Child and Adolescent Social Work Journal

, Volume 24, Issue 6, pp 509–522

Characteristics of Four Kinship Placement Outcome Groups and Variables Associated with these Kinship Placement Outcome Groups

Authors

  • Janet Chang
    • Department of Social Work, College of Social and Behavioral SciencesCalifornia State University, San Bernardino
    • Department of Social Work, College of Social and Behavioral SciencesCalifornia State University, San Bernardino
Article

DOI: 10.1007/s10560-007-0092-y

Cite this article as:
Chang, J. & Liles, R. Child Adolesc Soc Work J (2007) 24: 509. doi:10.1007/s10560-007-0092-y

Abstract

This study investigated factors that might be associated with the disrupted kinship care placements of abused and neglected children. Data were collected from face-to-face interviews of 130 kin caregivers who were randomly selected from lists provided by two child protection agencies and divided into four different outcome groups. Findings of the study revealed that characteristics of the children such as their health status, ages, and the extent to which they were getting into “trouble” were significantly associated with placement outcomes. The caregivers’ perceptions of the quality of relationships between themselves and the children in their care and the birth parents of those children were also associated with placement outcomes. Two other factors affecting outcomes were the frequency of contact between social workers and caregivers and the extent to which services plans were discussed. Practice implications related to these findings are discussed.

Keywords

Kinship careKinship placement outcome groupDisrupted kinship careChild protective services

Introduction

The placement of abused and neglected children with relatives rather than unrelated foster parents has increased substantially since the late 1980s. Nationally, the proportion of children in kinship care versus foster care peaked at 1997 with 29% of the 557,000 dependent children in out of home placements as the result of child abuse and neglect. Since then, it has leveled off and as of 2003, about 23% out of 523,300 dependent children were in “kinship” foster placements (U.S. Dept. of Health and Human Services 2005) rather than traditional foster homes. The proportion of children in kinship foster care varies by state from 1% to 48%. For example, in California, kinship foster care placement accounted for 48% of all out-of-home placements in 1998 and was the most common type of out-of-home placement. However, the proportion of children in kinship care in Utah and New Jersey were 1% and 2%, respectively (U.S. Dept. of Health and Human Services 2000). During 2004 in San Bernardino County, California, kin care placements constituted 35% of the total 5051 out-of-home placements (San Bernardino County Human Services 2005) of dependent children.

Several factors have contributed to the increase of kinship foster care. One factor is the lack of licensed foster family homes. For example, between 1985 and 1990, the number of children in foster care increased by 47%, while the number of available foster families decreased by 27% (Spar 1993). This disparity was assumed to be responsible, in part, for a growth in the use of kinship foster care as an alternative to more traditional foster placements. Other factors were important changes in legal precedents that encouraged the use of kin as foster parents. In Miller v. Youakim (1979), the Supreme Court ruled that relatives providing kinship foster care could not be excluded from receiving federal foster care payments. Other legal rulings and the Adoption and Safe Families Act of 1997 have moved states in the direction of placing more dependent children with family members (Gleeson and Hairston 1999).

The third factor contributing to the increase in kinship foster care includes growing support for “family-centered practice” in the child welfare system with kin caregivers often given priority over non-kinship providers (Shore et al. 2002). Child welfare agencies argue that children fare better in their own families and that relatives should be given priority when children need placement. Children who become dependents of the court because of abuse and/or neglect already know and may be attached to their kin to some degree or even significantly. Relatives of children who need placement are often already attached to these children and have a sense of love or familial obligation that provides a different motivation to provide care than an unrelated foster parent might have. It is thought that kinship care arrangements provide the best opportunity for stable, permanent placements for dependent children (Dore and Kennedy 1981; Pecora et al. 1992). Furthermore, placing dependent children with kin is also important because it allows children to more easily remain grounded in their own cultural heritage and community.

In recent years extensive research has been conducted on kinship care, however it has largely focused on the differences and similarities between the characteristics of kinship foster caregivers and non-kinship foster caregivers and their dependent children. Only a few studies have been conducted on the outcomes of kinship care placements. Thus, it is not well known why certain kinship placements succeed or fail and what factors are associated with differential kinship care placement outcomes. This study attempted to fill some of the gaps in kinship care research and attempted to determine what factors might be responsible for disrupted kinship placements.

