Differences in psychosocial functioning of foster children between kinship and non-kinship foster care:
A meta-analysis
Johanna Bakker
Student number: 10035370
Forensic Child and Youth Care Sciences University of Amsterdam
Supervisors: Prof. dr. G.J.J.M., Stams, Prof. dr. J. Vanderfaeillie, & Dr. M. Hoeve January 7, 2014
Abstract
This meta-analytic study examined risk and protective factors of kinship and non-kinship foster care and the degree to which these factors were associated with psychosocial functioning of the foster children. Methods A total of 35 studies, comprising 49,986
participants, were included. Effect sizes were computed from differences between kinship and non-kinship foster care in characteristics of the foster parents, biological parents and foster children as well as placement characteristics. Subsequently, these effect sizes were used to predict differences in social and cognitive development and psychopathology between foster children placed in kinship and non-kinship forster care. Results The results show that children placed in kinship foster care were exposed to less protective factors and more risk factors (e.g., unsafety, lower SES, ethnic minority background, less professional support) than non-kinship foster care. Despite these less positive findings for non-kinship care, more placement disruptions, higher rates of psychopathology and lower levels of social functioning were found with foster children placed in non-kinship foster care. Less employment and more absence of biological parents in kinship care compared to non-kinship care were associated with smaller differences in children’s psychopathology between kinship and non-kinship foster care. Conclusions Several risks are present in kinship foster homes and this asks, especially in situations with seriously disturbed foster children, for a careful matching process. Selective placement may explain less positive psychosocial outcomes in children placed in non-kinship foster care, and subsequently higher disruption rates.
Keywords: kinship foster care; non-kinship foster care; meta-analysis; psychosocial functioning
Introduction
The use of the family’s network in child and youth care is currently a widely supported practice, which for instance manifests itself in the Family Group Decision Making (FGDM) approach (Crampton, 2006; Frost, Abram, & Burgess, in press, Oosterkamp-Szwajcer, De Swart, & Gramberg, 2012; Pennell, Edwards, & Burford, 2010). A possible outcome of FGDM is to place a child at risk for out-of-home placement in a kinship foster home instead of a non-kinship foster home or (semi-) residential care. Notably, the use of kinship foster care, with family members or a member of the social network of the child, has shown a substantial increase in recent years (Washington, Gleeson, & Rulison, 2013), whereas placement in non-kinship foster care, with non-relative foster parents, has shown a decline (Honomichl & Brooks, 2009). Official placement in kinship foster care (i.e., involving the child welfare system or juvenile court) provides 24 percent of the children who are in out of home care (Koh, 2010; Pennel et al., 2010).
The question is in which circumstances kinship foster care is the best solution for children and their parents. In order to answer this question, it is imperative to have knowledge on the extant differences between kinship and non-kinship foster care with respect to placement characteristics (e.g., relationship between the foster child and its biological parents during placement, placement disruption, and professional support), characteristics of the biological parents (e.g., SES, mental health problems, substance abuse, child maltreatment), characteristics of the foster parents (e.g., SES, provision of safety, parenting quality) and characteristics of the foster children (e.g., gender, age, psycho-social functioning, and history of child maltreatment). In addition, it would be interesting to examine whether differences between kinship and non-kinship foster care can explain variation in psychosocial functioning (i.e., social and cognitive development as well as psychopathology) in children placed in kinship and non-kinship foster care. It should be noticed that many foster children are at risk for developmental problems, because they have parents with mental health problems and
experienced child abuse and neglect or poverty (Orme & Buehler, 2001; Orme et al., 2006). Oswald, Heil and Goldbeck (2010) found a broad spectrum of developmental problems and psychopathology in the foster children they examined, such as a high prevalence of comorbid mental disorders and externalizing and internalizing problems.
If we look at the type of foster placement, Koh and Testa (2011) and Vanschoonlandt, Vanderfaeillie, Van Holen, De Maeyer and Andries (2012) demonstrated that foster children who were placed in non-kinship foster homes were older and often had more serious psychopathology than foster children in kinship foster homes. Apparently, there is a reason to place the more seriously disturbed children in non-kinship foster care. It is plausible to suggest that in many cases risk factors for children’s psychopathology are also found in the family’s network, including relatives other than the child’s parents. There is empirical evidence showing that kinship foster parents are likely to be less educated, single, have a lower socioeconomic status and more often are from ethnic minority backgrounds compared to non-kinship foster parents (e.g., Cuddeback, 2004; Honomichl & Brooks, 2009). In addition, kinship caregivers tend to be older, because they are mainly grandparents or older aunts and uncles (Honomichl & Brooks, 2009), presenting an increased risk for somatic health problems. There is also some evidence showing that kinship foster parents show less empathy toward the foster child, have less appropriate expectations of the developmental stage of the foster child, and show less appropriate discipline strategies compared to non-kinship foster parents (Cuddeback, 2004; Honomichl & Brooks, 2009). Studies have found that kinship foster parents receive less training or professional support compared to non-kinship foster parents (Berrick et al., 1994; Cuddeback, 2004), even though Le-Prohn (1994) found that kinship foster parents felt significantly more responsible for the foster child and emphasized the high level of personal involvement compared to non-kinship foster parents. Kinship foster parents felt more need to do their best to create a healthy environment than non-kinship foster parents (Honomichl & Brooks, 2009).
Although kinship foster care has been shown to be suboptimal in some respects, it has also been found that kinship foster care provides a more stable home (i.e., less out of home disruptions) for foster children compared to non-kinship foster care (Cuddeback, 2004; Flynn, 2002; Koh & Testa, 2011; Orme et al., 2006; Strijker, Knorth, & Knot-Dickscheit, 2008; Usher, Randolph, & Gogan, 1999). The discussion is whether or not differences in problems that foster children experience are accounted for by selective placement, with the more seriously disturbed children placed in non-kinship foster care. It has also been argued that selective placement could partly explain the relatively high disruption rate or premature termination of the placement in non-kinship foster care (Oosterman et al., 2007; Strijker et al., 2008; Vanschoonlandt et al., 2012).
