Privatization, racial disproportionality and disparity in child welfare: Outcomes for foster children of color

Children and Youth Services Review 99 (2019) 125–131

Contents lists available at ScienceDirect

Children and Youth Services Review journal homepage: www.elsevier.com/locate/childyouth

Privatization, racial disproportionality and disparity in child welfare: Outcomes for foster children of color☆

T

⁎

Kimberly Y. Huggins-Hoyta, , Harold E. Briggsb, Orion Mowbrayb, Junior Lloyd Allenc a

Andrew Young School of Policy Studies, Georgia State University, 14 Marietta Street, NW, Atlanta, GA 30303, United States School of Social Work, University of Georgia, United States c School of Social Work, Wayne State University, United States b

A B S T R A C T

Purpose: This study examined the effect of privatization policy on the issue of racial disparity in the child welfare system. Method: Specific outcomes for N = 118,761 foster children across 10 states were compared to determine if the state system type [privatized vs. non-privatized] had any influence on disparities in outcomes by race. Results: A main effect emerged for race and in the interactions between system type and race. Discussion: The extent of disparity for children of color in foster care was supported, but for some outcomes, both groups fared better in privatized systems.

1. Introduction Over the past two decades, the U.S. Child Welfare System has struggled to achieve safety, permanency, and well-being outcomes for children and youth in foster care. Amidst a fluctuating foster care population while child welfare budgets have steadily declined, agencies have also been encumbered by the issues of disproportionate representation (disproportionality) and disparate treatment (disparity) of children and youth of color in care (Briggs, 2011; Harris, 2014; Hill, 2007; Roberts, 2008). In fact, according to the AFCARS Preliminary FY2015 Report (2016), the foster care population increased from 397,301 in 2012 to an estimated 427,910 in 2015; and for foster care services specifically, states reported allocating $3.2 billion of federal Title IV-E and $576.8 million in federal Title IV-B funds in 2014, a 7% and 6% decrease from 2012, respectively (Rosinsky & Connelly, 2016). Additionally, according to the 2013 report on Disproportionality Rates for Child of Color in Foster Care (Summers, 2015), African American/ Black and American Indian/Alaskan Native were 1.8 and 2.5 times more likely, respectively, to be represented in foster care than their Caucasian/White counterparts who were underrepresented along with Hispanic/Latino, Asian, and Hawaiian/Other Pacific Islander children. Consequently, child welfare stakeholders (advocates, judges, politicians, and the public) have called for reform of the system to improve effectiveness by increasing efficiency, accountability, cost-savings, as well as, reducing inequalities unique to the children and youth of color

in care. One such controversial yet popular reform strategy has been the implementation of privatization policy, which involves public child welfare agencies transferring or outsourcing the responsibility of delivering foster care case management services to the private sector, either partially or in full. While many states initiated efforts to relinquish case planning and decision-making authority to the private sector over the past decade (Meezan & McBeath, 2011), implementation actually declined from a reported 47 privatization initiatives in 29 states between 1997 and 2008 (Collins-Camargo, McBeath, & Ensign, 2011; Flaherty, Collins-Camargo, & Lee, 2008; Unruh & Hodgkin, 2004) to only 6 states with full privatized and 14 with partially privatized systems by 2012 (Coles, 2015). However, on one hand, there have been numerous empirical efforts to ascertain the effectiveness of privatization on outcomes for children and youth in foster care; but, on the other hand, examinations of the extent this policy initiative has impacted the issues of disproportionality and disparity remain absent from the literature. Thus, this study aimed to fill this void. 1.1. Racial disproportionality and disparity in foster care Historically, African American and American Indian/Alaskan Native children have been overrepresented in the foster care system, evidenced by both making up 24% and 2% of foster children but only 14% and 1% of the total child population, respectively (U.S. Department of Health and Human Services, 2015a, 2015b). Additionally, these children of

☆ ⁎

Special thanks to Dr. Michael J. Holosko for his guidance in the development of this research project. Corresponding author. E-mail address: [email protected] (K.Y. Huggins-Hoyt).

https://doi.org/10.1016/j.childyouth.2019.01.041 Received 12 November 2018; Received in revised form 29 January 2019; Accepted 30 January 2019 Available online 31 January 2019 0190-7409/ © 2019 Elsevier Ltd. All rights reserved.

Children and Youth Services Review 99 (2019) 125–131

K.Y. Huggins-Hoyt et al.

(privatized, partially privatized, non-privatized) on benchmark measures of effectiveness and efficiency. Unfortunately, devoid of the literature on privatization in child welfare, are explorations into its relationship with racial disproportionality and disparity. However, the Coles (2015) study opened the door for this important research focus by also finding that being African American was correlated with longer days in care, more placements, more removals, and previous stays in care; and in partially privatized states, African American children were more likely to be placed with non-African American foster families and age-out (turning age 18) of foster care. This study continues this exploration by aiming to determine if children of color fare better or worse than their counterparts in privatized vs public systems. The overarching goal is to inform existing child welfare policies and practices intended to protect and promote the overall welfare of our children. Specifically, the aim is to contribute to the knowledge-base on how the neoliberal reform initiative of privatization effects outcomes for children of color in the U.S. foster care system compared to their counterparts. Therefore, this study sought to ascertain the relationship between privatization, disproportionality, disparity in terms of these outcomes, across selected states by exploring this research question: Are there significant statewide differences between privatized and nonprivatized foster care systems in terms of specific outcome indicators for disproportionally represented children of color? However, drawing from the summation of the extant literature on privatization in child welfare that overall, privatized system performance is at par with that of non-privatized systems, this study hypothesized that the status-quo scenario—public or non-privatized delivery of case management services—will be maintained in terms of outcomes for children of color as well. In other words, the Researchers expected privatized systems to yield no significant difference in outcomes for the children of color that are disproportionately represented in the U.S. Foster Care System compared to those of the majority White/European American population.