Kinship Care Placement Outcomes

Kinship care placements appear to be more stable than non-kin care placements. Children with kinship care as a primary placement seem to have significantly more stable placement experiences than children placed with non-kin (e.g., Berrick et al. 1998; Dubowitz and Feigelman 1993; Iglehart 1994). According to Berrick et al. (1998), an initial placement in a kinship home reduced the number of subsequent placements. On the other hand, children in kinship care tend to remain in out-of-home care longer than those in non-kin foster care (Scannapieco et al. 1997; Courtney 1994). Pecora et al. (1999) found that youth in kinship care remained in out-of-home care for longer periods of time and reunified at a lesser rate than youth in non-kinship care. Courtney (1994) also suggested that initial foster care placement with kin was associated with a significantly slower rate of family reunification during first several months of foster care. However, Berrick (1997) indicated that the proportion of children ultimately reunified from kin and non-kin care was roughly similar for both groups after about four years.

Kinship care placements may not be as “safe” as non- kin care placements. Berrick (1997) assessed the quality of care in a 29 kin and 33 non-kin foster parents in a single county in California. Results indicated that family relations between children and their caregivers were similar for kin and non-kin, but non-kin homes were “safer” on number of measures including incidents of violence or threatened violence, problems with drugs or alcohol, and neighborhood conditions than kin homes. However, the study found the children in kinship homes were less emotionally traumatized by their past experiences than children in non-kin homes. The study also reported that kinship care promoted and maintained close relationships between foster children and their birth parents.

Factors contributing to the stability and duration of kinship foster care are not well known since relatively few studies have been conducted on the topic. Testa and Slack (2002) found in their five-year longitudinal study of 983 children in Cook County, Illinois that reunification and stability of kinship care were indeed dependent upon reciprocity, payment, empathy, and sense of duty. Reunification was more likely to occur if caregivers perceived birth parents as regularly visiting and working cooperatively. Also, children whose caregivers experienced a reduction in payment amounts were more likely both to be reunified than children with families who retained full foster care subsidy.

Permanency outcome for kinship foster care has been examined in terms of disruption rates of children placed with kin. Terling-Watt (2001) examined the disruption rate for the 875 children in Houston, Texas and found disruption rates of almost 50% for children placed in kinship care that did not become reunified with their families. A subsequent complementary qualitative study of 30 kin caregivers showed that the most common problem was the continued influence of the children’s birth parents and the kin caregivers’ inability or unwillingness to maintain the boundaries between birthparents and children as established by Child Protective Services. The second most common problem found was the difficulty relatives had understanding or believing the limitations of the parents and the dangers they represented to the children. Other reasons for the high disruption rates include adolescents’ difficulties adapting to structured placements, relatives’ uninformed and unrealistic ideas about their ability to address the problems of the children in their care, the age or general health of the relative, and relatives’ resistance to receiving supportive services.

Methods

Participants in this study were randomly selected from lists of relative caregivers (kin), provided by child protection agencies, whose dependent children were first placed with them during the periods of July 2002 through December 2002 in two large counties in California. Kin caregivers on the lists were divided into one of four outcome groups: (1) children who had already been reunified with their birth parents (reunification group); (2) children who were still moving toward reunification with their birth parents (reunification in progress group); (3) children who continued to be placed with kin after reunification with birth parents had failed (continued kinship placement group); (4) children whose kin care placement was discontinued prematurely and who were subsequently placed with non-relatives (disrupted kinship placement group). Data were collected through face-to-face interviews using structured interview schedules with a total of 130 kin caregivers from May 2004 through October 2005.

