How do academic and career services affect employment, education, and disability benefit receipt in the transition to adulthood for youth with disabilities who have aged out of foster care?

Abstract

The transition to adulthood is an important process with implications for inequality. Both those with disabilities and those who age of out of foster care are vulnerable during this transition. This project examines the intersection of these groups, exploring employment, education, and disability benefit receipt, the supports these youth receive, and how these supports may mitigate risk in this transition. Findings suggest that those with emotional or mental disabilities face employment risks in the transition to adulthood, and that services may mitigate some of this risk and present unique benefits for youth with disabilities. For example, the odds of employment are greater and the odds of receiving disability benefits are lower among those who receive academic supports (or_FTE=1.3, or_SSI/SSDI==0.7), post-secondary supports (or_FTE=1.3, or_SSI/SSDI=0.8), education financial assistance (or_PTE=1.3, or_SSI/SSDI=0.7), and career services (or_PTE=1.3, or_SSI/SSDI=0.8). These services should be prioritized for funding, especially education financial assistance as a lower proportion of youth with disabilities receive than their non-disabled peers.

The transition to adulthood is an important period of the life course marked by increasing independence and a density of challenges and opportunities on multiple fronts (Elder, 1998). Young people at this stage make decisions regarding education, employment, parenthood, and finances, which all have lifelong implications (Buchmann & Steinhoff, 2017; Hogan & Astone, 1986; Sawyer et al., 2018; Shanahan, 2000). Moreover, they navigate this time in the context of changing social roles and responsibilities (Schulenberg & Schoon, 2012; Schulenberg et al., 2004; Staff et al., 2010). Pathways to adulthood, and the timing and ordering of these decisions and transitions are also strongly linked with family and social context (Johnson & Benson, 2012; Kendig et al., 2014). Family resources, support, and guidance, or lack thereof, can constrain prospects or provide opportunities for individuals navigating the transition to adulthood (Fomby & Bosick, 2013; Lee & Mortimer, 2009; Melby et al., 2008; Schoeni & Ross, 2005).

One group that faces particular difficulty in the transition to adulthood are alumni of the foster care system (Courtney et al., 2001; Keller et al., 2007). These youth often navigate this transition abruptly and with receding support. Typically, adolescents rely on their parents for housing at various points during the transition to adulthood, particularly during the pursuit of higher education or as a safety net when shifting into new employment roles (Melby et al., 2008). Moreover, families often provide financial support and counsel during this stressful transition (Schoeni & Ross, 2005). However, youth who have aged out of foster care must often navigate this transition without such family assets and resources.

Among youth aging out of foster care, the transitional experience of those with disabilities merits particular attention (Harwick et al., 2017; Lee et al., 2022). Having a disability is associated with difficulty in the transition to adulthood in its own right, as those with disabilities are less likely to pursue higher education, encounter difficulty in securing employment, and struggle to gain independence (Janus, 2009). Youth with disabilities often need additional support during this time, including financial support, guidance in navigating discrimination and complex support systems, housing support, and educational support (Pascall & Hendey, 2004). In fact, one study which interviewed youth with disabilities during the transition to adulthood identified “exceptional parents” as key to finding success (Pascall & Hendey, 2004). However, not all youth with disabilities have access to such supports, compounding risk associated with emancipation from the foster care system (Harwick et al., 2017).

This study focuses on youth with disabilities who are aging out of the foster care system, investigating employment, education, disability benefit receipt, and service use in the transition to adulthood. This study uses linked administrative data available through the National Data Archive on Child Abuse and Neglect to estimate the association between disability and employment, and educational enrollment at age 21. Next, this study examines differences in service use during the transition to adulthood for those who do and do not have disabilities. Last, the potential effectiveness of boosting employment, educational enrollment, and disability benefit receipt are assessed for two types of services—those aimed at supporting a) education and b) employment. By examining these domains, this study will illuminate the context of employment, education, disability benefit receipt, and service use in the transition to adulthood for youth with disabilities who are aging out of foster care. Furthermore, the results of this study will provide insight as to the relationships between services aimed at supporting employment and education and employment, education, and SSI benefits at age 21 for youth with disabilities who have aged out of foster care. This will have important implications for service providers and policy makers who develop, prioritize, and implement services to help these youth.

Background

More than 650,000 youth were served by the foster care system in 2019 (The AFCARS Report, 2020). Estimates vary regarding the prevalence of disability among those involved in the child welfare system, but there is consensus that youth with disabilities are overrepresented in this system (The risk and prevention of maltreatment of children with disabilities, 2018). Children with disabilities are more than three times as likely to experience abuse or neglect than their peers (Jones et al., 2012; Sullivan & Knutson, 2000) and a third of youth in the foster care system have a disability (Slayter, 2016b). Additionally, youth with disabilities are more likely to experience serious injury because of maltreatment (Sedlak et al., 2010) and are more likely to be in restrictive placements, report higher levels of hopelessness, and experience more instability in placement (Hill, 2012; Lee et al., 2018; Slayter, 2016a; Slayter & Springer, 2011). Given the disproportionate risk of involvement in the child welfare system for those with disabilities, there is a need for expanded study of disability and child welfare involvement.

Life course theory

Informed by life course theory, which considers the sequence of socially defined events and positions across one’s life span, this study focuses on the transition to adulthood as a key moment of changing social roles and when a density of such events of occur (Elder, 1977). For all youth, decisions related to life course events and role transitions are rife with uncertainty and have lifelong consequences for wellbeing, socioeconomic standing, and success (Elder, 1998). For youth with disabilities who age out of foster care, these risks are greater as they do not necessarily have the same safety net as their peers. Further, life course theory emphasizes the importance of recognizing an individual’s historical context (Elder, 1998). For youth with disabilities who age out of foster care, their experiences with child protective services and in foster care are important to include in these analyses.

Disability and aging out

While the difficulty foster care youth face in the transition to adulthood has gained national attention, the unique experiences of those with disabilities who age out has received notably less consideration (Geenen & Powers, 2007; Lee et al., 2022). Youth with disabilities are often excluded from large-scale studies examining the transition to adulthood for foster care youth (for examples see: Courtney et al., 2007; Pecora, 2005). A few critical studies have narrowed in on this population, evidencing that youth with disabilities do face considerable difficulty in the transition to adulthood. One such study was an evaluation of the Independent Living Programs, which compared those with an identified disability to those without (Westat, 1991). While this evaluation is now quite dated, they found convincing evidence that youth with disabilities did worse than youth without on a variety of domains (including employment, educational attainment, social support, and independence) (Westat, 1991). More recent studies have found a consistent pattern—youth with disabilities, broadly defined, do worse in the transition to adulthood than their non-disabled peers (Cheatham et al., 2020; Geenen & Powers, 2007; Lee et al., 2022; McCauley, 2021).