color have historically received disparate treatment while in foster care compared to their counterparts, such that they have been known to experience longer stays in care, are re-victimized more while in care, and achieve positive permanency (reunification, guardianship, adoption) less often than their non-color counterparts (Harris, 2014; Roberts, 2002). Youth involved in child welfare, juvenile justice, or both systems are more likely to have increased educational, health (physical, mental, sexual), social, legal, and economic challenges than their non-system involved counterparts (Center for Juvenile Justice Reform, 2008; Herz et al., 2012; Leone & Weinberg, 2012; Morris & Freundlich, 2004). These youth often repeat grades, experiment with drugs and alcohol, engage in criminal activities, engage in risky sexual activities, early pregnancies, runaway, are sexually trafficked, become homeless, and become incarcerated (Child Welfare Information Gateway, 2016; Courtney & Skyles, 2003; Dworsky, Napolitano, & Courtney, 2013; Dworsky et al., 2010; Herz et al., 2012; Marshall & Haight, 2014). Child welfare-involved youth of color also experience longer stays in care, placement instability, achieve positive permanency less, more congregate care vs. residential settings, more re-entries or recidivism, and subsequent victimization while in care, increased crossover between systems (Harris, 2014; Herz et al., 2012; Marshall & Haight, 2014; Roberts, 2002). Consequently, researchers and government officials have posited that racial disproportionality and disparity in child welfare systems have led to state sponsored disruption, restructuring, and policing of minority families (Roberts, 2002), and tense relationships between communities and government (Roberts, 2008); thus, facilitating two different lived realities for whites and people of color in America, as intricately discussed in Hacker's (2003) book Two Nations: Black and White, Separate, Hostile, Unequal. 1.2. Privatization as foster care reform Privatization is a market-driven mechanism that is believed to facilitate increased efficiency, which is undergirded by market competition, to the delivery of human service (Meezan & McBeath, 2011). In other words, a pool of private non-profit agencies competing to deliver services would minimize inefficiency, lack of accountability, and fiscal waist; thus, yielding a more effective system overall. However, numerous empirical studies on privatization in child welfare have attributed its failings to counter the inefficiencies in the public institutional arrangement on the lack of organizational capacity and poor contract management and administration (Barillas, 2011; Burnett, 2011; Chuang, Collins-Camargo, McBeath, Wells, & Bunger, 2014; CollinsCamargo et al., 2011; Hubel, Schreier, Hansen, & Wilcox, 2013; McBeath, Jolles, Chuang, Bunger, & Collins-Camargo, 2014; Meezan & McBeath, 2011; Wells, Jolles, Chuang, McBeath, & Collins-Camargo, 2014). In fact, Collins-Camargo et al. (2011) found that most states that privatized foster care services reported significantly higher costs and less than optimal outcomes; thus suggesting that the market solution (privatization) to government inefficiencies does not seem to have panned out as expected by its proponents. As it relates to empirical comparisons of private versus public institutional arrangements to deliver case management services on state performance and individual child and youth outcomes, privatization was found to be no more effective or superior than public service delivery (Steen & Smith, 2012; Thornton & Cave, 2010). Comparing a myriad of variables for public and private agencies, Steen and Smith (2012) concluded [from their systematic review] that although there was a slight decline in safety measures for private agencies in totality, there were no significant differences across measures between private and public agencies. Also, in a recent national study examining case management service delivery, non-privatized or public systems were found to demonstrate higher rates of effectiveness and efficiency than fully and partially privatized systems (Coles, 2015). This study was the only national study this author could find that compared state systems

2. Methodology 2.1. Study design This secondary analysis employed a quasi-experimental, case-control design. Quasi-experimental designs are frequently used in studies that explore causal inferences with groups that cannot be randomly assigned (Creswell, 2009; Holosko, 2016; Rubin & Babbie, 2011). Casecontrol designs allow studies to compare these “groups of cases that have had contrasting outcomes and then collect retrospective data about differences that might explain the variances in outcomes” (Rubin & Babbie, 2011, p. 282). 2.2. Research criteria State foster care systems/agencies that have transferred case management service delivery functions to the private sector was analyzed and compared to selected state systems/agencies that maintain full responsibility for the delivery of these services on select outcomes for foster children by race. According to Collins-Camargo (2007), levels of privatization are operationalized as follows:

• Privatized (Group A): Private non- or for-profit organizations pro•

vide a full array of case management to the majority of service areas, with full case planning and decision-making authority. Non-privatized (Group B): Case management services are delivered through traditional arrangements in which state agencies sub-contract for services on an ad hoc basis, and state workers maintain primary case planning and management responsibilities. (p. 25)

Additionally, select racial groups of children for comparison was 126

Children and Youth Services Review 99 (2019) 125–131

K.Y. Huggins-Hoyt et al.

Non-Hispanic (NH) Black/African American, NH American Indian/ Alaskan Native, NH White/European American. As shown in the table, in privatized states, Black/African American children were 2 to 4 times more likely and American Indian/Alaskan Native children were 2 to 9 times more likely to be in foster care than their counterparts; and in non-privatized states, 2 to 3 and 2 to 6 times more likely to be in foster care than their counterparts, respectively. This is consistent with both historical and contemporary literature on disproportionality and disparity for Black/African American and American Indian/Alaskan Native children in the U.S. Foster Care System (Child Welfare Information Gateway, 2016; Crofoot & Harris, 2012; Cross, 2008; Dettlaff & Rycraft, 2008; Harris, 2014); and therefore, this study compared outcomes for these two specific racial groups of foster children only with those making up the majority in the U.S. population—White/European American children.