The sampling criteria used for the study were that the ages of the dependent children were 18 years or less, and both kin caretakers and children must be under the supervision of either the San Bernardino or Riverside County Child Protective Services (CPS) agencies. Caregivers also had to fit the current legal definition of “kin” in terms of child welfare placement. The final sample consisted of 31 selected from 184 caregivers of the reunified group, 30 from 181 caregivers of the reunification in progress group, 40 caregivers from 84 caregivers of the continued placement group, and 29 from 148 caregivers of the disrupted group with a total of 130 kin caregivers. Nearly 24% of the caregivers’ phone numbers from the lists of the four groups were either disconnected or incorrect and 8% refused to participate in the study. The study was approved by the Institutional Review Board of California State University, San Bernardino in 2003.

Structured interview questions included items assessing kin caregiver characteristics, dependent children’s characteristics, kin caregivers’ relationships with dependent children and birth parents, and kin caregivers’ relationship with social workers. Questions in the structured interview schedule were developed by the researchers based on the previous literature (e.g., Berrick 1997, and others cited above) related to kinship care, input from a graduate research assistant who had compiled a list of concerns expressed directly by kin caregivers in an agency meeting, the researchers’ practice experience in child welfare, consultation from a child welfare practitioner who was also conversant with the kinship literature, and the key research question of the study.

Questions related to kin caretaker characteristics included their relationships with the dependent children placed in their care, age, gender, ethnicity, income, health status, education, marital status, employment status, number of children in their care, self-reported overall health status, presence of health problems that might limit their ability to care for children, and use of alcohol. The dependent children’s characteristics included age, gender, ethnicity, health status, behavioral problems, special needs, and academic performance. Kin caregivers’ relationships with the dependent children in their care were assessed by measuring the frequency of contacts and perceived closeness with the children prior to placement. The quality of relationships with birth parents was examined by assessing the frequency of contact and perceived closeness between the birth parent(s) and kin caregiver during the child’s placement. Kin caregivers’ relationship with social workers was assessed by measuring the caregiver’s recollection on frequency of contact with social workers and the frequency with which they discussed case plans.

Quantitative data collected were entered into the Statistical Package for the Social Sciences (SPSS) and descriptive statistics including frequency distribution and measures of central tendency (e.g., mean, median) were used to describe the characteristics of the variables. Chi-square tests were used to assess the associations of variables with the four outcome groups.

Findings

Characteristics of Caregivers

Over 93% of the relative caregivers interviewed were female. In terms of caregivers’ filial relationships to the children in their care, approximately 44% were grandmothers, 33% were aunts, 7% were great aunts, and 5% were great grandmothers. While over half of the caregivers for both the reunification in progress group and the continued kinship placement group were grandmothers, over 62% of the caregivers for the disrupted group were aunts, uncles, or great aunts. Over a third of the respondents were white (35%), 28% were Hispanic/Latino, 25% were African American, and 13% were of other ethnicities. The age of the respondents ranged from 18 to 77 years (Mean = 48 years, SD = 13 years). Caregivers from the disrupted group tended to be younger than those from the other three outcome groups. However, the difference was not statistically significant.

The majority of the caregivers were married (55%), 18% were either divorced or separated, 12% were widowed, 10% were never married, and 6% were cohabitating. With respect to education, nearly half of the caregivers in the study completed high school, 21% had some college education, 20% had less than high school education, and 10% completed college or graduate school. Over 54% of the caregivers in the study reported to have a monthly income of less than $3,000, while 37% had monthly incomes of between $3,001 and $6,000, and 9% had monthly incomes of $6,001 or more. In terms of employment status, 59% were employed, 26% were unemployed, and 16% were retired. There were no significant differences in marital status, education, income, and employment status among the four outcome groups. Kin caregivers in the study reported to either have at the time of the interview or at the time of the previous placement to have had an average of 2.3 children in their care; 40% reported taking care of or having taken care of one child, 24%, two children, 22%, three children, and 15%, four children or more. There was no significant group difference in the numbers of children in current placements or during previous placements.

The great majority of the caregivers (83%) in the study reported their health status to be either “very good,” or “good.” Furthermore, 90% of the caregivers indicated that they did not have any health conditions limiting the ability to care for their dependent children. Again, there was no statistically significant difference in health status or the presence of health problems limiting their care among the four outcome groups. Twenty percent of the caregivers in the study reported using alcohol during the time when the children had originally placed in their care.

