Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Factors influencing foster placement outcomes for children in British Columbia Strubin, Tanya 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-ubc_2007-0249a.pdf [ 5.45MB ]
JSON: 831-1.0101267.json
JSON-LD: 831-1.0101267-ld.json
RDF/XML (Pretty): 831-1.0101267-rdf.xml
RDF/JSON: 831-1.0101267-rdf.json
Turtle: 831-1.0101267-turtle.txt
N-Triples: 831-1.0101267-rdf-ntriples.txt
Original Record: 831-1.0101267-source.json
Full Text

Full Text

FACTORS INFLUENCING FOSTER P L A C E M E N T OUTCOMES FOR CHILDREN IN BRITISH C O L U M B I A by T A N Y A STRUBIN B.S.W., The University of Northern British Columbia, 1998 A THESIS SUBMITTED FOR PARIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SOCIAL W O R K in THE F A C U L T Y OF G R A D U A T E STUDIES THE UNIVERSITY OF BRITISH C O L U M B I A January, 2007 © Tanya Strubin, 2007 Abstract An investigation of factors thought to contribute to the breakdown of foster homes for children in British Columbia was completed in this study. Data was collected from 22 foster parents across the province; all members of the British Columbia Federation of Foster Parents. Through the use of survey questionnaires, foster parents were asked to respond to questions regarding their experiences with a foster child aged 6 to 12 years formerly in their care. Data was collected to indicate the presence of difficult child behaviour, supportive services available to foster parents and parental attributions towards care giving outcomes. Family systems theory and social cognition theory provided a theoretical background to this research through the discussion of family dynamics and parental attributions. Variables under study in this project have been found in previous research to affect foster placement stability. Results revealed that verbally aggressive behaviour by children may affect placement stability while physical aggression and delinquent behaviour do not. Results also indicated that foster parent attributions regarding children in their care may relate to foster placement outcomes. Support from the foster care system appeared to have limited impact on the success of placement. However, foster parents reported an overall lack of available support. Throughout this study, the need for further research into foster placement outcomes is highlighted. Table of Contents Abstract i i Table of Contents i i i List of Tables v List of Figures vii Chapter 1 Introduction 1 Foster Care in British Columbia 1 Children in Care 2 Fostering Relationship 5 Chapter 2 Literature Review 7 Behaviour of Foster Children 7 Foster Parent Supports 9 Foster Parenting 11 Chapter 3 Theoretical Background 15 Family Systems Theory 15 Social Cognition Theory 19 Chapter 4 Hypotheses 23 Chapter 5 Methodology 25 Procedures 25 Sample Description 29 Measures 30 Child Behavior Checklist 30 Parent Attribution Test 31 Foster Parent Support Measure 32 Foster Placement Outcome Measure 33 Chapter 6 Results 35 Demographics 35 Descriptive Statistics 36 Hypotheses 1 37 n i Hypotheses 2 41 Hypotheses 3 44 Hypotheses 4 48 Chapter 7 Discussion 50 Chapter 8 Summary 54 Limitations 54 Conclusion ; 55 References 57 Appendix A Letter of Consent 62 Appendix B Demographic Measure 64 Appendix C Foster Parent Support Measure 65 Appendix D Foster Placement Outcome Measure 67 Appendix E Child Interaction Survey 69 Appendix F Child Behaviour Checklist 73 Appendix G Child Behaviour Checklist study scale items 75 Appendix H Ethics Approval Letter 76 iv List of Tables Table 1 Descriptive statistics 37 Table 2 Correlations between child behaviour, support, low perceived power and foster placement breakdown ; 38 Table 3 Summary of logistic regression analyses testing the relationship between Physical aggression behaviour, verbal aggression behaviour and delinquency and foster placement breakdown for model fit 39 Table 4 Summary of logistic regression analyses testing the relationship between physical aggression behaviour, verbal aggression behaviour and delinquency with foster placement breakdown 41 Table 5 Summary of logistic regression analyses testing the relationship between support and foster placement breakdown 42 Table 6 Summary of physical aggression, verbal aggression, delinquency and support as an interaction term with foster placement breakdown 42 Table 7 Summary of logistic regression analyses testing the effects of low perceived power with foster placement breakdown model fit 45 Table 8 Summary of logistic regression analyses of low perceived power with foster placement breakdown 46 Table 9 Summary of logistic regression analyses testing the effects of low perceived power with verbal aggression, physical aggression and delinquency on foster placement outcome 46 Table 10 Summary of logistic regression analyses testing the effects of support on low perceived power model fit 4: Table 11 Summary of logistic regression analyses testing the effects of support on low perceived power 4 vi List of Figures Figure 1 Relationship of variable under study Chapter 1 The relationship between foster parents and the foster children in their home is unique. It is a relationship that is formed when individuals in the community care for children unable to live with their primary caregivers (Ministry for Children and Family Development [MCFD], 2003). Foster parenting requires individuals to provide for the health, emotional, behavioural, physical and spiritual needs of children in their care (MCFD, 2003). This can prove to be a challenging task, and in many instances foster children are moved from one foster placement to another when their needs cannot be met in one home. However, experiencing multiple placements can have a detrimental effect on the emotional and social well-being of foster children (Newton, Litrownik, & Landsverk, 2000). In this, possible factors associated with the breakdown of foster home placements is explored in the hopes of identifying areas of improvement in the provision of stable foster homes for children in care. The behaviour displayed by foster children, the support provided to foster parents caring for them and parenting characteristics were all considered in this study. Family systems theory and social cognition theory provided a lens for a deeper examination of the dynamics among family members in a foster placement setting. Foster Care in British Columbia Currently, there are approximately 10,000 children and youth in British Columbia in care of the government under the authority of the Ministry of Children and Family Development (MCFD, 2006). More than half of these children and youth live in foster homes, while the remainder live in specialized group homes or on an independent living program. There are approximately 4000 foster and group homes around the province 1 providing care for children in need (MCFD, 2006). However, there is an ever-increasing demand for more foster homes as the number of children in care grows. Although it is difficult to recruit and retain foster parents, given the challenging nature of the position, many individuals agree to assume the role of a foster parent and care for children in need. By examining the experiences of foster parents, valuable information is provided on the current state of the foster care system and areas in need of improvement. There are three ways in which children enter the foster care system. A parent may voluntarily give M C F D parental authority over their child i f they are temporarily unable to care for them. Secondly, the special needs of a child that parents are unable to manage may necessitate a child coming into care. Thirdly, a child may be removed from the care of their home i f it is deemed that their safety and health are in danger and there is no other way to protect the child (MCFD, 2003). The amount of time children remain in foster care varies widely as it depends on what requirements are needed to return the child to their primary caregivers, i f they can return at all. As a result, some children only experience one or two foster homes, while others remain in the foster care system for years and may experience many placement changes. Children in Care The majority of children who enter the foster care system have experienced abuse or neglect in some manner. Many children have experienced emotional, physical or sexual abuse in their previous living environment (MCFD, 2003). Many of these foster children develop mental health disorders; suffer from Post Traumatic Stress Disorder, developmental delays or learning disabilities (Martin, 2000). Children experiencing these conditions often display difficult, oppositional, or even aggressive behaviours (Cavell, 2 2000). Children who display difficult, oppositional or aggressive behaviours as identified by their foster parents wil l be the focus of this study. Aggressive or difficult behaviour in children includes being physically combative towards others, such as hitting or shoving, destroying property, and disturbing others. Children may be verbally aggressive by threatening others, yelling and making derogatory remarks (Cavell, 2000). Children who display aggressive behaviours are more likely to have difficulty processing information from social interactions. As a result, they often misinterpret social cues, especially in stressful situations. In addition, these children are more likely to be impulsive, demonstrate a low level of frustration and lack empathy (Cavell, 2000; Fitzsimmons, 2003). Unfortunately, this behaviour is not easily extinguished. In fact, Cavell (2000) states that, "over time and with repeated episodes, aggressive behaviour thus becomes a more deeply rooted aspect of children's personalities" (p. 7). As children age, the severity of their behaviour tends to increase, along with the significance of their behaviour. This results in more serious consequences for the child and their community (Cavell, 2000). Armsden, Pecora, Payne and Szatkiewicz (2000) conducted a review of studies on child behaviour and foster care and found that the rate of acting out problem behaviours in foster children are several times greater than the rate found in the general population. Discerning trends among foster children and behaviour based on gender or age coming into care is difficult and consistent trends have not been established. However, the type of placement foster children experience does seem to correlate with their level of displayed difficult behaviour with children placed in residential treatment homes being twice as likely to exhibit difficult behaviour as children in regular foster homes. Clearly, difficult and aggressive behaviour 3 problems among foster children are a very real concern and appear to affect a large number of foster children of all ages and gender (Armsden et al., 2000). When children are placed in foster homes, it is done so to provide them with a stable living environment, either short or long-term (MCFD, 2003). However, many children who enter the foster care system engage in behaviours that are challenging to their caregivers (Fisher, 2000). Some foster parents may find a child's behaviour to be unacceptable or too difficult to manage and request the child be placed in another resource. Research indicates that children who display aggressive and acting out behaviours are more likely to be experience foster placement breakdown as a result of their behaviour (Newton et al., 2000; Fisher, 2000). Foster parents have identified foster children's behaviour problems as a reason for the termination of a child's residency in their home (Martin, 2000: Gilbertson & Barber, 2003). When children experience multiple disruptions in foster home placements, they are negatively affected and behaviour problems are likely to increase (Martin, 2000; Newton et al., 2000). When a foster child experiences multiple placements, problem behaviours related to school and social interactions are more likely thus increasing the likelihood that subsequent placements will fail (Martin, 2000, p. 188). When foster placements fail, the consequences are far reaching as it is not only the child who is affected but also the foster parents, social workers and foster care system as a whole (Martin, 2000; Barratt, 2002). Therefore, maintaining a stable foster placement for children is in the best interests of all involved (Newton et al., 2000). 4 Fostering Relationships Social workers are responsible for screening foster parents and evaluating placements, placing children in homes and providing support services for foster children and their foster parents (Barratt, 2002). As a result, social workers employed in child welfare are required to deal with the fallout of placement breakdowns. The experience may be traumatizing for the foster family and foster child and is costly in terms of time and money to the child welfare system (Barratt, 2002). Additional research on factors that help or hinder the relationship between foster parents and foster children could provide useful information for social workers to have. It may help inform practice decisions by increasing their understanding of the fostering relationship. Finding compatible matches between parents and children will likely aid in sustaining the length of foster placements (Martin, 2000). There are many instances where children are moved from their current foster home due to interactions between members of the foster family. Difficult behaviour displayed by foster children may be one such factor leading to a breakdown in the foster placement, although it is not likely the only factor contributing to the dissolution of the foster placement (Newton, 2000). Factors relating to the foster parent, such as the amount of support they receive from the foster care system, or from friends and family, may contribute to the outcome of a fostering relationship (Chamberlain & Moreland, 1992). The attributes or beliefs that parents hold about the nature of their relationship with a foster child could be another contributing factor. This might especially be the case when the foster children in the relationship display challenging behaviours (Bugental & Shennum, 1984). 5 An examination of contributors to the breakdown of foster placements is conducted in this study. Through a review of previous literature as presented in this study, many factors have been identified as influencing the outcome of foster placements for children. This study purported to further explore some of the difficulties present in providing children with stable foster placements by examining foster homes in BC. Data gathered from foster parents provided information on these challenges. Through an understanding of the reported experiences of foster parents, a vision for positive changes in the foster care system can be created. 