Table 1 Disproportionality indicesa (DI) and disparity ratiosbc (DR) for foster children in care on the last day of FFY2013c(N1 = 10 states, N2 = 118,761 cases). States

NH Black/African American

NH American Indian/ Alaskan Native

NH White/European American

DI

DR

DI

DR

DI

DR

Privatized Florida Hawaii Kansas Nebraska Wisconsin

1.6 0.8 2.1 3.0 3.8

1.5 0.9 2.2 4.1 6.4

1.5 2.0 1.1 9.2 4.6

1.4 2.4 1.2 12.6 7.8

1.1 0.8 1.0 0.7 0.6

0.0 0.0 0.0 0.0 0.0

Non-privatized Alabama Iowa New Jersey Pennsylvania Wyoming

1.2 2.9 3.0 3.3 2.8

1.5 3.6 5.2 5.8 2.9

0.0 6.0 0.0 2.0 0.5

0.0 7.5 0.0 3.5 0.5

0.8 0.8 0.6 0.6 1.0

0.0 0.0 0.0 0.0 0.0

2.3. Study sample

a Disproportionality Index (DI) = % of foster care population for a specific race ÷ % of total child population for the same specific race. b Disparity Ratio (DR) = DI for a specific minority population in foster care ÷ DI for the majority population in foster care. White children are considered the majority in the general and foster care populations. c DIs and DRs were calculated from population variables in the 2013 AFCARS foster care file.

The category of privatized states for this analysis was based on the Coles (2015) comparative study, which identified six different state systems that provided all foster care case management services through the private sector. These were: California, Florida, Hawaii, Kansas, Nebraska, and Wisconsin. To accomplish a relatively controlled matched group comparison, the 2013 foster care populations [as of the last day of the federal fiscal year (9/30/2013)] (2016b) of the six privatized systems were used to match and identify six comparable (< 5000 difference in populations) non-privatized state systems. California was an outlier with the largest foster care population of 56,947 in the United States and therefore, was excluded from the privatized group to prevent a disproportionally skewed comparison (Fig. 1). Table 2 shows the foster care population for the 2013 Federal Fiscal Year (FFY) [10/1/2012–9/30/2013], which consist of the number of children in care on the 1st day of FFY2013, the number of children that entered care during FFY2013, and the number of children in foster care on the last day of FFY2013.The resultant case–controlled matched sample of N1 = 10 state systems (5 privatized, 5 non-privatized) served a total of 69,750 foster children as of 9/30/2013. However, the resultant sample for analysis, N1 = 10 state systems and N2 = 118,761 foster children/cases (privatized n1 = 63,807, non-privatized n2 = 54,954), consisted of the number of children in care on 10/1/2012 and the number of children that entered care during FFY2013).

determined by the extent of their overrepresentation in the U.S. Foster Care System. Therefore, using the FFY2013 AFCARS foster care data, a disproportionality index (DI) and disparity ratio (DR) was calculated for each of the 10 states as displayed in Table 1. The formulas for these calculations are as follows (Harris, 2014):

DI =

the proportion of foster children of a certain race at each point the proportion of the same racial group in the child population

DR =

the rate of DI or outcome for one racial/ethnic group the rate of DI or outcome for another racial/ethnic group

In other words, racial disproportionality existed when the percentage of minorities (e.g., African American and American Indian/Alaskan Native) in a system was higher than their percentage in the general population; and racial disparity existed “when the rate of disproportionality, poor outcomes, and deficient services of one group (e.g., African Americans) exceeds that of a comparison group (e.g., European/White Americans)” (Harris, 2014, p. xv). Table 1 displays the DIs and DRs across the 10 states for each of the selected racial group: Selected States

2.4. Data collection The foster care population and outcome data for each state was

2013 Child Population

Foster care population on 9/30/2013

Privatized 1. Florida 2. Hawaii 3. Kansas 4. Nebraska 5. Wisconsin Totals

4,026,674 307,266 724,092 464,348 1,307,766 6,830,156

18,037 1,085 6,441 4,586 6,539 36,688

Non-Privatized 1. Alabama 2. Iowa 3. New Jersey 4. Pennsylvania 5. Wyoming Totals

1,111,481 724,032 2,022,117 2,715,645 137,679 6,710,954

4,524 6,341 6,946 14,270 981 33,062

Fig. 1. Sample of selected states based on the foster care population on the last day of FFY2013 (9/30/2013). 127

Children and Youth Services Review 99 (2019) 125–131

K.Y. Huggins-Hoyt et al.

Table 2 Foster care demographics by state for FFY2013a (N1 = 10 states, N2 = 118,761 casesb).a States

Table 3 Snapshot of foster care population demographics by state system type (N1 = 10 states, N2 = 118,761 cases). Privatized (n = 5) N (%)

Non-Privatized (n = 5) N (%)

63,807 (54%)

54,954 (46%)

545 8

539 9

Child Sex Male (52%) Female (48%)

33,011 (52%) 30,793 (48%)

28,882 (53%) 26,062 (47%)

Race and Ethnicity NH, White/European American (48%) NH, Black/African American (29%) NH, Am Indian/AK Native (1%) NH, Asian (0.6%) NH, Hawaiian/Other Pac Islander (0.4%) Hispanic (any race)/Latino (13%)

31,549 (49%) 16,617 (26%) 1396 (22%) 451 (0.7%) 454 (0.7%) 8401 (13%)

25,318 (46%) 17,996 (33%) 273 (0.5%) 253 (0.5%) 49 (< 0.1%) 7176 (13%)

In care on 1st day of FFY2013

Entered care during FFY2013

In care on last day of FFY2013

Demographics (% of total sample)