Characteristics of Dependent Children

There were a total of 291 children represented in the sample. Of these children, 75 children had already reunified with their birth parents, 65 children were currently placed in kinship care with reunification pending, 90 children were currently placed in kinship care after reunification efforts failed, and 61 children were placed in non-kinship foster care or in group homes after the failure of placement with the kin caregiver who were interviewed as part of this study. Gender of the children was almost evenly distributed with 49% female and 51% male in the study. The age of the dependent children ranged from one year to 21 years with an average of 8.6 years (SD = 5.3 years). There was a statistically significant age difference among the outcome groups (χ2 = 22.2, df = 9, p < 0.01). Older children (11 years or older) were overrepresented in both the disrupted and the continued placement groups (39% and 34%, respectively) compared to the reunified and the reunification in progress groups (16% and 13%, respectively; see Table 1).
Table 1

Demographic characteristics of dependent children

Variable

Reunified group n(%)

In current kin care with reunification pending n(%)

In current kin care after reunification failed n(%)

Disrupted group n(%)

Total

χ2

Gender of child

     

1.7

    Female

41(54.7)

32(40.2)

40(44.4)

29(45.5)

142(48.8)

 

    Male

34(45.3)

33(50.8)

50(55.6)

32(54.5)

149(51.2)

 

Age

     

22.2**

    1–5 years

10(32.3)

13(41.9)

36(40.9)

18(35.3)

77(38.3)

 

    6–10 years

16(51.6)

14(45.2)

22(25.0)

13(25.5)

65(32.3)

 

    11–15 years

5(16.1)

4(12.9)

18(20.5)

17(33.3)

44(21.9)

 

    16 or older

0(0.0)

0(0.0)

12(13.6)

3(5.9)

15(7.5)

 

Ethnicity

     

49.8***

    African American

26(34.7)

14(21.5)

21(23.3)

9(14.8)

70(24.1)

 

    Hispanic/Latino

12(16.0)

30(46.2)

17(18.9)

20(32.8)

79(27.1)

 

    White/Caucasian

19(25.3)

4(6.2)

14(15.6)

19(31.1)

56(19.2)

 

    Mixed

15(20.0)

16(24.6)

23(25.6)

10(16.4)

64(22.0)

 

    Other

3(4.0)

1(1.5)

15(16.7)

3(4.9)

22(7.6)

 

Health status

     

30.7**

    Very good

52(69.3)

47(72.3)

62(68.9)

27(44.3)

188(64.6)

 

    Good

15(20.0)

15(23.1)

25(27.8)

18(29.5)

73(25.1)

 

    Fair

4(5.3)

1(1.5)

0(0.0)

7(11.5)

14(4.8)

 

    Poor

1(1.3)

2(3.1)

0(0.0)

3(4.9)

6(2.1)

 

    Very poor

3(4.0)

0(0.0)

1(1.1)

6(9.8)

10(3.4)

 

Academic performance

     

10.9

    Above average

13(27.1)

15(36.6)

16(25.0)

11(26.8)

55(28.4)

 

    Average

22(45.8)

17(41.5)

35(54.7)

12(29.3)

86(44.3)

 

    Below average

13(27.1)

9(22.0)

13(20.3)

18(43.9)

53(27.3)

 

Got into trouble at school or home

     

12.3**

    Yes

13(21.3)

7(14.9)

21(23.6)

24(42.9)

72(25.7)

 

    No

48(78.7)

40(85.1)

68(74.4)

39(57.1)

197(74.3)

 

Child with special needs

     

1.5

    Yes

16(22.5)

16(30.8)

23(25.6)

17(40.6)

72(26.8)

 

    No

55(77.5)

36(69.2)

67(74.4)

39(59.4)

197(73.2)

 

*p < .05, **p < .01, ***p < .001

In terms of the dependent children’s ethnicity, 27% were Hispanic/Latino, 24% were African American, 22% were mixed, 19% were white, and 8% were other. There was a statistically significant association between the children’s ethnicity and the outcome groups (χ2 = 49.8, df = 12, p < 0.001). While over 46% of the children in the reunification in progress group were Hispanic/Latino, nearly half of the children (49%) in the continued placement group were either African American or mixed, however the great majority of the children in the disrupted group (64%) were either Hispanic/Latino or white.

