Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 1.
Published in final edited form as: Child Abuse Negl. 2024 Sep 17;157:107047. doi: 10.1016/j.chiabu.2024.107047

Economic Insecurity in Young Adulthood Among Children Maltreated During Early Adolescence

Michael Caniglia 1
PMCID: PMC11512671  NIHMSID: NIHMS2024264  PMID: 39293104

Abstract

Background:

Children living in foster care as teenagers often experience greater economic insecurity during adulthood than their peers. However, few studies examine the association between foster care entrance during early adolescence and later economic outcomes.

Objective:

Examine whether entrance into foster care in early adolescence is associated with employment, monthly earnings, and participation in the Supplemental Nutrition Assistance Program (SNAP) relative to an observationally similar counterfactual population.

Participants and Setting:

Using statewide administrative data from Wisconsin, I limited my study population to early adolescents exposed to alleged maltreatment between ages 10 through 13 who return to the home or achieve permanency by age 18. Economic security outcomes were examined at age 23.

Methods:

Propensity score matching generated a counterfactual group that was similar to children who entered foster care (N=1,252). Odds of employment and SNAP usage were modeled with logistic regressions, while earnings were estimated with zero-inflated negative binomial regressions.

Results:

On average, foster care entrance was not associated with later economic difficulties. Among children who entered foster care, longer stays predicted elevated likelihood of employment (OR 1.18, 95% CI 1.04, 1.33), while more placements within foster care were associated with greater odds of receiving SNAP (OR=1.38, 95% CI 1.01, 1.90).

Conclusion:

Overall, entrance into foster care during early adolescence was not associated with earnings, employment, or SNAP participation relative to the matched sample.

Keywords: Foster care, early adolescence, maltreatment, SNAP, earnings

1. Introduction

The child protective services (CPS) system acts as one of America’s most powerful institutions and is entrusted with responding to allegations of maltreatment or neglect involving minors (Turney & Wildeman, 2017). States remove children from their parents’ care and place them in foster care when officials determine that they face imminent risk of harm and that less intrusive interventions would not prevent further maltreatment (Font & Gershoff, 2020). In federal fiscal year 2021, more than 606,000 children experienced foster care stays (U.S. Children’s Bureau, 2023). The CPS system’s removal of children from their homes – and their ensuing experiences in foster care – may be associated with economic security outcomes in young adulthood (Stewart et al., 2014). The inability of young people to achieve economic security – defined here as attaining employment and earnings without using public assistance – during young adulthood may have far-reaching consequences. Economic setbacks during young adulthood are associated with diminished income and wealth later in life (Killewald & Bryan, 2018).

Early adolescence is a key moment of growth, and more than 64,000 early adolescents between the ages of 10 and 13 are in foster care (NASEM, 2019; U.S. Children’s Bureau, 2023). The time between childhood and middle-adolescence is an important period for the formation of future career expectations and may carry associations with wellbeing later in life (Cansler et al., 2012; Galambos & Sears, 1998; Johnson et al., 2020). Despite the developmental importance of early adolescence, few scholars have considered foster care entrance between ages 10 and 13 as a predictor of adults’ economic wellbeing (Bald, Doyle, et al., 2022; Gross & Baron, 2022; Warburton et al., 2014).

Entrance into foster care can expose children to a range of hazards, including separation from siblings and frequent moves between homes (Font & Kim, 2021). In light of the risks, CPS decisions focus on protecting children from immediate danger and do not consider outcomes later in the life course (Font & Gershoff, 2020). Yet, as journalists and academics debate whether foster care harms children, it remains important for researchers to examine the relationship between foster care entrance and later life outcomes (Bauer & Thomas, 2019; Cancel, 2021; Warburton et al., 2014). Rigorous evaluations of the association between foster care entrance and later wellbeing may offer perspective about how to improve interventions for children experiencing maltreatment (Bald, Doyle, et al., 2022).

This study uses a statewide administrative dataset to investigate the association between placement in foster care during early adolescence and economic security in young adulthood. The paper employs a matched sample pairing youth who enter foster care (and return home or achieve permanency by age 18) to those who remain at home following a maltreatment report. By analyzing the association between entrance into foster care during early adolescence and later economic security outcomes relative to a comparable counterfactual population, this paper informs ongoing policy debates.

2. Background

2.1. Economic security among young adults

Since the 1970s, the normative transition to adulthood in American society has become more protracted as young people delay marriages and careers (Arnett, 2014). Young adults often choose to live with their parents to save money while balancing education and work, and the affluent benefit from their parents’ financial resources well into their 20s (NASEM, 2019). Within this context, the transition to adulthood has become characterized by growing economic insecurities (Arnett, 2014).

The transition to adulthood is especially difficult for young people with histories of foster care involvement (Font et al., 2018). These young adults may not benefit from the family relationships that support other young people during the transition to adulthood (NASEM, 2019; Okpych et al., 2023). In addition, frequent moves within foster care may make it difficult for young people to succeed in school and build healthy attachments as adults (Goyette et al., 2021). These issues underscore the vulnerability associated with foster care experiences and the extent to which they may be associated with economic insecurity in young adulthood.

2.2. Economic Security and Living Situations

Families serve as key units for teaching social norms and expectations to children. In traditional settings, the biological family promotes children’s social learning (Akers, 1977). In healthy families, parents communicate norms connected to employment that influence their offspring’s views (Johnson et al., 2020). Early adolescents use information gathered from observing their parents’ work habits to form their own occupational goals (Mortimer, 2021). Early adolescents living in homes with working adults may prove more likely to participate in the labor force during adulthood than those with unstably employed parents (Mortimer, 2021).

Children who experience maltreatment face an array of challenges later in life. Children exposed to maltreatment confront elevated risks of emotional dysregulation, poor executive functioning, and worsened health (McEwen & McEwen, 2017). These deficits are associated with diminished economic outcomes later in life (Zielinski, 2009). Young people who experience maltreatment earn less money and are more likely to participate in safety-net programs compared to their peers after accounting for childhood socioeconomic status (Zielinski, 2009).

Foster care constitutes a diversity of living arrangements that differ from a child’s biological home. More than three-quarters of children in foster care live with other kin or with adults in non-relative foster family homes (U.S. Children’s Bureau, 2021). Other environments include congregate settings (i.e., group homes), pre-adoption placements, and trial reunifications with biological parents (Font & Gershoff, 2020). In the case of non-relative home-based foster care settings, CPS seeks to ensure the suitability of foster parents by evaluating their income stability, determining whether their living space can support a child, and mandating that they complete training (U.S. Children’s Bureau, 2018b). Kinship caregivers must meet similar guidelines, but requirements carry more leeway for relatives (U.S. Children’s Bureau, 2018b). The safety requirements mean that foster homes may prove less dangerous than those experienced by children who remain with their biological families after maltreatment (Font & Potter, 2019).

