No password is required. Text editing in the form

should be now allowed. For other versions of Word, or if you have any difficulties in

removing document protections, please contact ORI at hsrrc@pdx.edu or 725-2227.

Section II: Project Narrative (complete sections below)

A. Research Description:

1. Explain why, what, how, who and when.

i. Why: (i.e., describe specific study aims, research questions to be studied,

study goals and a brief description of the scientific background

After decades of incremental transformation, western society no longer

condones adult-perpetrated child abuse, nor interpersonal violence between adult

family members. While this progress is commendable, violence between siblings

remains conspicuously absent from discourse on family safety and wellbeing.

Labelled with trivializing terms such as “sibling rivalry” or dismissed as an

innocuous phenomenon that children will simply outgrow, research offers

compelling evidence to the contrary; child victims suffer numerous deleterious

effects across the lifespan (Hoffman, Kiecolt, and Edwards, 2005, Kramer and

Bank, 2005, Finkelhor et al., 2006, Button and Gealt, 2010, Graham-Bermann and

Cutler, 1992, as cited in Caffaro, 2011, Feinberg et al., 2012). These effects are of

great concern, given that sibling violence is now widely recognized as the most

common form of intra-familial abuse; various estimates garnered from the general

population suggest that 30 percent or more of children experience severe acts

violence inflicted by a sibling each year (Finkelhor and Dziuba-Leatherman, 1994;

Finkelhor, Ormrod, Turner, and Hamby, 2005; Straus and Gelles, 1990).

SIBLING VIOLENCE AND CHILD WELFARE 186

Numerous studies have drawn linkages between a child’s exposure to

violence and their propensity to inflict violent acts upon others. Children who

witness domestic violence between their parents are more likely to engage in violent

behavior with their siblings and peers; sibling abuse is also more prevalent in

families in which both spousal and child abuse are present (Button and Gealt, 2009;

Henning et al., 1997; Spaccarelli, Sandler, and Roosa, 1994; Wallace, 1999, Wiehe,

1997). Co-occurrence of family violence types suggests that sibling violence is

common among families served by the child welfare system, yet child welfare

caseworkers, with their focus on interventions to mitigate harms perpetrated by

adult caregivers, may miss indicators of sibling abuse. In cases where sibling

relationship problems are noted, case workers may neglect to intervene because

adult harms take precedence when family issues are triaged for intervention.

The proposed study will employ interview methods to gather qualitative data

from foster care alumni who have experienced physical and emotional sibling

violence. The research questions guiding the study are: (1) How do young people age

18-26 with a history of involvement in foster care describe their experiences with

physical and emotional sibling violence? (1.a.) To what extent are theories of sibling

violence represented in their descriptions? (1.b.) To what extent do the young people

perceive the violence as abusive? (2) How do adults who care for or work with

young people involved in the child welfare system (e.g., foster parents, case workers,

mental health providers, kinship caregivers, etc.) respond to sibling violence? (2.a.)

To what extent were responses perceived as helpful? (2.b.) What do foster care

alumni recommend to address sibling violence?

ii. What & How: (i.e., describe what the researchers and the participants will

be doing and how these activities will be accomplished.)

Study participants will be comprised of a convenience sample recruited

through FosterClub (https://www.fosterclub.com/_allstars/article/about-all-stars),

New Avenues for Youth, and Portland State University Better Futures. Recruitment

will begin with dissemination of the Recruitment Flyer (included as an attachment).

The flyer will be distributed by each of the agencies listed. Foster care alumni who

are willing to participate in the project will contact the researcher directly to learn

more about the study. They will also receive the contact information of three foster

care alumni who served as advisers for this study if they would like to hear about

their experiences working with me prior to contacting me about the study.

The interview data will be collected in-person, at a location determined by

each interview participant. I will only interview participants within a 1-day drive of

Portland, OR and will work with each respondent to identify a location at is safe

and private for the interview. Each interview will begin with the consent process,

discussed in subsequent sections. After completing the consent process, each study

participant will participate in an individual interview with the researcher, including

completion of the family system map (included as an attachment).

SIBLING VIOLENCE AND CHILD WELFARE 187

Upon conclusion of the interview, each participant will be asked to complete

the Background Information form.

Who: (i.e., describe who the participants are and how they will be identified.)

Study participants will be recruited to meet the following criteria: (1) foster care

alumni ages 18-26 years old with a history of involvement in the Child Welfare system; (2)

with at least one (self-defined) sibling; (3) and for whom the study participant describes their

sibling relationship as having been physically or emotionally violent in the period leading up to

or during their involvement in foster care. The researcher will recruit participants via a

convenience sampling process through agencies that serve foster care alumni.

iii. When: (i.e., describe the order of research activities in a timeline.)