Privatized Florida Hawaii Kansas Nebraska Wisconsin Totals

18,977 1043 5882 5056 6184 37,142

14,313 1022 3963 2697 4670 26,665

18,037 1085 6441 4586 6539 36,688

Foster carea Total foster care population (on 1st day and entered during FFY2013) Average stay (days) in foster care Average age of child (on last day of FFY or at child's date of exit)

Non-privatized Alabama Iowa New Jersey Pennsylvania Wyoming Totals

4375 6070 6682 13,167 885 31,179

3081 4500 5400 9789 1005 23,775

4524 6341 6946 14,270 981 33,062

Data Source: 2013 AFCARS Foster Care file. N2 includes number of children in care on 1st day of FFY2013 and the # of children entering care during FFY2013. a

b

a

obtained from the FFY2013 Adoption and Foster Care Analysis and Reporting System (AFCARS), Foster Care file [dataset] (U.S. Department of Health and Human Services, 2015a, 2015b). “AFCARS is a federally mandated data collection system intended to provide case specific information on all children covered by the protections of Title IV-B/E of the Social Security Act (Section 427)” (Bronfenbrenner Center for Translational Research, 2016). This data is collected electronically, under the auspices of the Children's Bureau, from all 50 states, the District of Columbia, and Puerto Rico twice a year (October 1 to March 31 and April 1 to September 30). AFCARS datasets encompass unrestricted case-level information for all children served by the foster care system; and are made available to the public annually by the National Data Archive on Child Abuse and Neglect (NDACAN), a project of the Bronfenbrenner Center for Translational Research located in the College of Human Ecology at Cornell University (Bronfenbrenner Center for Translational Research, 2016).

Data source for foster care demographics: FFY2013 AFCARS Foster care file.

analysis 2: 0 = White/European American, 1 = American Indian/ Alaskan Native; iii) comparative analysis 3: 0 = Black/African American, 1 = American Indian/Alaskan Native.

3. Results Comparison analyses were performed to determine if the state system type [privatized vs. non-privatized] had any influence on outcomes for overrepresented foster children of color [Black and American Indian/Alaskan Native], compared to their white counterparts. Table 3 shows the demographic statistics for N1 = 118,761 foster children (54% in privatized group and 46% in non-privatized group) included in this sample. Fifty-two percent of the total sample were male children; and total racial composition consisted of 48% NH White/European American, 29% NH Black/African American, 1% NH American Indian/ Alaskan Native, 1% NH Asian and Hawaiian/other Pacific Islander, 6% NH > 1 race, 13% Hispanic (any race), and 2% unknown. Also, the average age of children in the privatized group was M = 8 and M = 9 in the non-privatized group. Lastly, the average length of stay (days) children were in foster care for privatized states was M = 545 and M = 539 for non-privatized states. Preliminary analyses were conducted to determine correlations between the six outcome variables. Two variables were excluded from final analyses for the following reasons: i) missingness for the length of stay (days) in a previous FC episode variable was very high (83%), which suggests that the current FFY2013 episode was the first and only episode in FC for this percentage of sample participants; and ii) total days in FC across all episodes (6% missingness) and length of stay (days) since latest removal date (1% missingness) were highly correlated (r = 0.94). Therefore, since the total days in FC across all episodes variable had a greater percentage of missingness, it was excluded from final analyses along with the length of stay (days) in a previous FC episode variable. The remaining 4 variables were found to have low missingness (< 1%) and significantly low correlations. Table 4 shows the descriptive statistics for the remaining outcome variables included in final analyses. These preliminary analyses also found, for the remaining variables, < 1% missingness and significantly low correlation. Final analyses consisted of multiple 2 × 2 between-subjects Factorial ANOVA computations to determine the interacting effect of system type (privatized versus non-privatized) and race on outcomes for minority racial groups of children that are disproportionately represented in the U.S. Foster Care System compared to those of the

2.5. Measures The dependent variables, child welfare outcome indicators, were obtained from the archived 2013 AFCARS Foster Care file. The design of this study included 6 continuous outcome indicators, which were: total number of removals from home, number of placement (PLC) settings in the current foster care (FC) episode, length of stay (days) in FC since the latest removal date, length of stay (days) in the current episode foster care PLC setting, length of stay (days) in a previous FC episode, and total number days in FC across all episodes. These selected outcome variables were grounded in the extant literature as important measures for monitoring child welfare agency performance on achieving national safety, permanency, and well-being standards for children in foster care (Children's Bureau, 2016a, 2016b). Additionally, more in-depth information on how these measures were computed can be obtained via the AFCARS Foster Care file codebook at the NDACAN website (https:// www.ndacan.acf.hhs.gov/). While the AFCARS contains foster care data information for all 50 states, the data for this study was further configured, upon request of the researchers, by NDACAN to only include data for the 5 privatized and 5 non-privatized states included in analyses. A variable to capture the system type for each child/case was then added with coded values as 0 = non-privatized, 1 = privatized. Additionally, dummy variables for comparing the selected 3 racial groups were added to the data, which were coded as follows: i) comparative analysis 1: 0 = White/ European American, 1 = Black/African American; ii) comparative 128

Children and Youth Services Review 99 (2019) 125–131

K.Y. Huggins-Hoyt et al.

Table 4 Descriptive statistics by race and system type for foster care indicators (N1 = 10 states, N2 = 118,761 cases). Groups and outcome variables

Non-privatized N

Table 5 Factorial ANOVA by race and system type for foster care indicators (N1 = 10 states, N2 = 118,761 cases).