2.3. Heterogeneity in foster care

The variety of experiences within the foster care system underscores potential differences in the outcomes associated with it (Bald, Doyle, et al., 2022). For example, different types of placement settings may carry varying associations with long-term economic security. Children who spend most of their time in kinship care may be less likely to have a working adult in the household (Font & Potter, 2019). In addition, children who stay in congregate care settings tend to earn less in young adulthood compared to those who spend time in other foster care settings (Font et al., 2018).

How children exit foster care could also carry associations with economic security in young adulthood. For instance, youths who exit foster care via adoption are more likely to enter socioeconomically advantaged home environments and matriculate to college (Font et al., 2018; Potter & Font, 2019). Alternatively, returning to a biological family after a stay in foster care may carry risks if the home environment does not improve. Repeated exposure to maltreatment after reunification may cause greater trauma and hinder future wellbeing (McEwen & McEwen, 2017; Ryan et al., 2016). Although not included in this study, young people remaining in care at age 18 can opt to stay in foster care and connected to key supports until age 21 (Henderson, 2019). Given their access to tuition assistance and other services during the transition to adulthood, the experiences of young people in foster care at age 18 categorically differ from their peers who otherwise exit care (Hanson et al., 2023; Henderson, 2019).

2.4. Social learning and attachment theories

Social learning theory, or the perspective that interactions with peers and adults influence behavior, suggests that foster care placement could carry either positive or negative associations with economic security (Akers, 1977). On the one hand, leaving home settings characterized by maltreatment might reduce young people’s exposure to negative social learning experiences, such as witnessing criminal activity (Akers, 1977; Font et al., 2021). Placement into foster care settings could allow youth opportunities to absorb positive lessons, protect against risky behaviors and support the development of interpersonal relationships (Okpych et al., 2023; Potter & Font, 2019).

Existing research supports social learning as a mechanism through which foster care can improve adult outcomes. One study showed that children who entered non-relative family care were more likely to have an employed caregiver vis-à-vis individuals who remain in the biological home (Font & Potter, 2019). Other scholarship found that longer stays in foster care were associated with elevated odds of graduating from high school and improved earnings in young adulthood relative to those who exit (Font et al., 2018). These factors indicate that early adolescents who spend long periods of time in foster care may demonstrate greater odds of employment and reduced likelihoods of participating in public assistance programs.

On the other hand, foster care entrance can create obstacles to healthy development (Bald, Doyle, et al., 2022). Attachment theory, or the perspective that children need reliable adult relationships to feel safe, suggests that placement instability could harm development and later outcomes (Miranda et al., 2019). Frequent moves among foster homes could disrupt the construction of healthy relationships and make it more difficult to access important medical and behavioral health services (Goyette et al., 2021; Greiner et al., 2019). Young people who experience a higher number of placement moves within care are more likely to face academic challenges and exhibit externalizing disorders associated with poor attachment styles (Goyette et al., 2021; Mihalec-Adkins et al., 2020). In turn, children with multiple placements face higher risks of dropping out of high school (Goyette et al., 2021). Given elevated risks of leaving high school without a degree, experiencing multiple foster care placements may predict worsened employment and earnings outcomes among young people (Font et al., 2018).

Although Wisconsin’s CPS agencies take steps to reduce the number of placement moves a child experiences in care, individual outcomes vary (Henderson, 2019). CPS encourages foster families to help children prepare for young adulthood (U.S. Children’s Bureau, 2018a), but ongoing litigation accuses agencies of denying early adolescents the resources needed to navigate the transition to adulthood (V.R. & B.R. v. County of San-Diego, 2023). If children face frequent placement moves or experience under-resourced environments in care, they might not benefit from foster care.

2.5. The current study

The current study seeks to answer two key research questions: First, does entrance into foster care during early adolescence shape a child’s economic security during young adulthood relative to remaining in the home? Second, do experiences that occur within foster care predict economic outcomes during young adulthood? This study used a statewide linked administrative dataset to examine foster care placement during early adolescence as a predictor of economic security during the transition to adulthood. By generating a matched sample, the study examined whether outcomes among individuals with foster care experiences differed from similar peers who also experienced maltreatment allegations but did not enter care. The research design better identified the role of foster care itself in influencing later outcomes, while easing the threats posed by potential confounders.

The study also fills a gap in the literature by examining early adolescents – an age group infrequently studied in foster care. Although existing research investigates the association between foster care placements among young children and late adolescents (Bald, Doyle, et al., 2022; Gross & Baron, 2022), less information exists on the relationship between foster care entrance during early adolescence and later outcomes. Since early adolescence acts as a crucial moment of social learning, emotional development, and goalsetting (Cansler et al., 2012; Gavazzi & Lim, 2023; Goering & Mrug, 2023), I consider whether entrance into foster care during the period carries associations with later outcomes.

Based on research suggesting that foster care may offer protective effects relative to remaining in home environments following maltreatment reports (Font et al., 2018; Font & Potter, 2019; Ryan et al., 2016), I anticipate that youth who entered foster care during early adolescence are more likely to experience employment, generate greater earnings, and exhibit lower likelihoods of receiving food assistance at age 23 relative to a matched sample of early adolescents who were the subjects of CPS reports but who did not enter foster care. Given the findings from attachment and social learning theories (Akers, 1977; Bald, Doyle, et al., 2022; Font et al., 2021; Goyette et al., 2021), I predict that more time spent in foster care will predict higher odds of achieving economic security. I also expect an inverse association between the number of foster care placements and economic security (Goyette et al., 2021).

3. Methods

3.1. Study design and data

The study leveraged a large administrative dataset to conduct analyses using a propensity score matched sample. This analysis utilized the Wisconsin Administrative Data Core (WADC), a longitudinal administrative dataset managed by the University of Wisconsin-Madison’s Institute for Research on Poverty and received human subjects approval from the University of Wisconsin-Madison Institutional Review Board. The WADC serves as one of the nation’s most comprehensive longitudinal sources of administrative data on individuals interacting with state programs; this data features employment and public assistance records for all children who were subjects of child maltreatment reports or entered foster care in the state from 2004 through 2019.

The use of state administrative data circumvents the challenges associated with survey respondents underreporting program participation or earnings (Meyer & Mittag, 2019). Administrative data recording public assistance use and earnings are generally highly accurate because states have strong incentives to monitor recipients’ payments and taxpayers’ earnings (Meyer & Mittag, 2019). When creating the WADC, Wisconsin data administrators use deterministic and probabilistic matching techniques to create powerful linkages across systems (Brown et al., 2020). The data’s large size allows researchers to analyze infrequent events, such as entrance into foster care, while also using an array of covariates to account for potential confounders.