Recruitment will begin in February 2018 (pending study approval). Interview(s)

will occur from February through July 2018. The project will be complete 12-

months following the project start date.

Activity

Dates

Recruitment - distributes recruitment flyers

February-June, 2018

Interviews

February-July, 2018

Data Analysis & Synthesis

March-November, 2018

Study Completion – Notify PSU IRB

December, 2018

B. Study Design & Setting

1. Describe the study design:

The dissertation study will investigate abusive sibling violence among youth

involved in the child welfare system via in-depth interviews that include completion

of a family system map and completion of a background information form which

will capture basic demographic information. The study is grounded in a critical

realism (CR) philosophy of science, and is designed with the goal of building theory

of abusive sibling violence in the context of child welfare. by generating emergent

themes describing research participants’ attitudes and experiences. With

participant permission, the interview(s) will be recorded to accurately capture the

dialogue which takes place. I will transcribe qualitative data from the interview(s)

for review and analysis.

The data analysis process will be researcher and theory-driven, including three

key steps: (1) identification of demi-regularities; (2) abduction, and; (3) retroduction

(Fisher, 2017). The process is non-linear, moving from the concrete, to the abstract,

and then returning to the concrete (Sayer, 1992). Specifically, I will employ

Mayring’s (2000) deductive category application process, which works with

theoretically derived codes to organize textual data. Theories discussed in the

sibling violence literature will be the categories for the initial coding scheme, as will

SIBLING VIOLENCE AND CHILD WELFARE 188

the key critical realism concepts of structure, agency, and open systems. The

categories will be continuously refined through a constant comparison analysis

approach which will occur over the course of data collection. The codes that are

most prevalent (i.e., dominant) in the data will comprise the demi-regularities.

Once the demi-regularities have been identified, I will engage in abduction and

retroduction, thought exercises serving to elucidate meaningful causal theory that

incorporates not only underlying structures, but also the human actions which

fortify or dismantle them. Abduction will consist of a cross-walk between the demi-

regularities and the previously developed theories and CR concepts with the goal of

uncovering causal mechanisms of abusive sibling violence. I will use retroduction to

explore the contextual conditions necessary for a particular causal effect to manifest

at the empirical level. Retroduction focuses on theorizing about the social, cultural,

and/or historical conditions in which a causal mechanism takes effect (Fletcher,

2017; Lennox & Jurdi-Hage, 2017).

To establish trustworthiness I will maintain an audit trail throughout the study

such that it will be possible to follow the process of data analysis and synthesis back

to the original transcripts. I will engage a member of members of my dissertation

committee to review examples from the audit trail for guidance about data

reduction as needed. My committee will also serve as a panel of auditors who will

review the initial deductive categories, definitions, examples, and coding rules I

develop to guide the initial phase of analysis (i.e., identification of demi-regularities).

At appropriate junctures over the course of the data collection and analysis process

I will again engage the members of my committee to review the categories as they

are expanded and refined. Specifically, I will do this whenever I substantially revise

the initial categories or add numerous and highly divergent additional categories to

the analysis framework. Finally, I will carry out reflexive memoing throughout the

data collection and analysis phases to explore how my own conceptions bear upon

the process of interpretation. I will debrief with my academic adviser and student

peers to discuss the issues that arise in the memos in order to grow as a researcher

and to garner emotional support to engage with this sensitive topic.

2. Identify the sites or locations where the research/data analysis will be

conducted: To be determined

3. Describe the Principal Investigator’s experience conducting research at study

site(s) (or similar sites) and familiarity with populations and communities: The

student researcher has experience conducting in-depth interviews with

pregnant women and new mothers receiving home visiting services for low-

income women who self-report risk factors including IPV, drug/alcohol

problems, a history of abuse/neglect including Adverse Childhood

Experiences. For the study I conducted interviews with women in their

homes throughout the state of Oregon. In addition to this interviewing

experience, I have experience working with foster care alumni through two

past projects implemented in collaboration with FosterClub. First, I

SIBLING VIOLENCE AND CHILD WELFARE 189

conducted a pilot of this study with FosterClub interns (i.e., foster care

alumni participating in a leadership development program). Second, I

worked with four foster care alumni who served as “sibling study advisors”

providing consultation on this study via a series of conference calls. Finally,

as a mid-career evaluator, I have been involved in many evaluations of

health and human service programs including school readiness, child

maltreatment prevention, public health, and out of school time programs,

among others. These projects have included site visits and face-to-face

interviews and focus groups with a variety of program stakeholders

including parents, children, and students.