Recent studies focusing on youth with disabilities who have aged out of foster care found that those with disabilities face a greater risk of incarceration and homelessness in the transition to adulthood (Lee et al., 2022; McCauley, 2021). McCauley found differential experiences in the child welfare system accounted for some of the increased risks associated with emotional or mental related disabilities (2021). Lee and colleagues found that there were also specific resiliency factors, including employment and school enrollment, which reduced the risk of incarceration and homelessness (2022). Interestingly, both studies also found evidence of differential risk for youth with different disability types and authors argued that disability type needs to be disaggregated when examining risk and resiliency in the transition to adulthood (Lee et al., 2022; McCauley, 2021).

Taken together this work suggests that youth with disabilities face unique risks in the transition to adulthood but leaves unresolved how different services may mitigate or exacerbate these risks for those with different disability types. Building on this work, the present study will focus on employment, education, and disability benefit receipt in the transition to adulthood, examine service use among those who age out of the foster care system, and then examine the extent to which some of these services may boost employment, educational enrollment, and independence.

Employment, education, and disability benefit receipt

Three key areas for youth with disabilities in the transition to adulthood are employment, education, and disability benefit receipt. Employment has many benefits such as increasing independence, social connection, and self-confidence (Bal et al., 2017; Bevan et al., 2013; Saavedra et al., 2015), however, individuals with disabilities are more likely to be unemployed or to not transition into the labor market (Geenen & Powers, 2007). A recent study examined employment and education among this population and found that youth with disabilities, especially emotional disabilities, were less likely to be employed (Cheatham et al., 2020). The results of this study suggest that those with disabilities face risks related to employment and independence in the transition to adulthood.

Education is also important for the long-term success and wellbeing of youth with disabilities and foster care alumni. Youth with foster care history are less likely to enroll in and complete education (Geiger & Beltran, 2018). The same is true for youth with disabilities, and educational attainment can moderate the relationship between disability and employment, bolstering employment prospects (McCauley, 2019). A study examining education among youth with disabilities who had aged out of foster care found persistent evidence that those with emotional disabilities had lower educational attainment and were less likely to be engaged in education (Cheatham et al., 2020). Research has shown that additional supports and extended foster care services are associated with higher odds of college enrollment, although not completion (Okpych & Courtney, 2020). Boosting employment and education for those with disabilities would have positive implications for more than just income, it could also provide social connection and have health-promoting consequences for wellbeing.

The average age of disabled worker beneficiaries has decreased since the 1960s, and the number of youth receiving Supplemental Security Income (SSI) has increased over time, reaching over one million in 2019 (Fast Facts & Figures About Social Security, 2020). While outside the scope of this study, it is important to note that youth in foster care who are enrolled in disability benefits do not directly receive these benefits while in custody, they are used to off-set placement costs by the child welfare system. Once emancipated from custody, youth can directly receive benefits. However, many youths who receive disabilities benefits will live in poverty throughout their lives (She & Livermore, 2008). Younger awardees are likely to rely on SSA benefits for much of their working-age life and participate in the labor force only infrequently (Ben-Shalom & Stapleton, 2015).

The potential role of services aimed at supporting education and employment services

To help youth who are likely to age out of foster care the Foster Care Independence Act of 1999 authorized support for youth in the transition to adulthood (Fernandes-Alcantara, 2019). The Chafee statute was amended in the early 2000s to include Education and Training Voucher program which focused on boosting education and employment training (Fernandes-Alcantara, 2019). In exchange for receiving financial support for the John H. Chafee Foster Care Program for Successful Transition to Adulthood, states and territories collect and make available data on the services that youth use and their success in the transition to adulthood.

While youth with disabilities who have aged out of foster care face unique risks in the transition to adulthood, there are a number of resiliency factors that may help mitigate these risks. Harwick and colleagues have called for a strengths-based approach to understanding the transition to adulthood for this population (Harwick et al., 2017). A recent evaluation of an intervention to boost self-determination found equal effectiveness for youth who age out of foster care with and without disabilities, suggesting that programs which are designed to support the transition to adulthood for all youth who age out can hold promise for this sub-population (Blakeslee et al., 2020). The present study will examine which services youth with disabilities who age out use and the potential effectiveness of two types of these services: services aimed at supporting employment and education. This study narrows in on these two types of services due to their likely association with employment and education, key measures of success in the transition to adulthood which hold special importance for youth with disabilities.

The first set of services relate to supporting the education. These services are described in detail when discussing the data. In the general population, educational attainment can moderate the relationship between disability and employment, where those with higher levels of education experience a smaller decrease in the likelihood of employment due to having a disability (McCauley, 2019). Youth with disabilities are more likely to be enrolled in special education, however little is known about the experiences of these youth beyond the fact that they face complex challenges and high levels of uncertainty (Foley & Pang, 2006; Miller et al., 2020; Quest et al., 2012). However, if youth who age out of foster care were receiving special education services and had a committed adult supporting them, they described taking more confident action toward goals (Quest et al., 2012) and foster care alumni with mental health stressors attending college emphasized the importance of supports for their success (Miller et al., 2020).

The second set of services are related to supporting the employment prospects of youth. These services, which I will describe when discussing the data, are aimed at promoting work. The existing literature on the effectiveness of work-study type programs at promoting work for youth with disabilities is mixed. Vocational rehabilitative work programs for specific populations have demonstrated evidence of effectiveness (Cook & Rozzano, 2000; Dutta et al., 2008; Straaton et al., 1995). More broadly, however, these education and work promoting services have not been evaluated empirically at a population level.

A better understanding of the services aimed at supporting employment and education which youth with disabilities who age out of foster care use and the potential effectiveness of these services will provide clarity regarding the current state of support and point to types of services that promote employment and independence for this population. Supports and services in the transition to adulthood are important, yet access to and use of these supports vary by cultural and contextual factors (Arnau-Sabatés et al., 2021). Additionally, existing research has found that independent living programs aimed to support the transition to adulthood youth who age out of foster care often do not have disability-specific services, and therefore the unique needs of those with disabilities who age out of care are not met (Hill, 2010). This suggests the importance of evaluating the potential effectiveness of existing uniform programs for this population.