3.2. Sample inclusion

I restrict the initial population of WADC individuals with CPS involvement (n=402,166) to those born in 1994 and 1995 and who experience CPS interactions between the ages of 10 through 13 (n=10,081). CPS records feature allegations including physical abuse, sexual abuse, and neglect. I compare youth who entered foster care between the ages of 10 through 13 to cohort members with a child maltreatment investigation during the same age range, but who were not placed in foster care. Given that the National Academies of Science, Engineering, and Math (2019) classifies early adolescence as ages 10 through 13 and that foster care after age 13 often stems from juvenile justice referrals (Baglivio et al., 2016), I limited my sample to youth who are ages 10 through 13. Since comprehensive CPS records collection started in 2004, age 10 serves as the earliest moment in which maltreatment data exists for every child in the sample.

For children who entered foster care, the WADC records length of placement within kin, non-relative family, and congregate care settings. I exclude children who ever entered the state’s care via the juvenile justice system or voluntary removals (n=350), a unique population with experiences distinct from those who enter because of maltreatment (Baglivio et al., 2016). The study also excluded youth who remained in foster care at age 18 (n=121) because they qualified for special services and post-secondary education vouchers provided through the John H. Chafee Foster Care Program in Wisconsin (Henderson, 2019). The additional social supports provided via the Chafee program offered help not available to youths who reunited with their families or experienced adoption in early adolescence. I included Milwaukee County respondents in the study even though Milwaukee County did not report juvenile justice data to the WADC. Supplemental analyses that excluded Milwaukee County residents (n=2,481) mirrored the results presented here. (Results not shown but available upon request).

Finally, I excluded 88 individuals (less than 1 % of the sample) who died between ages 10 and 24. Group differences in the number of deaths among children who entered care during early adolescence relative to those who received a CPS report alone were not significant (p<0.64). The final sample eligible for matching includes 9,522 young people. Within this group of individuals with a report of child maltreatment filed during their early adolescent years, 663 entered foster care during the period from 2004 through 2009 and 8,859 did not.

3.2. Measures and covariates

Outcomes.

Since data records end in 2019, I maximized the window of observation by measuring outcomes during the 12 months after individuals turned 23. Young people born in 1994 turned 23 in 2017, while those born in 1995 turned 23 in 2018. First, I created a binary outcome to determine whether an individual ever received wage earnings during the 12 months in which they were 23 years old. This provides insight into whether a person was ever employed, a qualitatively different measurement of economic connection than the total amount earned during a year. A second outcome measures average gross monthly earnings when young people were 23. To account for inflation, I measure all earnings in 2019 dollars. WADC income data only includes sources that are taxed by Wisconsin; in some cases, in-kind payments or income from the informal economy may be excluded. Third, I captured public assistance usage with a dichotomous measure of whether the participant received SNAP benefits at age 23, given SNAP is one of the most frequently accessed social welfare programs (Vigil 2022).

Explanatory Variables.

The focal variable is entry into foster care between the ages of 10 to 13 years old (CPS report alone=0, entered foster care=1). Covariates include gender (Men=1), nativity (born in the U.S.=1) and childhood disability as measured by childhood receipt of SSI (=1). I account for race (non-Hispanic White=1, non-Hispanic Black=2, Hispanic=3, and Other=4) such that the “Other” group is made up of individuals identified as American Indian, Asian or Pacific Islander, and multiracial. Small sample sizes prohibited me from including specific indicators for smaller racial and ethnic groups. I also added a categorical variable that controls for the county in which a child’s first interaction with CPS occurred (Milwaukee County=1, other urban county=2, and rural county=3).

Models included indicator variables for the types of abuse alleged during early adolescence (physical, sexual, neglect, and other maltreatment) and whether an allegation was ever substantiated (=1). In addition, I adjusted for whether an allegation during early adolescence involved a biological, step-, or adopted parent (=1). Since statewide data collection of CPS allegations did not occur until 2004, complete information on child maltreatment histories did not exist for children in the sample prior to age 10 (Brown et al., 2020). To account for early life disadvantage, I included the percentage of months when the child was ages 0 to 10 years old that the mother received SNAP, mother’s age at her first child’s birth, and whether the child’s father was identified by the state via birth certificates (Font et al., 2018).

3.3. Analysis

My analysis proceeded in three phases. First, I implemented multiple imputation with chained equations to account for the 9% of observations with missing values. Second, I reviewed descriptive statistics as part of my preliminary analysis and to gain insight into the composition of the WADC caseload. Third, to answer the first research question regarding differences in outcomes between young people who entered care and those who did not, I used propensity score matching to identify a counterfactual population. Using the matched sample, models estimated whether young people with foster care experiences exhibited worse outcomes in young adulthood relative to similar peers who remained at home. Finally, I examined the study’s second research question and explored heterogeneity within foster care experiences.

Propensity Score Matching to Define Appropriate Comparison Group.

To better control for differences between youth with a CPS report who were and who were not placed in foster care, I sought to match each individual who entered foster care to a similar peer who was reported as a maltreatment victim but did not enter foster care. To do so, I predicted a child’s propensity to enter foster care during early adolescence using a logistic regression model that controlled for characteristics measured prior to a child reaching foster care (Rosenbaum & Rubin, 1983). In addition to the regression covariates, I used the following characteristics measured prior to age 10 to predict the likelihood of foster care entrance: maternal and paternal average monthly wages, maternal receipt of welfare assistance, indicators of maternal and paternal incarceration, and whether a young person was born in Wisconsin. I also controlled for forms of alleged and substantiated maltreatment that occurred during early adolescence but before a child entered care. (See Supplementary Table 1 for a list of variables used to complete matching and their association with foster care entrance.)

Next, I matched each foster care entrant’s probability of placement to that of a child who remained at home. The method dropped unmatched children who experienced a maltreatment report alone and had a low propensity to enter foster care. I also excluded 37 foster care entrants who could not be matched with a similar youth who received a CPS report alone. The Chi-squared test indicated no differences between youth with a CPS report alone and those who entered foster care (p>0.981). (See Supplementary Table 2 for additional information on sample balance.) The final matched sample featured 1,252 individuals – 626 youth with a CPS report alone and 626 who entered foster care in early adolescence.

First, I tested the association between early adolescent CPS involvement and employment and SNAP receipt during the transition to adulthood using a matched sample with logistic regression. Second, I predicted earnings using zero-inflated negative binomial regression models, which appropriately accounted for the over-representation of zero-values for earnings in the matched sample (30%) (Long & Freese, 2014). Bayesian information criterions indicated that the zero-inflated negative binomial models better fit the data than other approaches. When modeling estimates with the matched sample, I included covariates for demographic characteristics, allegations during early adolescence, childhood public assistance use, county, and birth cohort to reduce biases not fully addressed by propensity score models (Bai & Clark, 2019).

Examining Heterogeneity within Foster Care.