Cuddeback (2004) showed that the quality of contact that biological parents have with their children predicted the degree to which termination of the foster care placement was successful. Foster children in kinship foster placements had better relationships with their biological parents, and even some evidence was found that foster children who received visits from their biological parents during kinship foster placement showed more reunification (i.e., re-placement of the child with the biological parents) and had more successful outcomes, such as adoption, than children who did not have visits from their biological parents. Oosterman et al. (2007) found that foster children experienced more placement success in the foster home when they showed more ‘normal’ attachment behavior. This was especially so when foster children had experienced a positive attachment relationship with their biological parents before out of home placement. In situations where the foster children failed to attach to the foster parents, a positive and strong association was found with premature termination of the placement (Oosterman et al., 2007; Oosterman & Schuengel 2009). This means that there is a category of foster children who are at risk for serious attachment problems, which requires sufficient parenting skills of foster parents in order to facilitate attachment security in their foster children (Hamburg, 2012).
In sum, non-kinship foster care can be called a protective placement, because many necessary ingredients for a successful placement are involved. Kinship foster care appears to offer more stability and seems to provide close contacts with the biological parents, but also particular risks involved with kinship foster care placement have been documented.
The present study
The present meta-analytic study aims to map the differences between kinship and non-kinship foster care in placement characteristics, characteristics of biological parents, foster parents, and foster children as well as psychosocial functioning of the children. In addition, it was examined whether differences between kinship and non-kinship foster care can explain variation in social and cognitive development and psychopathology in foster children.
In the first part of this study, differences between kinship and non-kinship foster care will be examined in a series of meta-analyses. In the second part of this study, we conduct three meta-analyses to examine whether differences between kinship and non-kinship foster care established in the first part of the study can predict differences in social and cognitive development and psychopathology between foster children placed in kinship and non-kinship care. Effect sizes of differences between kinship and non-kinship foster care in characteristics of the biological parents, foster parents and foster children as well as placement characteristics (derived from the first part of this study) will now be used to predict differences in social and cognitive development and psychopathology between foster children placed in kinship and non-kinship fore care. Additionally, we examine publication characteristics (e.g., publication year, journal impact factor) and study characteristics (e.g., country, retrospective or prospective study design, measuring instrument, measurement moment, and response rate) that may explain differences between kinship and non-kinship foster care.
Based on the extant literature already alluded to in the introduction of our study, we hypothesize that children in non-kinship foster care have been exposed to a higher risk environment before out-of-home placement compared to children in kinship foster care. Therefore, we expect foster children in non-kinship care to show more psychosocial disturbances than foster children in kinship care. Related to this, it is plausible to suggest that premature termination of placement (i.e., disruption) is higher in non-kinship homes, and subsequently, duration of the placement longer in kinship foster homes. However, one could expect that a number of risks in the foster child’s biological family – such as low SES, unemployment, inadequate parenting, unsafe environment and ethnic minority background – are also found in kinship foster homes, predicting higher rates of disruption in kinship foster care. Notably, in the meta-analysis of disruptions in foster care by Oosterman et al. (2007), premature termination of foster care proved to be unrelated to the provision of kinship or non-kinship foster care, but this result was based on only six studies. Given that psychopathology proved to be the strongest predictor of placement breakdown in the meta-analysis by Oosterman et al. (2007), we expect that non-kinship foster care will show the highest disruption rate in the present meta-analytic study.
We expect the actual provision of professional support to be less in kinship foster care, because we assume that the less seriously disturbed children are placed in kinship foster homes. On the contrary, we expect the availability of mental health care to be less in non-kinship foster care, because Orme et al. (2006) showed that non-non-kinship foster parents are often seen as front-line service providers who do not need additional care. We expect quality of the relationship between foster children and their biological parents to be better in kinship foster care than in non-kinship foster care, and attitudes of the biological parents to placement to be more positive in kinship foster care because of the genetic connectedness (Hamilton, 1964), goodness of fit (Chess & Thomas, 1999) or valuing of the family bond (e.g., Boszormenyi-Nagy & Spark, 1973; Honomichl & Brooks, 2009). Based on family
connectedness, we assume there will be more voluntary than mandatory foster care placements, less caring allowances and less absence of biological parents in kinship foster care compared to non-kinship foster care.
Moreover, mandatory placement is more likely in non-kinship foster care because the most vulnerable children may need the highest level of protection and the most enriched environment. On the other hand, we expect that kinship foster families, due to a less enriched environment, have a reduced capacity to raise a large number of problematic foster children. Based on the expected reduced child-rearing capacity of kinship foster parents, we assume that foster children in kinship foster care are younger on average, because behavior problems in younger children may be less disruptive than in older children, putting less strain on the family. In the same vein, we hypothesize that boys will often be placed in non-kinship foster homes, because boys have been shown to experience more externalizing problem behavior than girls (Crijnen, Achenbach, & Verhulst, 1997; Stacks & Goff, 2006).
Finally, we expect kinship foster placement to end more often in guardianship and less often in reunification with the biological parents (Leslie, Landsverk, Horton, Ganger, & Newton, 2000), residential care, independency, or adoption. We expect permanency planning (i.e., adoption) for infants will be more prevalent in non-kinship foster homes (Smith, Rudolph, & Swords 2002).
Method
Selection of studies
We conducted a comprehensive search for empirical studies from 1994 until 2012 comparing kinship foster care with non-kinship foster care. The following databases were searched: Google Scholar, PsychINFO and Science Direct. The search terms were “kinship foster care”, “non-kinship foster care”, “kinship and non-kinship” and “relative and non-relative foster care”. Further reviews, books, chapters and reference sections were searched for qualifying
studies. The search yielded 62 studies of which 35 studies (N=49,986 participants) met the inclusion criteria of our meta-analytic study.
Inclusion criteria
Three inclusion criteria were used: 1) comparison of formal kinship foster care with non-kinship foster care, 2) availability of unmatched data and 3) report on demographic foster care characteristics (i.e., placement characteristics, characteristics of foster parents, biological parents and foster children) and/or psychosocial characteristics of foster children (i.e., cognitive development, social development and psychopathology). We found 8 studies that only measured demographic characteristics, 2 studies that reported psychosocial characteristics of foster children and 25 studies measuring both psychosocial characteristics of foster children and demographic characteristics. One study (Winokur, Crawford, Longobardi, & Valentine, 2009) used a matched group design controlling for demographic variables from the foster children. In this case, we only coded duration of placement, number of placements and permanency outcomes among the foster placements (i.e., reunification, adoption, and guardianship).
Analysis
In the first part of the results section a series of meta-analyses were conducted to examine differences between kinship and non-kinship foster care in characteristics of the foster parents, biological parents and foster children as well as placement characteristics. In the second part of the results section, we conducted three meta-analyses of differences in social and cognitive development and psychopathology between foster children placed in kinship and non-kinship fore care. In order to explain these differences in psychosocial problems between foster children in non-kinship and kinship foster care, we derived effect sizes for differences in characteristics from the first part of the results section to be used as moderators
in the second part of the study. Effect sizes (Cohen’s d) were based on means and standard deviations, proportions and chi-square and F test statistics.