Privatized

Variables

Overall modela

System type ∗ raceb

F

F

M

SD

N

M

SD

NH White/European American 25,316 Total # of removalsa # of PLC settings 25,151 # of days in foster care 25,309 # of days in current PLC 25,174

1.29 2.40 524.78 273.11

0.70 2.47 642.81 375.15

31,507 31,485 31,544 31,506

1.28 2.68 492.60 275.18

0.60 4.00 547.30 330.88

NH White/Eur Am vs. NH Black/Afr Am Total # of removals 201.41⁎⁎ # of PLC settings 265.94⁎⁎ # of days in foster care 291.70⁎⁎ # of days in current PLC 188.71⁎⁎

95.15⁎⁎ 71.81⁎⁎ 10.35⁎ 122.20⁎⁎

NH Black/African American Total # of removalsa 17,993 # of PLC settings 17,839 # of days in foster care 17,965 # of days in current PLC 17,891

1.43 2.74 666.47 353.50

0.93 2.98 848.42 492.66

16,604 16,576 16,611 16,596

1.32 3.47 604.20 296.69

0.64 5.59 772.89 384.34

NH White/Eur Am vs. NH Am Indian/AK Native Total # of removals 25.41⁎⁎ # of PLC settings 32.32⁎⁎ # of days in foster care 17.40⁎⁎ # of days in current PLC 10.48⁎⁎

4.81⁎ 1.50 2.80 26.81⁎⁎

NH Am Indian/Ak Native Total # of removalsa # of PLC settings # of days in foster care # of days in current PLC

1.35 2.61 525.48 184.60

0.67 2.49 689.92 322.35

1391 1391 1396 1393

1.43 2.61 559.59 309.13

0.78 3.02 661.61 416.06

NH Black/Afr Am vs. NH Am Indian/AK Native Total # of removals 57.19⁎⁎ # of PLC settings 85.34⁎⁎ # of days in foster care 22.83⁎⁎ # of days in current PLC 57.14⁎⁎

12.36⁎⁎ 6.00⁎ 3.17 37.09⁎⁎

273 270 273 270

⁎

a Total # of removals from home with primary caregiver and placed in foster care.

⁎⁎ a b

majority racial group (NH White/European American vs NH Black/ African American, NH White/European American vs NH American Indian/Alaskan Native, NH Black/African American vs NH American Indian/Alaskan Native). This statistical approach versus performing a MANOVA analysis, was taken for 2 reasons. First, this study did not aim to analyze and compare the effect of privatization on child welfare outcomes across all racial groups, but rather, for selected racial groups. Secondly, according to Weinfurt (1995), a MANOVA analysis requires that dependent or outcome variables be statistically correlated indicating an empirical relationship between the measures; and the variables for this study were found to have significantly low correlations. For instance, total number of removals was significantly correlated with number of placement settings in current FC episode (r = 0.075, p < .001) and length (days in current placement setting (r = −0.042, p < .001). Number of placement settings in current FC episode was also significantly correlated with length (days) in current placement setting (r = −0.037, p < .001) and length (days) since latest removal date (r = 0.501, p < .001). Lastly, length (days) in current placement setting and length (days) since latest removal date were also significantly correlated (r = 0.580, p < .001). Thus, Table 5 shows the results of the final Factorial ANOVA analyses, which found statistical significance for all models, overall. More specifically, the ANOVA results found significant main effects by system type and race individually and by their interaction, which suggests that the effect of the system type was influenced by the race of the foster child. Regardless of system type, Black/African American children, compared to their White/European American counterparts, experienced significantly more removals from their homes, F(1,3) = 95.15, p < .01, more placement settings, F(1,3) = 71.81, p > .01, longer stays in care overall, F(1,3) = 10.35, p < .01, and in their current placement settings, F(1,3) = 122.20, p < .01; but they also fared significantly better on these indicators in privatized systems, except for the number of placement settings. These children had an average of 3.47 placement settings in privatized systems versus 2.74 in non-privatized systems. For American Indian/Alaskan Native children, they had significantly more removals from their homes than both White/European American and Black/African American children in privatized systems, F (1,3) = 4.81, p < .05 and F(1,3) = 12.36, p < .01, respectively. They also had longer stays in their current placement settings than both White/European American and Black/African American children in privatized systems, F(1,3) = 26.81, p < .01, and F(1,3) = 37.09,

p < .05. p < .01. Overall model: df = 3. System type ∗ race: df = 1.

p < .01. However, American Indian/Alaskan Native children only had more placement settings than their Black/African American counterparts in privatized systems, F(1,3) = 6.00, p < .05.

4. Discussion The purpose of this study was to add to the body of research that has examined the effectiveness of privatization on child welfare outcomes compared to the status quo public institutional arrangement to foster care case management service delivery. The findings of this study also extends the comparative analyses of system types (privatized vs nonprivatized or public) on safety (Huggins-Hoyt, Mowbray, Briggs, & Allen, 2018) and permanency outcomes (Huggins-Hoyt, Mowbray, Briggs, & Allen, 2019) for foster children, and overall effectiveness and efficiency of these systems (Coles, 2015) by examining the relationship between privatization and disparity in outcomes for children who are disproportionately represented in foster care. While results support the extent of disparity for both Black/African American and American Indian/Alaskan Native children they also indicate that for some outcomes both groups may fare better in privatized versus non-privatized state systems, which partially conflicts with the study hypothesis that children fare no better in privatized vs non-privatized systems. However, in terms of the length of stay in the current PLC settings, there are two additional points to consider regarding these racial groups faring better in privatized states, which are: i) that these children were in their current PLC settings longer because they were in foster care longer, and ii) that their longer stays were due to having fewer PLC disruptions. While the former indication implies a worse permanency outcome for children in terms of timeliness to reunification, adoption, or another positive permanency option, the latter implies that children maintain some stability in placement while in care. In other words, children continuing to linger in foster care is not an optimal outcome, but for as long as they have to be in care, placement instability (i.e., experiencing placement disruptions causing placements in multiple foster homes or facilities) can exacerbate negative effects on child well-being (Whitelaw-Downs, Moore, McFadden, Michaud, & Costin, 2004). Another consideration that may explain the marginally better results for Black/African American children in privatized systems, is that they are less populated in these states, than in non-privatized states. 129

Children and Youth Services Review 99 (2019) 125–131

K.Y. Huggins-Hoyt et al.

updated data is made available can also create feasibility and accuracy issues. Fortunately, having the ongoing technical support of the archiving institution to configure these large data, minimized the potential for additional feasibility issues common in accessing large up-todate government data (Rubin & Babbie, 2011).