Average effects may mask important variation in foster care experiences (Bald, Doyle, et al., 2022). I also estimated a series of models wherein the sample is restricted to only youth placed in foster care and predict whether and how various foster care experiences are correlated with young adults’ economic security. First, to better understand the role of foster care experiences in predicting economic outcomes, I tested the amount of time in foster care, measured as the logged total days an individual spends in the state’s custody, as a predictor of different economic outcomes. Second, I controlled for the number of foster care placements the child experienced as the logged number of placements.

I also included a variable that disaggregated foster care into specific living situations. Based on the foster care arrangement in which the child spent the most time during their stays after age 10, I categorized youth into 1) nonrelative, 2) congregate, and 3) kinship care settings. Nonrelative family care included traditional home environments headed by foster parents who were not biologically related to the child, congregate care represented group home environments, and kinship care reflected children living with biological family members outside of their immediate household.

Finally, I accounted for whether and how the child leaves foster care. Children within this sample exit foster care through reunification with their family, adoption, or other permanency. The latter category includes individuals leaving care by entering detention in state prisons or other institutions, running away, or a family member assuming guardianship. Nationwide, most children exiting foster care via the other permanency category do so through guardianship, while much smaller numbers exit by becoming incarcerated or running away (U.S. Children’s Bureau, 2023). By allowing for the child’s final exit type, the models consider possible associations between how a person leaves foster care and economic security outcomes (Font et al., 2018; Ryan et al., 2016).

3.5. Sample composition

Table 1 illustrates demographic information for both the complete and matched samples. Those retained in the matched sample were more likely to be Black, have their parent named as the perpetrator, and have the allegation of maltreatment substantiated. In the matched sample, 40% were non-Hispanic White, 32% were non-Hispanic Black, 15% were Hispanic, and 13% were of an other race. Regarding the type of maltreatment alleged, neglect (54%) was most common, followed by physical abuse (32%), other maltreatment (17%), and sexual abuse (12%) allegations. For 88% of youth in the matched data, a parent was listed as a perpetrator.

Table 1.

Descriptive statistics for the complete and matched samples by level of child protective services involvement during early adolescence.

Matched Sample
Original WADC caseloada Full sample CPS report alone subsampleb Foster care subsample
Proportion/Mean SD Proportion/Mean SD Proportion/Mean SD Proportion/Mean SD
Entered foster care 0.07 0.50 0.00 1.00
Demography
 Gender
  Male 0.47 0.49 0.48 0.50
  Female 0.53 0.51 0.52 0.50
 Race and ethnicity
  Non-Hispanic white 0.54 0.40 0.39 0.42
  Non-Hispanic Black 0.22 0.32 0.33 0.31
  Hispanic 0.12 0.15 0.15 0.14
  Other 0.11 0.13 0.13 0.13
 Born in U.S. 0.92 0.96 0.95 0.96
 Age of mother at first child’s birth 20.79 4.62 20.10 4.42 20.06 4.50 20.15 4.35
 Father identified 0.88 0.94 0.95 0.93
Public assistance use
 Received SSI as childc 0.12 0.20 0.21 0.19
 Proportion of childhood mother received SNAP benefits prior to child reaching age 10d 0.30 0.39 0.40 0.39
Allegations during early adolescence
 Parent perpetrator 0.77 0.88 0.88 0.88
 Sexual abuse 0.22 0.12 0.12 0.12
 Physical abuse 0.36 0.32 0.32 0.31
 Neglect 0.38 0.54 0.54 0.54
 Other maltreatment 0.13 0.17 0.17 0.17
Substantiated allegations during early adolescence
 Sexual abuse 0.08 0.06 0.06 0.06
 Physical abuse 0.03 0.09 0.09 0.09
 Neglect 0.04 0.19 0.18 0.20
 Substantiated maltreatment 0.14 0.33 0.31 0.34
Birth cohort
 1994 0.51 0.51 0.51 0.51
 1995 0.49 0.49 0.49 0.49
Observations 9,522 1,252 626 626
Open in a new tab
a

Wisconsin Administrative Data Core

b

Child protective services (CPS)

c

Supplemental Security Income (SSI)

d

Supplemental Nutrition Assistance Program (SNAP

4. Results

4.1. Descriptive Statistics

Unadjusted descriptive statistics provide insight into the experiences and outcomes of young people included within the study. As noted in Figure 1, most sample participants had low economic security at age 23. Children who entered foster care earned $850.80 per month, while those who received a CPS report alone earned $856.10. At age 23, approximately 40% of both groups received SNAP and about two-thirds of each group were employed.

Figure 1.

Figure 1.

Open in a new tab

Matched sample’s unadjusted means of receiving SNAP receipt, obtaining employment, and average monthly earnings at age 23.

4.2. Regression Results

In Table 2, I examined variation in the odds of employment by type of CPS involvement in early adolescence while accounting for key covariates. Entrance into foster care during early adolescence did not predict SNAP receipt (OR 0.95, 95% CI 0.74, 1.22) or employment (OR 1.09, 95% CI 0.85, 1.38) relative to no placement. Models also suggested no earnings differences (IRR 1.06, 95% CI 0.84, 1.36) between those in the matched sample who entered care and those who did not.

Table 2.

Regression models estimating the association between child protective services (CPS) involvement and economic security outcomes at age 23 using a matched sample.

Logistic regression models ZINB model a
SNAPb Employment Earnings
OR [95% CI] OR [95% CI] IRR [95% CI]
CPS involvement in early adolescencec
 Child maltreatment report alone (Reference)
 Foster care 0.95 [0.74, 1.22] 1.09 [0.85, 1.38] 1.06 [0.84, 1.36]
Allegations during early adolescence
 Parent perpetrator 0.69 [0.45, 1.05] 0.88 [0.58, 1.36] 1.44 [0.94, 2.21]
 Sexual abuse 0.8 [0.49, 1.32] 1.15 [0.70, 1.89] 1.27 [0.75, 2.16]
 Physical abuse 0.8 [0.53, 1.21] 0.81 [0.55, 1.20] 1.05 [0.80, 1.36]
 Neglect 0.63* [0.42, 0.94] 0.73 [0.50, 1.06] 0.95 [0.72, 1.26]
 Other maltreatment 0.93 [0.62, 1.40] 1.02 [0.70, 1.49] 1.13 [0.88, 1.43]
 Substantiated maltreatment 0.97 [0.73, 1.29] 1.12 [0.85, 1.47] 1.06 [0.86, 1.30]
Demographics
 Gender
  Female (Reference)
  Male 0.36*** [0.28, 0.47] 0.87 [0.68, 1.11] 1.24 [0.99, 1.55]
 Race and ethnicity
  Non-Hispanic white (Reference)
  Non-Hispanic Black 1.49* [1.03, 2.16] 0.71 [0.49, 1.03] 0.75 [0.54, 1.03]
  Hispanic 1.48 [0.99, 2.22] 1.00 [0.66, 1.50] 0.86 [0.66, 1.11]
  Other 1.32 [0.88, 1.97] 0.75 [0.51, 1.11] 0.75 [0.55, 1.02]
 Bio father identified 1.07 [0.63, 1.82] 2.42*** [1.46, 4.01] 1.2 [0.75, 1.91]
 Age of mother at first child’s birth 0.96** [0.93, 0.99] 1.00 [0.97, 1.03] 0.99 [0.97, 1.02]
Public assistance use
 Received SSI as childd 1.67** [1.21, 2.29] 0.43*** [0.32, 0.59] 0.7 [0.44, 1.09]
 Percentage of months mother received SNAP prior to child reaching age 10 2.83*** [1.80, 4.44] 1.60* [1.02, 2.51] 1.26 [0.88, 1.82]
County
 Milwaukee County (Reference)
 Other urban county 0.78 [0.56, 1.08] 1.16 [0.84, 1.61] 1.16 [0.89, 1.51]
 Non-urban county 0.77 [0.52, 1.14] 1.17 [0.80, 1.73] 0.99 [0.74, 1.32]
Birth cohort
 1994 (Reference)
 1995 0.72* [0.56, 0.93] 0.84 [0.66, 1.07] 1.19 [0.97, 1.45]
Lnalpha 1.06 [0.85, 1.30]
Observations 1,252 1,252 1,252
Open in a new tab