In terms of health status of the children, nearly two thirds of the children (64%) were reported to be in “very good” health, while 25% indicated to be in “good” health and 11% reported to be in either “fair,” “poor,” or “very poor” health. There was a statistically significant difference in health status of the children among the outcome groups (χ2 = 30.7, df = 12, p < 0.01). Approximately 70% of the children in three of the groups, the reunified, the reunification in progress, and the continued placement groups, were reported to be in “very good” health, while only 44% of the children in the disrupted group were indicated to be in such a health status.

In terms of academic performance, 44% of the children were reported by their caregivers to be doing “average,” 28% were reported to be “above average,” and 27% were indicated to be “below average” overall. Kin caregivers were asked about whether the children in their care were or had been “getting into trouble” either at school or home. The great majority of the children (73%) were reported to not have gotten into trouble, while 27% were reported to have been in trouble at home or in school. Interestingly, nearly 43% of the children in the disrupted group were reported as having gotten into trouble, while 24% or less of the children from the other three groups were reported to have gotten into trouble. This was a statistically significant difference (χ2 = 12.3, df = 3, p < 0.01). The great majority of the children in the study were reported not to have any special needs (73%). However, children in the disrupted group tended to have more special needs compared to the other three groups, but the difference was not statistically significant.

Caregivers’ Relationship and Frequency of Contact with Birth Parents

As shown in Table 2, there was a significant association between caregiver’s perceived closeness and frequency of contact with birth parents and differential placement outcomes. For example, caregivers from the reunified and reunification in progress groups were likely to have had a more positive relationship with the dependent’s child’s father than caregivers from the continued placement group and the disrupted group while the children were still in their care (χ2 = 30.9, df = 15, p < 0.01). In terms of caregivers’ frequency of contact with the birth father, again, caregivers from the reunified and reunification in progress groups tended to have more frequent contacts with the child’s birth father than those from the disrupted and continued placement groups (χ2 = 60.7, df = 15, p < 0.001). The great majority of the caregivers from the discontinued group (74%) reported to have no contact with the child’s birth father during placement.
Table 2

Caregiver’s relationship and frequency of contact with birth parents

Variable

Reunified group n(%)

In Current kin care with reunification pending n(%)

In current kin care after reunification failed n(%)

Disrupted group n(%)

Total

χ2

Relationship with child’s birth father

     

30.9**

    Very good

16(21.3)

13(20.0)

10(11.2)

9(15.0)

48(16.6)

 

    Good

15(20.0)

22(33.8)

14(15.7)

14(23.3)

65(22.5)

 

    Fair

20(26.7)

8(12.3)

15(16.9)

4(6.7)

47(16.3)

 

    Poor

6(8.0)

2(3.1)

11(12.4)

5(8.3)

24(8.3)

 

    Very poor

0(0.0)

2(3.1)

5(5.6)

2(3.3)

9(3.1)

 

    No relationship

18(24.0)

18(27.7)

34(38.2)

26(43.3)

96(33.2)

 

During placement, Frequency of contact With birth father

     

60.7***

    Never

23(31.9)

16(28.1)

49(43.6)

42(73.7)

130(47.4)

 

    Once per year or less

1(1.4)

6(10.5)

6(6.8)

0(0.0)

13(4.7)

 

    A few times per year

3(4.2)

1(1.8)

7(8.0)

4(7.0)

15(5.5)

 

    Once a month

12(16.7)

5(8.8)

10(11.4)

3(5.4)

30(10.9)

 

    Once a week

12(16.7)

17(29.8)

8(9.1)

4(7.0)

41(15.0)

 

    More than once a week

21(29.2)

12(21.1)

8(9.1)

4(7.0)

45(16.4)

 

Relationship with Child’s birth mother

     

40.4***

    Very good

14(18.7)

7(11.1)

11(12.2)

11(18.0)

43(14.9)

 

    Good

21(28.0)

14(22.2)

39(43.3)

11(18.0)