The direction of the effect size was based on hypotheses described in the introduction of this study. We included published studies of varying quality. In order to assess quality, we examined publication characteristics (e.g., journal impact factor) and study characteristics (e.g., retrospective or prospective study design, measuring instrument, and response rate) that could moderate differences between kinship and non-kinship foster care. We used the program MLwiN (Hox, 2002) for conducting multilevel meta-analysis, testing the significance of combined effect sizes by means of a z-test. If there was heterogeneity, we conducted moderator analyses to explain differences in effect sizes. Moderators were publication characteristics (e.g., publication year, journal impact factor) and study characteristics (e.g., country, retrospective or prospective study design, measuring instrument, measurement moment, and response rate) in both parts of this meta-analytic study. In the second part of this study, we used the combined effect sizes from part one (i.e., placement characteristics, characteristics of foster parents, biological parents and foster children) to predict differences in psychosocial functioning of children between kinship and non-kinship foster care. To examine possible publication bias, we calculated the fail-safe number, which is an estimation of the number of unpublished studies, presumably reporting null results, that are needed reduce a significant overall effect size to non-significance. Publication bias is unlikely if the fail-safe number is larger than 5 times the number of studies plus 10 (Rosenthal, 1995).
Coding of characteristics
Characteristics in four domains were coded: placement characteristics, and characteristics of foster parents, biological parents and foster children.
Placement characteristics were worrisome placement (i.e., load of the placement, problematic placement and lack of satisfaction), out of home disruption, duration of
placement, safety (i.e., harsh discipline in parenting, threat of drugs and alcohol use or dealing, safety neighbourhood, overcrowded household), attitude biological parents about the placement, caring allowance, voluntary or mandatory placement, professional support, satisfaction about professional support, availability of mental health care, and children’s living arrangement following foster care (ended in residential placement, independent living, reunification with the biological parents, adoption, guardianship.
Characteristics of foster parents were age, number of foster children, educational level, ethnicity, (un)employment, rent or own house, poverty, single parenthood, parenting skills (i.e., supporting the social and emotional development of foster children, quality of social interaction, attachment, empathy, harsh discipline, using the time-out method), stimulation of contact with biological parents, and somatic illness.
Characteristics of biological parents were absence, ethnicity, poverty, single
parenthood, death and parental problems (i.e., somatic illness, substance abuse, imprisonment or convicted of offenses, mental health problems, prostitution, teenage motherhood, parents in care, that is, involvement of the child welfare system as a child).
Characteristics of foster children were child related problems before foster placement
(i.e., bullied before, and child related problems no further defined), child maltreatment before foster placement (i.e., abuse, domestic violence and neglect), placed as infant in foster care, age at placement (i.e., child < 12 year and adolescent > 12 year), gender and relationship quality between the child- and its biological parents during foster placement (i.e., relationship quality, frequency of contact). Psychosocial characteristics of the foster children were
cognitive functioning (i.e., school performance and special education), social functioning (i.e.,
social adjustment, autonomy, self efficacy, moral development, social activities, view about themselves and total competence) and psychopathology (i.e., sex problems, internalizing problems, externalizing problems, substance abuse, thought problems and attention problems).
Results
This meta-analytic study included 35 studies (N = 32.229 kinship foster care placements and
N = 23.255 non-kinship foster care placements). The sample size ranged from 22 kinship and
18 kinship foster placements (Mosek et al., 2001) until 21.914 kinship and 10.108 non-kinship foster placements (Koh & Testa, 2008). The effect sizes showed no outliers. Most of the combined effect sizes presented in Table 1 and 2 (i.e., placement characteristics, characteristics of the foster parents, biological parents and foster children) were homogeneous. We conducted moderator analyses if effect sizes were heterogeneous.
Part 1. Differences between kinship and non-kinship foster care
In the first part we examined differences between kinship and non-kinship foster care in placement characteristics, characteristics of the biological parents, foster parents, and foster children (Table 1 and 2).
Placement characteristics
Large significant differences were found for safety, professional support, and whether foster placement ended in guardianship, indicating that safety and professional support were less in kinship care than in non-kinship care and that kinship care ended more often in guardianship. Small to medium effects were found for caring allowance, satisfaction about professional support, and availability of mental health care, indicating less caring allowance and more positive attitude about professional support in kinship foster families. Small effect sizes were found for worrisome placement and the duration of placement, indicating more worrisome placements in kinship care (unexpected) and longer duration of placement in kinship care. No significant differences were found in attitude of parents about placement, absence of biological parents and the portion voluntary versus mandatory foster placements.
A large significant effect was found for adoption, which was moderated by measuring moment (z = 2.250, p < .05). If the measuring moment was later during the foster placement, the foster placement ended more often in adoption. Our hypothesis that kinship foster homes would experience less out of home disruptions compared to non-kinship foster homes was confirmed (medium effect). However, the combined effect size proved to be moderated by measuring moment (z = 4.600, p < .001). If the measuring moment was later during foster care placement, differences in out of home disruptions were smaller between kinship and non-kinship foster care.
Characteristics of foster parents
We found large differences in educational level, ethnicity and house ownership, indicating that kinship foster parents were less educated and more often rented a house compared with non-kinship foster parents. Ethnic minority status was overrepresented in kinship care if compared with non-kinship care. Small to medium differences were found the number of foster children, somatic illness and single parenthood, which indicated that kinship foster parents had less foster children, experienced more somatic illness and were more often single parents than non-kinship foster parents. Small differences were found in age and parenting skills, indicating that kinship foster parents were older and had less parenting skills. No significant differences were found in employment and contact with biological parents.
Finally, we found a large and significant effect size for poverty, indicating more poverty in kinship care, which proved to be moderated by impact factor of the journal (z = 3.037, p < .01). A higher impact factor was associated with larger differences in poverty between kinship and non-kinship foster care.
Small differences were found in ethnicity and poverty, indicating that biological parents with children in kinship foster homes mostly belonged to the ethnic minority groups and experienced less poverty compared to biological parents with children in non-kinship foster homes. No significant effect sizes were found for single biological parents, death and parental problems.
Characteristics of foster children
We found a medium difference in relationship quality between the foster child and his/her biological parents, indicating that the quality of this relationship was better in kinship foster care. There were no significant differences in child related problems and child maltreatment before foster care placement, percentage infants, age and gender.