6 Chapter 2: Literature Review Previous researchers have attempted to identify factors leading to the success or failure of foster home placements by interviewing foster parents, examining the foster care system and programs in place to support them (Armsden et al., 2000; Gilbertson & Barber, 2003). In a review of the literature on foster placement outcomes disruptive behaviour displayed by foster children is identified. (Newton et al., 2000; Fisher, 2000; Gilbertson & Barber, 2003), the support foster parents receive (Kalland & Sinkkonen, 2001; Chamberlain & Moreland, 1992; Gilbertson & Barber, 2003), and parenting practices utilized by foster parents (Tripp De Robertis & Litrownik, 2004; Barratt, 2002; Fisher, 2000) all impact the outcome of foster placements. Behaviour of Foster Children Using a longitudinal framework and standardized measures with a sample of 415 children in foster care, Newton, Litrownik and Landsverk (2000), conducted a study examining the relationship between behaviour problems in children and the number of placement changes they experience. The Child Behavior Checklist (Achenbach, 1991), was used to assess behaviour problems in children involved in the study. Internalizing and externalizing behaviours and total behaviour problems were examined. Data was collected from the children's case records to assess the number of placement changes children experienced during anl8 month period following entry to the foster care system. By conducting hierarchical regression analyses on collected data, Newton et al. (2000) were able to conclude that children who engage in behaviour deemed to be disruptive, aggressive or dangerous are likely to be moved from their current foster home. The study also revealed that foster children who experience multiple placements are likely to be 7 emotionally and behaviourally affected in a negative manner by these experiences. The study highlights not only the impact that disruptive behaviour has on placement outcomes, but also the vicious circle that is created as children who experience multiple placements display increasing levels of behaviour problems and are then more likely to experience additional placement disruption. Gilbertson and Barber (2003), employed qualitative methods to interview 19 Australian foster parents who terminated foster placements in their home due to the behaviour of children in their care. This study aimed to gain the foster parent's perspective on the process of placement breakdown and to determine interventions that may have been useful in preventing the breakdown. The study revealed that for youth whose placements had experienced breakdown were often in care due to a history of abuse, neglect, or parents' inability to care for them (Gilbertson & Barber, 2003). Foster parents described the youth as displaying " . . . multiple problem behaviours including animal torture . . . physical assaults, repeated acts of property damage, verbal abuse, self-harm, running away, school refusal... and threats of harm to the carer" (p. 332). The majority of foster parents in this study ended the foster placement due to safety concerns related to the behaviour of their foster child. However, foster parents were able to identify factors they felt would aid in the stability of foster placements, such as better placement preparation, and support for managing difficult behaviours. This study highlights the importance of creating foster placements that bring stability to the lives of troubled children. Limiting the adverse and cumulative effects of multiple placement breakdowns will benefit both foster children and their carers (Gilbertson & Barber, 2003). 8 Foster Parent Supports Foster parents are required to manage children's behaviour problems as they arise within the context of the fostering relationship (Martin, 2000). Several factors are seen to contribute to the ability of foster parents to deal with these difficult situations in a competent and appropriate manner. Several studies have identified that the support foster parents receive from the foster care system has a significant impact on rates of placement disruption (Barratt, 2002; Kalland & Sinkkonen, 2001; Chamberlain & Moreland, 1992). Chamberlain and Moreland (1992) designed a controlled study to assess the benefits of support services for foster parents. One group of foster parents participated in a weekly support group focused on behaviour management strategies, received telephone contact three times a week with a support worker and a monthly stipend of $70. Results were compared with a group of foster parents who only received the $70 monthly stipend and a control group (foster parents who received neither support nor the stipend). Results revealed that retention rates improved significantly for the foster parents who were provided with regular and on-going support regarding child management techniques compared with those receiving only the stipend or not receiving any intervention. In fact, the drop out rates for foster parents receiving support services was cut by almost two-thirds over a two-year period when compared to the control group (Chamberlain & Moreland, 1992). Kalland and Sinkkonen (2001) evaluated factors precipitating placement breakdown or stability for foster children in Finland by conducting a retrospective study during a five-year period in the 1990's. This study revealed many factors that lead to stability in long-term foster placements. In particular, the study notes that this stability 9 was often associated with support from relatives. In addition, there was a correlation between placement stability and the amount of training foster parents received (Kalland & Sinkkonen, 2001). This study supports previous research on the benefit of training for foster parents, but also reveals the importance of acknowledging informal support systems accessed by foster parents, such as family and friends. Fisher (2000) conducted a study with high-risk preschool children to determine the effectiveness of an early intervention foster care program aimed at promoting placement stability. A n intensive early intervention program was designed to promote a therapeutic relationship between foster parents and their foster children with an aim to increase pro-social behaviours in children and reduce behaviour problems. Foster parents involved in the project were provided with behaviour management training, daily phone contact, home visits and weekly support groups. Children involved in the early intervention program generally displayed behaviour problems, had a history of abuse and were referred due to their unlikely success in a regular foster home. The behavioural adjustment to a new foster home for these children was compared to that of children newly placed in regular foster homes and to a group of children in the community living with their natural family. Children in the regular foster homes and children in the community group did not receive the intervention. The sample size for this study was small with only 10 participants in each of the three groups (Fisher, 2000). Children's behaviour problems, parenting strategies and (foster) parent stress levels related to the child's behaviour were measured shortly after the child entered a new foster placement, and 12 weeks later. Results indicated that although children involved in the intensive intervention program initially displayed more disruptive behaviour, they 10 show an improvement over time while the converse was true for children placed in regular foster homes. Fisher (2000) concluded that the increase in behaviour problems for children in regular foster homes was likely due to inadequate resources available to these homes. Fisher also revealed that parents who received intensive training and support are less likely to engage in negative patterns of interaction with their foster child. This then reduces the stress level experienced by the parent and helps to create a more stable placement for the child. Foster Parenting Parenting practices utilized by foster parents as a contributor to disruptive behaviour problems in foster children was examined in a study conducted by Tripp De Robertis and Litrownik (2004). The study assessed 70 children who were in foster care and obtained information from both the child and the foster parents. Data from caregivers was collected through the use of structured interviews and measurement tools, including the C B C L Aggressive Problem Scale (Achenbach, 1991). Tripp De Robertis and Litrownik used regression analyses to evaluate the relationship between harsh discipline and child aggressiveness. Results revealed that, " . . . as caregivers increased their endorsement of harsh disciplinary practices . . . there was almost a three fold increase in children generating aggressive solutions to social problems" (p. 98-9). This study highlights the importance of examining the way in which foster parents discipline children in their care. Harsh disciplinary measures are likely to increase behaviour problems in foster children and this in turn may affect the stability of the placement. Foster parents play an important role in the social development of the children in their care and the provision of adequate support programs to assist foster parents with this is 11 necessary. The importance of providing foster parents with adequate support programs is again highlighted in this study (Tripp De Robertis & Litrownik, 2004). Although the research projects reviewed above clearly outline the importance of providing foster parents with supportive services such as education and training to enhance foster parent - child interactions, Barratt (2002) reveals the need to focus on another aspect of this relationship. Barratt discusses the importance of examining the past experiences of foster parents in relation to their parenting style. She states that the manner in which foster parents themselves were raised, along with their experience with children will impact the relationship they develop with foster children. Adults own experiences with caregivers in their life form a basis for knowledge structures in regards to parenting. These experiences are stored in long-term memory and act to organize how parents respond to caregiving situation with children in their care (Bugental, Johnson & Silvester, 1998). When difficulties arise within a foster care placement the behaviours of the child are often deemed to be the problem. As a result, an inclination to focus on the behaviour of the child negates focus on the relationship between the child and the carers. This may lead to further placement breakdown for the child and an increased sense of failure that they wil l continue to move through placements without the opportunity to experience a long-term placement. In addition, Barratt states that foster parents may not feel comfortable discussing personal issues in the context of support services provided to them, which are professional in nature. Providing support for foster families that extends beyond child management techniques, such as family therapy, could have the benefit of increasing placement stability. Addressing concerns that arise in the foster parent/child 12 relationship as they occur will likely enhance placement stability for the child, and aid foster parents in their care of future foster children (Barratt, 2002). A review of current literature reveals a number of factors related to the stability or breakdown of foster placements. Although each study examined different variables, several findings were consistent with one another. Evidence revealed that disruptive behaviour exhibited by foster children is a significant contributor to the breakdown of foster placements (Newton et al., 2000; Fisher, 2000). In addition, the manner in which children are parented in a foster home also has an impact on the outcome of the placement (Barratt, 2002; Tripp De Robertis & Litrownik, 2004). Child behaviour management training and support from professionals greatly enhances parenting strategies used by foster parents when caring for foster children who exhibit difficult behaviour (Kalland & Sinkkonen, 2001; Chamberlain & Moreland, 1992). Informal support networks, such as family and friends, have also been shown to have a positive impact on foster parents (Fisher, 2000). These supportive factors in turn help to maintain stability in foster placements (Kalland & Sinkkonen, 2001). An additional factor seen to contribute to the success or failure of foster placements is the manner in which foster parents manage the behaviour of foster children in their care. Barratt (2002) discusses the need to consider how the beliefs foster parents hold in relation to parenting affects the outcome of foster placements. Despite the contributions of the research cited in this paper, empirical research regarding foster care and foster child outcomes leaves many unanswered questions (Tripp De Robertis & Litrownik, 2004; Fisher, 2000). Continuing research should explore what changes are needed in the foster care system to ensure safe, stable and successful foster 13 placements for children in need. Although there are barriers to doing research with foster families, such as obtaining access to an adequate sample, selection bias, and demographics, it is vitally important to continue exploring the needs of foster children and their foster parents. Research in this area is necessitated by the detrimental impact that numerous foster placement breakdowns has on children in care and the foster parents who care for them (Barratt, 2002; Newton et al , 2000). 