85(29.4)

 

    Fair

23(30.7)

22(34.5)

18(20.0)

8(13.1)

71(24.6)

 

    Poor

5(6.7)

6(9.5)

3(3.3)

12(19.7)

26(9.0)

 

    Very poor

10(13.3)

5(7.9)

10(11.1)

12(19.7)

37(12.8)

 

    No relationship

2(2.7)

9(14.3)

9(10.0)

7(11.5)

27(9.3)

 

During placement, Frequency of contact With birth father

     

42.9***

    Never

7(9.7)

12(21.8)

10(11.1)

15(25.9)

44(16.0)

 

    Once per year or less

0(0.0)

2(3.6)

1(1.1)

5(8.6)

8(2.9)

 

    A few time per year

3(4.2)

2(3.6)

12(13.3)

7(12.1)

24(8.7)

 

    Once a month

13(18.1)

8(14.5)

28(31.1)

13(22.4)

62(22.5)

 

    Once a week

24(33.3)

20(36.4)

18(20.0)

12(20.7)

74(26.9)

 

    More than one a week

25(34.7)

11(20.0)

21(23.3)

6(10.3)

63(22.9)

 

*p < .05, **p < .01, ***p < .001

The caregivers’ relationships with the children’s birth mothers appeared to be somewhat different from their relationships with birth fathers. Caregivers from the continued placement group tended to have more positive relationship with the child’s birth mother in comparison to those from the other outcome groups (χ2 = 40.4, df = 15, p < 0.001). However, in terms of their frequency of contact with the birth mother, caregivers from the reunified and reunification in progress groups tended to have more frequent contact than those from the continued and disrupted groups (χ2 = 42.9, df = 15, p < 0.001).

Caregivers’ Relationship and Frequency of Contact with Child Prior to Placement

Prior to placement caregivers from the reunified, reunification in progress, and continued placement groups were likely to have had more positive relationships with their dependent children compared to those from the disrupted group(χ2 = 30.0, df = 9, p < 0.01). Prior to placement, 87% of the caregivers from reunified group, 83% from the reunification in progress group, 82% from the continued placement group reported their relationship with the child was either “very good” or “good,” while only 70% of the caregivers from the disrupted group indicated similar pre-placement relationships (see Table 3).
Table 3

Caregiver’s relationship and frequency of contact with child prior to placement

Variable

Reunified group n(%)

In current kin care with reunification pending n(%)

In current kin care after reunification failed n(%)

Disrupted group n(%)

Total

χ2

Prior to placement, frequency of contact with child

     

22.6

    Never

3(4.1)

1(1.6)

2(2.3)

6(9.8)

12(4.2)

 

    Once a year or less

4(5.4)

7(10.9)

3(3.4)

5(8.2)

19(6.6)

 

    A few times a year

9(12.2)

7(10.9)

7(8.0)

12(19.7)

35(12.2)

 

    Once a month

8(10.8)

6(9.4)

6(6.8)

6(9.8)

26(9.1)

 

    Once a week

14(18.9)

20(31.3)

23(26.1)

11(18.0)

38(23.7)

 

    More than once a week 36(48.6)

23(35.9)

47(53.4)

21(34.4)

127(44.3)

  

Prior to placement, Relationship with child

     

30.0**

    Very good

39(52.0)

43(66.2)

63(70.0)

26(42.6)

171(58.8)

 

    Good

28(37.3)

11(16.9)

11(12.2)

17(27.9)

67(23.0)

 

    Fair

5(6.7)

4(6.2)

10(11.1)

12(19.7)

31(10.6)

 

    No relationship

3(4.0)

7(10.8)

6(6.7)

6(9.8)

22(7.6)

 

Making the decision to care For child

     

30.8***

    Very difficult

0(0.0)

0(0.0)

3(3.4)

2(3.3)

5(1.8)

 

    Difficult

10(13.3)

2(3.4)

22(24.7)

22(36.1)

56(19.7)

 

    Easy

23(30.7)

24(40.7)

24(27.0)

17(27.9)

88(31.0)

 

    Very easy

42(56.0)

33(55.9)

40(44.7)

20(32.8)

135(47.5)

 

*p < .05, ** p< .01, ***p < .001

Also prior to placement, caregivers from three outcome groups, reunified, reunification in progress, and continued placement, were likely to have had more frequent contacts with their dependent children than those from the disrupted group. For example, 68% of the caregivers from the reunified group, 66% of the caregivers from the reunification in progress group, and 80% of the caregivers from the continued placement group reported to have contact at least once a week as opposed to 52% of the caregivers from the disrupted group.