Child maltreatment before foster placement proved to be moderated by year of study (z = 2.034, p < .01). More recent studies yielded larger differences in child maltreatment between kinship and non-kinship foster children.
Part 2. Differences in psychosocial functioning of foster children between kinship and non-kinship foster care
In the second part of the study, differences between kinship and non-kinship foster care in characteristics of the biological parents, foster parents and foster children as well as placement characteristics (derived from the first part of this study) were used to predict differences in social and cognitive development and psychopathology between foster children placed in kinship and non-kinship fore care (Table 3, 4 and 5).
Psychopathology
We found a small difference in psychopathology, which indicated that foster children in kinship care showed less psychopathology. However, the overall effect size proved to be
moderated by measuring instrument (z = 2.267, p < .01), in that the Child Behaviour Check List yielded relatively small differences in psychopathology between kinship and non-kinship foster children, and type of psychopathology, with the largest differences in total problems and thought and attention problems. Also employment of foster parents (z = 3.299, p < .001) and absence of biological parents (z = 2.615, p < .01) were significant moderators, indicating that less employment and more absence of biological parents were associated with smaller differences in psychopathology between kinship and non-kinship foster care.
Social functioning
A small significant difference was found in social functioning, indicating that foster children in kinship homes showed a better social development compared to foster children in non-kinship homes.
Cognitive functioning
No significant differences were found in cognitive functioning. However, moderator analyses showed significant differences in year of publication, measuring moment, infant, out of home disruption, duration of the foster placement, availability of mental health care and response rate (Table 3).
Cognitive functioning, proved to be moderated by year of the study (z = 2.750, p < .01). Newer studies were associated with larger differences in cognitive functioning between kinship and non-kinship foster children. A later measuring moment (z = 2.000, p < .05) was also associated with greater differences between kinship and non-kinship foster care in cognitive functioning. Less placements of infants in foster care (z = 2.406, p < .05), were associated with larger differences in cognitive functioning between kinship and non-kinship foster care. Cognitive functioning also proved to be moderated by out of home disruption (z = 2.181, p < .05) in that more out of home disruptions were associated with larger differences in
cognitive functioning between kinship and non-kinship care. Shorter duration of placement (z = 2.417, p < .05) was associated with larger differences in cognitive functioning between kinship and non-kinship care. Larger differences in the availability of mental health care (z = 3.571, p < .001) were associated with larger differences in cognitive functioning between kinship and non-kinship care. Finally, higher response rates (z = 2.667, p < .01) proved to be associated with larger differences in cognitive functioning.
Publication bias
To examine publication bias, we calculated the fail-safe number for each overall mean effect size. We found indications of possible publication bias in about 80% of the cases. No
indication of publication bias was found for out of home disruption, availability of mental health care, educational level, rent or own a house, poverty, somatic illness, psychopathology and social functioning.
Discussion
This meta-analysis examined risk and protective factors of kinship and non-kinship foster care and the degree to which these factors are associated with psychosocial functioning of the foster children. The results show that kinship foster families received hardly any professional support from the public child welfare system compared to the non-kinship foster families. Remarkably, kinship foster parents were satisfied about the professional support they received, while kinship foster care did offer a relatively unsafe environment compared to non-kinship care, with more risk factors for the child. In addition, more non-kinship foster families from lower socioeconomic status had lower educational levels and an ethnic minority background compared to non-kinship foster families. No differences in employment were found between kinship and non-kinship care. However more single parenthood in kinship care is likely to increase the risk for poverty.
Although children placed in kinship foster care are exposed to more risk factors than children placed in non-kinship care, results showed better relationships between foster children and their biological parents in kinship foster families and longer duration of placement, with less out of home disruptions. Greater stability of kinship placements may be related to feelings of more responsibility and personal involvement with the social and emotional development of the foster child in kinship foster parents (Honomichl & Brooks, 2009; Le-Prohn, 1994). Orme and Buehler (2001) found that social and emotional adjustment of children in the general population were positively related with higher levels of parental acceptance, which may partly explain the positive results for social development of children in kinship foster families in this meta-analytic study.
Possible lack of parenting skills of kinship foster parents relative to non-kinship foster parents may be considered a serious problem (Berrick & Skivenes, 2012). Moreover, kinship foster parents tend to be older, single, to suffer more often from a somatic illness, and to experience the foster placement as relatively worrisome, which additionally increases the risk for poor developmental outcomes of the foster children (see Orme & Buehler, 2001).
On the other hand, the question arises why non-kinship foster parents, who have better parenting skills, struggle with a higher number of failed foster placements (i.e., out of home disruption) given relatively more protective factors and less risk factors compared to kinship foster parents. A part of the explanation may be that non-kinship foster parents have to deal with the more seriously disturbed children, given the results showing higher rates of psychopathology compared to foster children raised by kinship foster parents. Moreover, the results showed that foster children in non-kinship care experienced more serious poverty before foster care placement (i.e., when living with their biological parents). Poverty has been shown to predict oppositional and conduct disorder (Costello, Compton, Keeler, & Angold, 2003), which may lead to a higher number of failed foster placements (Newton, Litrownik, & Landsverk 2000).
Foster children in non-kinship foster care proved to have worse contact during the foster placement with their biological parents, while absence of biological parents was associated with smaller differences in psychopathology between kinship and non-kinship foster care. McWey, Acock and Porter (2010) showed more frequent contact with the biological mother during the foster placement to be associated with lower levels of depression and less externalizing problem behavior, whereas in particular acting-out behavior can result in placement disruptions. In addition, if there are problems between the foster child and the biological parents, these problems may be less easily solved, which may be a reason for premature termination of the foster care placement. A negative spiral can occur, with premature termination of the foster placement contributing to externalizing problem behavior and disruptive mental health problems, which are the strongest predictors of placement breakdown (Newton, Litrownik, & Landsverk 2000; Strijker, Knorth, & Knot-Dicksheit 2008).
It is important to observe that problem behavior of foster children may lead to parenting stress. Foster parents, regardless of kinship or non-kinship care, have been shown to display relatively high levels of negative control and lack of parental support (Vanderfaeillie, Van Holen, Trogh, & Andries 2011; Vanderfaeillie, Van Holen, Vanschoonlandt, Robberechts, & Stroobants 2013). More negative parenting proved to be associated with higher rates of problem behavior, whereas supportive parenting was associated with lower rates of problem behavior (Vanderfaeillie et al., 2013). Moreover, this meta-analytic study confirmed that the availability of mental health care was less in non-kinship care compared to kinship foster care and associated with larger differences in cognitive functioning. Worse contact with the biological parents, parenting stress and lack of mental health care may cause that non-kinship foster families, even if they offer a healthy environment with sufficient parenting skills, struggle with higher numbers of out of home disruption compared to kinship foster families. Our meta-analytic results showed that out of home disruption and shorter
duration of the placement were associated with larger differences in cognitive functioning in foster children.