14 Chapter 3: Theoretical background Family systems and social cognition theories provide a lens for analysing the relationship between foster parents and the foster children they care for. Family systems theory provides for an analysis of how families incorporate a new member and how the behaviours of family members affect overall family functioning (Brown & Christensen, 1999). As well, family systems theory aids in the interpretation of how families interact with individuals, and groups outside of the family, including social support networks (Andreae, 1996). Social cognition theories and attribution theory also provide an analysis of how the perceptions of individuals play a role in the interaction between family members (Bugental & Johnson, 2000). These theories will be used in the interpretation and analysis of the following research study. Family Systems Theory Family systems theory provides for an understanding of the family that focuses on the interactions between members rather than on the actions of individual members (Nichols, 1998). In this manner, the behaviour of each family member is related to and dependent upon the behaviour of all other family members. An examination of problems within a family system reveals that they are caused and perpetuated by the interactions between members (Nichols, 1984). Family systems theory is useful in the examination of problems that result in the breakdown of foster placements as it moves beyond focusing solely on the behaviour of the foster child. Family systems theory provides a view of how interactions between family members and interactions between the family and the outside world influence the stability of a foster placement. 15 Family systems theory recognizes that families do not exist by themselves but function in relation to systems outside the family as well. For foster families, these other systems include community institutions such as school, government and social service agencies (Andreae, 1996). To some extent, family boundaries control the influence and interaction members have with these environments. Boundaries function to ". . . control the flow of energy, information, and people between the system and the environment and within the system itself (Leslie, 1988, p. 54). Boundaries regulate the information and people who enter the family system. This is done by controlling the flow of energy and information within the family system and its interaction with the surrounding environment (Andreae, 1996: Leslie 1988). When families decide to foster a child, they are inviting a new member into their family. However, foster children bring with them the other individuals involved in their life, including natural family members, school staff, social workers and other professionals. Families are expected to receive input from these various sources in relation to the foster child (Martin, 2002). The ability of families to respond to the integration of a new member, and the services that come with the child, will impact how the family system functions (Andreae, 1996). Although evidence reveals that overall, enhanced support from the foster care system aids in the stability of foster placements, some barriers may exist in the reception of these services (Fisher 2000; Barrat, 2002). Foster may parents receive conflicting messages from different sources about their role as parents. In addition, services provided may not be meeting the specific needs of the foster child or family. As Barratt (2002) states: "The expectation that carers continue to maintain a safe family environment for their birth and foster children while managing the requirements of birth families and the professional system may be an 16 important factor in foster care breakdown" (p. 165-6). Additional stress may be placed on the foster family as it attempts to negotiate increasing demands from outside systems (Barratt, 2000). As a result, the manner in which foster families are provided with services, and the impact these services have on the family, is an important consideration in regards to placement success or failure. When a family accepts a new member into their home, the family system is changed by the arrival of the new member and the family's history, structure and style of interacting are imprinted on the new member (Koman & Stechler, 1985). When the new member is a foster child, changes to the family unit can be significant as the foster child may come from a family whose values, expectations, and experiences are vastly different than those in their new home (Martin, 2000). The child may display behaviours and interact with foster carers in a manner that is contrary to the values and practices of the family. As previously stated, children who come into care often exhibit challenging behaviours (Newton et al., 2000). When foster children enter a new home, they are doing a great deal of learning about the day-to-day functioning of the foster family, including family roles, expectations and the manner in which the home operates. It is a confusing time for foster children as they struggle with being apart from their primary caregivers and adjust to being part of a new family and the social service system (Martin, 2000). During this time, behaviour problems may be amplified, causing stress on the family system. Children will likely test the boundaries of their new foster home by challenging parental limits and control in the home (Barratt, 2002; Fisher, 2000). The ability of a family to adjust to the integration of a foster child, and the child's ability to adjust to the family, will have an impact on the subsequent functioning of the 17 family and the success of the placement (Martin, 2000). Foster families need to adapt in a manner that allows the child to be integrated as a participating member of the family (Koman & Stechler, 1985). This can be a difficult balance to achieve and could lead to further problems in the family i f not accomplished. Family systems theory does not view problems as originating from any specific cause. Instead it is understood that all interactions are multiply determined by forces operating within and upon the family system. As a result, no individual or interaction can be blamed (Koman & Stechler, 1985). Neither the foster child nor parent would be held responsible for presenting problems. Rather, problematic interactions are a signal that the family is having difficulty adjusting to demands. When a family is not adjusting well to the integration of a new member, communication patterns among family members may become pathological. Unless this pattern is recognized and addressed as a function of the family system, the dysfunction may be attributed to the presence of a foster child (Koman & Stechler, 1985). A request for the foster child to be removed from the family home could result. Family systems theory anticipates this possibility in considering the ways in which an existing system may resist significant changes that may require the relinquishment or realignment of roles and beliefs. A system confronted with new demands for which it is ill-prepared may resist changes to its beliefs and associated patterns by rejecting the integration of the new or challenging member. Family systems theory provides a lens for viewing family functioning in relation to the success or failure of foster placement. This theory does not look at individual characteristics, but rather the interplay between family members. However, foster families present a unique situation and an examination is required that extends beyond 18 the systems view. For instance, the compatibility or 'goodness of fit' between foster children and the families they are placed with is an important factor in the success of the relationship. The easier a child's behaviour and temperament fits within the tolerance level of the parents, the greater likelihood that the children wil l settle well into the home (Martin, 2000). As a result, individual characteristics should be considered when matching children and families (Ramey, 2002). Social cognition theory wil l be used to address this area of concern, which has not been met by family systems theory. Social Cognition Theory Social cognition theory studies the role of thoughts and belief structures on individual's experiences with the social world. In social cognition research, the study of families focuses on " . . . the origins, consequences, and perpetuation of stable, knowledge-based cognitive structures in family life. Particular attention is often directed to the role of cognitive structures in the emergence and maintenance of behavioural and emotional problems within the family" (Bugental & Johnson, 2000, p. 319). Attribution theory is one area of social cognition research that can be applied to parent-child relations. Attributions theory examines how individuals view the causes of event and how these causes are attributed to the behaviour of other people and themselves (White, 1983). Social cognition theories can be applied to how children acquire aggressive behaviours and respond to parenting relationships (Huessmann, Moise-Titus, Podolski & Eron, 2003). As well, attribution theory can be applied to parents as a way of examining how the causal beliefs parents hold about care giving outcomes influences their parenting behaviour (Bugental & Blue, 1989). 19 Attribution theory emphasizes that individuals engage in a process of causal analysis of all interactions in daily life, those significant and trivial. Ideas about how people perceive causality of events is an important concept to consider when examining relationships within a family system. Attribution theory examines the construction of perceptions from information in the social environment when the causality of an event is inferred. It distinguishes whether the locus of causality is perceived to be in a person or in the environment. How people perceive the intentions of others' actions is an important concept within this construction. When individuals judge another's behaviour to be intentional they reach the conclusion that it reflects on an underlying stable quality in the person. These perceptions thus developed allow one to predict the behaviour of others (Fiske & Taylor, 1984). Many children who enter the foster care system have a history of acting out and aggressive behaviour. Social cognitive theories state that the early acquisition of aggressive behaviours may be due to children imitating what they see in their environment. When young children witness aggressive behaviours, the likelihood of acquiring these behaviours themselves increases (Huesmann et al., 2003). Children may witness violence between family members, individuals in their community, or in the media. It is not uncommon for young children (2 to 4 years of age) to engage in aggressive behaviours as they move through normal developmental phases. However, children are impressionable at this age and "the observation of specific aggressive behaviors . . . leads to the acquisition of more coordinated aggressive scripts for social problem solving and counteracts environmental forces aimed at conditioning the child out of aggression" (Huesmann et al., 2003, p. 202). The longer children are exposed to 20 violence, the more likely they are to attribute hostility to the actions of others', and react by responding in an aggressive manner. Thus, for children, exposure to violence has long-term effects on their beliefs about aggression and their skills in managing it. (Huesmann et al., 2003). Children will carry these behaviours and beliefs with them when they are removed from the environment in which they acquired them as their view of the social world has been altered by their experiences. The above description fits for many children who enter the foster care system. As a result, foster parents may be caring for a child who attributes hostility to the foster parent's actions and thus responds in an aggressive manner (Fitzsimmons, 1998). To fully understand the parent-child relationship, attribution theory must also be applied to the role of caregivers in this relationship. Researchers believe that the ". . . ways in which parents explain care giving events [has] important implications for their immediate emotional and behavioural responses, as well as for the long-term quality of family relationships" (Bugental et al., 1998, p. 459). Particular attention has been given to an examination of parental attributions in relation to children who display difficult behaviour. It has been found that mothers of children who display chronic aggressive behaviour tend to interpret their children's action in a negative manner (Bugental et al., 1998). Parents who attribute caregiving outcomes to the behaviour of children in their care may be considered to demonstrate low perceived power in the relationship. Low perceived power is potentially an important moderator of negative parental affect in response to children's behaviour (Bugental & Blue, 1989). These individuals wil l see children as being in control of negative behaviour patterns in the family. This is 21 characteristic of adults who perceive child behaviour problems as being intentional and a threat to them as a parent (Bugental, Lyon, Krantz & Cortez, 1999). As a result, these parents see themselves as having little control in care giving situations with children who display difficult behaviour. Parents with a low level of perceived power are more likely to engage in more coercive parenting practices, which may in fact escalate the child's behaviour. There has been a link established between low power parents and instances of child abuse (Bugental & Blue, 1989). Although foster parents have been known to abuse foster children in their care, the number of documented cases is small (Martin, 2000). In the case of foster parents, an escalation of the child's behaviour problems may instead lead a foster parent with low perceived power to request transfer of the child from their home, resulting in foster placement breakdown. 22 Chapter 4: Hypotheses The interplay among many factors appears to influence the fostering relationship between parents and children. Family systems theory provides an analysis of the importance of examining how family work as systems that are irrevocably changed by the introduction of a new foster child. Attribution theory allows insight into the individual relationship between parents and children in regards to parenting practices with children who display difficult behaviour. A literature review on the subject of foster placement breakdown provides insight into this subject, but also poses more questions. In this study, an attempt is made to further explore factors related to the disruption of foster placements for pre-adolescent children in care. Hypotheses (see figure 1) 1. Difficult behaviour displayed by foster children leads to the breakdown of their foster placements. 2. Perceived support from the foster care system wil l act as a moderator in the relationship between foster children and placement breakdown. Families who perceive the foster care system as providing them with adequate support wil l be better equipped to successfully parent foster children and placement breakdown will occur in less often. 