4. Is the research conducted outside the United States? Yes No

a. If yes, describe site-specific regulations or customs affecting the

research, local scientific and ethical review structure: N/A

5. Are there any permissions that have been, or will be, obtained from

cooperating institutions, community leaders, or individuals, including approval

of an IRB or research ethics committee? Yes No

a. If yes, provide a list of the permissions (also include copies with the application,

if available): Permission has been obtained via email communication from

Celeste Bodner, FosterClub Chief Executive Director, Founder, Non-Voting

Board Member. I will seek approval from New Avenues for Youth and the PUS

RTC for Pathways to Positive Futures once this amendment is approved.

6. Does the research require approval from other PSU compliance committees?

(e.g., Radiation Safety Committee (RSC), Institutional Animal Care and Use

Committee (IACUC), and Institutional Biosafety Committee (IBC), etc.)

Yes

If yes, the PI is responsible for seeking approval from the other committees required

for this research. Work cannot start until final approval is received from all

appropriate committees. List each compliance committee review required: N/A

7. Provide an approximate number of subjects to be enrolled and justify the sample size:

Interview participants will count 6-12 foster care alumni living within a 1-day round-trip

drive from Portland, OR. The sample size is small, per the CR philosophy and constructivist

epistemology grounding the study. In addition, this is an unfunded dissertation project and so

is subject to resource constraints.

(Provide information for each subject group, as defined in the sections 8A and 8B

below. For example, minors’ #, crime victims’ #s, etc.):

8. Approximate total number of subjects to be recruited: 6-12

a. Please identify subjects that will be recruited by checking all that apply in 8A

and 8B. Submit additional materials as required.

A. Children or Adult: Check all that apply

Age

Consent/Permission /Assent Required

SIBLING VIOLENCE AND CHILD WELFARE 190

Birth to 3 years

Parental Permission Form

4-7 years

Parental Permission Form and Verbal

Child's Assent

8-17 years

Parental Permission Form and Child's

Written Assent

18 & over

Written Consent

B. Potentially Vulnerable Populations: If potentially vulnerable populations will be

recruited, identify these groups by checking below.

Neonates/Fetuses

Children (Complete Addendum 4 and include in application.)

Prisoners (Complete Addendum 5 and include in application. If using prisoner

data sets collected for other than research purposes complete Addendum 5a and

include in application.)

Pregnant women

Decisionally impaired (for groups not already identified on this list)

HIV/AIDS patients

Native American Tribes

Crime victims

Substance abusers

Persons living outside the U.S.

Non-English speaking

Terminally ill

Institutionalized individuals

College Students

Other: Young adults who have aged out of the foster care system

9. Are there groups of people purposefully being excluded? Yes No

A. If yes, identify the groups that are being excluded [Check all that apply in 9A and

explain the reasons for exclusion in 9B below]:

Ethnic/racial groups Non-English speaking

Adults 65 or older Sexual orientation

Children (under 18) Marital status

Pregnant women Religion

Males Other:

SIBLING VIOLENCE AND CHILD WELFARE 191

Females

B. Explain the reasons for the exclusion criteria identified in #9A: I am mono-lingual so

cannot conduct interviews in languages other than English. The pilot is not funded

so does not include budget for language interpretation services.

10. Describe safeguards to protect the rights and welfare of vulnerable

populations:

Participants will represent a ‘vulnerable’ population. Considerations for sense

of safety are therefore paramount to ensuring a positive participant experience and

obtaining meaningful data. I will address these key social dynamics in a number of

ways which are informed by the literature and the guidance of the sibling study

advisors. First, I have built information about myself into the data collection

instrument and I will use this rapport-building process to foster a two-way

relationship that models a save level of self-disclosure about my own experiences

with sibling violence and may serve to disrupt the research-participant power

imbalance. Second, I will explicitly convey that the interviewee holds the power to

guide the process in that they can decide where we meet, which questions to answer,

and that they can stop participating at any time. Third, throughout my engagement

with each participant I will emphasize an ethic of care that privileges empathy and

mutual respect over capturing data. For example, if a participant demonstrates a

desire to discuss a topic unrelated to the focus of the inquiry, I will allow time and

space for this as a demonstration that their needs are important, valid, and are

equal to my own.

The alumni who participate in the research face a risk of psychological distress-

this could occur during the interview(s), or after data collection is complete. I

understand that talking about personal experiences can make some people feel

uncomfortable, angry, sad, guilty, or embarrassed. I will work to mitigate these

sources of discomfort by emphasizing the importance of avoiding over-disclosure

and to not say anything that they are not comfortable sharing. Study participants

will be reminded that their participation is voluntary and that they can take a

break, not respond to some questions, or choose not to continue with the pilot at any

time. I will interact in a calm, respectful manner and I will be alert to instances of

distress. If needed, I will pause the interview and inquire about their desire to

continue (or need for a break) if it appears that any participant is becoming

distressed.