While research has explored educational supports and services for youth with disabilities and youth who age out of foster care, the intersection of these experiences has received comparatively less attention (for recent exceptions see (Cheatham et al., 2020; Lee et al., 2022; McCauley, 2021)) despite the particularly high risk that these youth face (Blakeslee et al., 2013; Powers et al., 2012). This project will help shed light on what services this vulnerable group uses and how those various services are associated with employment, educational enrolment, and disability benefit receipt in the transition to adulthood. This is particularly important given the steady rise in disability receipt and decline in the employment rate for those with disabilities (Autor & Duggan, 2010; Mann & Stapleton, 2011). Better understanding how those who face the highest risk can achieve employment and independence in the transition to adulthood can inform policy decisions regarding what types of programs to support and help practitioners reach the most at-risk youth. Additionally, better understanding how to support the employment and independence of these marginalized youth during this pivotal transition time can have long term implications for health, independence, and social inclusion across the life course.

Research objectives

• What is the association between disability type and the odds of employment, educational enrollment, and disability benefit receipt at age 21?

• What services do youth with disabilities who are likely to age out of foster care receive? And are there differences in service receipt between those with and without disabilities?

• What is the association between receiving education and employment related services during the transition to adulthood and the odds of employment, educational enrollment, and disability benefit receipt at age 21 for youth with disabilities who are likely to age out of foster care?

Materials and methods

I use data from the National Youth in Transition Database (NYTD) from the National Data Archive on Child Abuse and Neglect (NDACAN) to explore employment, education, and disability benefit receipt in the transition to adulthood for youth with disabilities who emancipate from the foster care system. These data consist of two datasets, the Services File, and the Outcomes File. The Services File is an administrative data set collected at 6-month intervals and reflects which services or programs youth participate in during that 6-month period. The NYTD data are a cohort based longitudinal survey of youth who are identified as likely to age out of foster care without finding permanent placement. The first wave is collected when the youth are 17 years old (2011) and follow-up surveys are conducted when youth are 19 (2013) and 21 years old (2015). NDACAN has two additional administrative databases which are used in this study to develop covariates and one of the independent variables; 1) the National Child Abuse and Neglect Data System Child File (NCANDS) which has data on children’s child protective service (CPS) history, and 2) the Adoption and Foster Care Analysis and Reporting System (AFCARS) which has data on children’s foster care histories. Using a unique identifier, these four datasets are linked to connect youth who age out of foster care to their service use, their histories in the foster care system, and their child protective histories.

Independent Variables

Disability

The independent variable in the first and second set of analyses are the disability status and disability type of the youth. These indicators are developed from the administrative AFCARS database. Youth are considered to have a disability if their caseworker responded yes to the question “has this child ever been clinically diagnosed with a disability?” as collected in AFCARS.

Disability type is a categorical variable which was developed from a series of questions related to disability. It has five categories: no disability, emotional and/or intellectual disability, physical and/or sensory disability, both types of disability, and other disability. This is a similar categorization as a recent study of a different scope using much of the same data to explore social risk factors in the transition to adulthood for youth with disabilities who have aged out of the foster care system (McCauley, 2021). The disability language used in this manuscript is that used in the administrative data collection unless otherwise noted (for full text see Administration for Children and Families’ Information Memorandum). For more information about the disability classifications refer to the AFCARS Foster Care Code Book (AFCARS Foster Care Annual File, 2019).

Emotional or intellectual disability refers to youth who have intellectual disabilities and/or emotional disabilities. The following text outlines the disabilities which fall into this category and is sourced from the Administration for Children and Families’ Information Memorandum and AFCARS Foster Care Code Book. Intellectual disabilities refer to a youth with significant subaverage general cognitive and motor functioning existing concurrently with deficits in adaptive behavioral manifested during the developmental period that adversely affects a child’s/youth’s socialization and learning. This includes those diagnosed with Downs Syndrome, Borderline Intellectual Functioning, Hydrocephalus, Microcephaly, and all degrees of Intellectual Disability¹. And emotional disability refers to a condition exhibiting one or more of the following characteristics over a long period of time and to a marked degree: an inability to build or maintain satisfactory interpersonal relationships, inappropriate types of behavior or feelings under normal circumstances, a general pervasive mood of unhappiness or depress, or a tendency to develop physical symptoms or fears associated with personal problems. This includes those diagnosed with adjustment disorders, Attention Deficit and Disruptive Disorders (including ADD, ADHD, Conduct Disorders, and Oppositional Defiant Disorder), anxiety disorders (Agoraphobia, Obsessive Compulsive Disorder, Panic Disorder, Phobias, Post-Traumatic Stress Disorder, Separation Anxiety Disorder), Eating Disorders, Impulse Control Disorder, Mood Disorders (Bipolar Disorder, Cyclothymic Disorder, Depressive Disorders, Dysthymic Disorder), Personality disorders (such as Antisocial Personality Disorder, Borderline Personality Disorder, Paranoid Personality Disorder, and Schizoid Personality Disorder), Reactive Attachment Disorder, Schizophrenic and other Psychotic Disorders (such as Delusional Disorder, Psychotic Disorder, and Schizoaffective Disorder), Somatoform Disorder, and Tourette Syndrome.

Physical or sensory disability refers to youth with a visual impairment that may significantly affect educational development or performance, a hearing impairment, whether permanent or fluctuating, that adversely affects educational performance, or a physical condition that adversely affects the child’s day-to-day motor functioning. The following text outlines the disabilities which make up this category and the text is sourced from the Administration for Children and Families’ Information Memorandum and AFCARS Foster Care Code Book. This includes being diagnosed with Blindness and Low Vision, Cataracts, congenital anomaly of the eye, Glaucoma, Diabetic Retinopathy, Retinal Detachment and Defects, Deaf, and Hearing Loss. This also included Arthritis, Brittle Bones/Osteogenesis Imperfects, Cerebral Palsy, Chronic Motor Tic Disorder, Club Foot, Diplegia, Multiple Sclerosis, Muscular Dystrophy, Myasthenia Gravis, Paralysis (including Paraplegic, Quadriplegic, Diplegic), Poliomyelitis, Rheumatoid arthritis, and spina Bifida. Both refers to individuals with both sensory or physical disabilities and emotional or mental disabilities.

Other disability refers to youth with conditions other than those that fall into the aforementioned disability categories but that which require special medical care including chronic illnesses. This category was not developed by the author, it is the designation in the source data and unfortunately cannot be parsed or disaggregated.. Again, this text is sourced from the Administration for Children and Families’ Information Memorandum and AFCARS Foster Care Code Book. This includes Acquired Immunodeficiency Syndrome (AIDS), Aplastic Anemia, Asthma, Autistic Spectrum Disorders (Asperger’s Syndrome, Pervasive Developmental Disorder not otherwise specified, Autistic Disorder), blood disorders which require hospitalization once a month, cancers, Childhood Disintegrative Disorder, Chronic Granulomatous Disease, Cleft palate, Coagulation Defects, Congenital cystic lung, congenital heart anomaly. Crohn’s Disease, Cushing’s Syndrome, Cystic Fibrosis, Diabetes, Immune Disorders, Encephalopathy, Epilepsy, Fetal Alcohol Syndrome, Fetal Drug Addiction, Heart murmur which curtails vigorous activity, heart disease, hemophilia, Hypertension, Human immunodeficiency Disease (HIV), Human T-Cell Lymphotropic Virus-III, Immunodeficiency, Kidney Disease, Klinefelter’s Syndrome, Learning Disability, Leukemia, Liver Disease, Lupus, Malignant Neoplasms, Misplaced facial feature, Organic Brain Syndrome, Pancreatic Disease, Rett Disorder, Sarcomas, Seizure Disorder, Sickle Cell Anemia, Shaken Infant Syndrome, Late Effects of Tuberculosis, and Nutritional deficiency.