In addition, selective placement may explain less positive psychosocial outcomes in children placed in non-kinship foster care and subsequently higher disruption rates. Several risks are present in kinship foster homes and this can lead, in situations with seriously disturbed children, even to more serious damage compared to non-kinship foster homes (Newton, Litrownik, & Landsverk 2000; Strijker, Knorth, & Knot-Dicksheit 2008). Given the particular risks involved with kinship foster care, the social network may not be capable of coping with higher rates of psychopathology in the foster child.
In situations where kinship foster parents experience many risk factors that are not compensated for by protective factors, the foster child with (serious) psychopathology can be a heavy burden. However, this does not imply that kinship foster parents are not capable of raising seriously disturbed foster children. On the contrary, placement of children within their own kinship network should always be the first consideration. First, as shown in this meta-analytic study, kinship foster parents raise on average a lower number of foster children in their homes compared to non-kinship foster parents, which may reduce the burden that is put on kinship foster parents relative to non-kinship foster parents. Second, kinship foster parents should receive sufficient professional support. However, this meta-analytic study showed that kinship foster parents received less professional support compared to non-kinship foster parents. If nevertheless, a child should be placed in a non-kinship family, careful matching between the foster child and his or her foster family and sufficient availability of professional support and mental health care may become the solution for a successful placement (Brown, George, Sintzel, & Arnault, 2009; Buehler, Rhodes, Orme & Cuddeback 2006).
A possible solution to help the children with higher levels of psychopathology is combining the strengths of both kinship and non-kinship foster care. Figure 1 summarizes the
results of the present meta-analysis comparing kinship with non-kinship foster care, and figure 2 depicts the ideal situation.
Figure 1.
Figure 2.
Note depending on the level of what foster parents can handle. This may also prove insufficient. feel at home comparable background feel at home comparable background educated offer safety non-kinship more safety non-kinship feel at home comparable background *educated offer safety kinship unsafety kinship
Static and dynamic protective factors
Static and dynamic risk factors
Figure 1 shows that there is a missing circle in non-kinship foster care, indicating that foster children in non-kinship care tend to live in families that have a different cultural and socioeconomic background than their own family, and may therefore feel less at home. Meanwhile, a number of necessary preconditions for a successful foster care placement are missing in kinship foster care, such as dealing with unsafety and a number of ‘relatively static’ sociodemographic risk factors, inadequate parenting and lack of professional support (dynamic or changeable risk factors). The non-kinship foster care can be called a protective placement, because many necessary ingredients for a successful placement are involved. Still the out of home disruption is significantly higher with an estimated 28 percent more disruptions (d = .50), which shows that non-kinship foster care offers less continuity. Still non-kinship care ended more frequently in adoption and less frequently in guardianship compared with kinship care. Possibly, apart from selective placement of the less behaviorally disturbed children in kinship families, a positive relationship between the foster child and its biological parents, the family bond (e.g., Boszormenyi-Nagy & Spark, 1973; Chess & Thomas, 1999) and a comparable background (e.g., feel at home) all protect against disruption.
Figure 2 shows the combined strengths of both foster care types to create a healthy and stable home in kinship foster care as well as in non-kinship foster care. Given the lack of a comparable background in many non-kinship foster families, it seems of paramount importance that the identity of children in foster care be respected based on ethnic, religious and cultural background (article 20 of Convention on the rights of the child; Brown, George, Sintzel, & Arnault, 2009). These conditions should be included as most important 'match' in the element of comparable background. As a result conflicts can be minimized between foster parents and biological parents. The child feels safe where beliefs were shared and a greater chance of placement success follows (Brown et al., 2009).
Parents’ dedication reflects parental involvement and indicates the degree to which the parent is interested in the child’s life (Grolnick & Ryan, 1989), which is an important precondition to become a kinship or non-kinship foster parent. Berrick and Skivenes (2012) showed that a loving and nurturing parent provides high quality foster care. Kinship foster parents have been shown to feel more responsible and to be more personally involved to do their best (Honomichl & Brooks, 2009; Le- Prohn, 1994). The motivation of non-kinship foster parents to continue foster care mainly depends on professional support (e.g., confidential relationship with service agency, communication, and teamwork) and appropriate training and services (e.g., receiving necessary information to handle foster children). Even economic support was part of the motivation of non-kinship foster parents (Mullins Geiger, Hayes, & Lietz, 2013), which has links with the significant result of caring allowance in this meta-analytic study.
Cuddeback (2004) confirmed the differences in parenting skills between kinship and non-kinship parents that we found in this meta-analytic study, and showed that non-kinship foster parents compared to kinship foster parents praise their foster children more and provide greater psychical affection. Non-kinship foster parents also tend to provide more verbal and behavioral stimulation of cognitive and social development. Apparently, different forms of motivation and approach exist. By screening the foster parents based on empirically supported selection criteria (Orme et al., 2006; Vanderfaeillie, Van Holen, & Vanchoonlandt, 2013), the motivation and parenting skills (effective parenting: Berrick & Skivenes, 2012) can be measured. Clear guidelines should be used for assessing the suitability of foster parents, to provide the right match with the foster child, and to establish the degree to which professional support and care are necessary.
Training and professional support from the public child welfare system in both types of foster care seems of great importance. Recommended by Vanderfaeillie et al. (2011), specific training and support is necessary to handle foster children with problem behavior.
The present meta-analytic study showed that kinship foster families received hardly any support, whereas the actual provision of mental health care proved to be less available in non-kinship foster care. Kinship foster parents seem to be satisfied about the professional support as it is now, but with the increase of this type of foster care, we argue that the availability of sufficient professional support should be guaranteed. This applies to both kinship and non-kinship foster care, especially when the foster child suffers from high rates of psychopathology and/or serious attachment problems (Hamburg, 2012; Oosterman & Schuengel, 2009). Finally, foster parents must be able to protect the foster children from an unsafe environment, which appears to be a risk factor in kinship foster homes.