There is also a small risk that information shared during data collection will be

seen by those other than me. During the consent process I will emphasize the

following points which will be included in the consent form:

1. I will not talk to anyone about individuals who participate in the study

2. Participant information will be kept private including names or other

potentially identifying information

SIBLING VIOLENCE AND CHILD WELFARE 192

3. All study information will be stored in locked cabinets and on password-

protected computers

4. Recordings will be transported in a locked briefcase and uploaded onto a

secure server

5. I (Katie Winters) and my dissertation supervisor are the only people

who will see participant information

In addition to these risks, by participating in the alumni will be at increased

risk of being subject to mandatory reporting if they share that they plan to hurt

themselves or others, or report that they are in imminent danger. In such case I will

call the appropriate authorities. I will provide detailed information about my role as

a mandatory reporter during the consent procedures.

(See Additional Requirements for Research with Vulnerable Populations for

guidance regarding children, prisoners and participants who become

incarcerated after enrolling. Contact ORI for guidance regarding human

fetuses and neonates.)

C. Data Collection Methods

Check all method(s) to be used (Include copies of all the data collection methods

checked in Survey/Questionnaire or Interview sections below, including

translations, if applicable.):

1. Survey/Questionnaire – Identify modality(ies)

In person Web-based E-mail Postal mail

Telephone

Other:

2. Interview – Identify modality(ies)

One-on-one Focus group Oral history Other:

3. Observation of Public Behavior – Identify modality(ies)

Classroom Public meetings Other:

4. Examination of Archived Data/Secondary or Records

Briefly describe the records to be examined:

5. Taste Evaluation

Wine/alcohol *Non-wholesome food Genetically altered food

*Wholesome food may meet Category 6 exemption. Fill out Exempt form.

6. Examination of Human Pathological or Diagnostic Tissue Specimens (e.g.,

blood, bodily fluids)

SIBLING VIOLENCE AND CHILD WELFARE 193

7. Unproven or Untested Procedures

Biomedical Psychological Other:

If any checked, describe:

8. Recordings – Identify type(s)

Voice Video Photograph/Image

Check Method of recording: Analog Digital

Check the purpose of the recordings: For transcription Other

If checked ‘Other’ explain: (For example, recorded for speech pattern analysis,

archiving purposes, presentation at the meetings, etc.)

9. Internet:

10. Social Media:

11. Other:

D. Recruitment Methods

Does the study involve the recruitment of participants? Yes No

1. Describe recruitment/advertising methods:

Participants will be initially recruited through the FosterClub, New Avenues for Youth, and

the PSU RTC for Pathways to Positive Futures, which will disseminate the recruitment flyer

with study information through their networks. Those who are interested in learning more

will have the opportunity to hear about the purpose and procedures for the project directly

from the researcher prior to opting in to the project, as well as to talk to sibling study advisers

who have agreed to serve as foster care alumni references for the study. Recruitment will be,

at minimum, a two-step process. Potential participants will express interest with the

recruitment flyer. The consent process will serve as the second step in the recruitment

process.

Check all that apply and attach all recruitment materials that will be used:

Person to person Media (TV, newspaper, radio, Web site)

Phone Social Media

Postal mail Other:

E-mail

2. How will potential subjects be identified and how will potential subjects be approached

SIBLING VIOLENCE AND CHILD WELFARE 194

to participate? (Answer for each subject group)

Explain in detail: Potential subjects will include foster care alumni who are connected

to FosterClub, New Avenues for Youth, and the PSU RTC for Pathways to Positive

Futures The agencies will disseminate the recruitment flyer information through their

networks.

3. Who will obtain consent/assent and when will that be done? (Answer for each subject

group) Katie Winters, M.A.

Explain in detail: I will obtain consent on the day that the interview occurs. I will provide

a hard copy of the consent form and verbally review it in detail.

4. What screening procedures or tools will be used? (Answer for each subject group)

Explain in detail: The consent process will occur on the day that the interview is

conducted. I will provide a consent form to the potential study participant which specifies

what the participant will be asked to do, how the data will be used, associated risks and

protections, and that participation is voluntary. I will review the consent form in its entirety

and answer questions that participant has about the form or study procedures. I will

emphasize that the participant can revoke consent at any time throughout the pilot process. I

will ask permission for the discussion(s) to be recorded. I will describe the purpose of audio-

recording, including the added ability to capture accurate information when the recording is

transcribed, and to be able to practice respectful and active listening during the discussion(s).

Participants may refuse to be recorded. In this case, I will respect this decision and take notes

during the discussion(s).

