Abstract
Research has consistently documented poor employment outcomes for young adults with autism spectrum disorder (ASD). Vocational rehabilitation (VR) services provide substantial federal and state commitments to individuals with disabilities to obtain and maintain employment. To date, little research has examined the relationship between VR services and employment outcomes of clients with ASD. The purpose of this descriptive study was to better understand employment outcomes of individuals with ASD. Data spanning 10 years from the Rehabilitation Services Administration (RSA) 911 database were analyzed to identify characteristics of VR clients with ASD and the services they received. The percent of individuals who achieved competitive employment averaged 37% across the 10 years. Those who achieved an employment outcome participated in twice as many services as those who did not achieve an employment outcome.
The Rehabilitation Act was passed in 1973 to “ . . . empower individuals with disabilities to maximize employment, economic self-sufficiency, independence and inclusion and integration into society, and to ensure that the federal government plays a leadership role in promoting employment of individuals with disabilities” (29 U.S.C. §701). Since 1973, millions of individuals with disabilities have received vocational rehabilitation (VR) services designed to help them achieve and retain employment. Other legislative mandates have also addressed employment-related issues for individuals with disabilities of all ages (e.g., School-to-Work Opportunities Act of 1994; Workforce Innovation and Opportunity Act of 2014).
Research has consistently documented poor employment for young adults with autism spectrum disorder (ASD; Hendricks & Wehman, 2009; Taylor & Seltzer, 2011). The National Longitudinal Transition Study–2 (NLTS2) found that at the time of interview, 45.2% of youth with autism had a paid job outside the home compared with 71% of same-age peers in the general population (Sanford et al., 2011). Taylor and Seltzer (2011) found 18% (n = 12) of the 66 individuals with ASD were employed either with or without supports.
To date, little research has investigated the relationship between VR services and employment outcomes of clients with ASD by analyzing VR data. Since 1990, when ASD was first federally recognized as a disability category, only three studies analyzed VR data on individuals with ASD (Burgess & Cimera, 2014; Cimera, Burgess, & Wiley, 2013; Schaller & Yang, 2005) and one focused on developmental disabilities (DD), including ASD (Wehman, Chan, Ditchman, & Kang, 2014). Burgess and Cimera (2014) analyzed Rehabilitation Services Administration (RSA) data from 2002 to 2011 and found that while the number of ASD clients increased, their employment outcomes—percent achieving employment closure, wages earned, and hours worked—did not improve over that time. Those with ASD were significantly more likely than the general VR population to achieve employment closure; however, the latter worked more hours and earned higher wages on average. Cimera et al. (2013) analyzed RSA data from 2006 to 2009 and found young adults with ASD who received transition services at age 14 were significantly more likely to achieve employment than those who received transition services at age 16. Schaller and Yang (2005) analyzed RSA data for 2001 and found clients with ASD had significantly higher closure rates in supported employment (SE) than in competitive employment. Clients’ average weekly earnings were significantly higher in competitive employment, but the average cost of VR services were significantly higher (slightly more than double) for SE cases. Furthermore, 80% of those in competitive employment and 76% of those in SE were Caucasian. Wehman et al. (2014) examined employment outcomes of youth with intellectual and developmental disabilities (e.g., ASD) who received SE services within VR for 2009. Their findings indicated SE was an effective service for achieving employment outcomes for individuals with significant disabilities.
Although previous research has analyzed RSA data to study VR outcomes for individuals with ASD, the connection between client characteristics (e.g., demographics), the specific types of VR services, and case outcomes remains sparse. Cimera and Cowan (2009) examined the cost of services and outcomes of VR clients with ASD. They found the average cost of services was $3,213, second highest only to individuals with sensory impairments. The 40.8% of individuals with ASD who were employed at case closure “worked far fewer hours (18.65) and earned lower wages (mean = $146.65) per week than most groups studied” (p. 294). Examining one federal fiscal year ([FFY], 2011) of RSA-911 data, Chen, Sung, and Pi (2015) found those aged 18 and younger received six services—assessment, college or university training, occupational/vocational training, miscellaneous training, information and referral services, and other services—significantly more than those aged 25 and older. Furthermore, individuals with ASD ages 25 and older received five services—diagnosis and treatment of impairment, job search assistance, job placement assistance, on-the-job training support, and transportation services—significantly more than those aged 18 and younger.