Odds ratios (OR) and incident rate ratios (IRR) shown.

*

p<.05

**

p<.01

***

p<.001

a

Zero-inflated negative binomial (ZINB)

b

Supplemental Nutrition Assistance Program (SNAP)

c

Child Protective Services (CPS)

d

Supplemental Security Income (SSI)

Note: Coefficients for born in U.S. suppressed as required by Wisconsin Administrative Data Core license.

In Table 3, I estimated odds of SNAP receipt and employment within a sample restricted to youth who entered foster care during early adolescence. Foster care duration predicted an elevated likelihood of employment (OR 1.18, 95% CI 1.04, 1.33), but it did not have significant associations with either odds of SNAP usage (OR 1.07, 95% CI 0.94, 1.23) or earnings (IRR 0.97, 95% CI 0.90, 1.05). Meanwhile, a higher number of total placements was associated with a greater risk of receiving SNAP (OR 1.38, 95% CI 1.01, 1.90).

Table 3.

Regression models estimating the association between child protective services (CPS) involvement and economic security outcomes at age 23 among young people who entered foster care.

Logistic regression models ZINB model a
SNAPb Employment Earnings
OR [95% CI] OR [95% CI] IRR [95% CI]
Characteristics of time in foster care
 Log total time in care 1.07 [0.94, 1.23] 1.18** [1.04, 1.33] 0.97 [0.90, 1.05]
 Log total placements 1.38* [1.01, 1.90] 0.79 [0.58, 1.08] 0.84 [0.69, 1.01]
Foster care exit type
 Adopted (Reference)
 Reunified 1.68 [0.78, 3.61] 1.18 [0.58, 2.41] 0.88 [0.58, 1.34]
 Other permanency 1.08 [0.47, 2.51] 0.95 [0.43, 2.07] 0.97 [0.60, 1.55]
Modal placement type
 Non-relative foster home (Reference)
 Kinship care 0.92 [0.59, 1.42] 1.68* [1.10, 2.56] 1.23 [0.94, 1.59]
 Congregate care 0.96 [0.49, 1.90] 1.64 [0.87, 3.12] 0.88 [0.58, 1.35]
Allegations during early adolescence
 Parent perpetrator 0.54 [0.29, 1.02] 0.66 [0.35, 1.23] 1.25 [0.89, 1.74]
 Sexual abuse 1.02 [0.48, 2.15] 0.58 [0.28, 1.22] 1.07 [0.70, 1.64]
 Physical abuse 0.65 [0.36, 1.18] 0.57* [0.32, 0.99] 1.00 [0.70, 1.42]
 Neglect 0.69 [0.38, 1.24] 0.71 [0.41, 1.24] 0.9 [0.64, 1.26]
 Other maltreatment 1.09 [0.60, 1.98] 0.85 [0.48, 1.49] 1.15 [0.80, 1.66]
 Substantiated maltreatment 1.12 [0.73, 1.70] 0.89 [0.60, 1.33] 1.14 [0.90, 1.45]
Demographics
 Gender
  Female (Reference)
  Male 0.38*** [0.26, 0.57] 0.80 [0.55, 1.16] 1.05 [0.84, 1.30]
 Race and ethnicity
  Non-Hispanic white (Reference)
  Non-Hispanic Black 1.42 [0.80, 2.50] 0.88 [0.51, 1.51] 0.71* [0.52, 0.98]
  Hispanic 1.85* [1.00, 3.40] 1.18 [0.64, 2.20] 0.96 [0.68, 1.36]
  Other 1.23 [0.67, 2.28] 0.88 [0.49, 1.56] 0.72 [0.50, 1.02]
 Bio father identified 0.85 [0.40, 1.80] 1.63 [0.80, 3.32] 1.07 [0.63, 1.83]
 Age of mother at first child’s birth 0.94* [0.89, 0.99] 1.04 [1.00, 1.09] 0.99 [0.97, 1.02]
Public assistance use
 Received SSI as childc 1.31 [0.80, 2.15] 0.59* [0.37, 0.93] 0.92 [0.68, 1.25]
 Percentage of months mother received SNAP prior to child reaching age 10 3.86*** [1.94, 7.69] 2.03* [1.03, 3.99] 0.91 [0.61, 1.37]
County
 Milwaukee County (Reference)
 Other urban county 0.83 [0.50, 1.38] 1.71* [1.05, 2.78] 0.92 [0.69, 1.23]
 Non-urban county 0.74 [0.40, 1.38] 1.90* [1.05, 3.44] 0.9 [0.63, 1.28]
Birth cohort
 1994 (Reference)
 1995 0.59** [0.40, 0.86] 0.73 [0.51, 1.05] 1.09 [0.87, 1.36]
Lnalpha 1.06 [0.94, 1.20]
Observations 626 626 626
Open in a new tab
a

Zero-inflated negative binomial (ZINB)

b

Supplemental Nutrition Assistance Program (SNAP)

c

Supplemental Security Income (SSI)

Odds ratios (OR) and incident rate ratios (IRR) shown.

*

p<.05

**

p<.01

***

p<.001

Note: Coefficients for born in U.S. suppressed as required by Wisconsin Administrative Data Core license.