Therefore, privatized state systems may have more opportunities to produce better outcomes for these children simply because they serve fewer of them than do non-privatized systems. Also, beyond the more polarized market-based arguments for and against privatization in child welfare, there is a community-based rationale of proponents of privatization that exposes the potential practice benefits for children and families served by child welfare agencies, particularly as it relates to disparate outcomes for Black/African American and American Indian/ Alaskan Native children in the foster care system. This notion asserting that children and families are better served by smaller institutions (private child welfare agencies) spatially located in and connected to communities/neighborhoods has both a historical and theoretical base in child welfare and social work (i.e. settlement houses, community/ neighborhood centers) (Melton, Thompson, & Small, 2002; Pecora, 2000; Whitelaw-Downs et al., 2004). Thus, it is believed that community-based child welfare agencies have greater capacity and flexibility to provide a more effective, culturally, and linguistically competent service system (Kahn & Kamerman, 1999; Melton et al., 2002; Smith & Lipsky, 1993). For an example, in an assessment of privatized child welfare services, Freundlich and Gerstenzang (2003) stated

5. Conclusion As can be seen from the information provided, this study fills a void in the literature of large-scale comparative studies on national performance standards for safety, permanency, and well-being outcomes in child welfare. Too few of these types of studies have been published, and according to these findings, further examination of the role of privatization as a policy practice intervention for addressing the protracted nature of racial disproportionality and racial disparity in child serving systems is warranted. Particularly, there is a dearth of scholarly publications that present a well-reasoned research investigation encompassing an analysis of privatization as a policy practice intervention in child welfare systems for improving outcomes for overrepresented children of color. Also, the data on racial disproportionality and racial disparities highlighted in this paper points to the need for theory-driven research that investigates the benefits and drawbacks of applying microeconomic theory in designing social programs for children of color in public child welfare systems. However, this study sought to enhance the groundwork for such a research agenda to further explore this political economic policy initiative and its relationship to race, racial disproportionality and racial disparities. Thus, future theory-driven research of this type is needed to assess the extent to which private versus public child welfare systems mitigate or exacerbate the manifestations of inter-institutional practices of structural inequality, in terms of safety, permanency, and well-being outcomes. Chief among future research questions to study may include: (1) Are there significant differences between system types and race in terms of specific well-being outcomes for disproportionately represented foster youth? (2) How are relationships between individual, family, case, and community-level factors and well-being outcomes for older youth (ages 17–21) moderated by system types and race? Studies of these type, along with exploring effects of system-level factors (i.e., child welfare expenditures, practice model implementation), state-level factors (i.e., child poverty rate), and spatial characteristics (i.e., urban vs rural) are needed to advance the prudent practices that promote safety and permanency for all foster children of color, but even more acutely advance well-being outcomes among older youth of color in child welfare systems of care. Conflict of interest declaration. The authors declare no conflicts of interests in conducting or publishing this research.

…for other proponents, the value of privatization rests in its capacity to strengthen communities…for example, viewed governments as “alienating megastructures” and argued that government's role should be to “empower” community organizations, churches, voluntary associations and other less formal institutions to “mediate” between government and individuals. This emphasis downplays the rationales of efficiency and cost savings often advanced as the basis for privatization and highlights, instead, the value of local and smaller scale providers of social services… (p. 7) Additionally, Reed (1989), quoting a study conducted by Stephen Moore of the Heritage Foundation, noted: Throughout their small scale, non-bureaucratic nature, local knowledge and personal relationships, neighborhoods, families, churches, and voluntary associations can respond rapidly, accurately and in a more acceptable manner to local and individual needs in ways that large formal institutions such as government agencies cannot. (p. 1) Taken all together, the aforementioned statements promote local private providers as a sort of socio-spatial structure—a public or centralized place or space serving as a locale for facilitating social interactions, supports, development, and services in communities (Svendsen, 2010). Offering foster care services in such a localized sociospatial context may have great potential for diminishing both disproportionality and disparity. Overall, these findings warrant an even deeper examination into the extent race coupled with system type may influence the impact child/case-level factors have on the likelihood state systems meet federal safety, permanency, and well-being performance outcome standards as determined and evaluated by the U.S. Department of Health and Human Services, Administration for Children and Families, Children's Bureau via the Child and Family Services Review (CFSR). However, there were several general limitations to using pre-existing archived secondary datasets, as was done in this study. According to Rubin and Babbie (2011), these may include: i) missing data, ii) problems with validity, iii) problems with reliability, iv) inadequate documentation, and v) feasibility issues. While this study did encounter an issue with moderate to exorbitant percentages of missingness for variables, the study researchers could not resolve this issue with the archival data they did not collect, and therefore, had to exclude these variables from final analyses. Furthermore, the extent to which these data are reliable is also a limitation as often, secondary researchers cannot ensure that the data originally collected was accurate or had fidelity (e.g., how well trained the data collectors were in identifying inaccuracies). Also, changes in existing data over time and when