Service use

The second and third set of analyses examine service use. The NYTD data include information about which services funded by the Chafee Independence Program youth use. These categories reflect the categories in the data and were not developed by the author. This text is sourced from the Administration for Children and Families’ Information Memorandum and the NYTD Services Code Book The proportion of youth who have used each of the available services by disability status is examined in the second set of analyses. These services include educational attainment supports, career supports, training or education related interventions, financial support, independent living program services, and other services. The following text about what falls into these services categories is sourced from the NYTD Services Code Book and the Administration for Children and Families’ Information Memorandum and additional details about these services are available there (NYTD Services File Code Book, 2018).

Educational attainment supports include special education services, academic supports, and post-secondary supports. Special education services refer to specifically designed instruction to meet the unique needs of a child with a disability. Academic support services are supports designed to help youth complete high school or their General Equivalency Degree (GED), including academic counseling, preparing for GED, tutoring, help with homework, study skills training, literacy supports, and help accessing educational resources. Post-secondary educational supports are services designed to help a youth enter or complete post-secondary education including entrance exam test preparation, counseling about college, information about financial aid and scholarship, help completing college or loan applications, and tutoring during college.

Career supports include career services and employment supports. Career services refers to services that develop a youth’s ability to find, apply for and retain employment, including vocational and career assessment, career exploration and planning, job seeking and job placement support, employment preparation support (such as writing resumes and practice interviewing), job retention support and job coaching, learning to work with employers and other employees, understanding workplace values, and understanding authority and customer relationships. Employment programs or vocational training are programs designed to build a youth’s skills for a specific trade, vocation, or career through classes or on-site training, including apprenticeships, internships, employment programs.

Training or education interventions include budgeting education, housing education, and health education programs. Budget education and financial management includes training on how to live within a budget; how to open and use a checking and savings account; how to balance a checkbook; how to develop consumer awareness and smart shopping skills; how to access information about credit, loans and taxes; and filling out tax forms. Housing education includes training in locating and maintaining housing, food preparation, laundry, housekeeping, living cooperatively, meal planning, grocery shopping, and basic maintenance. Health training programs provide information on hygiene; nutrition; fitness and exercise; first aid; medical and dental benefits; health care resources and insurance; maintaining personal medical records; sex education, abstinence education, and HIV prevention (including family planning); and education and information about the effects and consequences of substance use (including substance avoidance).

Financial support programs include housing financial assistance, education financial assistance, and other financial assistance. Housing financial assistance is a payment that is provided by the State agency for room and board (including rent deposits, utilities, and other start up housing expenses). Education financial assistance is provided by the State agency for education or training including support to purchase textbooks, uniforms, computers, or supplies; tuition assistance; scholarships; payment for educational preparation and support services (such as tutoring); or payment for GED and other education related tests. Other financial assistance programs include any other payments made or provided by the State to help youth live independently.

Independent living services include the independent living assessment and the independent living program. Independent living needs assessments are a systemic procedure to identify youths needs and risks, and then pair them with an independent living program. Independent living programs refer to programs where the youth are living independently under a supervised arrangement by the state. The monitoring is less than 24 hours a day and the youth assume increasing responsibilities over time (such as paying bills, assuming leases, and working with a landlord).

Lastly, other services include mentoring and family services. Mentoring programs are where a youth is connected with an adult for a one-on-one relationship. These relationships may be short-term or long-term, and are often run through school, work, or family. Family support services include information and education about safe and stable families, healthy marriages, spousal communication, parenting, responsible fatherhood, childcare skills, teen parenting, and family violence prevention.

Next, in the third set of analyses, the study narrows in on how services aimed at supporting employment and education are associated with employment, educational enrollment, and disability benefit receipt for those with disabilities. These services are special education services, academic supports, post-secondary supports, career services, employment supports, and education financial assistance. All services are measured as binary variables, with 1 indicating that the service was used and 0 indicating that the service was not used. This study examines services used prior to age 21 on outcomes at age 21.

Dependent variables

There are four dependent variables in this study, employment (full-time and part-time), educational enrollment, and disability benefit receipt. In this study, I examine the odds that youth experience these outcomes at age 21 (Wave 3). Dependent variables are binary, with 1 indicating yes and 0 indicating no. Participants are asked during Wave 3 (age 21) if they were employed in a full-time position (meaning they were employed for a position more than 35 hours a week). If participants answered that they are employed for a position at 35 or more hours a week, they are considered to have FTE. Participants are asked during Wave 3 (age 21) if they were employed in a part-time position (meaning they were employed for a position less than 35 hours a week). If participants answered that they are employed for a position at 35 or fewer hours a week, they are considered to have PTE. Participants are asked about their current enrollment in an educational program. If participants respond affirmatively, they are considered to be enrolled in education. Participants are also asked at Wave 3 if they are receiving disability benefits directly or as a dependent beneficiary at the date of the interview. Disability benefits refer to Supplemental Security Income (SSI) and Social Security Disability Insurance (SSDI). All outcomes, including FTE and PTE, are not exclusive. This means that, for example, participants may have PTE, FTE, and be enrolled in college. It is also important to note that these outcomes, while explored in independent models in this study, are related. For example, having FTE may be more difficult or impossible for those who are enrolled in college or who are receiving disability benefits. More information about these outcomes is available in the NYTD Outcomes Code Book (NYTD Outcomes File Code Book, 2016). Participants who are missing data on the dependent variables will be dropped from the sample for that model.

Covariates

Three levels of control variables are used in this study. First, demographic control variables will be included. This information comes from Wave 1, or the baseline wave, of the NYTD dataset. The baseline of NYTD is collected at age 17. Control variables from NYTD are male gender, race (non-Hispanic White, non-Hispanic Black, Hispanic, non-Hispanic other), and highest grade completed. Also, from the NYTD data, all models include control variables to reflect the outcome at the Wave 1 baseline (i.e., the models estimating the association between services and FTE at age 21 include a control for FTE at age 17) and a control variable to reflect educational enrollment at age 21, as enrollment in an educational program may limit time available for employment.