4.4. Sensitivity Tests

Sensitivity tests revealed that these findings were robust to a variety of specifications. First, I tested whether the results changed if I excluded those who either spent time in state prisons or who received SSI benefits at age 23; results were similar. (See Supplementary Tables 3 and 4.) Second, I restricted the sample to those with substantiated allegations and the results were largely unchanged. (See Supplementary Table 5.) Third, the results for economic wellbeing remained similar when using listwise deletion. (See Supplementary Tables 6 and 7.) Finally, I determined that accounting for high school graduation and GED completion did not change the null association between foster care entrance and economic security outcomes later in life relative to remaining in the home. (See Supplementary Table 8.) Among youth who entered foster care, high school graduation and GED completion attenuated the relationships between placement moves and quantity of time spent in care with later economic outcomes. (See Supplementary Table 9.) The latter finding suggests that high school graduation and GED completion may mediate the associations between foster care experiences and later economic outcomes.

5. Discussion

Overall, young adults with a history of child abuse allegations and contact with CPS had poor economic outcomes: Average earnings within the matched sample only reached $850.80 for those who entered foster care as early adolescents and $856.10 for those who remained in the home after a CPS report. Foster care placement did not predict lower risk of SNAP receipt, greater odds of employment, or improved earnings in the matched sample. Among youth with a history of foster care, more placements predict greater risk of SNAP usage at age 23, while more time in foster care predicts higher odds of employment.

Multiple factors may explain the association between placement instability and economic security. The placement changes may disconnect young people from needed services. For instance, placement changes may disrupt coordination of medical care and behavioral health supports (Greiner et al., 2019). Second, placement moves that require children to change schools can inhibit academic progress and undermine young people’s educational attainment (Mihalec-Adkins et al., 2020). Academic challenges could lessen young people’s abilities to access good-paying jobs and harm earning potential (Goyette et al., 2021). Finally, frequent placement moves may make it more difficult for children to practice the attachment and relationship skills needed in the workplace (Goyette et al., 2021).

Longer stays in care may reduce exposure to unhealthy environments and offer improved settings for social learning (Font et al., 2021). As Okpych and colleagues (2023) note, long-term relationships between children in foster care and compassionate adults may act as an important mechanism linking extended durations of time in care with improved economic security in young adulthood. The duration of time and the number of placements experienced during stays in care play a key role in predicting later outcomes.

This paper’s null findings show no evidence of variation in economic outcomes among young people who enter care in early adolescence and a matched sample who did not. The results could serve as an important bridge between works suggesting positive later-life outcomes among children who enter foster care in early childhood and less positive outcomes among those placed into foster care in late adolescence (Bald, Doyle, et al., 2022; Gross & Baron, 2022). Scholars suggest that youth who enter foster care in early childhood achieve better education outcomes than those who remain in the home following a maltreatment investigation, while other work depicts late adolescents as facing worse economic prospects after entering care (Bald, Chyn, et al., 2022; Gross & Baron, 2022; Warburton et al., 2014). The null findings among young people who enter care in early adolescence provide additional support for a gradient in which older children might not experience the same benefits from care as their younger peers.

Foster care outcomes may reflect other policy interventions with heterogeneous associations based on a child’s age. For instance, children benefit most when government-supported residential moves from low- to high-opportunity neighborhoods occur prior to age 13 (Chetty et al., 2016). Based on existing research, foster care entrance in early life may carry stronger, more positive associations with later economic security than at other ages (Bald, Doyle, et al., 2022; Gross & Baron, 2022).

5.2. Limitations

Despite using a statewide administrative dataset to track employment, earnings, and SNAP participation at age 23, this study has several limitations. The administrative data only records earnings taxed by the state and the records cannot differentiate between full-time and part-time employment (Brown et al., 2020). Young adulthood remains a period of rapid growth and transition within the labor force. Although earnings trajectories can change as young adults complete postsecondary training, early income deficits typically widen over the life course (Killewald & Bryan, 2018).

Though the WADC only features statewide CPS records beginning in 2004, preventing the use of controls accounting for early life maltreatment reports prior to age 10, other variables measure early life experiences (Brown et al., 2020). Samples are matched on an array of characteristics, including maternal and paternal earnings prior to the child reaching age 10. Even after matching, the main models control for maternal receipt of SNAP prior to the child reaching age 10 and whether officials could identify the child’s biological father, further reducing possible bias (Bai & Clark, 2019).

Next, these administrative data do not have information on sample members’ enrollment in post-secondary education. Future research should examine how foster care entrance during early adolescence is associated with graduation from two-year community colleges, technical schools, and four-year universities. It seems likely that post-secondary enrollment mediates the association between CPS involvement in early adolescence and later life earnings and public assistance use.

This study only examines associations between early adolescent CPS involvement and later outcomes among a cohort of Wisconsin children who entered foster care compared to a similar group of CPS-involved youth who stayed at home. At the cost of generalizability, however, the study can better answer policy relevant questions about the economic corollaries of foster care using highly detailed, though geographically narrow data. Wisconsin’s child welfare system might not reflect those of other states or countries (Font et al., 2018) and the observational nature of the study prohibits identifying the findings as causal.

6. Recommendations and Conclusions

This study indicates that young adults with histories of CPS involvement during early adolescence face adverse economic outcomes during young adulthood. I use multivariate regression models to identify variations in employment status, earnings, and SNAP participation during the transition to adulthood within a matched sample of 1,252 Wisconsin youth who interacted with CPS as early adolescents. Children who entered foster care exhibited few differences in employment, earnings, or SNAP receipt relative to those who were the subjects of a CPS report alone. Although heterogeneity characterized individual experiences in foster care, overall, findings suggest that foster care places early adolescents in its care at neither greater advantage nor disadvantage for young adult economic outcomes relative to a highly similar counterfactual population.

This study provides insights for policymakers and CPS practitioners. First, the negative association between total placements and economic security outcomes underscores the importance of implementing evidence-based practices to reduce unnecessary placement changes (Akin et al., 2021; Goyette et al., 2021). Although children with multiple placements often have greater behavioral health needs relative to their peers with fewer placements (Akin et al., 2021), CPS should work to limit the number of changes. Agencies should screen children for behavioral health challenges when they first enter care and offer prompt referrals to specialists (Akin et al., 2021). Officials should also require foster parents to participate in regular trainings on trauma-informed care (Lotty et al., 2020). The instruction can help foster parents constructively navigate a child’s behavioral challenges and limits the risk of placement changes. Evidence suggests that trainings for foster parents are most effective when programs provide instruction over extended periods (Lotty et al., 2020; Roberts et al., 2016). Even after trainings, agencies should continue to ensure that foster parents have access to community resources to support both themselves and the children in their care (Lotty et al., 2020). Foster parents who report greater levels of social support also indicate they experience less parenting stress, a key predictor of placement changes (Leathers et al., 2019). Promoting placement stability during early adolescence may not only avoid disruptions in schooling and medical care but also help youth build attachment skills later used on the job (Goyette et al., 2021; Greiner et al., 2019).

Second, states might consider broadening eligibility for services typically directed toward foster youth who exit at age 18. The programs include special access to Medicaid and tuition vouchers (Bullinger & Meinhofer, 2021; Hanson et al., 2023). The initiatives provide support as young people leave foster care and enter adulthood. In the future, states should look to expand the programs to other youth with CPS involvement during adolescence (Ryan et al., 2016). These actions may improve young people’s wellbeing at age 23.