3. A low level of perceived power in foster parents will act as a moderator in their ability and willingness to provide foster care for children who display difficult behaviour. Parents with low perceived power will be more likely to be involved in fostering relationships that have lead to breakdown. 23 4. A relationship exists between parents' perceived level of support from the foster care system and low perceived power. Low power parents who perceive the foster care system as providing them with adequate support wil l have fewer instances of placement breakdown than parents with low power who do not perceive the foster system as providing them with adequate support. Figure 1. Relationship of variables under study Child behaviour Perceived support from the foster care system Foster placement breakdown Parental low perceived power 24 Chapter 5: Methodology Procedures By employing a retrospective observational research design, and collecting data from foster parents via surveys, it was believed that some factors associated with the breakdown of foster placements for children would be uncovered. This method allowed foster parents to provide information about the nature of past relationships they have had with foster children, and the outcome of those relationships. Logistical and linear regression analyses were used in the analysis of this data. The independent variable in the study was behaviour displayed by children aged 6 - 12 years of age. The dependent variable was breakdown of the child's foster placement as defined by the foster parents' decision to end the placement. Foster parents' perceived support from the foster care system and parental attributes of caregiving outcomes were examined as moderators in this relationship. A moderator model was used, instead of a main effect model, as it is recognized that foster parent-child relationships are complex and influenced by many factors (Bugental & Shennum, 1984). As a moderator, parental perceived power is particularly useful when looking at difficult or uncertain family situations (Bugental & Johnson, 2000). Parental perceived power was used as a moderator in this study to help qualify the reactions of foster parents to events in the family involving foster children with difficult behaviour. By including this as a moderator, the extent to which parental attributions play a role in the breakdown of foster placements was examined. Perceived social support also acts as a moderator in this model. This provides for an examination of the degree to which foster parents perceived social support affects the outcome of a foster placement in their home. 25 Using SPSS Version 14.0 computer software, binary logistic regression method was used to analyse data obtained from the sample population under exploration. Specifically, binary logistic regression was used to analyse the effect of child behaviour, perceived social support and parental perceived power on the dependent variable, foster placement outcome. Foster placement breakdown was measured by the item "foster parents request the child to be removed from the home" from the Foster Placement Outcome Measure. This is a binary response variable examining whether or not foster parents choose for their foster child to be removed from their home. Due to the small sample size, caution must be used in interpreting results due to a risk of Type II errors. Type II errors occur when the study fails to reject the null hypothesis when it is false. Logistic regression modelling is used when the response variable is binary and the explanatory variables are quantitative or qualitative. Logistic regression is used to predict a dependent variable on the basis of independent variable(s). It also allows for the interpretation of the percent of variance in the dependent variable as explained by the independent variable(s) and allows for the assessment of effects of interaction terms in the model (Garson, 2004). Logistic regression is applicable to this study as there is a binary response variable with two possible outcomes, foster placement breakdown or foster placement stability. Logistic regression provides us with the probability that a foster home placement breaks down in response to behaviour displayed by a foster child in the home. As well, the effect of the moderating independent variables, perceived social support and perceived parental power, was assessed by including them in the model as interaction terms (interact support x child behaviour and interact low power x child behaviour). Binary logistic regression was also used to examine the relationship 26 between perceived support and parental low power as parental low power is also a binary response variable. Logistic regression determines the probability of an event occurring by transforming the dependent into a logit variable (the natural log of the odds of the dependent occurring or not and then applying maximum likelihood estimation (MLE). M L E seeks to maximize the log likelihood, L L , which reflects the odds that the observed values of the dependent are predicted from the observed values of the independents. The model chi-square test, or likelihood ratio test, uses the L L and provides a test of significance indicating the goodness-of-fit for the logistic regression model. The significance test measures the improvement in fit that the independent variables make compared to the null model (Garson, 2004). The model-chi square wil l be used instead of the Wald statistic as it provides a more powerful test for small to moderate samples and there is a decreased risk of Type II errors (ie. false negatives (Agaresti & Finlay, 1999; Garson, 2004). To determine the strength of the association between the variables, the Nagelkerke R- square statistic, which ranges from 0 to 1, will be considered. This statistic attempts to measure the strength of association between the variables and is useful when the sample may be too small for a goodness-of-fit test. Requirements for logistic regression are not as stringent as linear regression as it does not assume a normal distribution or a linear relationship between the independent and dependent variables (Garson, 2004). The sample population involved in this research project consisted of individuals (male or female) who were current and former foster parents in the province of British Columbia. Individuals who responded to the survey were all members of the British 27 Columbia Federation of Foster Parent Associations (BCFFPA) as the study was advertised through this organization to its members. The B C F F P A is a registered non-profit society that is run by foster parents for foster parents who care for children in homes approved by M C F D (BCFFPA, 2004). To date, the organization provides support, educational opportunities, and training to approximately 1000 foster parents (BCFFPA, 2004). Membership includes foster parents residing in 5 regional districts across the province, each represented by a BCFFPA regional delegate (BCFFPA, 2004). Study respondents were asked to complete survey questionnaires that measuring the constructs outlined above. The study was advertised at an Annual General Meeting held by the British Columbia Federation of Foster Parents Associations (BCFFPA) in Kamloops, BC. Individuals at the conference included foster parents, M C F D social workers and community members. The study was presented at the conference and a total of 40 surveys were distributed to interested participants. Snowball sampling was used as many conference members took away survey packages to distribute to foster parents in their community. The study was also advertised via e-mail on the B C F F P A mailing list. Willing participants were asked to phone or send an e-mail to state their interest in the study. A total of 25 survey packages were sent to interested respondents who requested a study package by phone or email. The packages were distributed via Canada Post, along with stamped return envelope for the completed forms. Another 10 survey packages were sent to a regional BCFFPA delegate who stated interest in distributing the packages to BCFFPA members in their community. In total, 75 packages were distributed to potential participants. 28 Due to the limited time and scope of this research project, data was not collected from children in care or social workers involved in the foster care system. However, foster parents are believed to be a substantial source of information on this topic. Foster parents experience the support services available to them and are in a position to personally assess their effectiveness. Foster parents spend considerably more time with foster children than do social workers and experience first hand behavioural difficulties foster children may display. As a result, foster parents are often the individuals making decisions about the fate of foster children in their care. Volunteer participants were told that this study sought to understand factors associated with foster placement breakdown for children aged 6 - 1 2 years who displayed difficult behaviours. Participants were asked to report on their experiences with the most recent former foster child in their care. They were asked to think of a child who was between the ages of 6 - 12 years at the time the foster parent cared for them. Foster parents who chose to participate in this study were directed not to reveal the identity of the former foster child they were reporting on. Foster parents were also assured that their responses to survey questions would remain confidential and would in no way be released to the public. Participants had the option of withdrawing from the study at any time. Sample Description Participants were twenty-two foster parents who had experience fostering children aged 6-12 with whom they identified as displaying difficult behaviours. The sample consisted of individuals who voluntarily offered to participate in the study and they were not financially compensated for their time. Each family interested in participating 29 received one set of questionnaires and the individual who identified him/her self as the primary foster parent was asked to complete the questionnaires. Only one foster parent from each foster family was included in the final sample. Demographic information, as well as questions asking participants their about their fostering experiences were asked. In addition, four measures were included in the study and the estimated completion time of the survey was approximately 20 minutes. Of the 75 survey packages distributed to interested foster parents, 22 completed survey packages were returned (29% response rate). Initial interest in the project was strong and the completed surveys where returned in a timely fashion. However, subsequent advertisements of the survey and the final mail out of 10 survey packages did not result in additional respondents. Surveys were distributed to potential participants over a 3 month period of time. Measures Measures used in this study included; the Child Behaviour Checklist 6-18 (CBCL) Parent Report version for school aged children, the Parent Attribution Test, the foster parent support measure, and the foster placement outcome checklist. Child Behaviour Checklist The C B C L is an empirically based measure commonly used in academic research to identify behavioural and emotional problems in children (Newton et al., 2000; Tripp De Robertis & Litrownik, 2004). The C B C L is a measure consisting of 118 items testing for a range of child behavioural problems. The measure includes items to identify three competence scales for activities, socializing and school, as well as a total competence scale. Internalizing behaviours are determined as the sum of subscales for withdrawn, 30 somatic complaints and anxious depressed. Externalizing behaviours are sum of subscales for delinquent and aggressive behaviour and Total Problems is the sum of all problem behaviour items (Achenbach & Rescorla, 2001; Armsden et al., 2000). The latest version of the C B C L includes DSM-oriented scales, but wil l not be used as a diagnostic tool for this purpose in this study. Foster parent participants completed a two-page form consisting of 118 items on the measure. The C B C L uses a three-point Likert scale with respondents having the option to choose; "0 = not true" (as far as you know), 1 = "somewhat or sometimes true" and 2 = "very true or often true" in regards to how often their foster child engages in the behaviours described in the items. Although this measure contains 118 items, the questions are short and the completion time is approximately 10 minutes. For the purpose of this study, only scales used to determine aggressive and delinquent behaviour were used in the analysis as these were the behaviours of interest in the study. However, participants were asked to complete the full checklist of items to control for a possible experimental effect in asking only problem oriented questions. The aggressive behaviour scale was further separated into two scales with one for physical aggression and one for verbal aggression. Items from the aggressive behaviour scale were sorted on to either the physical or verbal aggression scales as created for this study. This was determined by whether or not the individual items reflected either physically or verbally aggressive behaviours as determined by the author. To estimate how reliable the scales were in measuring the construct under study, a Cronbach's alpha was performed. The alpha score for the physical aggression scale was .85, for the verbal aggression scale, .78 and the delinquency scale it was .78 indicating adequate to good reliability for these scales. 