Caregivers from two outcome groups (reunified and reunification in progress groups) tended to report that their decisions to take in the children was easier than those from the continued and disrupted placement groups. For example, 87% of the caregivers from the reunified group and 97% of the caregiver from the reunification in progress groups reported their decision to care for child was either “very easy,” or “easy” compared to 71% of the caregivers from the continued placement group and only 60% of the caregivers from the disrupted group (χ2 = 30.8, df = 9, p < 0.001).

Caregivers’ Frequency and Type of Contact with Social Workers

As seen in Table 4, caregivers from the disrupted group reported having had less frequent contact with their social workers than those from the other three outcome groups (χ2 = 28.7, df = 9, p < 0.01). In answering a question about whether social workers had discussed a services plan with the caregiver, 69% of the caregivers responded in an affirmative way, while 31% responded in a negative way. Again, caregivers from the disrupted group were likely to report that their social workers had not discussed their dependent children’s service plans with them as compared to those from the other three outcome groups (χ2 = 10.9, df = 3, p < 0.05). Approximately 53% of the caregivers from the disrupted group indicated that their social workers had discussed a services plan with them as compared to 72% from the reunification group, 79% from the reunification in progress group, and 72% from the continued placement group. The great majority of the caregivers (88%) reported they had never received any foster parent or related training. In their answers to a question about whether they wanted to receive some “foster” parenting training from social service agencies, 74% of the caregivers responded “no, while 26% answered “yes.” Interestingly, nearly half of the caregivers (49%) from the disrupted group indicated that they wanted to receive some type of foster parenting training, while far fewer caregivers from the other three groups indicated they wanted to receive such training (χ2 = 32.0, df = 3, p < 0.001).
Table 4

Caregiver’s contact with social workers

Variable

Reunified group n(%)

In current kin care with reunification pending n(%)

In current kin care after reunification failed n(%)

Disrupted group n(%)

Total

χ2

Frequency of contact with Social workers

     

28.7**

    Less than once a month

6(8.0)

14(21.5)

9(10.0)

8(13.1)

37(12.7)

 

    Once a month

41(54.7)

34(52.3)

65(72.2)

27(44.3)

167(57.4)

 

    More than once a week

19(25.3)

15(23.1)

11(12.2)

13(21.3)

58(19.9)

 

    Other

9(12.0)

2(3.1)

5(5.6)

13(21.3)

29(10.0)

 

Social worker discuss service plan With caregiver

     

10.9*

    Yes

54(72.0)

51(78.5)

64(71.9)

31(52.5)

200(69.4)

 

    No

21(28.0)

14(21.5)

25(28.1)

28(47.5)

88(30.6)

 

Receive foster parenting training

     

4.8

    Yes

9(12.0)

4(6.2)

16(17.8)

7(11.5)

36(12.4)

 

    No

66(88.0)

14(21.5)

74(82.2)

54(88.5)

855(87.6)

 

Want to receive foster parenting Training from Social Services agency

     

32.0***

    Yes

4(5.6)

17(26.2)

24(27.3)

29(49.2)

74(26.1)

 

    No

68(94.4)

48(73.8)

64(72.7)

30(50.8)

210(73.9)

 

*p < .05, **p < .01, ***p < .001

Discussion

Demographic characteristics of kin caregivers appear to be similar across the four placement outcome groups. Caregivers in the sample were predominantly female (93%) and the majority of the caregivers are either grandmothers or aunts to their dependent children. It is interesting to note that grandmothers were overrepresented in the continued group (58%) and reunification in progress group (53%), while aunts or great aunts were overrepresented in the disrupted group (63%) and reunified group (57%). It appears that children who are placed in a grandmother’s care are more likely to stay longer than those who are placed in an aunt’s or great aunt’s care.