For each participant who opts to sign the consent form and proceed with the pilot, I

will retain a signed copy which will be kept in a locked file cabinet throughout the study

period. I will provide a participant copy to each participant who opts in.

E. Consent Process

Choose all that apply and attach appropriate forms to this application. (See

Informed

Consent or Waiver of Consent Checklists for guidance.)

1. Adult(s) Children Parent(s) Guardian(s)/legally authorized

representatives

Written

A consent, assent, or permission form that contains

all of the required elements of informed consent.

Alteration of

Informed

Requesting IRB approval for waiver of some or all

of the elements of informed consent, assent, or

SIBLING VIOLENCE AND CHILD WELFARE 195

Consent/Assent

process

permission (i.e. medical record review, deception

research, or collection of biological specimens).

If checked, complete Addendum 1 and submit with

the application.

Waiver of

Documentation

of Informed

Consent/Assent

Requesting IRB approval for waiver of the

requirement for documentation of informed

consent, assent, or permission (i.e. telephone

survey or mailed survey, internet research, or

certain international research).

If checked, complete Addendum 2 and submit with

the application.

Waiver of

Informed

Consent/Assent

Process

Requesting IRB approval for waiver of the

requirement for the informed consent, assent, or

permission process (i.e. medical record review,

deception research, or collection of biological

specimens).

If checked, complete Addendum 3 and submit with

the application.

2. What steps have been taken to prevent potential coercion or undue influence in

recruiting subjects and obtaining consent or assent? (For example, if the project

involves students of the PI or a product developer who will be testing the

product, a neutral third party must be engaged in these processes.) Explain in

detail: The first phase of recruitment will be carried out by FosterClub, New

Avenues for Youth, and the PSU RTC for Pathways to Positive Futures in

that they will disseminate the recruitment flyer through their networks.

Interested participants will have the option of talking/emailing/texting with a

sibling study adviser to hear the perspective of a peer about their experiences

working with me. Interested participants will then contact me by telephone

or email to schedule a study screening telephone call in which I will share

more about the study and confirm that they have direct experience with

physical and/or emotional sibling violence. I will emphasize the voluntary

nature of the study during the screening call and again during the face-to-

face consenting process. For those who agree to be interviewed, the final

phase of the consent process will occur face-to-face prior to the interview.

The participants will receive a $20 gift card to Amazon.com, as an expression

of appreciation for their contribution to the study.

F. Study Procedures

1. Describe any study procedures that have not been described elsewhere:

SIBLING VIOLENCE AND CHILD WELFARE 196

2. Does the study involve the collection of data/specimens (including the use of

existing data/specimens)? Yes No

a. If yes, indicate how, when, where and from whom specimens or data will

be obtained and what data or specimens will be collected: N/A

3. Is there a data and safety monitoring plan (required for greater than minimal

risk studies)? Yes No

a. If yes, describe the plan: All data collection procedures will be HIPAA-

compliant. Data will be stored on a password-protected computer and

locked file cabinet. Any breach of participant data will be reported to the

IRB immediately.

4. Are there any anticipated circumstances under which participants will be

withdrawn from the research without their consent? Yes No

a. If yes, describe the circumstances, as well as any associated procedures to

ensure orderly termination: N/A

G. Risks/Benefits

1. Potential risks to participants (check all that apply):

Invasion of privacy to the subject or family

Breach of confidentiality

Physical harm or discomfort

Psychological/emotional discomfort or distress

Psychological effect that is more than discomfort or distress

Social stigmatization

Economic (e.g., employment, insurability)

Legal

Any study related activity which subjects might consider sensitive,

offensive, threatening, or degrading?

Withholding standard care and procedures

Significant time or inconvenience

Other:

2. Does the study pose risk to individuals other than the participants?

Explain in detail: N/A

3. Indicate the risk category that most accurately describes the risk level for the risks

identified in Section G, questions 1 & 2 above:

SIBLING VIOLENCE AND CHILD WELFARE 197

Not greater than minimal risk

Greater than minimal risk, but presenting the prospect of direct benefit to

individual subjects

Greater than minimal risk, no prospect of direct benefit to individual subjects,

but likely to yield generalizable knowledge about the subject’s disorder or

condition

Research not otherwise approvable which presents an opportunity to

understand, prevent, or alleviate a serious problem affecting the health or

welfare of subjects

4. How will these potential risks be minimized in order to protect subjects' rights and

welfare? (See Additional Requirements for Research with Vulnerable Populations

for

guidance regarding children, prisoners and participants who become incarcerated after

enrolling. Contact ORI for guidance regarding human fetuses and neonates.)