The purpose of the present study was to better understand employment outcomes of individuals with ASD by analyzing 10 years of RSA data for client characteristics and services received. We examined three a priori research questions regarding VR clients with ASD: (a) What are their demographic and service characteristics; (b) What are their case outcomes; and (c) What are their differences in service and case employment outcomes over time?
Method
Data Collection
The data source for this study was the RSA-911 database, a federal data collection on VR services and clients from VR agencies in all 50 states and the U.S. Territories and Entities. For this analysis, we examined data from the 50 states and District of Columbia and excluded Territories and Entities with different political and economic structures. We obtained RSA-911 data from fiscal years 2003 to 2012. These data were mailed to the authors in text data (.txt) format on CD-ROMs.
Defining Variables
RSA provided a data-variable dictionary, the RSA-911Reporting Manual (2008), listing all variables and their operational definitions for the annual 911 data set. The variables we used in this analysis were defined according to the RSA-911 Reporting Manual (see Tables 1 and 2). We also created two variables—time with VR and improved education level (see Table 1).
Variable Definitions.
Note. RSA = rehabilitation services administration; VR = vocational rehabilitation.
Definitions of the 22 VR Services for Clients.
Note. VR = vocational rehabilitation; RSA = rehabilitation services administration.
Data Analysis
Only clients coded with ASD as their primary disability were included in the statistical analyses. We organized the RSA data for each fiscal year in SPSS statistical software v.21 (IBM-SPSS, 2013) for Windows and screened these data for keystroke errors and missing values. Descriptive statistics were run for demographics—ethnicity, gender, highest level of education attained at client’s application, individual education program (IEP) status, and employment at application and case closure (Table 3). Statistical tests were conducted using t tests and chi-square tests of independence to empirically ascertain differences in employment outcomes among VR clients with ASD based on gender, ethnicity, IEP status, improved education, and total services received (Tables 4 to 8). A priori statistical significance was set at p < .05.
Demographics of VR Clients With ASD.
Note. Percentages rounded to the nearest whole number. VR = vocational rehabilitation; ASD = autism spectrum disorder; AA = associate degree.
Percent of Clients With ASD Receiving VR Services.
Note. ASD = autism spectrum disorder VR = vocational rehabilitation.
Rates of VR Case Closure Type (%) by Year.
Note. VR = vocational rehabilitation; IPE = individual plan for employment. aConsidered the only “successful” VR closure.
Achieved Employment Outcome by Gender Chi-Square.
Note. *Statistically significant at p < .05.
Achieved Employment Outcome by Ethnicity Chi-Square.
Note. *Statistically significant at p < .05.
Achieved Employment Outcome by IEP Status Chi-Square.
Note. IEP = individual education program.
Statistically significant at p < .05.
Results
Clients with ASD (n = 49,623) constituted less than 1% of the 6,048,847 total client cases in VR across all states and D.C. from Fiscal Years 2003 to 2012. We found less than 1% of all data missing for clients with ASD. Cases with any missing data were excluded from further statistical analyses. Results are organized by research question.
What are the Demographic and Service Characteristics of VR Clients With ASD?
Table 3 shows the percentage of cases with ASD in six demographic characteristics (a) gender, (b) ethnicity, (c) IEP status, (d) education attainment at the time of application, (e) employment at application, and (f) employment at closure. The majority of clients with ASD were male, ranged from 81% in 2003 to 84% in 2012. The majority of clients identified as White ranged from 84% in 2003 to 88% in 2012. The percent of clients identified as Black ranged from 11% to 12% until 2009 when the percentage dropped and remained at 10% through 2012. The percent identified as Hispanic/Latino ranged from 3% in 2004 to 6% in 2011. Clients identified as Asian ranged between 2% and 3% across the decade. Combined, less than 2% of clients identified as American Indian/Alaska Native or Hawaiian/Pacific Islander.