No single ethnicity for caregivers or children emerged as the major group in this study. It appears that the ethnic makeup of kin caregivers and their dependent children in the sample is representative of the population of children in kin care and their caregivers under the supervision of the two child protective services agencies (Child Welfare Research Center 2006). Some additional research attention may need to be given to the group of “mixed ethnicity” children and the group of kin caregivers and children who do not share the same ethnicity.

The study indicated that kin caregivers’ perceived quality of relationship and frequency of contact with birth parents were significantly associated with differential placement outcomes. Furthermore, the study found that the caregiver’s perceived closeness to their dependent children prior to placement were associated with various placement group outcomes. Particularly, caregivers in the disrupted group had less positive relationships and less frequent contact prior to placement with the children who were later placed with them when compared to those from the other three outcome groups. Again, this suggests that in an effort to increase the probability of successful and stable placements, it seems critical for child welfare workers to carefully assess caregivers’ levels of attachment and/or the quality of the relationships with the children who might be placed in their care before placement decisions are made and to perhaps develop programs or services to increase attachment between children and their relative caregivers once a placement is made.

While the characteristics of kin caregivers do not appear to be different across the four outcome groups, children from the disrupted group appear to be dissimilar from the other dependent children in several respects. In terms of health status, children in the disrupted group were likely to be less healthy (as reported by their caregivers) than those from the other three groups. Furthermore, children from the disrupted group were reported to have gotten into trouble either at school or home more frequently than those from the other groups. In addition, children from the disrupted group were likely to be older than their counterparts from the other three groups. It is possible that the children’s health status, emotional and behavior problems, as well as the fact that they, as a group, were somewhat older played critical roles in contributing to the prematurely discontinued placements represented in this study. These risk factors need to be carefully assessed prior to placement and subsequent appropriate interventions or services for both children and their caregivers need to be planned and provided to lower the number and rate of disrupted placements.

Another finding that deserves attention is that aunts or great aunts are overrepresented in the disrupted group. It is very likely that aunts and/or great aunts have needs for professional support and services that are somewhat different from the needs of grandparents in kinship care arrangements. One need that was verbalized by this group was the need for more foster or kinship care related parent training. County child protective services agencies and/or contract agencies might consider responding to this need by developing and offering specialized “auntie” kinship care training groups and/or ongoing support groups.

Another major finding in this study was that caretakers in the disrupted group reported fewer contacts with agency social workers and also reported fewer instances in which case services plans were discussed with social workers. If this finding is replicated in future studies it suggests that practice with kinship caregivers and their dependent children could be improved by using some or all of the above findings in pre-placement assessments, increasing the numbers of contacts between social workers and kin caregivers in which some constellation of the above risk factors are found, developing case plans that may include special services such as additional caregiver training, and increasing the discussions of case plans between social workers and caregivers.

Limitations

One of the limitations of this study is the small sample size, which limits the validity and generalizability of the findings. Although a systematic random sampling was attempted in selecting study participants from the lists obtained from the two protective services agencies, quite a number of caregivers on the lists were excluded due to outdated or incorrect contact information. The fact that so many potential respondents were unavailable might be a source of sampling error and compromise the generalizability of the findings. It may be that kin caretakers who move more often or who do not have a current telephone number may have different characteristics than those that were contacted for the study. Thus, findings from this study must be interpreted with some caution.

Another limitation of this study is that the data depended on the self reports of kin caregivers. It may be that some of the differences found in the prematurely discontinued or “disrupted” group are related to the feelings experienced by kin care providers (such as frustration at having “failed”) rather than the result of objective phenomenon (such as less contact with social workers). For example, it would be interesting to conduct a study that used child welfare services case management system (CWS/CMS) data as another source of data in an effort to determine if a large group of “disrupted” kin care providers actually had less contact with child welfare services social workers than other kin care providers.

Acknowledgment

This study was funded by the California Social Work Education Center (CalSWEC).

Copyright information

© Springer Science+Business Media, LLC 2007