Explain in detail: Great care will be taken to minimize risk to participants

who choose to engage in the research. Participants will be encouraged to

skip questions that they do not want to answer, to take breaks if needed,

and will be told that they can leave the conversation at any time, and for

any reason. To further reduce the potential that participants will

experience emotional distress during the discussions, I will closely

monitor participant affect and if any participant appears upset or

agitated, I will check in with them about their wellbeing.

5. In the event that any of these potential risks occur, how will it be handled (e.g.

compensation, counseling, etc.)?

Explain in detail: Participants demonstrating discomfort during the

interview(s) will be encouraged to take a break or discontinue

participation. All participants will receive general assistance information

which will be included on the consent form (e.g., 211 info number).

Breaches of participant confidentiality will be reported to the PSU IRB

and any recommended actions will be executed immediately to remedy

harm done through the breach.

6. Is it probable that a subject's previously unknown physical or psychological

condition will be discovered (e.g. disease, depression, genetic predisposition,

illegal activity etc.) as a result of the study activities? Yes No

a. If yes, what would types of conditions could be discovered and how will

Minimal risk” means that the probability and magnitude of harm or discomfort anticipated in the research are not

greater in and of themselves from those ordinarily encountered in daily life or during the performance of routine

physical or psychological examination or tests. 45 CFR 46.102(i)

SIBLING VIOLENCE AND CHILD WELFARE 198

these situations be handled?

Explain in detail:

7. Describe the expected benefits of this project (NOTE: compensation is not

considered a benefit):

a. To the individual subjects:

Explain in detail: Study participants may benefit from the knowledge

that others have also experienced sibling violence, and so they are not

alone in that regard.

b. To society:

Explain in detail: Society will benefit from increased knowledge of a

common, yet understudied and poorly understood phenomenon.

8. Explain how, in your assessment, benefits of this study outweigh the risks. (e.g.

risk/benefit ratio): Previous studies of sibling violence have not included youth

voice. By engaging youth in conversations about their experiences with sibling

violence, the study has far-reaching potential to inform child welfare practice

and policy on a national scale.

H. Available Resources

1. Are there research staff members, in addition to the Principal Investigator/Student

Investigator?

No (If no, skip to 3)

Yes

a. If yes, outline training plans to ensure that research staff members are

adequately informed about the protocol and study-related duties: N/A

2. If necessary to the research, describe the minimum qualifications for each

research role (e.g., RN, social worker), their experience in conducting research,

and their knowledge of study sites and culture(s): N/A

3. Briefly describe how the research facilities and equipment at the research

site(s) support the protocol’s aims (e.g., private rooms available for interview,

etc.): I will work with each participant to agree on a private location for the

interview.

4. Are there provisions for medical and/or psychological support resources (e.g., in the

event of incidental findings, research-related stress)?

Yes No N/A (not

needed)

a. If yes, describe the provisions and their availability: All participants will receive

general assistance information which will be included in the consent form (e.g.,

211 info number).

I. Reportable Events

SIBLING VIOLENCE AND CHILD WELFARE 199

Outline plans for communicating reportable events (e.g.

adverse events or unanticipated

problems involving risks to participants or others, breach of confidentiality, child abuse, and

suicidal ideation): Adverse events or unanticipated problems such as loss of data files or

extreme emotional responses to participation in the pilot will be reported as soon as

possible to the PSU IRB with the Unanticipated Problems/Adverse Event Report Form. Any

mandatorily reportable events will be reported to the appropriate authorities.

J. Research Related Injuries

1. Does this research involve greater than minimal risk to participants? Yes No

2. If yes, are there provisions for medical care and compensation for research-related

injuries?

Yes No

a. If yes, outline these provisions (Medical treatment should be available including

first aid, emergency treatment and follow-up care as needed. If the research plan

deviates from this policy, provide appropriate justification. Compensation for

physical injuries that result from study participation is not generally required): N/A

K. Participant Privacy

Describe provisions to protect participants’ privacy (their desire to control access of others to

themselves, e.g., the use of a private interview room) and to minimize any sense of intrusiveness that

may be caused by study questions or procedures.

In addition to carrying out the interview(s) in a private room, participants will

be strongly encouraged to avoid over-disclosure. The primary purpose of the study

is to learn about child welfare-involved youth’s experiences with sibling violence.

Participants will be encouraged to focus on these aspects of their experiences and to

disclose the details of their violent interactions only to the extent that they feel will

helpful to understanding the circumstances surrounding their experience.

L. Data Confidentiality

1. Will the information obtained be recorded in such a manner that participants can be

identified, either directly or through identifiers linked to the participants?