31 Parent Attribution Test The perceived cause of caregiving success or failure in the relationship between foster parents and foster children was measured using the Parent Attribution Test (PAT) Parent Form B - short form version. According to Bugental (1998), the PAT is thought to measure "the perceived balance of power or control within a relationship i.e., the amount of power or control attributed to self versus the amount of power or control attributed to children" (p. 1). The measures provides hypothetical situations in which parents are asked to respond on a 7 - point scale ranging from 1 = "not important at all" to 7 = "very important". The PAT short form version asks participants to respond to a series of questions related to each of two hypothetical situations involving a relationship with a child. Indicators for the construct low perceived power are: low attributed control to adults for unsuccessful outcomes (ACF), and high attributed control to children for unsuccessful outcomes (CCF). Parents who score low on the A C F and high on the CCF are seen as having low perceived power (Bugental, 1998). The PAT was used to explore whether or not low perceived power was a moderator in the relationship between foster children and the outcome of their foster placement. The coefficient of reliability or Cronbach's alpha for this scale was .88. Foster Parent Support Measure The support foster parents receive from the foster care system was measured with the foster parent support measure, a survey created specifically for this study. Foster parents were asked to rate the level of support they felt they received from various potential sources of support. A three point scale was used with the following options regarding each support service listed: 0 = "not helpful or not offered", 1 = "sometimes 3 2 helpful or somewhat helpful", and 2 = "very helpful or often helpful". The scale was devised to include various formal support services accessed by foster parents. For example, parents were asked how supportive they found training workshops, respite services and contact with M C F D social workers. This survey also asked respondents about informal support networks they access and the level at which they have found them to be helpful. Informal support networks include family, friends and other foster parents accessed on a casual basis (ie. not through foster parent support groups etc.). Space provided for parents to write additional comments. This measure was developed by the author. The Cronbach's alpha score for the support measure was .76, which indicates a reliable measure of support. Foster Placement Outcome Measure The foster placement outcome survey was developed to ascertain the occurrence of foster placement disruption. Foster parents were asked to identify the outcome of the last foster child in their foster placement in their home in regards to a child they cared for. Foster parents were asked whether or not they initiated the placement change, or i f it was initiated by an involved social worker. If a foster placement with a child was terminated, foster parents were asked to identify, to the best of their knowledge, where next the child was moved (ie. another foster home, residential program, psychiatric hospital inpatient unit, or return to family). In cases where the child was returned to their primary family, this did not constitute a placement breakdown. The other options were considered as evident of a foster placement breakdown. The questionnaire was in a check-mark format, with space provided for parents to write additional comments. This measure was developed by the author for the purpose of this study. 33 Although quantitative data was collected via the questionnaires outlined above, all surveys allowed for written comments by participants. This provided an element of qualitative analysis parents were invited to share their thoughts and experiences. This data will be included in the analysis to provide further depth of understanding of foster parents' lived experiences with children in their care. In addition, due to a small sample size, written comments provide for a greater range of data from participants. 34 Chapter 6: Results Demographics Demographic information collected on survey participants indicates that survey participants were predominately female (n=20, male n=l and missing, n=l) and ranged in aged from 40 years to 57 years with the median age being 50 years old. Basic demographic information was collected from participants in order to understand their experiences as foster parents. To gain perspective on the level of experience by the foster parents in the sample, they were requested to indicate how long they had been a foster parent and how many foster children they had cared for during this time. Responses revealed that these were experienced foster parents who had been fostering children for many years. The length of time indicated ranged from 5.5 years to over 20 years (median=9 years, mean=10.2 years). The level of experience of these foster parents was also indicated by the number of children they had fostered during these years. Responses for this questions ranged from 2 children to 60 plus children (median=30). The majority of foster parents had cared for approximately 20 - 40 children thus far. Most respondents indicated that they operated a levelled foster home (1, 2 or 3), designed for foster children with more challenging behaviour or special needs. Twenty-nine percent (n=6) operated a level one home, 38% (n=8) operated a level 2 home and 29% (n=6) operated a level 3 home. One respondent operated a regular designated foster home. Foster children are placed in different level homes based on an assessment of their needs and behaviours. Children placed in regular foster homes are considered to require "a general level of care," while children in a level 1 home "have average to moderately challenging behaviour" (MCFD, 2005). Foster parents with a 35 level one designation can provide care for up to six children at one time. Level 2 designation is reserved for homes "for up to three children who have moderate to severely challenging behaviour and/or a moderate to severe degree of risk to self, others or property" (MCFD, 2005). Level 3 homes are the most specialized with only two children residing in the home at one time. Children in these homes "may be a danger to themselves or others and require additional support and supervision" (MCFD, 2005). To ascertain the range in length of stay for foster children in these homes, respondents were asked to indicate the shortest duration of time a foster child was in their home, as well as the longest duration of time. Responses for the shortest duration of time ranged from less than one day to 7 months (median=14 days, mean= 37 days). Responses for the longest duration of time ranged from 3 months up to 9 years (median=3 years, mean=3.9 years). Participants were also asked i f they had other children in their home, such as natural, adoptive or step children. Responses indicated that 42% of these foster parents did not have other children living in their home, while 58% did have either adoptive, natural or stepchildren living with them. Descriptive Statistics Descriptive statistics (range, mean and standard deviation) for the variables under study are presented in Table 1. The variables for child behaviour and support are derived from 3-point likert scales and show a larger range and means compared to low power and placement breakdown which are binary response variables with a range of only one point. 36 Table 1 Descriptive statistics for variables under study (n=22) Range Mean Standard Deviation Delinquency 1.81 1.09 0.43 Verbal Aggression 1.33 1.35 0.46 Physical Aggression 2.00 1.39 0.53 Support 1.56 1.15 0.45 Low Perceived Power 1.00 0.68 0.48 Placement Breakdown 1.00 0.64 0.49 Hypotheses 1 Due to the large number of items on the child behaviour checklist (118), only items relating to the behaviour under study where used in the final analysis. Specifically, 18 items which compile the aggressive behaviour scale and the 11 items on the delinquency scale were included in the analysis. The aggressive behaviour scale items were then further divided into two scales, with 9 items on each. Items were sorted due to their relation with either verbal or physically aggressive behaviour (see Appendix G). This was done in order to provide further analysis of whether or not different types of aggressive behaviour are more or less likely to have an effect on the foster caring relationship. In the final analysis, physical aggression, verbal aggression and delinquency were measured as scale variables computed from the average raw scores of responses for each item on the scales. Binary logistic regression was able to capture not 37 only the relationship between child behaviour and placement breakdown, but the impact of these three different types of behaviour on the dependent variable. Table 2 Correlations between support, child behaviour, low perceived power and placement breakdown 1 2 3 4 5 6 1. Foster placement breakdown _ -.25 -.31 .20 -.24 70** 2. Support — .28 .049 .18 -.34 3. Delinquency — .19 .57** -.17 4. Verbal aggression .66** -.21 5. Physical aggression _ -.39 6. Low perceived power _ **p-value >.01 In the examination of a relationship between child behaviour and placement breakdown, the physical aggression scale, verbal aggression scale and delinquency scale were entered into the logistic regression model as independent variables with placement breakdown as the dependent variable. The fit of the model wil l be examined, in addition to the significance of the independent variables in predicting an outcome. In examining the classification tables for the model, it is indicated that there is minimal improvement in the model's ability to correctly predict the outcome when the variables, physical, verbal and delinquency are added. The model with the constant wil l predict correctly 63.6% of the time; while the model with the independent variables will predict correctly 77.3% of 38 the time (see Table 3). The Nagelkerke R square, which provides a score between 0 and 1 to identify the strength of association between the variables, is .39 and the -2 Log likelihood is 21.54. The Omnibus Tests of Model Coefficients provides a Chi-square statistic of 7.3 with a degree of freedom of 3 and p-value = .063. This provides a measurement of the improvement in the model when the independent variable is added compared to the null model. Since the p-value is greater than .05, we cannot reject the null hypothesis that adding the independent variable makes no difference. These statistics indicate that although the model improves with the addition of the independent variables, this improvement is moderate. Table 3 Summary of logistic regression analyses testing the relationship between physical aggression, verbal aggression and delinquency and foster placement breakdown for model fit Percentage correct Chi-square -2 Log likelihood Nagelkerke R square Constant 63.6 Model 1 77.3 73 21.54 _39 p-value = .063 An examination of the resulting statistics for the each independent variable reveals that none of the results reveal statistical significance in a relationship with the dependent variable. The variable, verbal aggression resulted in a p-value of .053 which indicates that this relationship nears statistically significance but, does not meet the standard test of significance for a p-value of > .05. Table 4 indicates that for the variable verbal 39 aggression the odds ratio Exp(B) explains that a for a one unit increase in verbal aggression the odds of a placement breakdown versus no placement breakdown, increases by a factor of 80.34. Results from this model indicate that verbal aggression is more likely to be related to foster placement breakdown than physical aggression or delinquency, but this result is not statically significant. P-values for physical aggression and delinquent behaviour are not near significant levels and do not indicate in an increase in the likelihood of placement breakdown. In fact, negative Beta scores (Table 4) reveal that lower levels of physical aggression and delinquency are related to placement breakdown. Given that foster parents reported physical aggression in particular as a concern (see comments below) this finding raises questions about the negative Beta scores in regards to the data collected. As well, 14 of the 22 respondents in the study indicated that they had requested that the foster child be removed from the home indicating that it is not uncommon for a foster parent to make this request. Although physical aggression was not determined to have a statistically significant effect on placement outcomes, written comments from foster parents in the study indicated that fear for their safety and the safety of other children in the home was a serious concern. This provides some descriptive information about foster parents' experiences with these children. One parent stated that s/he "requested that the child be removed from home because other children are at risk with his presence." Another parent described a child in her care as being a "large boy and extremely violent. Behaviour was worsening and I could not defend myself." Another parent describes her concerns for the future of a child in her care who "had many assessments and was diagnosed with several mental health problems. He had no control and with the rapid 40 growth and hormonal changes, was getting too dangerous to myself and other children in the home. Due to the lack of facilities for this type of child, he wil l just continue being moved until he is of age to be put away." Clearly physically aggressive behaviour by foster children had an effect on the stability of some foster placements in the study. Table 4 Summary of logistic regression analyses testing the relationship between physical aggression, verbal aggression, and delinquency with foster placement breakdown Variable B S.E. Exp(B) P-value Verbal aggression 4.39 2.27 80.34 .053 Physical aggression -4.16 2.49 0.02 .095 Delinquency -1.26 1.70 0.28 .458 Hypothesis 2 The second hypothesis in this study sought to examine the relationship between the support for foster parents and foster placement breakdown. The intent was to examine whether or not parents who perceived a greater amount of support were less likely to request a foster child to be moved from their home. The support scale was created from the average raw scores for the items on the support measure. Support was entered into logistic regression as an interaction term with verbal and physical aggression and delinquency to test the effect of support as a moderator in the relationship between child behaviour and placement breakdown. This relationship was not statistically significant. Support was also tested as an independent variable with placement breakdown as the dependent variable and this relationship was also not statistically significant. 