Second, control variables reflecting youth’s foster care histories are included (AFCARS). These control variables include the total number of foster care placements, the total number of days the youth spent in foster care, if the youth was ever adopted, the caretaker family structure, and if the foster family was receiving disability benefits on behalf of the child. The number of removal times, and the total numbers of days in care are measured as continuous variables. Caretaker family structure is a categorical variable with married couple, unmarried couple, and single parent as the options. Ever adopted, if the family received Medicaid, and if the foster family received disability benefits on behalf of the child are measured as binary variables.

Last, control variables reflecting CPS history will be included (NCANDS). These controls will be the reason for the most recent report (parent related risk, child related risk, abuse, neglect, other), the total number of reports, the total number of victimizations, and if the parent was ever a perpetrator. The number of reports and victimizations are measured as continuous variables and if the parent was ever a perpetrator is measured as a binary variable.

Data management

Prior to analysis, the analytic sample was created. The base of the analytic sample used in this study consists of youth who completed the NYTD data collection (n = 7,077). This represents a portion of the original baseline sample in NYTD (n = 29,104). First, I linked the NYTD outcomes file to the services file. Then, I cleaned and reshaped the AFCARS and NCANDS datasets to link these data to the NYTD dataset for youth who appear in NYTD (i.e. youth who age out of foster care). This created a linked data set that is organized by the participant (wide format) and included information about participant’s outcomes at age 17 and 21, and their foster care and CPS histories. Youth with disabilities often have different CPS and foster care histories, such as differential reunification rates and more restrictive placements (Schmidt et al., 2013; Slayter, 2016a). This makes the ability to adjust for foster care and CPS history important in reducing the threat of unobserved heterogeneity.

Next, I examined missingness in the dataset I created (n=7,077). Youth were excluded from the sample if they had missing data related to their disability status and type (n=165). For the covariates, urbanicity was available for all youth and other covariates had missingness ranging from less than 1% to 24%. I used Stata to conduct chained multiple imputation for missing covariate data using observed variables for disability, urbanicity, and NYTD baseline data from age 17 on history of incarceration, history of homelessness, history of substance abuse treatment, and prior childbearing. I used 10 imputations, which is at the top of the threshold at which time estimator efficiency loses little relative to infinite imputations (Rubin, 1987).

Analytic strategy

This study consists of three types of analyses; 1) an inferential analysis examining how disability type shapes the odds of FTE, PTE, and educational enrollment at age 21, 2) a descriptive analysis examining which services youth with disabilities who age out of foster care use between the ages of 17 and 21, and 3) an inferential analysis estimating the association between two groups of services (those aimed at supporting education and employment) and the odds of FTE, PTE, educational enrollment, and SSI/SSDI disability benefit receipt among those with disabilities at age 21. Analysis one employs a similar approach as another recent study examining disability in the transition to adulthood (McCauley, 2021). For the inferential analyses (analyses one and three), separate logistic regression models are employed for each outcome because the outcome variables are binary. The summary tables of these analyses are presented, and all analyses presented will include all covariates described above as well as a covariate reflecting the outcome measured at baseline (i.e. the model examining PTE at age 21 adjusts for PTE at age 17). Sample sizes vary between the models due to missing data for individual outcomes (FTE, PTE, educational enrollment, and SSI/SSDI disability benefit receipt, n varied by outcome). The descriptive analysis (analysis two, Table 3) shows the proportion of youth with and without disabilities who receive various services during the transition to adulthood as well as the p-value for a mean difference test between the groups. Independent samples t-tests were used to examine these mean differences. This work was granted exemption from the Cornell Institutional Review Board (protocol 2002009384).

Table 3.

Descriptive table showing the proportion of youth with and without disabilities who participated in various services during the transition to adulthood.

	No Disability	Disability	P-Test of Difference
Educational Attainment
Special Education Services	0.17	0.34	0.00
Academic Supports	0.57	0.58	0.58
Post-Secondary Supports	0.43	0.43	0.59
Career Support
Career Services	0.52	0.55	0.01
Employment Supports	0.36	0.39	0.01
Training/Education
Budgeting Education	0.51	0.51	0.98
Housing Education	0.51	0.56	0.00
Health Education	0.48	0.50	0.12
Financial Support
Housing Financial Assistance	0.24	0.21	0.01
Education Financial Assistance	0.39	0.37	0.07
Other Financial Assistance	0.45	0.54	0.00
Independent Living
Independent Living Assessment	0.53	0.62	0.00
Independent Living Program	0.19	0.22	0.01
Other
Mentoring Services	0.41	0.45	0.00
Family Services	0.27	0.32	0.00

Notes: N=6,912. Services are from the NYTD data and refer to the time period between 17 and 21.

Results

Nearly half of the sample of youth aging out of foster care without finding a permanent placement have a disability, as seen in Table 1. The most common disability is an emotional or intellectual disability, with 34% of the sample reporting just an emotional or intellectual disability and another six-percent reporting both an emotional or intellectual disability and a physical or sensory disability. A quarter of the sample reported full time employment (FTE) at age 21, nearly 30% reported part time employment (PTE), a third were enrolled in an educational program, and 13% reported SSI/SSDI disability benefit receipt. These outcomes are not mutually exclusive, meaning that individual participants can report multiple outcomes or no outcomes. In fact, a quarter of the sample reported no PTE, FTE, educational enrollment, or disability benefit receipt, nearly half reported one outcome, and a quarter reported two or more outcomes. The demographic characteristics of the sample, which come from the NYTD data, show that just under half of the sample are male (45%) and the majority are from urban spaces (82%). The sample is 43% White, over a third Black (34%), and just under a fifth Hispanic (18%).

Table 1.

Means and proportions for the sample.

		Mean / Proportion	Standard Error
Disability		0.48
	Emotional/Mental	0.34
	Physical/Sensory	0.03
	Both	0.04
	Other	0.06
Outcomes
	Full Time Employment	0.25
	Part Time Employment	0.28
	Educational Enrollment	0.32
	SSI /SSDI	0.13
Demographics
	Male	0.45
	Race
	White	0.43
	Black	0.34
	Hispanic	0.18
	Other	0.05
	Urbanicity
	Urban	0.82
	Non-Metro	0.16
	Rural	0.02
Foster Care History
	Number of Placements	6.59	0.08
	Total Days in Foster Care	1835	18
	Ever Adopted	0.08
	Medicaid Receipt	0.97
	Parent Marital Status
	Married Couple	0.22
	Unmarried Couple	0.09
	Single Parent	0.56
	Unknown	0.12
	SSI Benefits	0.12
Child Protective History
	Reason for Report
	Parent Related Risk	0.22
	Child Related Risk	0.34
	Abuse	0.21
	Neglect	0.44
	Other	0.31
	Number of Reports	3.23
	Number of Victimizations	1.08
	Parent Perpetrator	0.30

Notes. N=6,229. Outcomes are from the NYTD data, foster care history is from the AFCARS data, and child protective history (CPS) is from the NCANDS child file data. SSI is Supplemental Security Income and SSDI is Social Security Disability Insurance.