If these preconditions, as shown in Figure 2, could be accomplished for both types of foster care, kinship foster care would theoretically even be able to handle seriously disturbed children. Although kinship foster parents are probably in an earlier stage unable to provide sufficient foster care (i.e., static risk factors), which ensures that kinship foster care is different from non-kinship foster care.
Limitations
The results of this meta-analytic review should be considered in the light of some limitations. First, as the fail-safe number analyses indicated possible publication bias in the majority of comparisons between kinship and non-kinship foster care, the results of this meta-analytic study need to be interpreted with caution. It must be noted, however, that the fail-safe number, as a test of possible publication bias, is related to the magnitude of the effect sizes and the number of effect sizes (Rosenberg, 2005). As effect sizes become larger, the fail-safe number rapidly increases to exceed Rosenthal’s (1995) critical value. In addition some effect sizes (e.g., safety, worrisome placement, number of foster children) were based on only a small number of effect sizes and/or studies. Second, multiple comparisons between kinship
and non-kinship foster care were made in this meta-analytic study, unduly increasing the chance of significant findings.
The majority of studies reported ethnic minority groups to provide kinship foster care. With the increase of using the own network kinship foster care, it is likely that more (Caucasians) ethnic majority foster homes will become available. It is important to get insight into the static and dynamic risk factors in the light of selective placement (i.e., seriously disturbed foster children placed in non-kinship foster care). Followed further research is desired to improve our understanding about the differences in psychosocial problems between foster children in non-kinship and kinship foster care to provide the right match (i.e., foster parents with appropriate parenting skills for the specific (problem) behavior of the child).
Results from this meta-analytic study are important for politicians, policy makers and professionals in child welfare to frame the use of the own network interventions. Of great importance in the professional field is a careful matching process between the child and the foster family. By sharing the same values, beliefs and traditions as much as possible, the identity of children in foster care will be respected. As a result conflicts can be minimized between foster children, foster parents and biological parents. To protect foster children and provide a healthy environment, sufficient training and a confidential relationships with the agency service professionals are of great importance to continue fostering.
References
*Beeman, S.K., Kim, H. & Bullerdick, S.K. (2000). Factors affecting placement of children in kinship and nonkinship foster care. Children an d youth services review, 22, 37-54. *Benedict, M.I., Zuravin, S. & Stallings, R.Y. (1996). Adult functioning of children who
lived in kin versus nonrelative family foster homes. Child welfare, 75, 529-549. *Berrick, J.D. (1997). Assessing quality of care in kinship and foster family care. Family
relations, 46, 273-280.
*Berrick, J.D. (1998). When children cannot remain home: foster family care and kinship care. The future of children, 8, 72-87
*Berrick, J.D., Barth, R.P. & Needell, B. (1994). A comparison of kinship foster homes and foster family homes: Implications for kinship foster care as family preservation.
Children and youth services review, 16, 33-63.
Berrick, J.D. & Skivenes, M. (2012). Dimensions of high quality foster care: Parenting plus. Children and youth services review, 34, 1956-1965.
Boszormenyi-Nagy, I., & Spark, G. (1973). Invisible loyalties: Reciprocity in
intergenerational family therapy. New York: Harper & Row.
*Brooks, D. & Barth, R.P. (1998). Characteristics and outcomes of drug-exposed and non drug-exposed children in kinship and non-relative foster care. Children and youth
services review, 20, 475-501.
Brown, J.D., George, N., Sintzel, J. & St. Arnault, D. (2009). Benefits of cultural matching in foster care. Children and youth services review, 31, 1019-1024.
Buehler, C., Rhodes, K., Orme, J. & Cuddeback, G. (2006). The potential for successful family foster care: conceptualizing competency domains for foster parents. Child
welfare, 85, 523-558.
*Carpenter, S.C., Clyman, R.B., Davidson, A.J. & Steiner, J.F. (2001). The association of foster care or kinship care with adolescent sexual behaviour and first pregnancy.
Pediatrics, 108(3), E46. PubMed PMID: 11533364.
Chess, S., & Thomas, A. (1999). Goodness of fit: Clinical applications from infancy through
adult life. Routledge.
Convention of the rights of the child (1990). United nations human rights, advance access September 2013 http://www.ohchr.org/en/professionalinterest/pages/crc.aspx Costello, E.J., Compton, S.N., Keeler, G. & Angold, A. (2003). Relationships between poverty and psychopathology. American Medical Association, 290, 2023-2029.
Crampton, D. (2007). Research review: Family group decision-making: a promising practice in need of more programme theory and research. Child and family social work, 12, 202-209.
Crijnen, A.M., Achenbach, T.M. & Verhulst, F.C. (1997) Comparisons of problems reported by parents of children in 12 cultures: total problems, externalizing, and internalizing.
Journal of the American academy of child & adolescent psychiatry, 9, 1269-1277.
Cuddeback, G.S. (2004). Kinship family foster care: a methodological and substantive synthesis of research. Children and youth services review, 26, 623-639.
*Del Valle, J.F., Lopéz, M., Montserrat, C. & Bravo, A. (2009). Twenty years of foster care in Spain: profiles, patterns and outcomes. Children an youth services review, 31, 847-
853.
*Ehrle, J. & Geen, R. (2002). Kin and non-kin foster care findings from a national survey.
Children and youth services review, 24, 15-35.
*Farmer, E. (2009). How do placements in kinship care compare with those in non-kin foster care: placement patterns, progress and outcomes?. Child and family social work, 14, 331-342.
Flynn, R. (2002). Research review: Kinship foster care. Child and family social work, 7, 311- 321.
and future research. Child & Family Social Work, advance access December 2012 doi:10.1111/cfs.12049
*Gebel, T.J. (1996). Kinship care and non-relative family foster care: a comparison of caregiver attributes and attitudes. Child welfare, 75(1), 5-18.
Hamburg, P.D. (2013). Attach me if you can. Bellevile, Ontario: Essence publishing.
Hamilton, W.D. (1964). The genetical evolution of social behavior. Journal of Theoretical
Biology, 7, 1-52.
Hoeve, M., Semon Dubas, J., Eichelsheim, V.I., Van der Laan, P.H., Smeenk, W. & Gerris, J.R.M. (2009). The relationship between parenting and delinquency: A meta-analysis.
J Abnorm child psychol, 37, 749-775.
*Holtan, A., Rønning, J.A., Handegård, B. H. & Sourander, A. (2005). A comparison of mental health problems in kinship and nonkinship foster care. European child &
adolescent psychiatry, 14, 200-207.
Honomichl, R.D. & Brooks, S. (2009). Fostering family connections: pursuing permanence
for children and youth in foster care. California.