Yes No

2. Will data be made public? Yes No

a. If no, describe provisions to maintain confidentiality at each phase of the

data in the research. If engaging in internet or social media research,

provide copies of the sites privacy policy and include an explanation of

how approval is obtained for performing research activities that include

these sites or explain why approval is not required: During travel to/from

the program site, all study materials will be kept with the interviewer

at all times. Upon return to Portland, I will transfer any identifiable

paper documents to a locked cabinet. I will download recordings to a

password protected computer, make a back-up copy on a USB flash

SIBLING VIOLENCE AND CHILD WELFARE 200

drive (which I will keep in a locked cabinet), and then delete the

recording from the digital recorder. All data will be maintained in a

fashion that is in compliance with HIPAA regulations.

b. If yes, verify by checking “yes” that participants will be informed of what

data will be public and this information is included in the consent/assent

form/processes.

Yes

3. Confidentiality of Data Collection Instruments

Instructions: List all data collection instruments covered in this IRB application. For each

instrument, enter the letter designating the level of confidentiality for this instrument at each

data stage. Use the following Confidentiality codes:

A= Anonymous (No identifiers that link the data to a specific subject)

U=Unlinked-Confidential (Collected with identifier or code, but all identifiers & codes are

removed)

C= Coded-Confidential (Linked to a specific subject by a code, not by a direct identifier)

I=Intentionally Identified (Personal identifiers and research data are stored together in

one file)

Instrument

Data Stage

Collection

Analysis

Storage

Dissemination

Example: Teacher Survey

Example: Teacher Interview

1. Interview Protocol

2. Participant Background

Information

4. Method(s) of protection and location of data storage: (Check all that apply)

Locked office

Locked cabinet

Coded to a master list

Other:

When coded to a master list, check the appropriate description of how the

master list will be kept separate from the data:

SIBLING VIOLENCE AND CHILD WELFARE 201

Restricted Computer

Password Protected

Locked Private Office

Encrypted Data

Fire Wall System

Other: Location of data:

Building and room number: 1924 SE 11

Ave., #7, Portland, OR 97214

Electronic storage location: 1924 SE 11

Ave., #7, Portland, OR 97214

5. How long will research materials be stored, and when will they be destroyed,

including voice/video/digital/images? (PSU guidelines require all research

materials (consent forms, surveys etc.) to be kept for a minimum of three years

after completion of the study.) The research materials will be kept for three

years and then will be destroyed.

6. Will the data be transmitted? Yes No

a. If yes: i. How long will data be transmitted and stored? N/A

ii. What are the plans for the data at the end of the storage period (how

will it be destroyed, or will it be returned to data provider)? Paper

forms will be shredded. Digital recordings will be deleted.

7. How will research team members and/or other collaborators have access to

information about study participants? I (Katie Winters) and my dissertation supervisor

are the only people who will need access to the information about study participants.

It will not be shared.

M. Costs and Payments

1. Identify any costs that participants may incur during the study, including

transportation, costs, childcare, or other out-of-pocket expenses: None.

2. Will subjects be compensated for these costs? Yes No

a. If yes, what is the compensation, how much will the subject be offered, and how

will they receive it? (i.e., money or gift certificate, extra credit, etc.) N/A

3. Are there any OTHER payments, compensations or reimbursements that participants

may receive during the study that are not related to participant incurred costs?

Yes No

If yes, specify the amount, method and timing of disbursements: Participants will

each receive a $20 Amazon gift card to thank them for their contribution to the

study.

4. Will compensation be extra credit?

Yes No

a. If yes, students must be able to complete an alternative assignment for extra

credit, should they choose not to participate in the research. This assignment must

SIBLING VIOLENCE AND CHILD WELFARE 202

be comparable, with respect to time and effort, as the participation in research.

Describe the alternative assignment: N/A

5. When will the participants be compensated?

Before the study Installments during the study

Withdraw/complete the study

N. Multi-site Study Management

1. Does the study involve multiple sites? Yes No

a. If yes, describe plans for communication among sites regarding adverse events,

interim results, protocol modifications, monitoring of data, etc.: N/A

O. Investigational Drug, Biologic or Device

1. Does the study does involve an investigational Drug, Biologic or Device? Yes

2. Identify and describe the drug/biologic/device (e.g., marketing status): N/A

3. Is there an IND/IDE, classification of a device as significant vs. non-significant risk?

Yes No

4. Describe its administration or use: N/A

5. Compare the research drug/biologic/device to the local standard of care: N/A

6. Describe plans for receiving, storage, dispensing and return (to ensure that they will be

used only for participants and only by authorized investigators): N/A

7. If proven beneficial, describe anticipated availability and cost to participants post-

study, and plans (if applicable) to make available: N/A

P. HIPAA Privacy Protections

1. Are HIPAA privacy protections required? Yes No

(Protected Health Information obtained from a Covered Entity [e.g. a hospital or

community health center] requires these protections. PSU is not a Covered Entity.)