41 Table 5 Summary of logistic regression analyses testing the relationship between support and foster placement breakdown Variable B S.E. Exp(B) P-value Support -1.21 1.07 .30 .257 Table 6 Summary of physical aggression , verbal aggression, delinquency and support as an interaction term with foster placement breakdown Variable B S.E. Exp(B) Constant 0.56 0.44 1.75 Support x physical x Verbal x delinquency -.028 0.250 0.97 Participants were asked to provide comment when completing the Foster Parent Support Measure. Comments provided by participants reveal that it may not be the support services themselves that parents do not find useful, but rather a lack of services available to assist them. For example, one participant stated that "Initially (sic) very difficult to access supports for both the child and myself. It took a full year before I was able to find someone interested or with the knowledge to provide support and assistance." Another parent stated that "You must make your own connections for support as there is very little available through the ministry (if any)." Other parents reported that a good 42 support team is very important when fostering a child with difficult behaviours, but that this was difficult to access. One parent indicated that "extreme budget slashing, and therefore cuts in services, are making it harder to continue to care for today's difficult children in care." This was echoed by another participant who addressed the issue of time limited support services and stated that: "Several months ago we lost our home support worker. This has created some problems". The lack of statistical results regarding foster parent support and the comments provided above indicated that overall, foster parents do not perceive the current support system to be meeting their needs. Hypothesis 3 The third hypothesis in this study examines the relationship between the variable low perceived power and foster placement breakdown, and low perceived power as a moderator in the relationship between child behaviour and placement breakdown. Low perceived balance of power over care giving failure is a binary response variable with participants scoring as either fitting the criteria for low perceived power or not fitting the criteria for low perceived power. This construct was scored according to the method outlined by short form version of the Parent Attribution Test. A Pearson correlation was performed to examine possible correlations between variables (Table 2). Low power and placement breakdown resulted in a significant correlation at the p <.01 level. No correlation was found between child behaviour and placement breakdown or child behaviour and low power. Low perceived power was entered into the binary logistic regression model as an independent variable to assess the likelihood of a relationship between low perceived power and foster placement breakdown (Model 1). In Model 2, low power was entered 43 with the constructs for child behaviour in a main effects model and as an interaction term to assess its effect as a moderator. Model 1 was found to be significant in that the resulting p-value is <.05. However, the interaction terms for Model 2 did not produce statistically significant results. Results provided for Model 2 reflect the main effects model. In examining the output for Model 1, it is evident that low power likely has an impact on the outcome of foster placements. In examining the statistics which determine the fit of the model (Table 7), it is evident that the model fit improves with the introduction of the child behaviour constructs. Table 7 indicates the Nagelkerke R square for Model 1 and Models 2 are showing range in the strength of association for the different models (.55 for Model 1 vs .61, .83, .56 for Models 2) with low power and verbal aggression showing the greatest association. The -2 Log likelihood is at its smallest value (8.45) for low power and verbal aggression compared to the other results (17.52, 15.84, & 17.39). The classification tables provide information that compared to the constant, adding the variables to the equation allows for the model to correctly predict the data more often (percent correct for constant 63.6%, percent correct for Model 1 and model 2 is 86.4%). As well, in comparing the results of the analyses of the percentage correct and the -2 Log likelihood from hypothesis 1 (table 3) for child behaviour, it is clear that adding low power with the child behaviour variables improves the predictive power of these variables, but not to a significant level. 44 Table 7 Summary of logistic regression analyses testing the effects of low perceived power with foster placement breakdown model fit Percent correct Chi-square -2 Log likelihood Nagelkerke R square Constant 63.6 Model 1 86.4 11.32* 17.52 .55 Model 2 (a) 86.4 13.00* 15.84 .61 Model 2(b) 86.4 20.39* 8.45 .83 Model 2 (c) 86.4 11.46* 17.39 .56 (a) = delinquency and low power (b) = verbal aggression and low power (c) = physical aggression and low power * P-value < .05 Examining the p-values resulting from analysis the variables low power and low power as an interaction term with delinquency, verbal and physical aggression on placement breakdown indicates that only low power produces a statistically significant result (Table 8). This is consistent with results from Hypothesis 1 that provided evidence that child behaviour as independent variables did not have a relationship with placement breakdown. Although low power is significant in the relationship with placement breakdown, it does not act as a moderator between child behaviour and placement breakdown. 45 Table 8 Summary of logistic regression analyses of low power with foster placement breakdown Variable B S.E. Exp(B) Low power -3.66* 1.32 0.03 * P-value <.05 Table 9 Summary of logistic regression analyses testing the effects of low perceived power with verbal aggression, physical aggression and delinquency on foster placement breakdown Variable B S.E. Exp(B) P-value Low power -3.83* 1.46 0.02 .009 Delinquency -2.15 1.79 0.12 .231 Low power -15.31 9.74 0.00 .118 Verbal aggression 15.31 10.19 4471108.8 .133 Low power -3.91* 1.53 0.02 .011 Physical aggression 0.69 1.88 2.00 .712 * P-value < .05. Comments from foster parents indicate some of the challenges in establishing successful fostering relationship and suggest that, at times, the success or failure of the relationship is seen as a result of the child's behaviour. One parent commented that: "Another very important issue with foster children is the dynamics of the children in your home - because these children are so fragile emotionally to begin with. If you have a bad 46 mixture of children in your home, it makes it very difficult to establish any relationship between the children and because you're always busy "putting out fires" your relationship with the children is jeopardized as well. Unfortunately while you are fostering this happens quite often because the children brought into care are not always known by the social workers or they might have developed attitudes at another home as a survival technique while there." Another parent commented that: "Fostering is not like parenting a child of your own, because foster children have not grown up with your family's ethics already internalized. It's more like taming wild horses. I've had to learn to balance expectations and reality and sorrow - and reserve judgment for later." Despite the limited confirmation of hypotheses in this small sample research, these remarks suggest that fostering children is a challenging endeavour and that the perception of the causes of success or failure in these relationships plays a factor in the stability of these unions. Hypothesis 4 The final hypothesis of this study aimed to examine the relationship between the foster parents perceived level of support and low perceived power. Logistic regression analyses was performed for this purpose as low perceived power is a binary response variable. Pearson correlation, as identified in Table 2, indicates the variables low power and support are not significantly correlated with one another. Parental perceived support was entered into the logistic regression analysis as an independent variable with low perceived power as the dependent variable. Results from this model revealed that this relationship is not statistically significant (Table 10). Resulting statistics (Table 10) 47 reveal a small Nagelkerke R square (.15) and a large -2 Log likelihood (24.96) indicating a poor fit to this model. Table 10 Summary of logistic regression analyses testing the effects of support on low perceived power model fit Percent correct Chi-square -2 Log likelihood Nagelkerke R square Constant 68.2 Model 63.6 2.56 24.96 .15 P-value=.110 Table 11 Summary of logistic regression analyses testing the effects of support on low perceived power Variable B S.E. Exp(B) Support -1.75 1.17 0.17 P-value= .134 48 Chapter 7: Discussion The present study attempted to test the hypothesis that difficult behaviour exhibited by foster children aged 6-12 years of age affects the outcome of foster placements. As well, it attempted to determine the moderating and direct effects of parental low perceived power and perceived support from the foster care system on this relationship. The results provided evidence that in this sample the type of difficult behaviour displayed by foster children may be a consideration. Specifically, there was a limited relationship between verbal aggression and placement outcome and evidence was lacking for the effect of physical aggression or delinquency on this relationship. A relationship was determined to exist between low perceived power in caregiving relationships and foster parents' decision to end a foster placement. The hypothesis examining the role of parental perceived support as factor likely to prevent foster placement breakdown was not supported in this study. However, comments by foster parents reveal that they often do not feel that they have access to adequate support, rather than support not being useful to them. This study provided limited confirmation of earlier empirical findings suggesting that externalizing child behaviour has a direct effect on foster placement outcomes (Newton et al., 2000). This study also suggests the importance of examining the type of behaviour a child exhibits as a predictor of placement outcome. Other studies have examined externalizing versus internalizing behaviour as a predictor of placement breakdown (Newton et al., Fisher, 2000). This study provided further analysis that it is important to consider the behaviour displayed, such as verbal aggression, physical aggression or delinquency. In this case, analysis revealed that verbal aggression by 49 children was the behaviour most likely to be associated with the outcome of foster placements; however this finding was not statistically significant. Verbal aggression is likely to be a more common behaviour displayed by children than physical aggression or delinquency, especially given the age of participants. Previous research clearly indicates that physical aggression is often a factor in placement breakdown (Newton et al., 2000; Fisher, 2000; Gilbertson & Barber, 2003). However, due to the small sample size in this study, it is possible that this was not captured in the results. Given the total number of foster parents who requested a child to be moved from their home (n= 14/22) and written comments outlining aggressive behaviour exhibited by the foster children, it would seem likely that this would be a concern. The fact that this was not found to be statistically significant may be due in part to length of time foster parents have cared for difficult children (mean=10.2 years) thus making them more likely to be able to manage these children in their home. As all parents in the study, with the exception of one, operated levelled foster home designated to care for children with difficult behaviour. It is possible that foster parents with extensive experience may report differently on difficult child behaviour compared to that of parents without this experience. Overall, findings do suggest that further investigation into specific child behaviours may reveal more information about the nature of foster placement stability. The most significant finding in this study resulted from the analysis of the construct low perceived power and its relationship with placement breakdown. Results of this analysis were statistically significant revealing that parental attributions may play an important role in the relationship between foster carers and their foster children. Results indicate that when foster parents perceive themselves as having low power or 50 little control over the behaviours of children in their care they are more likely to request the child removed from their home. This is a significant factor in terms of promoting foster placement stability. There may be implications for the screening of foster parents in terms of their ability to manage difficult behaviours. As well, such screening may assist in more appropriate matching of foster parents to foster children. As results indicate, the level of control foster parents feel in the caregiving relationship is an important factor. Assisting to facilitate this by examining the dynamics of the parent/child relationship could benefit foster placement stability long-term (Baker et al., 2000; Gilbertson & Barber, 2003). Perceived support from the foster care system did not appear to have a significant effect overall on the outcome of foster placements in this study according to the statistical results. However, written comments provided by parents revealed that many foster parents did not perceive support programs as available to them rather than not finding them to be helpful. In the absence of support services, it is possible that foster parents may be more likely to terminate placements in their home. Left without resources or respite when working with behaviourally challenged children, is likely to negatively impact the fostering relationship over time (Baker, Gibbs, Sinclair & Wilson, 2000). Comments by foster parents revealed a level of frustration in the absence of supportive services that they requested but were not available or had been withdrawn. Despite the absence of statistical results indicating that support is helpful in this study, comments by foster parents reveal that they wish more supports would be available to them. This is an important factor to consider in the interpretation of results. In this study, foster parents were asked whether or not a program was offered or found to be helpful. Unfortunately, 51 this questioning does not provide clear information about whether or not parents had access to these supportive services. To further explore these questions, it would be beneficial to ask parents about support availability and the effectiveness of supports as separate constructs. Several empirical studies have found supportive services to be very beneficial to the stability of foster placement. Many of these studies examine the role of specific support programs by studying a cohort of foster parents and comparing a control group with an experimental group. These studies attempt to address specific needs of foster families by providing intensive support programs (Chamberlain & Moreland, 1992; Fisher, 2000). This study attempted to ascertain the level of support felt by foster parents as available in the foster care system at large in BC. Not all services would be available or appropriate for all study subjects. In addition, many study participants who responded were from rural or northern communities in BC which may have less access to support services. As a result, it is difficult to gain a clear idea which supports participants have found to be helpful. 