Finally, economic security outcomes offer only one set of benchmarks with which to gauge young people’s wellbeing. Future scholarship should consider the association between foster care entrance during early adolescence and education outcomes, criminal justice involvement, and health in young adulthood. Additional evidence could inform the development of new services for children who experience maltreatment.

Supplementary Material

1

Acknowledgements

This material is based upon work supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01HD095946, T32HD007514, P50HD551411, and P2CHD041025). I thank the Wisconsin Departments of Children and Families, Corrections, Health Services, Public Instruction, and Workforce Development for the use of data but acknowledge that these agencies do not certify the accuracy of the analyses. I am grateful to Sarah Font, Molly Martin, Andrew Fenelon, Alex Chapman, and participants at the Population Association of America’s 2022 conference for comments on earlier drafts of this paper.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

CRediT Authorship Contribution statement

Michael Caniglia – Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing, Writing – original draft

References

  1. Akers RL (1977). Deviant behavior: A social learning approach (2d ed.). Wadsworth Publishing Co. [Google Scholar]
  2. Akin BA, Collins-Camargo C, Strolin-Goltzman J, Antle B, Nathan Verbist A, Palmer AN, & Krompf A (2021). Screening for trauma and behavioral health needs in child welfare: Practice implications for promoting placement stability. Child Abuse & Neglect, 122, 105323. 10.1016/j.chiabu.2021.105323 [DOI] [PubMed] [Google Scholar]
  3. Arnett JJ (2014). Emerging Adulthood: The Winding Road from the Late Teens Through the Twenties (2nd ed.). Oxford University Press. [Google Scholar]
  4. Baglivio MT, Wolff KT, Piquero AR, Bilchik S, Jackowski K, Greenwald MA, & Epps N (2016). Maltreatment, Child Welfare, and Recidivism in a Sample of Deep-End Crossover Youth. Journal of Youth and Adolescence, 45(4), 625–654. 10.1007/s10964-015-0407-9 [DOI] [PubMed] [Google Scholar]
  5. Bai Y, & Clark MH (2019). Propensity Score Methods and Applications. SAGE Publications, Inc. https://methods.sagepub.com/book/propensity-score-methods-and-applications [Google Scholar]
  6. Bald A, Chyn E, Hastings J, & Machelett M (2022). The Causal Impact of Removing Children from Abusive and Neglectful Homes. Journal of Political Economy, 130(7), 1919–1962. 10.1086/719856 [DOI] [Google Scholar]
  7. Bald A, Doyle JJ, Gross M, & Jacob BA (2022). Economics of Foster Care. Journal of Economic Perspectives, 36(2), 223–246. 10.1257/jep.36.2.223 [DOI] [Google Scholar]
  8. Bauer L, & Thomas JL (2019, December 15). Throwaway Kids: Star investigation reveals stark outcomes for America’s foster care children. The Kansas City Star. https://www.kansascity.com/news/special-reports/article238206754.html [Google Scholar]
  9. Brown P, Thornton K, Ross D, Smith JA, & Wimer L (2020). Technical Report on Lessons Learned in the Development of the Institute for Research on Poverty’s Wisconsin Administrative Data Core. Institute for Research on Poverty. https://www.irp.wisc.edu/wp/wp-content/uploads/2020/08/TechnicalReport_DataCoreLessons2020.pdf
  10. Bullinger LR, & Meinhofer A (2021). The Affordable Care Act Increased Medicaid Coverage Among Former Foster Youth. Health Affairs, 40(9), 1430–1439. 10.1377/hlthaff.2021.00073 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cancel S (2021, September 16). I Will Never Forget That I Could Have Lived With People Who Loved Me. The New York Times. https://www.nytimes.com/2021/09/16/opinion/foster-care-children-us.html [Google Scholar]
  12. Cansler E, Updegraff KA, & Simpkins SD (2012). Mexican American Seventh Graders’ Future Work and Family Plans: Associations With Cultural Experiences and Adjustment. The Journal of Early Adolescence, 32(3), 313–339. 10.1177/0272431610397660 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chetty R, Hendren N, & Katz LF (2016). The Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment. American Economic Review, 106(4), 855–902. 10.1257/aer.20150572 [DOI] [PubMed] [Google Scholar]
  14. Font SA, Berger LM, Cancian M, & Noyes JL (2018). Permanency and the Educational and Economic Attainment of Former Foster Children in Early Adulthood. American Sociological Review, 83(4), 716–743. 10.1177/0003122418781791 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Font SA, Berger LM, Slepicka J, & Cancian M (2021). Foster Care, Permanency, and Risk of Prison Entry. Journal of Research in Crime and Delinquency, 00224278211001566. 10.1177/00224278211001566 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Font SA, & Gershoff ET (2020). Foster Care and Best Interests of the Child: Integrating Research, Policy, and Practice. Springer International Publishing. 10.1007/978-3-030-41146-6 [DOI] [Google Scholar]
  17. Font SA, & Kim HW (2021). Sibling Separation and Placement Instability for Children in Foster Care. Child Maltreatment, 10775595211012482. 10.1177/10775595211012482 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Font SA, & Potter MH (2019). Socioeconomic Resource Environments in Biological and Alternative Family Care and Children’s Cognitive Performance*. Sociological Inquiry, 89(2), 263–287. 10.1111/soin.12262 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Galambos NL, & Sears HA (1998). Adolescents’ Perceptions of Parents’ Work and Adolescents’ Work Values in Two-Earner Families. The Journal of Early Adolescence, 18(4), 397–420. 10.1177/0272431698018004004 [DOI] [Google Scholar]
  20. Goyette M, Blanchet A, Esposito T, & Delaye A (2021). The role of placement instability on employment and educational outcomes among adolescents leaving care. Children and Youth Services Review, 131, 106264. 10.1016/j.childyouth.2021.106264 [DOI] [Google Scholar]
  21. Greiner MV, Beal SJ, Dexheimer JW, Divekar P, Patel V, & Hall ES (2019). Improving Information Sharing for Youth in Foster Care. Pediatrics, 144(2), e20190580. 10.1542/peds.2019-0580 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gross M, & Baron EJ (2022). Temporary Stays and Persistent Gains: The Causal Effects of Foster Care. American Economic Journal: Applied Economics, 14(2), 170–199. 10.1257/app.20200204 [DOI] [Google Scholar]