If yes, fill out the HIPAA Application Supplemental form.

If unsure, refer to the HIPAA Application Supplemental form for guidance, or

call ORI for assistance.

Q. Human Data and Human Specimen Banking

(These are repositories established by PSU investigators for the purpose of storing data and/or

specimens for future research purposes. Data banking includes electronic data files and

databases.)

1. Does the study include Specimen Banking? Yes No

2. Does the study include Data Banking? Yes No

3. Identify what will be collected and stored, and what information will be associated with

the specimens: N/A

SIBLING VIOLENCE AND CHILD WELFARE 203

4. Describe where and how long the data/specimens will be stored and whether

participants’ permission will be obtained to use the data/specimens in other future

research projects: N/A

5. Identify how and who may access data/specimens: N/A

6. Will specimens and/or data be sent to OR from research collaborators outside of PSU?

Yes No

a. If yes, describe the plan: N/A

R. Sharing Study Results

1. Is there a plan to share study results with individual participants? Yes No

a. If yes, describe the plan:

2. Is there a plan to disseminate aggregate results to the community where the research

is conducted?

Yes No

a. If yes, describe the plan:

S. Disclosure of Financial Interests

Does the PI, Co-PI, or any other person responsible for the design, conduct, or

reporting of this research have an economic interest in, or act as an officer or

director of, any outside entity whose financial interest would reasonably appear to

be affected by the results of the study? Yes No

If yes, complete below:

a. Name of the person with a potential financial conflict of interest (COI): N/A

b. Explain the potential financial conflict of interest: N/A

c. Explain how the potential financial conflict of interest will be managed: (If the

financial interest is a “significant financial interest” as defined in PSU’s Financial

Conflict of Interest Policy, submit the management plan established with the

Financial Conflict of Interest Committee.) N/A

T. Regulatory Compliance

This section is for documenting compliance with other regulatory requirements.

1. Are student records being used? Yes No

a. If yes, describe how compliance will be maintained with the Family Educational

Rights and Privacy Act (FERPA

): N/A

2. Does this project have funding from any of the following federal agencies? (Check all

that apply)

SIBLING VIOLENCE AND CHILD WELFARE 204

Department of Defense (DOD)

Department of Education

Department of Energy

Department of Justice

Department of Navy

Environmental Protection Agency (EPA)

National Institute of Health

National Science Foundation (NSF)

If any of the above are checked, describe the plan to comply with the regulations

required by that agency: (See

Regulatory Compliance Required by Federal Funding

Agencies for a list of these regulations.) N/A

Section III: Appendices

a. Informed Consent/Assent/Permission forms

(See Informed Consent or Waiver of Consent Checklists for guidance.)

b. Training and Experience

All staff engaged in human subjects’ interaction and intervention, or working with

identifiable human data or private information about live human subjects activities are

required to complete training. The submission packet must include documentation of

training for all personnel listed in the protocol, including student investigators and PI’s.

It is the PI’s responsibility to ensure that all other staff (not listed on the protocol)

complete this training and keep documentation. The IRB may request documentation of

training at any time as part of a post approval monitoring activities.

IRB applications received without training certificates are considered incomplete. The

effective application receipt date will be when the complete application (including

training) is received by ORI.

Training is available online through the Collaborative Institutional Training Initiative

(CITI): https://www.citiprogram.org/

. When signing up for the training, please select the

Social & Behavioral Research Investigators learner group and complete either the Social

& Behavioral Research - Basic/Refresher course OR the Social and Behavioral

Responsible Conduct of Research course. Submit a copy of the completion certificate

electronically to the IRB office. Alternatively, a completion certificate of the National

Institute of Health’s Protecting Human Research Participants course may be submitted.

Please note, we are unable to verify NIH training electronically, so completion

documentation must be provided directly to the IRB at

hsrrc@pdx.edu.

In addition to the CITI or NIH training, please describe any specialized training,

education, or experience that would help to minimize the risks, particularly if working

SIBLING VIOLENCE AND CHILD WELFARE 205

with vulnerable populations and/or sensitive topics. If the researcher will be advised by

an expert or on-site mentor, note this information in the application.

c. Recruitment Materials (Posters, Flyers, Scripts)

d. Data Collection Instruments (Interviews, Surveys, Focus Group Questions)

e. Expedited Checklist (optional)

The IRB makes the final determination of whether a non-exempt project is eligible for

review under expedited or full board review. If you believe that the research is non-

exempt and eligible for expedited review, you may fill out the expedited checklist and

attach to this application.

f. Addendums as appropriate

Please submit completed applications by email to hsrrc@pdx.edu

DATA COLLECTION CANNOT BEGIN UNTIL IRB APPROVAL IS GRANTED.