In terms of the foster care histories of the sample, which come from the AFCARS data, youth had an average of six placements and spent 1,835 days in foster care. Nearly all were receiving Medicaid, just over a tenth were receiving SSI benefits (12%), and the majority were from single-parent households (56%) with married couples being the second most common caretaking arrangement (22%). Last, when looking at the child protective histories of the sample, which comes from the NCANDS data, the most commonly included report reasons given were neglect (44%) and child related risk (34%). On average, the sample had more than three reports for each youth, one confirmed victimization, and the youth’s parent was listed as a perpetrator for 30% of youth.

Table 2 examines the association between disability type (specifically, an emotional or intellectual disability, a sensory or physical disability, both types of disability, or another type of disability) and outcomes (FTE, PTE, and educational enrollment) at age 21 controlling for the above-described demographic characteristics, foster care histories, and CPS histories. These models also adjust for the reported outcome at baseline (age 17).

Table 2.

The association between disability type and employment and educational enrollment at age 21 among youth who age out of foster care using logistic regression.

	FTE	PTE	Educ Enroll
	Odds Ratio	Odds Ratio	Odds Ratio
Disability Type (comparison group is no dis)
Emotional/Mental Disability	0.76*** (0.06)	0.83* (0.06)	0.90 (0.06)
Sensory/Physical Disability	1.08 (0.20)	0.97 (0.17)	1.25 (0.21)
Both Disabilities	0.68 (0.12)	0.75 (0.13)	1.01 (0.16)
Other Disability	0.88 (0.12)	0.87 (0.11)	1.18 (0.14)
Controls
Demographics	Yes	Yes	Yes
Foster Care History	Yes	Yes	Yes
Child Protective History	Yes	Yes	Yes
N:	5,270	6,476	6,544

Notes: Comparison group for disability type is no disability. Outcomes are from the NYTD data, foster care history is from the AFCARS data, and child protective history (CPS) is from the NCANDS child file data. Sample size differences are a result of missing outcome data. FTE is full time employment, PTE is part time employment, and educ enroll is educational enrollment. Odds Ratio presented with standard error in parenthesis. All models adjust for control variables, including a control for outcome variable at baseline (i.e., model estimating FTE at age 21 includes a control variable for FTE at baseline).

⁺p<0.10.

^*p<0.05,

^**p<0.01,

^***p<0.001.

The odds of having FTE at age 21 for those with emotional or intellectual disabilities are 0.8 times the odds of FTE for those without a disability (p<0.00). The odds of having FTE at age 21 for those with both emotional or mental disabilities and sensory or physical disabilities are 0.7 times the odds of FTE for those without a disability (p=0.04). This means that the odds of full time employment are 20% lower for those with emotional or intellectual disabilities and 30% lower for those with both emotional or intellectual disabilities and sensory or physical disabilities than for those without disabilities. Having just sensory or physical disabilities or having a disabilities categorized as other are not significantly associated with the odds of FTE. Similarly, those with emotional or mental disabilities (or=0.8, p=0.01) and those with emotional or mental disabilities and physical or sensory disabilities (or=0.8, p=0.09) have lower odds of PTE. There are no significant associations between sensory or physical disability or other disability types and the odds of part time employment. Last, this table examines educational enrollment. Having any type of disability is not significantly associated with the odds of educational program enrollment.

Next, the proportion of youth with and without disabilities who receive various services and supports funded by the Chafee Independence Program are examined. Table 3 shows the proportion for both groups as well as the results of a P-value of the difference test between the means. Overall, a higher proportion of youth with disabilities are receiving services than youth without disabilities from that population. The primary exception is related to financial support, where a significantly smaller proportion of youth with disabilities are receiving housing financial assistance and education financial assistance. In many cases, the differences in the proportion of youth receiving services between those without and with a disability are significant. A significantly higher proportion of youth with disabilities are receiving special education services, career services, employment supports, housing education, other financial assistance, independent living, mentoring, and family services. There is no statistically significant difference in the proportion of youth receiving services between those with and without disabilities in terms of academic supports, post-secondary supports, budgeting education, and health education.

Next, the association between receiving services aimed at supporting employment and education and the outcomes (FTE, PTE, educational enrollment, and disability benefits) among those with disabilities who have aged out of foster care are examined in Table 4. Among those who have disabilities and age out of foster care, the odds of FTE for those who receive special education services are 0.5 times the odds of those who do not (p<0.00). This means that the odds of having FTE are 50% lower for those who receive special education services. Alternatively, the odds of having FTE for those who receive academic support are 1.3 times greater than the odds of those who do not (p=0.05) and for those who receive post-secondary supports the odds of FTE are 1.3 times the odds of those who do not receive those supports (p=0.05). This means that the odds of having FTE are 30% greater for those who receive academic supports and for those who receive post-secondary supports compared to those who do not receive those respective services. Educational financial assistance, career services, and employment supports are not significantly associated with the odds of FTE among those with disabilities.

Table 4.

The association between services aimed at supporting employment and education during the transition to adulthood and employment and SSI benefit receipt at age 21 among those with disabilities who age out of foster care using logistic regression.

	FTE		PTE		Educ Enroll		SSI/SSDI
	OR	p	OR	p	OR	p	OR	p
Services
Special Education Services	0.57*** (0.06)	<0.00	0.89 (0.09)	0.23	0.79* (0.8)	0.02	3.46*** (0.43)	<0.00
Academic Supports	1.29* (0.16)	0.05	0.86 (0.10)	0.23	1.09 (0.13)	0.46	0.71* (0.11)	0.03
Post Secondary Supports	1.27* (0.16)	0.05	1.26* (0.15)	0.05	1.08 (0.13)	0.50	0.78+ (0.11)	0.10
Education Financial Assistance	0.92 (0.11)	0.51	1.31* (0.14)	0.01	1.24* (0.13)	0.04	0.68** (0.10)	0.01
Career Services	0.88 (0.13)	0.37	1.27* (0.15)	0.05	0.99 (0.12)	0.46	0.67* (0.11)	0.01
Employment Supports	1.04 (0.12)	0.72	1.12 (0.13)	0.35	1.09 (0.13)	0.46	1.24 (0.19)	0.17
Controls
Demographics	Yes		Yes		Yes		Yes
Foster Care History	Yes		Yes		Yes		Yes
Child Protective History	Yes		Yes		Yes		Yes
N	3,058		3,106		3,110		3,002

Notes. Outcomes are from the NYTD data, foster care history is from the AFCARS data, and child protective history (CPS) is from the NCANDS child file data. SPED is special education, CPS is child protective history, FTE is full time employment, educ enroll is educational enrollment, SSI is Supplementary Security Income and SSDI is Social Security Disability Insurance. All models adjust for control variables, including a control variable for the outcome at baseline (i.e., model estimating FTE at age 21 includes a control variable for FTE at baseline).