52 Chapter 8: Summary Limitations Several limitations exist for this research project due to the type of study that was undertaken and the research methods employed. Although there was a great deal of support from the B C F F P A for the project, only a small number of foster parents completed and returned survey packages. As a result, it is difficult to generalize the statistical results to the larger population of foster parents. Due to limited time and funds available to conduct this study, a retrospective design was employed. Limitations exist, however, in doing so. For instance, participants were asked to complete surveys based on experiences with children no longer in their care. As a result, memories may be altered, thus creating less accurate responses on survey questions and increasing the margin of error. Observational studies allow researchers to compare participants who happen to have experienced the conditions under study. However, causation cannot be inferred from study results. Survey methods were chosen in an attempt to reach a large number of foster parents across the province. However, there are restrictions to using survey methods. Due to the fact that participants have volunteered to complete the surveys, self-selection bias is a potential source of sampling error. It is possible that foster parents with more negative or positive experiences with foster children were likely to respond, thus creating bias in the results. Bias may have been extenuated by the fact that all participants were members of the BCFFPA. In addition, only practising foster parents were included in the sample. Individuals who have decided that they no longer want to provide this service were automatically excluded as this study did not have access to them. Former foster 53 parents could provide potentially significant data, particularly i f their reason to stop being a foster parent was related to difficulties encountered when caring for foster children. This limits the ability of the results to be generalized to the larger population of foster parents under study. Due to limits in the time and scope of this research project, the sample size for this study was small and restricted to the experiences of foster parents only. As a result, the perspective of foster children and the social workers involved in these foster relationships could not be examined. This too restricted the generalizability of results, as it is possible that different data would have been collected had the sample also consisted of foster children and social workers. Strengths Research studies on the foster care system in Canada are limited, making this study one of a few looking at the foster care system in BC. This study provided a voice for foster parents to express their experiences of being a foster caregiver in this province. Survey questions regarding foster children in their care and the support they receive from the foster care system provide a lens for viewing what is occurring in foster homes today. Foster parents from across the province were included, providing a perspective from both rural and urban centres. Examining the role of low perceived power for caregiving outcomes for foster placements expands beyond focus on the child or the foster care system to include characteristics of foster parents themselves. This also provides a unique perspective on the outcome of foster placements. This study may provide a model for future research on the foster care system as it identifies areas of concern and the need for further investigation. Although all hypotheses were not support statistically, written comments and descriptive data about the sample would indicate the need for further 54 research on these variables. This study also reflects the commitment of foster parents in the province to continue to care for the difficult children in their care in the absence of well developed support system. Given the many challenges foster parents face, it is clear from this sample that many parents continue to foster for years and in doing so provide care to a number of children in need. A longitudinal study of foster placements examining low perceived power, child behaviour and support on placement outcomes could provide beneficial information on our current foster care system and areas in need of improvement. In particular, a focus on the experiences of foster parents in rural versus urban centres is needed to examine i f differences exist in outcomes for foster placements due to the availability of support services and other factors. Conclusion When there is a breakdown in a foster placement, it is not only the child who is affected but also the foster family, the child welfare system and society as a whole (Martin, 2000). As a result, it is imperative that factors associated with foster placement breakdown are identified and analysed. Research in this area has implications for the development and implementation of child welfare policies regarding the foster care system (Newton et al., 2000). Included in this are the selection, preparation and training of individuals willing to assume the role of foster parent. It is imperative that social workers and other professionals in the child welfare system are aware of the challenges faced by foster parents in caring for difficult children. Foster parenting relationships are complex and that it is likely a combination of many factors that influences the outcome of placements, rather than one definable variable. As identified in this study, the attributions of foster parents towards caregiving and the type of behaviour displayed by 55 children are important areas to consider. In addition, research in this study and others, indicates that it is important to consider the 'fit' between members in a fostering relationship. Baker et al., (2000), concluded in their study of fostering relationships that: "The best carers could fail with some children. The most difficult children could succeed in some families" (p. 7). Matching children with appropriate foster parents and providing services that appropriately support the fostering relationship may contribute to the stabilization of foster homes and thus the lives of vulnerable children (Baker et al., 2000; Gilbertson & Barber, 2003). Research can provide an opportunity for foster parents, foster children, social workers and other professionals in the child welfare system to describe their experiences in an attempt to steadily improve the current system. Closing words from a participant in the study: Foster parenting is: rewarding, exhausting, challenging, draining. Thanks for caring enough to ask. 56 References Achenbach, T. M . (1991). Manual for the Child Behavior Checklist 4/18 - and 1991 profile. Burlington, VT: University i f Vermont Department of Psychiatry. Achenbach, T. M . , & Rescorla, L. A . (2001). Manual for theASEBA school-age forms & profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth & Families. Agresti, A . & Finlay, B. (1999). Statistical methods for the social sciences (3 r d ed.). New Jersey: Prentice Hall. Andreae, D. (1996). Systems theory and social work treatment. In F. J. Turner (Ed.), Social work treatment (4 t h ed.) (pp. 601-616). Toronto: The Free Press. Armsden, G., Pecora, P. J., Payne, V . H. , & Szatkiewicz, J. P. (2000). Children placed in long-term foster care: A n intake profile using the Child Behavior Checklist/4-18. [Electronic Version]. Journal of Emotional and Behavioral Disorders, 8, 49-64. Baker, C , Gibbs, L, Sinclair, I., & Wilson, K . (2000). Supporting Foster Placements. Social Work Research and Development Unit, University of York. Retrieved July 4, 2006 from: Baron, R. M . , & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research. Journal of Personality and Social Psychology, 51, 1173-1182. Barratt, S. (2002). Fostering care: The child, the family and the professional system [Electronic Version]. Journal of Social Work Practice, 16, 163-173. British Columbia Federation of Foster Parents Associations. (2004). B C F F P A info. Retrieved May 05, 2004 from: 57 Brown, J. H. , & Christensen, D. N . (1999). Family therapy: Theory and practice (2 n d ed.). Toronto: Brooks/Cole Publishing Company. Bugental, D. B., & Shennum, W. A. (1984). "Difficult" children as elicitors as targets of adult communication patterns: An attributional-behavioral transactional analysis. Monographs of the Society for Research in Child Development, 49 (1, Serial No. 205). Bugental, D. B. (1998). Parent attribution test. Retrieved online December 1, 2003 from: Bugental, D. B., & Blue, J. (1989). Perceived control over caregiving outcomes: Implications for child abuse [Electronic version]. Developmental Psychology, 25, 532-539. Bugental, D. B., Lyon, J. E., Krantz, J., & Cortez, V . (1999). Who's the boss: Differential accessibility of dominance ideation in parent-child relationships [Electronic version]. Developmental Psychology, 36, 1367-1378. Bugental, D. G., & Johnson, C. (2000). Parental and child cognitions in the context of the family [Electronic version]. Annual Review of Psychology, 51, 315-344. Bugental, D. B., Johnson, C , New, M . , & Silvester, J. (1998). Measuring parental attributions: Conceptual and methodological issues [Electronic version]. Journal of Family Psychology, 12, 459-480. Cavell, T. A. (2000). Working with parents of aggressive children. Washington, DC: American Psychological Association. Chamberlain, P., & Moreland, S. (1992). Enhanced services and stipends for foster parents: Effects on retention rates and outcomes for children [Electronic version]. 58 Child welfare, 71, 387-402. Fiske, S. T., & Taylor, S. E. (1984). Social cognition. New York: Random House. Fisher, P. A. (2000). Preventive intervention for maltreated preschool children: Impact on children's behavior, neuroendocrine activity, and foster parent functioning [Electronic Version]. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 1356-1364. Garson, D. (2004). Logistic regression. Retrieved May 04, 2004 from: Gilbertson, R. & Barber, J. (2003). Breakdown of foster care placements: carer Perspectives and system factors [Electronic Version]. Australian Social Work, 56, 329-339. Huesmann, L . R., Moise-Titus, J., Podolski, C-L., & Eron, L. D. (2003). Longitudinal relations between children's exposure to T V violence and their aggressive and violent behavior in young adulthood: 1977 - 1992 [Electronic Version]. Developmental Psychology, 39, 201-221. Kalland, M . & Sinkkonen, J. (2001). Finnish children in foster care: Evaluating the Breakdown of long-term placements [Electronic Version]. Child Welfare, 80, 513-528. Koman, S. L., & Stechler, G. (1985). Making the jump to systems. In Mirkin, M . P., & S. L. Koman (Eds.), Handbook of adolescent psychotherapy and family therapy (pp. 3-42). New York: Gardner Press. Leslie, L. A . (1988). Cognitive-behavioral and systems models of family therapy: How compatible are they? In N . Epstein, S. E. Schlesinger & W. Dryden (Eds.), 59 Cognitive-behavioral therapy with families (pp. 49-83). New York: Braner Mazel. Ministry for Children and Family Development. (2003). Foster care in British Columbia. Retrieved November 10, 2003 from: Ministry for Children and Family Development. (2006). Foster care in British Columbia. Retrieved November 4, 2006 from: Martin, J. A . (2000). Foster family care: Theory and practice. Toronto: Allyn and Bacon. Newton, R. R., Litrownik, A . J., & Landsverk, J. A . (2000). Children and youth in foster care: Disentangling the relationship between problem behaviors and number of placements [Electronic version]. Child Abuse and Neglect, 24, 1363-1374. Nichols, M . P. (1984). Family therapy: Concepts and methods. New York: Gardner Press, Inc. Ramey, S. L. (2002). The science and art of parenting. In J. G. Borkowski, S. L. Ramey & M . Bristol-Power (Eds.) Parenting and the child's world: Influences on academic, intellectual, and social-emotional development. New Jersey: Lawrence Erlbaum, Associates. Tripp De Robertis, M . & Litrownik, A . J. (2004). The experience of foster care: Relationship between foster parent disciplinary approaches and aggression in a sample of young foster children [Electronic Version]. Child Maltreatment, 9, 92-102. 