References Barillas, K. H. (2011). State capacity: The missing piece in child welfare privatization. Child Welfare, 90(3), 111–127. Briggs, H. E. (2011). The road to equity: Transforming Oregon's child welfare system through culture, science, and collaboration. Final report to the oregon legislature by the state of Oregon's governor's task force on equity in child welfare. Retrieved from https:// www.oregon.gov/DHS/DHSNEWS/CWIndependentReview/gov-taskforce-report-311.pdf. Bronfenbrenner Center for Translational Research (2016). National data archive on child abuse and neglect. Retrieved from http://www.ndacan.acf.hhs.gov/about/aboutmission.cfm. Burnett, J. (2011). Success & challenges in privatizing child services. Capitol Ideas, 54(5), 38–39. Center for Juvenile Justice Reform (2008). Bridging two worlds: Youth involved in the Child Welfare and Juvenile Justice Systems. Retrieved from http://cjjr.georgetown. edu/wp-content/uploads/2015/03/BridgingTwoWorlds_2008.compressed.pdf. Child Welfare Information Gateway (2016). Racial disproportionality and disparity in child welfare. Washington, DC: U.S. Department of Health and Human Services, Children's Bureau. Children's Bureau (2016a). Child and family services reviews (CFSRs). Retrieved from http://www.acf.hhs.gov/cb/monitoring/child-family-services-reviews.

130

Children and Youth Services Review 99 (2019) 125–131

K.Y. Huggins-Hoyt et al.

Marshall, J. M., & Haight, W. L. (2014). Understanding racial disproportionality affecting African American Youth who cross over from the child welfare to the juvenile justice system: Communication, power, race and social class. Children and Youth Services Review, 42, 82–90. https://doi.org/10.1016/j.childyouth.2014.03.017. McBeath, B., Jolles, M. P., Chuang, E., Bunger, A. C., & Collins-Camargo, C. (2014). Organizational responsiveness to children and families: Findings from a national survey of nonprofit child welfare agencies. Children and Youth Services Review, 38(0), 123–132. https://doi.org/10.1016/j.childyouth.2014.01.018. Meezan, W., & McBeath, B. (2011). Moving toward performance-based, managed care contracting in child welfare: Perspectives on ctaffing, financial management, and information technology. Administration in Social Work, 35(2), 180. Melton, G. B., Thompson, R. A., & Small, M. A. (2002). Toward a child-centered, neighborhood-based child protection system: A report of the Consortium on Children, Families, and the Law. Westport, CT: Praeger. Morris, L., & Freundlich, M. (2004). Youth involvement in the child welfare and juvenile justice systems: A case of double jeopardy? Washington, DC: CWLA Press. Pecora, P. J. (2000). The child welfare challenge: Policy, practice, and research (2nd ed.). New York: Aldine de Gruyter. Reed, L. W. (1989). Privatization works for social services too. Retrieved from https:// www.mackinac.org/217. Roberts, D. E. (2002). Shattered bonds: The color of child welfare. New York, NY: Basic Books. Roberts, D. E. (2008). The racial geography of child welfare: Toward a new research paradigm. Child Welfare, 87(2), 125–150. Rosinsky, K., & Connelly, D. (2016). Child welfare financing SFY 2014: A survey of federal, state, and local expenditures. Retrieved from http://www.childtrends.org/ wp-content/uploads/2016/10/2016-53ChildWelfareFinancingSFY2014.pdf. Rubin, A., & Babbie, E. (2011). Research methods for social work (7th ed.). Belmont, CA: Cengage Learning. Smith, S. R., & Lipsky, M. (1993). Nonprofits for hire: The welfare state in the age of contracting. Cambridge, Mass: Harvard University Press. Steen, J. A., & Smith, S. (2012). An organizational view of privatization: Is the private foster care agency superior to the public foster care agency? Children and Youth Services Review, 34(4), 851–858. https://doi.org/10.1016/j.childyouth.2012.01.016. Summers, A. (2015). Disproportionality rates for children of color in foster care (Fiscal Year 2013). Retrieved from Reno, Nevada: http://www.ncjfcj.org/sites/default/ files/NCJFCJ%202013%20Dispro%20TAB%20Final.pdf. Svendsen, G. L. H. (2010). Socio-spatial planning in the creation of bridging social capital: The importance of multifunctional centers for intergroup networks and integration. International Journal of Social Inquiry, 3(2), 45–73. Thornton, J., & Cave, L. (2010). The effects of organizational form in the mixed market for foster care. Annals of Public & Cooperative Economics, 81(2), 211–245. https://doi. org/10.1111/j.1467-8292.2010.00410.x. U.S. Department of Health and Human Services, Administration of Children and Families, Administration on Children, Youth and Families, Children's Bureau (Administration of Children and Families, Administration on Children, Youth and Families, Children's Bureau, 2015a). Adoption and foster care analysis and reporting system (AFCARS) foster care file, FFY2013 [Data File]. Retrieved from: http://www.ndacan.acf.hhs. gov/datasets/dataset-details.cfm?ID=187. U.S. Department of Health and Human Services, Administration of Children and Families, Administration on Children, Youth and Families, Children's Bureau (Administration of Children and Families, Administration on Children, Youth and Families, Children's Bureau, 2015b). The AFCARS report: Preliminary FY 2014 estimates as of July 2015. Retrieved from http://www.acf.hhs.gov/sites/default/files/cb/afcarsreport22.pdf. U.S. Department of Health and Human Services (2016). Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau. The AFCARS Report #23. Retrieved from https://www.acf.hhs.gov/sites/default/files/cb/ afcarsreport23.pdf. Unruh, J. K., & Hodgkin, D. (2004). The role of contract design in privatization of child welfare services: The Kansas experience. Children and Youth Services Review, 26, 771–783. https://doi.org/10.1016/j.childyouth.2004.02.020. Weinfurt, K. P. (1995). Multivariate analysis of variance. In L. G. Grimm, & P. R. Yarnold (Eds.). Reading and understanding multivariate statistics (pp. 245–276). Washington, DC: American Psychological Association. Wells, R., Jolles, M., Chuang, E., McBeath, B., & Collins-Camargo, C. (2014). Trends in local public child welfare agencies 1999–2009. Vol. 38, Elsevier Ltd93–100. Whitelaw-Downs, S., Moore, E., McFadden, E. J., Michaud, S. M., & Costin, L. B. (2004). Child welfare and family services: Policies and practice (7th ed.). Boston: Pearson Allyn & Bacon.