*Iglehart, A.P. (1995). Readiness for independence: comparison of foster care, kinship care, and non-foster care adolescents. Children and youth services review, 17, 417-432. *Keller, T.E., Wetherbee, K., Le Prohn, N.S., Payne, V., Sim, K. & Lamont, E.R. (2001). Competencies and problem behaviors of children in family foster care: variations by
kinship placement status and race. Children and youth services review, 23, 915-940. Koh, E. (2010). Permanency outcomes of children in kinship and non-kinship foster care:
testing the external validity of kinship effects. Children and youth services review, 32, 389-398.
*Koh, E. & Testa, M.F. (2011). Children discharged from kin and non-kin foster homes: Do the risks of foster car r-entry differ?. Children and Youth Services Review, 33, 1497-
*Koh, E. & Testa, M.F. (2008). Propensity score matching of children in kinship and nonkinship foster care: do permanency outcomes still differ?. Social work research,
32(2), 105-116.
*Leslie, L.K., Gordon, J.N., Meneken, L., Premji, K., Michelmore, K.L. & Ganger, W. (2005). The physical, developmental and mental health needs of young children in child welfare by initial placement type. Journal of developmental & behavioral
pediatrics, 26, 177-185.
*Leslie, L.K., Landsverk, J., Ezzet-Lofstrom, R., Tschann, J.M., Slymen, D.J. & Garland, A.F. (2000). Children in foster are: factors influencing outpatient mental health service use. Child abuse & neglect, 24, 465-476.
Leslie, K., Landsverk, J., Horton, M.B., Ganger, W. & Newton, R.R. (2000). The heterogeneity of children and their experiences in kinship care. Child welfare, 3, 315- 334.
*Le Prohn, N.S. (1994). The role of the kinship foster parent: A comparison of the role conceptions of relative and non-relative foster parents. Children and youth services
review, 16, 65-84.
*Metzger, J. (2008). Resiliency in children and youth in kinship care and family foster care.
Child welfare, 87, 115-140.
*Mcmillen, J.C., Zima, B.T., Scott, L.D., Auslander, W.F., Munson, M.R., Ollie, M.T. & Sptiznagel, E.L. (2005). Prevalence of psychiatric disorders among older youths in the foster care system. Journal of the American academy of child adolescent psychiatry,
44, 88- 95.
McWey, L.M., Acock, A. & Porter, B.E. (2010). The impact of continued contact with biological parents upon the mental health of children in foster care. Children and
youth services review, 32, 1338-1345.
foster care. International social work, 44(2), 149-162.
Mullins Geiger, J., Hayes, M.J. & Lietz, C.A. (2013). Should I stay or should I go? A mixed methods study examining the factors influencing foster parents decisions to continue
or discontinue providing foster care. Children and youth services review, 35, 1356- 1365.
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.
Child abuse & neglect, 24, 1363-1374.
Oosterkamp-Szwajcer, E.M., de Swart, J.J.W. & Gramberg, P.J. (2012). Succesvol op Eigen
Kracht. Een onderzoek naar de resultaten van Eigen Kracht-conferenties in Nederland van januari 2011 tot en met augustus 2012, advance access Februari 2013
http://www.canonsociaalwerk.eu/1993_FamiliesFirst/121203%20EKC%20Rapport%2 0Saxion%20Succesvol%20op%20Eigen%20Kracht%20DEF.pdf
Oosterman, M. & Schuengel, C. (2009). Gehechtheid van pleegkinderen in relatie tot gedragsproblemen en sensitiviteit van pleegouders. Kind en adolescent, 30, 95-107. Oosterman, M., Schuengel, C., Slot, N.W., Bullens, R.A.R. & Doreleijers, T.A.H. (2006). Disruptions in foster care: A review and meta-analysis. Children and youth services
review, 29, 53-76.
Orme, J.G. & Buehler, C. (2001). Foster family characteristics and behavioural and emotional problems of foster children: A narrative review. Family relations, 50, 3-15.
Orme, J.G., Buehler, C., Rhodes, K.W., Cox, M.E., McSurdy, M. & Cuddeback, G. (2006). Parental and familial characteristics used in the selection of foster families. Children
and youth services review, 28, 396-421.
Pennel, J., Edwards, M. & Burford, G. (2010). Expedited family group engagement and child permanency. Children and youth services review, 32, 1012-1019.
calculating fail-safe numbers in meta-analysis. Evolution, 59, 464-468.
Rosenthal, R. (1995). Writing meta-analytic reviews. Psychological bulletin, 118, 183-192. *Scannapieco, M., Hegar, R.L. & McAlpine, C. (1997). Kinship care and foster care: a comparison of characteristics and outcomes. Families in society, 78(5), 480-488. *Shore, N., Sim, K.E., Le Prohn, N.S. & Keller, T.E. (2002). Foster parent and teacher
assessments of youth in kinship and non-kinship foster care placements: are behaviors perceived differently across settings?. Children and youth services review, 24, 109- 134.
Smith, C. J., Rudolph, C. & Swords, P. (2002). Kinship care: issues in permanency planning. Children and youth services review, 24, 175-188.
Stacks, A. M., & Goff, J. (2006). Family correlates of internalizing and externalizing behavior among boys and girls enrolled in Head Start. Early Childhood Development and Care,
176, 67–85.
*Strijker, J., Knorth, E.J. & Knot-Dickscheit, J. (2008). Placement history of foster children: A study of placement history and outcomes in long-term family foster care. Child
welfare, 87, 107-123.
*Strijker, J., Zandberg, T. & Van der Meulen, B.F. (2003). Kinship foster care and foster care in the Netherlands. Children and youth services review, 25, 843-862.
*Sykes, J., Sinclair, I., Gibbs, I. & Wilson, K. (2002). Kinship and stranger foster carers: how do they compare?. Adoption and fostering, 26(2), 38-48.
*Tarren-Sweeney, M. & Hazell, P. (2006). Mental health of children in foster and kinship care in new south wales, Australia. Journal of paediatrics and child health, 42, 89-97. Usher, C.L., Randolph, K.A. & Gogan, H.C. (1999). Placement patterns in foster care. Social
services review, 73, 22-29.
Vanderfaeilie, J., Van Holen, F., Trogh, L. & Andries, C. (2011). The impact of foster children’s behavioural problems on Flemish foster mothers’ parenting behaviour.
Child and family social work, 17, 34-42.