⁺p<0.10.

^*p<0.05,

^**p<0.01,

^***p<0.001.

When examining part time employment, or PTE, there are positive significant associations with post secondary supports, education financial assistance, and career services. Among those with disabilities who age out of foster care, the odds of PTE for those who receive post secondary supports, education financial assistance, and career services are all 1.3 times greater than the odds of those who do not (p=0.05, p=0.01, and p=0.05 respectively). There are no significant relationships between special education services, academic supports, or employment supports and the odds of PTE among those with disabilities who have aged out of the foster care system.

When examining educational enrollment, those who receive special education services have 0.8 times the odds of educational enrollment at age 21 compared to those who do not receive special education services (p=0.02), meaning those with special education have 20% lower odds of educational enrollment. On the other hand, receiving educational financial assistance is associated with 1.2 times the odds of educational enrollment compared to not receiving those services (p=0.04). This means that the odds of educational enrollment are 20% greater for those who receive financial assistance for their education.

Last, the association between receiving services aimed at supporting employment and education and SSI/SSDI disability benefits among those with disabilities who have aged out of foster care are examined (again, in Table 4). On average and holding all else constant, the odds of receiving SSI/SSDI disability benefits for those who receive special education services are 3.5 times the odds of those that do not (p<0.00) among those with disabilities who age out of foster care. However, the opposite pattern is true for those who receive other services. The odds of SSI/SSDI benefit receipt for those who receive academic supports are, on average, 0.7 times the odds of those that do not (p=0.01) among those with disabilities who age out of foster care. Similarly, the odds of SSI/SSDI disability benefit receipt for those who receive post-secondary supports are 0.8 times the odds of those that do not (p=0.06) with marginal significance and the odds of SSI/SSDI disability receipt for those who receive educational financial support are 0.7 times the odds of those that do not (p=0.01). This means that the odds of receiving SSI/SSDI disability benefits are 30% lower for those that receive academic supports, 20% lower for those that receive post-secondary supports, and 30% lower for those that receive educational financial assistance than those that do not receive those respective services. For those who received career services, the odds of SSI/SSDI disability benefit receipt are 0.7 times the odds of those that did not (p=0.01). However, receiving employment supports are not significantly associated with the odds of SSI/SSDI benefit receipt among those with disabilities who age out of foster care.

Discussion

The results of this study suggest that youth with emotional or intellectual disabilities face particular risk in the transition to adulthood. For youth with emotional or intellectual disabilities and youth with both emotional or intellectual disabilities and sensory or physical disabiltiites have lower odds of FTE and PTE than their non-disabled counterparts. For those with only sensory or physical disabilities or the other disability type, the odds of FTE, PTE, and educational enrollment are statistically not different for those with and without disabilities who age out of care. However, youth who are alumni of the foster care system generally have lower employment and educational enrollment rates (Blome, 1997; Day et al., 2011; Gypen et al., 2017; Stewart et al., 2014), so this lack of difference does not suggest high levels of employment. By providing a clearer picture of the risk these youth face, this study calls attention to the need to support the employment, education, and independence of youth with disabilities who age out of foster care, especially those with emotional or mental related disabilities. Moreover, this study also takes steps to understand which supports this population is accessing from a national perspective and seeks to evaluate the association between a subset of these services and employment, education, and SSI disability benefit receipt for those with disabilities. This provides insight to the potential benefit of these services, which can inform a rigorous test of their effectiveness using a more robust and costly design like a randomized control trial for researchers and decisions by practioners and policy makers about which programs to offer and prioritize for funding.

In examining service receipt among youth with disabilities who have aged out of foster care this paper provides three primary conclusions; 1) for youth who age out of foster care an equivalent or higher proportion of youth with disabilities are accessing most services, 2) the exception is among financial assistance services, and 3) the services aimed at bolstering education and career services have the potential to improve the odds of employment, increase educational enrollment, and decrease the odds of SSI/SSDI benefit receipt net of employment, education, and benefit receipt at age 17.

Largely, youth with disabilities who age out of care are accessing services aimed to support their transition to adulthood. A recent review of research on the transition to adulthood for youth in foster care found that services aimed at increasing youth developmental skills fostered positive outcomes (Kothari et al., 2020). Relative to youth without disabilities who age out of care, a higher proportion of youth with disabilities are receiving special education services, career services, employment supports, housing education training, other financial assistance, independent living assessments and services, mentoring services and family services. However, there are two services where a lower proportion of youth with disabilities are receiving services—housing financial assistance (21% compared to 24%) and education financial assistance (37% compared to 39%). It is important to understand why a smaller proportion of youth with disabilities are receiving these services, and future research should examine the root of this disparity.

Services aimed at supporting employment and education are thought to be two important avenues to bolster employment and educational enrollment, and reduce early reliance on government benefits in the transition to adulthood. The results of this study support this hypothesis, finding encouraging results related to most services. Youth with disabilities who age out of care who receive academic supports, post secondary supports, education financial assistance, and career services have higher odds of employment and lower odds of disability benefit receipt compared to those who do not receive these services. Given the ongoing task of financial solvency at the Social Security Administration, whose 2020 report projected fund depletion in 2035 (A Summary fo the 2020 Annual Reports, 2020), reducing disability benefit in this population among those who could be independent through prioritizing serviced aimed at supporting education could prove an important tool to help reduced costs for the SSA.

It is important to note that FTE and independence from disability benefits may not be feasible or the goal for all youth. Scholars have argued that part-time employment can act as an important form of workplace accommodation for those with work-limiting disabilities, allowing them access to the workforce and its associated benefits for health and wellbeing (DeLeire, 2001; Jones, 2007). Furthermore, those who need disability benefits should have them to rely on. However, FTE and independence from disability benefits for those who can, especially at this young age, has a host of benefits including reducing poverty, increasing social connection, and improving health (Bal et al., 2017; Bevan et al., 2013; Saavedra et al., 2015; She & Livermore, 2008).