60 White, M . C. J. (1982). Consistency in cognitive social behaviour. Boston: Routledge & Kegan Paul. Wild, C. J., & Seber, G. A . F. (2000). Chance encounters: A first course in data analysis and inference. Toronto: John Wiley & Sons, Inc. 61 Appendix A : Letter of Consent T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A Research Project Title: Parental attributions and foster care support: Examining foster placement outcomes for children displaying difficult behaviour This research project will be conducted by Tanya Strubin, Master of Social Work student at The University of British Columbia as partial requirement for completing a graduate thesis. Dr. Richard Sullivan, UBC School of Social Work and Family Studies, will be the Principal Investigator and supervise the study. The BC Federation of Foster Parents Associations has kindly agreed to support this research project. As a current practicing foster parent, who has identified yourself as having cared for foster children who display difficult behaviours in your home, you are invited to participate in this research project. You will be asked to report on a child formerly in your care between the ages of 6 -12 years. Please report on the last child you fostered who is no longer in your care. Your contribution to this study is greatly appreciated. Foster children who act in a difficult manner pose many challenges for their caregivers. This study seeks information from foster parents regarding factors that may play a role in how foster parents face these challenges and care for these children. The questionnaires you are asked to complete reflect the issues under study. It is hoped that completion of this study will provide valuable information that could benefit both foster children and foster parents. We are aware that participation in this study may bring up feelings of discomfort or memories that are difficult for you. Should this occur, you are invited to contact Tanya Strubin or Richard Sullivan to discuss your concerns. (Version #3 cover letter) School of Social Work 2080 West Mall Vancouver, B.C. Canada V6T 1Z2 Tel: 604-822-2255 Fax: 604-822-8656 Page 2 of 2 62 Appendix B: Demographic Measure Dear foster parent, In order to understand who you are I would like to ask a few questions. Remember, this information is anonymous and will be kept confidential. Please do not include any identifying personal information (ie. name, birth date, address etc.). Please fill in a response for each question provided. Your Gender: Female Male Your Age: How long you have been a foster parent: The approximate number of foster children you have cared for: On average, the number of foster children in your home at one time: Are you designated as a regular foster home, or a level 1, 2 or 3 foster home ? In order to get an idea of the range of time children stay in one foster home, please indicate the duration of time foster children have stayed with you. The shortest duration of time The longest duration of time Do you have other children besides foster children in your home (ie. natural, adoptive or stepchildren)? 64 Appendix C: Foster Parent Support Measure FOSTER PARENT SUPPORT MEASURE Please indicate the level of support you feel you have received in your role as a foster parent from the services listed below. Please answer specifically in relation to your experiences working with the child you are reporting on for this survey (ie. the last foster child formerly in your care). 0 = Not Offered or Not Helpful 1 = Somewhat Helpful or Sometimes helpful 2 = Very helpful or Often helpful 1. Training workshops and education seminars 0 1 2 2. Respite services (ie. Child stays overnight at a respite home) 0 1 2 3. Child care worker (ie. the child is taken out for activities) 0 1 2 4. Before or after school daycare 0 1 2 5. Contact with the resource social worker 0 1 2 6. Contact with a guardianship or child protection social worker 0 1 2 7. Consultation with a child behaviour consultant 0 1 2 8. Consultation with the child's mental health therapist or counsellor 0 1 2 9. Foster parent support group 0 1 2 Continued on the next page 65 0 = Not Offered or Not Helpful 1 = Somewhat Helpful or Sometimes helpful 2 = Very helpful or Often helpful 10. Please list any other supports that you have utilized (ie. Friends, other foster parents, family etc. - do not include names please). a) 0 1 2 b) 0 1 2 c) 0 1 2 11. Other (please specify): 0 1 2 Additional Comments: 66 Appendix D: Foster Placement Outcome Measure Please indicate: Your foster child's gender: Male or Female Foster child's age when they lived with you: FOSTER PLACEMENT OUTCOMECHECKLIST Please respond to questions in terms of your experiences with the child you are reporting on for this survey (ie. the last foster child formerly in your care). At the beginning of the placement, how long were you told that the child would live in your home? How long did the child live in your home? **Please do not include the name of the foster child you are reporting on. Please respond to the following questions by indicating YES or NO. YES NO 1. Did you ever request that the foster child be removed from your home? 2. Did a social worker request that the foster child be moved from your home? 3. Did the child move from you home to live in a residential setting (ie. a group home, assessment / treatment program, or a hospital inpatient unit)? 4. Did the foster child move from your home to live with their natural family or an adoptive family? 5. Did the child move to live in a long-term foster placement? 6. Did the foster child move from your care for reasons other than those listed above? If so, please provide the reason: 6 7 Additional comments: 68 Appendix E: Child Interaction Survey When completing the following Child Interaction Survey, do not report on your foster child. Instead, report on your experiences with children in general. Thank you. CHILD INTERACTION S U R V E Y In this questionnaire, we want to know how important you believe different factors might be as potential causes of successful and unsuccessful interaction with children. We are interested in discovering the way people think about children-there are no right or wrong answers. For this survey, do not report on your former foster child. Instead, respond according to the instructions given below. Place a circle around a number. Pick one of the bigger numbers if you think this factor is important, and a smaller number if you think it is not important. Example: If you were teaching a child an outdoor game and he or she caught on very quickly, how important do you believe these possible causes would be? not at all very important important a. How good she is in sports in general. 1 2 3 4 5 6 7 b. How good a teacher you are. 1 2 3 4 5 6 7 c. How easy the game is. 1 2 3 4 5 6 7 Answer the following questions by making ratings in the same way as shown on the previous page. 1. Suppose you took care of a neighbour's child one afternoon, and the two of you had a really good time together. How important do you believe the following factors would be as reasons for such an experience? not at all very 69 important important a. whether or not this was a "good day" for the child, 1 2 3 4 5 6 7 e.g., whether there was a TV show s/he particularly wanted to see (or another special thing to do). d. how lucky you were in just having everything 1 2 3 4 5 6 7 work out well. e. how much the child enjoys being with adults. 1 2 3 4 5 6 7 f. how pleasant a disposition the child had. 1 2 3 4 5 6 7 g. how well the neighbour had set things us for you 1 2 3 4 5 6 7 in advance. h. whether the child was rested. 1 2 3 4 5 6 7 The next question asks about BAD experiences with children. Reasons for good interactions are not necessarily the same as those for unsuccessful ones. So please think about this situation without regard for the way you answered the first question. 2. Suppose you took care of a neighbour's child one afternoon, and the two of you did not get along well. How important do you believe the following factors would be as possible reasons for such an experience? not at all very important important b. how unpleasant a disposition the child had. 1 2 3 4 5 6 7 c. whether the child was tired or not feeling well. 1 2 3 4 5 6 7 d. whether or not you really enjoy children that much. 1 2 3 4 5 6 7 not at all very important important f. whether or not this was a bad day for the child, 1 2 3 4 5 6 7 e.g., whether there was nothing good on TV, whether it was raining and s/he couldn't go outside. i. whether you used the wrong approach for this child. 1 2 3 4 5 6 7 70 j. the extent to which the child was stubborn and 1 2 3 4 5 6 7 resisted your efforts. k. how you get along with children in general. 1 2 3 4 5 6 7 m. what kind of mood you were in that day. 1 2 3 4 5 6 7 q. how hungry the child was. 1 2 3 4 5 6 7 t. how little effort the child made to take an interest 1 2 3 4 5 6 7 in what you said or did. u. the extent to which you were not feeling well that day. 1 2 3 4 5 6 z. whether or not this was a bad day for you in general. 1 2 3 4 5 6 7 Additional Comments: Is there anything else that I need to know about you and your experiences as a foster parent? 71 Thank you so much for completing the surveys. I know that you are busy and I really appreciate your time. 7 2 Appendix F: Child Behaviour Checklist Please print. Be sure to answer all items. Below is a list of items that describe children and youths. For each item that describes your child now or within the past 6 months please circle the 2 if the item is very true or often true of your child. Circle the 1 if the item is somewhat or sometimes true of your child. If the item is not true of your child, circle the 0. Please answer all items as well as you can, even if some do not seem to apply to your child. 0 = Not True (as far as you know) 1 = Somewhat or Sometimes True 2 = Very True or Often True 0 1 2 1. Acts too young for his/her age 0 1 2 32. Feels he/she has to be perfect 0 1 2 2. Drinks alcohol without parents' approval 0 1 2 33. Feels or complains that no one loves him/her (describe): 0 1 2 34. Feels others are out to get him/her 0 1 2 35. Feels worthless or inferior 0 0 1 1 2 2 3. 4. Argues a lot Fails to finish things he/she starts 0 0 1 1 2 2 36. 37. Gets hurt a lot, accident-prone Gets in many fights 0 0 1 1 2 2 5. 6. There is very little he/she enjoys Bowel movements outside toilet 0 0 1 1 2 2 38. 39. Gets teased a lot Hangs around with others who get in trouble 0 0 1 1 2 2 7. 8. Bragging, boasting Can't concentrate, can't pay attention for long 0 1 2 40. Hears sounds or voices that aren't there (describe!: 0 1 2 9. Can't get his/her mind off certain thoughts; obsessions (describe): 0 1 2 41. Impulsive or acts without thinking 0 1 2 10. Can't sit still, restless, or hyperactive 0 0 1 1 2 2 42. 43. Would rather be alone than with others Lying or cheating 0 0 1 1 2 2 11. 12. Clings to adults or too dependent Complains of loneliness 0 0 1 1 2 2 44. 45. Bites fingernails Nervous, highstrung, or tense 0 0 1 1 2 2 13. 14. Confused or seems to be in a fog Cries a lot 0 1 2 46. Nervous movements or twitching (describe): _ 0 1 2 15. Cruel to animals 0 1 2 16. Cruelty, bullying, or meanness to others 0 1 2 47. Nightmares 0 1 2 17. Daydreams or gets lost in his/her thoughts 0 1 2 48. Not liked by other kids 0 1 2 18. Deliberately harms self or attempts suicide 0 1 2 49. Constipated, doesn't move bowels 0 1 2 19. Demands a lot of attention 0 1 2 50. Too fearful or anxious 0 1 2 20. Destroys his/her own things 0 1 2 51. Feels dizzy or lightheaded 0 1 2 21. Destroys things belonging to his/her family or 0 1 2 52. Feels too guilty others 0 1 2 53. Overeating 0 1 2 22. Disobedient at home 0 1 2 54. Overtired without good reason 0 1 2 23. Disobedient at school 0 1 2 55. Overweight 0 1 2 24. Doesn't eat well 56. Physical problems without known medical 0 1 2 25. Doesn't get along with other kids cause: 0 1 2 26. Doesn't seem to feel guilty after misbehaving 0 1 2 a. Aches or pains (not stomach or headaches) 0 0 1 1 2 2 27. 28. Easily jealous Breaks rules at home, school, or elsewhere 0 0 0 1 1 1 2 2 2 b. c. d. Headaches Nausea, feels sick Problems with eyes (not if corrected by glasses) 0 1 2 29. Fears.certain animals, situations, or places, (describe): other than school (describe): 0 0 1 1 2 2 e. f. Rashes or other skin problems Stomachaches 0 1 2 30. Fears going to school 0 1 2 9- Vomiting, throwing up 0 1 2 31. Fears he/she might think or do something bad 0 1 2 h. Other (describe!: P A G E 3 Be sure you answered all Items. Then see other side. J 73 Please print. Be sure to answer ail items. 0 = Not True (as far as you know) 1 = Somewhat or Sometimes True 2 = Very True or Often True 0 1 2 57. Physically attacks people 0 1 2 84. Strange behavior (describe): 0 1 2 58. Picks nose, skin, or other parts of body (described 0 1 2 85. Strange Ideas (describe): 0 1 2 59. Plays with own sex parts in public 0 1 2 86. Stubborn, sullen, or irritable 0 1 2 60. Plays with own sex parts too much 0 1 2 87. Sudden changes in mood or feelings 0 1 2 61. Poor school work 0 1 2 88. Sulks a lot 0 1 2 62. Poorly coordinated or clumsy 0 1 2 89. Suspicious 0 1 2 63. Prefers being with older kids 0 1 2 90. Swearing or obscene language 0 1 2 64. Prefers being with younger kids 0 1 2 91. Talks about killing self 0 1 2 65. Refuses to talk 0 1 2 92. Talks or walks in sleep (describe): 0 1 2 66. Repeats certain acts over and over; compulsions (describe): 0 1 2 93. Talks too much 0 1 2 94. Teases a lot 0 1 2 67. Runs away from home 0 1 2 95. Temper tantrums or hot temper 0 1 2 68. Screams a lot 0 1 2 96. Thinks about sex too much 0 1 2 69. Secretive, keeps things to self 0 1 2 97. Threatens people 0 1 2 70. Sees things that aren't there (describe): 0 1 2 98. Thumb-sucking 0 1 2 99. Smokes, chews, or sniffs tobacco 0 1 2 71. Self-conscious or easily embarrassed 0 1 2 100. Trouble sleeping (describe): 0 1 2 72. Sets fires 0 1 2 101. Truancy, skips school 0 1 2 73. Sexual problems (describe): 0 1 2 102. Underactive, slow moving, or lacks energy 0 1 2 103. Unhappy, sad, or depressed 0 1 2 74. Showing off or clowning 0 1 2 104. Unusually loud 0 1 2 75. Too shy or timid 0 1 2 105. Uses drugs for nonmedical purposes (don't 0 1 2 76. Sleeps less than most kids I n r i n r i R alrnhnl n r tnhflfim) (describeY. 0 1 2 77. Sleeps more than most kids during day and/or night (describe): 0 1 2 106. Vandalism 0 1 2 78. Inattentive or easily distracted 0 1 2 107. Wets self during the day 0 1 2 79. Speech problem (describe): 0 1 2 108. Wets the bed 0 1 2 109. Whining 0 1 2 80. Stares blankly 0 1 2 110. Wishes to be of opposite sex 0 1 2 81. Steals at home 0 1 2 111. Withdrawn, doesn't get involved with others 0 1 2 82. Steals outside the home 0 1 2 112. Worries a 1 2 83. Stores up too many things he/she doesn't need 113. Please write in any problems your child has that (describe!: were not listed above: 0 1 2 0 1 2 0 1 2 P A C E 4 Please be sure you answered all items. Appendix G: Child Behaviour Checklist study scale items Verbal aggression scale items: 3. Argues a lot 7. Bragging, boasting 16. Cruelty, bullying, meanness to others 19. Demands a lot of attention 68. Screams a lot 86. Stubborn, sullen or irritable 93. Talks too much 94. Teases a lot 104. Unusually loud Physical aggression scale items: 20. Destroys his/her own things 21. Destroys things belonging to his/her family or others 22. Disobedient at home 23. Disobedient at school 27. Easily jealous 37. Gets in many fights 57. Physically attacks people 74. Showing off or clowning 97. Threatens people Delinquency scale items: 26. Doesn't seem to feel guilty after misbehaving 39. Hangs around with others who get in trouble 43. Lying or cheating 63. Prefers being with older kids 67. Runs away from home 72. Sets fires 81. Steals at home 82. Steals outside the home 90. Swearing or obscene language 96. Thinks about sex too much 101. Truancy, skips school 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items