Vanderfaeillie, J., Van Holen, F., & Vanschoonlandt, F. (2012). Op weg met pleegzorg:
kansen en risico’s. Leuven: Acco, pp. 250-254.
Vanderfaeillie, J., Van Holen, F., Vanschoonlandt, F., Robberrechts, M. & Stroobants, T. (2013). Children placed in long-term family foster care: A longitudinal study into
development of problem behavior and associated factors. Children and youth services
review, 35, 587-593.
*Vanholen, F., Vanderfaeilie, J. & Haarsma, Ch. (2007). Gezinsbelasting en
gedragsproblemen in netwerk- en bestandpleegzorg. Verslag van een Vlaams onderzoek. Tijdschrift voor orthopedagogiek, 46, 129-139.
*Vanschoonlandt, F., Vanderfaeillie, J., Van Holen, F., De Maeyer, S., & Andries, C. (2012). Kinship- and non-kinship foster care: Differences in contact with parents and foster
child’s mental health problems. Children and Youth Services Review, 34, 1533-1539. *Warde, B. (2008). Role perceptions of foster care in African American kinship and
nonkinship foster parents: a quantitative analysis. Journal of family social work, 11(3), 272-286.
Washington, T., Gleeson, J.P., & Rulison K.L. (2013). Competence and African children in informal kinship care: The role of family. Children and youth services review, 35, 1305-1312.
*Winokur, M.A., Crawford, G.A., Longobardi, R.C., & Valentine, D.P. (2008). Matched comparison of children in kinship care and foster care on child welfare outcomes.
Table 1
Characteristics in kinship and non-kinship foster care
Overall effects # Studies # ES Mean d
Z Heterogeneity Fail Safe
Placement Characteristics
Worrisome placement (less with kin) 3 3 -0.259 3.408*** 0 8.366 Out of home disruption (less with kin) 16 16 0.500 6.494*** 2.394* 789.417+ Duration of placement (kin longer) 12 12 0.199 1.932* 2.212* 26.640 Safety (less safety with kin) 3 3 0.939 3.726*** 1.000 9.149 Attitude parents about placement (more
positive with kin)
5 5 0.150 0.904 1.264 -4.087
Absent parents (less absent with kin) 3 3 -0.038 0.792 0.000 -0.200 Caring allowance (less payment with kin) 4 4 0.474 1.749* 1.340 1.001 Voluntary or mandatory (more voluntary
with kin)
3 3 0.186 1.120 1.059 -1.848
Professional support (less with kin) 6 6 1.373 2.814** 1.709 22.851 Satisfaction about the professional
support (positive with kin)
4 4 0.423 2.431* 1.186 1.740
Availability of mental health care (more with kin)
4 4 0.391 6.860*** 0.000 76.154+
Ended residential (less with kin) 3 3 0.573 1.628 1.177 -0.253 Ended independent (less with kin) 2 2 0.482 1.071 0.924 -1.176 Ended guardianship (more with kin) 7 7 0.790 2.792** 1.827 22.349 Ended reunification (less with kin) 7 7 0.142 0.732 1.782 -4.488 Ended adoption (less with kin) 9 9 0.562 2.882** 2.043* 39.565
Characteristics fosterparents
Age (higher age kin) 11 11 0.278 4.344*** 1.333 59.050
Number of foster children (less foster children kin)
Educational level (lower at kin) 8 8 0.813 4.811*** 1.809 74.405+ Ethnicity (less white at kin) 7 7 0.757 2.992** 1.760 23.246
Employed (less with kin) 6 6 0.203 1.103 1.542 -1.387
Rent or own house (more rent with kin)
5 5 0.841 6.893*** 1.109 129.601+
Poverty (more poverty with kin) 11 11 0.728 4.550*** 2.174* 105.943+
Single (more with kin) 12 12 0.564 4.700*** 2.113* 66.735
Parenting skills (less with kin) 7 7 0.255 1.962* 1.381 2.213 Stimulate relationship with bio-
parents (more with kin)
4 4 0.111 0.216 1.350 -3.947
Somatic illness (more ill with kin) 3 3 0.614 11.164*** 0.000 138.450+
Characteristics biological parents Ethnicity biological family (less white with kin)
18 18 0.235 2.975** 2.632** 92.820
Poverty biological family (less with kin)
5 5 0.334 2.474** 1.431 9.982
Single (more single with kin) 3 3 -0.161 0.817 1.042 -1.582
Death (more with kin) 3 3 0.300 0.993 1.005 -0.942
Parental problems (less with kin) 6 6 -0.040 0.460 1.231 -4.923
Child characteristics
Child related problems before (less with kin)
5 5 0.147 1.246 1.477 -3.406
Child maltreatement (less with kin) 10 10 -0.255 1.401 2.142* -25.893 Percentage of infants (less babies
with kin)
8 8 0.109 1.225 1.844 -6.214
Age (younger children with kin) 17 17 0.020 0.282 2.448* -16.417 Gender foster child (more girls with
kin)
# Studies = number of independent studies; # ES = number of effect sizes, Z = significance of d; mean d = mean effect size (d); heterogeneity = within class heterogeneity (Z), * p < 0.05; ** p < 0.01; *** p < 0.001. + Fail safe = no indication of publication bias.
Relationship quality child-parents (better with kin)
Table 4
Psychopathology
Moderator variables # ES β0 (SE) β1 (SE) Z Heterogeneity Fit
Employment 5 .242 (.056) .640 (.194) 3.299*** 0 7.819** Absent parents 6 .219 (.055) 3.942 (1.507) 2.615** 3.9*** 5.248* Note. Z = Significance of moderator. Fit = differnece with model without moderators.
Table 5
Cognitive functioning
Moderator variables # ES β0 (SE) β1 (SE) Z Heterogeneity Fit
Year 15 .109 (.059) .033 (.012) 2.750** 1.667 6.328*
Measuring moment Infant foster child Out of home disruption Duration of placement Availability of mental health Respons rate kinship Resons rate non-kinship
9 4 9 9 6 13 13 .025 (.067) .049 (.060) .130 (.063) .074 (.078) .048 (.044) .051 (.065) .118 (.065) .014 (.007) -.900 (.374) .530 (.243) -.539 (.223) 1.95 (.546) .008 (.003) .008 (.003) 2.000* 2.406* 2.181* 2.417* 3.571*** 2.667** 2.667** 1.278 0 .867 1.560 0 1.579 1.619 4.193* 5.215* 4.687* 5.508* 9.399*** 6.909** 5.866* Note. Z = Significance of moderator. Fit = difference with model without moderators.