Children's Bureau (2016b). Child welfare outcomes 2010–2013: Report to Congress. Retrieved from Washington, DC: http://www.acf.hhs.gov/sites/default/files/cb/ cwo10_13.pdf. Chuang, E., Collins-Camargo, C., McBeath, B., Wells, R., & Bunger, A. (2014). An empirical typology of private child and family serving agencies. Children and Youth Services Review, 38(0), 101–112. https://doi.org/10.1016/j.childyouth.2014.01.016. Coles, D. C. (2015). Identifying differences between privatized, partially privatized, and nonprivatized state foster care systems: A comparative study examining efficiency and effectiveness. Doctoral Dissertation Virginia Commonwealth University, VCU Scholars Compass. Retrieved from http://scholarscompass.vcu.edu/cgi/viewcontent.cgi? article=4830&context=etd. Collins-Camargo, C. (2007). Private Delivery. State News (Council of State Governments), 50(1), 23–25. Collins-Camargo, C., McBeath, B., & Ensign, K. (2011). Privatization and performancebased contracting in child welfare: Recent trends and implications for social service administrators. Administration in Social Work, 35(5), 494. Courtney, M., & Skyles, A. (2003). Racial disproportionality in the child welfare system. Children and Youth Services Review, 25, 355–358. 10.1016/S0190-7409(03)00025-2. Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). Thousand Oaks, CA: SAGE Publications. Crofoot, T. L., & Harris, M. S. (2012). An Indian Child Welfare perspective on disproportionality in child welfare. Children and Youth Services Review, 34(9), 1667–1674. https://doi.org/10.1016/j.childyouth.2012.04.028. Cross, T. L. (2008). Disporportionality in child welfare. Child Welfare, 87(2), 11–20. Dettlaff, A. J., & Rycraft, J. R. (2008). Deconstructing disproportionality: Views from multiple community stakeholders. Child Welfare, 87(2), 37–58. Dworsky, A., White, C. R., O'Brien, K., Pecora, P., Courtney, M., Kessler, R., & Hwang, I. (2010). Racial and ethnic differences in the outcomes of former foster youth. Children and Youth Services Review, 32(6), 902–912. http://dx.doi.org/10.1016/j.childyouth. 2010.03.001. Dworsky, A., Napolitano, L., & Courtney, M. (2013). Homelessness during the transition from foster care to adulthood. American Journal of Public Health, 103, S318–S323. 10. 2105/AJPH.2013.301455. Flaherty, C., Collins-Camargo, C., & Lee, E. (2008). Privatization of child welfare services: Lessons learned from experienced states regarding site readiness assessment and planning. Children and Youth Services Review, 30(7), 809–820. https://doi.org/10. 1016/j.childyouth.2007.12.009. Freundlich, M., & Gerstenzang, S. (2003). An assessment of the privatization of child welfare services: Challenges and successes. Washington, D.C.: CWLA Press. Hacker, A. (2003). Two nations: Black and white, separate, hostile, unequal. New York, NY: Scribner. Harris, M. S. (2014). Racial disproportionality in child welfare. New York: Columbia University Press. Herz, D., Lee, P., Lutz, L., Stewart, M., Tuell, J., & Wiig, J. (2012). Addressing the needs of multi-system youth: Strengthening the connection between child welfare and juvenile justice. Retrieved from http://cjjr.georgetown.edu/pdfs/msy/ AddressingtheNeedsofMultiSystemYouth.pdf. Hill, R. B. (2007). An analysis of racial/ethnic disproportionality and disparity at the national, state, and county levels. Casey-CSSP alliance for racial equity in child welfare. 27. Holosko, M. J. (2016). Introduction to the special issue on intervention research. Research on Social Work Practice, 26(1), 5–7. https://doi.org/10.1177/1049731515581015. Hubel, G. S., Schreier, A., Hansen, D. J., & Wilcox, B. L. (2013). A case study of the effects of privatization of child welfare on services for children and families: The Nebraska experience. Children & Youth Services Review, 35(12), 2049–2058. https://doi.org/10. 1016/j.childyouth.2013.10.011. Huggins-Hoyt, K. Y., Mowbray, O., Briggs, H. E., & Allen, J. L. (2018). Privatization and child welfare systems: A comparison analysis of safety performance outcomes. (Manuscript submitted for publication). Huggins-Hoyt, K. Y., Mowbray, O., Briggs, H. E., & Allen, J. L. (2019). Privatization and child welfare systems: A comparison analysis of permanency outcome performance. (manuscript in preparation). Kahn, A. J., & Kamerman, S. B. (1999). Contracting for child & family services: A missionsensitive guide. Retrieved from Baltimore, MD http://ezproxy.gsu.edu/login?url= http://search.ebscohost.com/login.aspx?direct=true&db=eric&AN=ED450179& site=eds-live. Leone, P., & Weinberg, L. (2012). Addressing the unmet educational needs of children and youth in the Juvenile Justice and Child Welfare Systems. Retrieved from Washington, DC: http://cjjr.georgetown.edu/wp-content/uploads/2015/03/ EducationalNeedsofChildrenandYouth_May2010.pdf.

131