  23. Hanson DA, Pergamit M, Packard Tucker L, Thomas KA, & Gedo S (2023). Do Education and Training Vouchers Make a Difference for Youth in Foster Care? Child and Adolescent Social Work Journal, 40(2), 163–176. 10.1007/s10560-022-00896-8 [DOI] [Google Scholar]
  24. Henderson W (2019). Child and Family Services Plan 2020–2024: The Final Five-Year Report. Wisconsin Department of Children and Families. https://dcf.wisconsin.gov/files/cqi-cfsr/pdf/plans/cfsp.pdf
  25. Johnson MK, Mortimer JT, & Heckhausen J (2020). Work Value Transmission From Parents to Children: Early Socialization and Delayed Activation. Work and Occupations, 47(1), 83–119. 10.1177/0730888419877445 [DOI] [Google Scholar]
  26. Killewald A, & Bryan B (2018). Falling Behind: The Role of Inter- and Intragenerational Processes in Widening Racial and Ethnic Wealth Gaps through Early and Middle Adulthood. Social Forces, 97(2), 705–740. 10.1093/sf/soy060 [DOI] [Google Scholar]
  27. Leathers SJ, Spielfogel JE, Geiger J, Barnett J, & Vande Voort BL (2019). Placement disruption in foster care: Children’s behavior, foster parent support, and parenting experiences. Child Abuse & Neglect, 91, 147–159. 10.1016/j.chiabu.2019.03.012 [DOI] [PubMed] [Google Scholar]
  28. Long JS, & Freese J (2014). Regression Models for Categorical Dependent Variables Using Stata (Third). Stata Press. [Google Scholar]
  29. Lotty M, Bantry-White E, & Dunn-Galvin A (2020). The experiences of foster carers and facilitators of Fostering Connections: The Trauma-informed Foster Care Program: A process study. Children and Youth Services Review, 119, 105516. 10.1016/j.childyouth.2020.105516 [DOI] [Google Scholar]
  30. McEwen CA, & McEwen BS (2017). Social Structure, Adversity, Toxic Stress, and Intergenerational Poverty: An Early Childhood Model. Annual Review of Sociology, 43(1), 445–472. 10.1146/annurev-soc-060116-053252 [DOI] [Google Scholar]
  31. Meyer BD, & Mittag N (2019). Using Linked Survey and Administrative Data to Better Measure Income: Implications for Poverty, Program Effectiveness, and Holes in the Safety Net. American Economic Journal: Applied Economics, 11(2), 176–204. 10.1257/app.20170478 [DOI] [Google Scholar]
  32. Mihalec-Adkins BP, Christ SL, & Day E (2020). An exploration of placement-related psychosocial influences on school engagement among adolescents in foster care. Children and Youth Services Review, 108, 104616. 10.1016/j.childyouth.2019.104616 [DOI] [Google Scholar]
  33. Miranda M, Molla E, & Tadros E (2019). Implications of Foster Care on Attachment: A Literature Review. The Family Journal, 27(4), 394–403. 10.1177/1066480719833407 [DOI] [Google Scholar]
  34. Mortimer JT (2021). Agency, Linked Lives, and Historical Time: Evidence from The Longitudinal Three Generation Youth Development Study. Longitudinal and Life Course Studies : International Journal, 13(2), 195–216. 10.1332/175795921X16398283564306 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. NASEM. (2019). The Promise of Adolescence: Realizing Opportunity for All Youth. National Academies of Sciences, Engineering, and Medicine. 10.17226/25388 [DOI] [PubMed] [Google Scholar]
  36. Okpych NJ, Park S (Ethan), Powers J, Harty JS, & Courtney ME (2023). Relationships That Persist and Protect: The Role of Enduring Relationships on Early-Adult Outcomes among Youth Transitioning Out of Foster Care. Social Service Review, 000–000. 10.1086/724736 [DOI] [Google Scholar]
  37. Potter MH, & Font SA (2019). Parenting influences on adolescent sexual risk-taking: Differences by child welfare placement status. Children and Youth Services Review, 96, 134–144. 10.1016/j.childyouth.2018.11.038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Roberts R, Glynn G, & Waterman C (2016). ‘We know it works but does it last?’ The implementation of the KEEP foster and kinship carer training programme in England. Adoption & Fostering, 40(3), 247–263. 10.1177/0308575916657956 [DOI] [Google Scholar]
  39. Rosenbaum PR, & Rubin DB (1983). The central role of the propensity score in observational studies for causal effects. 70(1), 41–55. [Google Scholar]
  40. Ryan JP, Perron BE, & Huang H (2016). Child Welfare and the Transition to Adulthood: Investigating Placement Status and Subsequent Arrests. Journal of Youth and Adolescence, 45(1), 172–182. 10.1007/s10964-015-0305-1 [DOI] [PubMed] [Google Scholar]
  41. Stewart CJ, Kum H-C, Barth RP, & Duncan DF (2014). Former foster youth: Employment outcomes up to age 30. Children and Youth Services Review, 36, 220–229. 10.1016/j.childyouth.2013.11.024 [DOI] [Google Scholar]
  42. Turney K, & Wildeman C (2017). Adverse childhood experiences among children placed in and adopted from foster care: Evidence from a nationally representative survey. Child Abuse & Neglect, 64, 117–129. 10.1016/j.chiabu.2016.12.009 [DOI] [PubMed] [Google Scholar]
  43. U.S. Children’s Bureau. (2018a). Helping Youth Transition to Adulthood: Guidance for Foster Parents [Factsheet for Families]. U.S. Department of Health and Human Services, Administration for Children and Families. [Google Scholar]
  44. U.S. Children’s Bureau. (2018b). Home Study Requirements for Prospective Foster Parents [Factsheet for Families]. U.S. Department of Health and Human Services, Administration for Children and Families. [Google Scholar]
  45. U.S. Children’s Bureau. (2021). Adoption & Foster Care Statistics. U.S. Department of Health and Human Services: Administration for Children and Families. https://www.acf.hhs.gov/cb/research-data-technology/statistics-research/afcars [Google Scholar]
  46. U.S. Children’s Bureau. (2023). Adoption and Foster Care Analysis and Reporting System (AFCARS) FY 2021 data (29). U.S. Department of Health and Human Services, Administration for Children and Families. https://www.acf.hhs.gov/sites/default/files/documents/cb/afcars-report-29.pdf [Google Scholar]
  47. V.R. & B.R. v. County of San-Diego, No. 318594 (Juvenile Court of the State of California: County of San Diego 2023). http://www.caichildlaw.org/Misc/FY_Survivor_Benefit_Complaint.pdf
  48. Warburton WP, Warburton RN, Sweetman A, & Hertzman C (2014). The Impact of Placing Adolescent Males into Foster Care on Education, Income Assistance, and Convictions. Canadian Journal of Economics/Revue Canadienne d’économique, 47(1), 35–69. 10.1111/caje.12064 [DOI] [Google Scholar]
  49. Zielinski DS (2009). Child maltreatment and adult socioeconomic well-being. Child Abuse & Neglect, 33(10), 666–678. 10.1016/j.chiabu.2009.09.001 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2024264-supplement-1.docx (94.2KB, docx)

RESOURCES