Alternatively, receiving employment supports are not significantly associated with the odds of FTE, PTE, educational enrollment, or receiving SSI/SSDI disability benefits, as seen in Table 4. Jointly, this suggests that the services aimed at supporting employment may not be as promising for bolstering education and reducing reliance on disability benefits for youth with disabilities who age out of foster care compared to the services aimed at supporting education or career services. These analyses do not consider the timing of these services within the age 17 to age 21 span. Given the dense transitions and life course decisions which occur during this age span, future research should examine how the timing of these services may have differential returns and even alter the implications of these services.

It is important to consider the results about relative service use by those with disability and the potential effectiveness of services for those with disability together. While services which support education hold promise for bolstering employment and reducing SSI benefit receipt for those with disabilities (with the notable exception of special education, which is discussed in the limitations section), the proportion of those with disabilities who receive these services (academic supports and post-secondary supports) are the same as those without disabilities or, in the case of education financial assistance, lower than those without disabilities. Employment supports do not have the same positive consequences for the odds of employment, educational enrollment, or SSI/SSDI benefit receipt among those with disabilities. However, a higher proportion of those with disabilities are receiving these employment support services than those without disabilities. This suggests that policymakers should prioritize funding programs and services to support education (academic supports, post secondary supports, and education financial assistance) and career services for those with disabilities who age out of foster care to reduce reliance on disability benefits, increase educational enrollment, and boost FTE.

Taken together, this study provides a clearer picture of the transition to adulthood for youth with disabilities who age out of foster care and provides insight into how policy reform and practice can mitigate risk for these vulnerable youth. Policy makers should prioritize programs and services which support the educational achievement and attainment of youth with disabilities who are likely to age out of foster care. Additionally, practitioners working with this population should incorporate educational support into care. Last, and most importantly, youth with disabilities who are likely to age out of foster care should be receiving educational and financial support on par with or in a higher proportion than those without disabilities.

Limitations

The response rate for the NYTD data is low, threatening the generalizability of these findings. Given this limitation, these results may provide a national perspective while not necessarily being national representative given attrition. Overall, youth who age out of foster care are an extremely vulnerable population and a lower response rate is expected. The NYTD data does include weighting to adjust for non-response resulting from attrition, however analyses conducted by Children’s Bureau found little difference between the weighted and non-weighted analyses suggesting that the threat of non-response bias is limited for observed variables (Children’s Bureau, 2014). Moreover, this project uses data from the first cohort of youth for whom transition to adulthood data are collected. However, it is important to note that youth who are aging out of foster care are typically excluded from national surveys examining employment and disability despite being a potentially high risk and high receipt population. While these findings may be limited to providing insight to those who participated in the survey, as opposed to generalizable findings which we can confidently extend to the entire population aging out foster care, this is among the first studies to specifically look at disability, education, employment, and independence for this population and begin to understand these relationships and the potential role of education and career related services in supporting employment, education, and reductions in disability benefits at this young age. Regardless, the limited generalizability of these findings is important to note.

A notable exception to the positive association between education-based services and the odds of employment and independence, is receiving special education supports. The odds of FTE and educational enrollment are lower and the odds of SSI/SSDI benefit receipt are considerably higher for those with disabilities who receive special education services. This is potentially due to unobserved heterogeneity and selection into special education services. One limitation of this study is that it is not possible to adjust for the severity of symptoms for disability. It is possible that those with disabilities who meet the eligibility criteria to receive special education services may have, on average, more severe limitations which would in turn explain the negative association between special education services and the odds of employment and independence for those with disabilities. It is not necessarily that the services that special education itself provides lead to lower odds of employment and higher odds of SSI/SSDI disability benefit, but rather the population that selects into special education services likely have lower odds of employment and higher odds of SSI/SSDI benefits to begin with. This result is consistent with an existing study which found that involvement in the child-welfare system and the special education system had multiplicative risks (Geenen & Powers, 2006). Geenen and Powers found that foster care status and special education status were both associated with academic difficulties, but that youth at the intersection of these statuses faced compounding risk (2006). Future research should examine the intersectional risks of child welfare involvement and special education service receipt to elucidate the potential mechanisms leading to decreased employment and increased SSI/SSDI receipt.

Another related limitation to consider is that these data do not include information about the severity of disability symptoms. It is important to note that individuals with more severe disability symptoms may be eligible for specific services or overrepresented in the population which received specific services (such as special education). This selection effect limits our interpretation of these results, especially in terms of how service receipt is associated with the outcomes of interest. While the results for special education show that the design of the study does not address unobserved heterogeneity related to disability severity, the other results do adjust for special education service receipt which may proxy disability severity. This study design is associational, not causal, and the results should be interpreted as such.

One final limitation to note is that this study relies largely on administrative data. While these data provide unique strengths, such as a large sample size allowing for the disaggregation of disability type and a national perspective, they also introduce notable limitations. These data do not provide depth. The service use categories reflect the available categories in the data and could refer to substantively different programs across states and jurisdictions, and more detailed data on the texture and quality of service use are not available. Moreover, the use of administrative data necessitates the use of disability categories which are not ideal. As noted in the Methods section, the “other” disability category contains a variety of disabilities that the author would not categorize together, however it is not possible to alter the level of detail in the data. Furthermore, this measure may miss youth who have a disability but were never diagnosed or youth who developed or were first diagnosed with a disability between leaving foster care at age 18 and when data collection was completed at age 21. While it is important to interpret the results in light of these limitations, this study does provide an in-depth exploration of service use, disability disaggregated by type, and their relationships to employment, education, and independence from a national perspective.

Conclusion

This study provides initial empirical evidence linking disability and receiving education and career-based services to employment, education, and disability benefit receipt in the transition to adulthood for youth with disabilities who age out of foster care. This study finds that those with emotional or intellectual disabilities face the highest risk in the transition to adulthood, with lower odds of employment (FTE and PTE) at age 21. Importantly, this study finds that receiving services aimed at supporting education and employment may mitigate some of this risk and provide the potential for supporting the successful transition to adulthood for those with disabilities who age out of foster care. Youth with disabilities who received academic supports, post-secondary supports, education financial assistance, and who received career services had greater odds of employment and lower odds of disability benefit receipt. Practitioners and policy makers should prioritize providing academic and post-secondary services, providing financial assistance, and providing career services to youth with disabilities who age out of foster care.

Biography

Erin J. McCauley is an Assistant Professor of Sociology in the Department of Social and Behavioral Sciences at the University of California San Francisco. She is also an affiliated faculty in the Philip R. Lee Institute for Health Policy Studies. Her work interrogates the causes and consequences of involvement in the child welfare and criminal legal systems for children and families. She earned her doctorate in Sociology with a minor in Demography from the Department of Sociology and the Department of Policy Analysis and Management from Cornell University in 2021.

Data Availability:

These data are available through the National Data Archive on Child Abuse and Neglect.