Clients with ASD who had less than a high school education ranged from 41% in 2003 to 33% in 2012, with a high of 43% in 2004 and 2005. The majority of clients who received VR services were recorded in the database as regular high school students who graduated or received an equivalency certificate; this group ranged from a high of 43% in 2003 to 41% in 2012, with a low of 38% in 2005. An additional 26% in 2003 to 17% in 2012 of the cases were recorded as having received a special education certificate of completion, diploma, or being in attendance. This classification included individuals who were still in school and receiving special education services, those who received special education services and earned a certificate of completion or high school diploma, and those who received special education services but did not receive a certificate or diploma. Clients with ASD who had some postsecondary education or training without earning a degree ranged from 8% in 2003 to 14% in 2012. Those with an associate degree or vocational technical certificate ranged from 4% to 6%. Those with a bachelor’s degree ranged from 3% to 4%, and those with a master’s degree or higher ranged from less than 1% to 1%. Clients who were employed at the time of VR application ranged from 9% to 14%.
Table 4 shows the percent of VR clients with ASD who received services across all 10 years. On average, 63% of these clients received an assessment to determine their eligibility for VR services and an average of 52% received VR counseling. On average, 38% received a service related to diagnosis and treatment of impairment, and 23% to 31% received job-related services such as job placement assistance (31%), on-the-job supports (28%), and/or job search assistance (23%). Across the decade, on average, fewer than 20% of these clients were reported as receiving other services (e.g., occupational licenses, tools and equipment, initial stocks and supplies; 18%), transportation services (16%), and job readiness training (14%), and information and referral services (15%) and miscellaneous training (e.g., GED or high school training leading to a diploma).
Across the decade, on average, fewer than 10% of clients with ASD received occupational vocational training (6%), university training (6%), on-the-job training (5%), and maintenance services (5%). Furthermore, 2% or fewer were reported as receiving disability-related augmentative skills training (2%), basic academic training (1%), rehabilitation technology (1%), and personal attendant services (1%), and technical assistance service, reader, or interpreter services were reported for less than an average of 1% of the clients with ASD. On average, 87% of these clients participated in VR for 3 years or less. Yearly, approximately one third of the VR clients with ASD improved their education level at least one reporting level (e.g., not having high school diploma to having a high school diploma).
What Are the Outcomes of VR Clients With ASD?
Only an exit with an employment outcome is considered a successful case closure by VR (RSA, 2008). Although the number of cases of clients with ASD steadily increased from 1,541 in 2003 to 10,347 in 2012, successful case closure rate did not, averaging 37.5%, with a low of 33% in 2010 to a high of 42% in years 2004, 2006, and 2007 (Table 5). On average, a quarter (25%) of those who sought VR services exited without employment after receiving services, ranging from a low of 23% in 2008 to a high of 28% in 2010 and 2011. Still other clients with ASD exited VR before receiving services (e.g., at application, after eligibility determination), a low of 32% in 2004 to a high of 39% in 2010 (Table 5). The percent of individuals who achieved competitive employment averaged 37%, fluctuating from 39% to 41% through 2008 before declining to a low of 32% in 2010 and then increasing to 36% in 2012.
What Are the Differences in Employment Service and Case Outcomes Among VR Clients With ASD Over Time?
Across all 10 years, a total of 31,004 of the 49,623 (i.e., 62%) cases of clients with ASD did not achieve an employment outcome. These clients were mostly male (78% to 83%), White (83% to 87%), who had an IEP (62% to 75%), and had at least a high school diploma at the time of their application for VR services (54% to 59%).
To examine who achieved an employment outcome and whether differences were statistically significant, we conducted chi-square tests of independence between the discrete outcome variable, achieved employment outcome (yes/no), and each of four dichotomous predictor variables: (a) gender (male/female), (b) ethnicity (White/non-White), (c) IEP status (yes/no), and (d) improved education (yes/no). In all 10 years, male clients with ASD were significantly more likely than female clients with ASD to achieve an employment outcome (see Table 6). In 8 of 10 years, White clients with ASD were significantly more likely than non-White clients with ASD to achieve an employment outcome (see Table 7). In 5 of the 10 years, clients with ASD who had an IEP were significantly more likely to achieve an employment outcome (see Table 8). In 9 of 10 years, clients who increased their education level from application to exit were significantly more likely to achieve an employment outcome (see Table 9).
Achieved Employment Outcome by Improved Education Chi-Square.
Note. *Statistically significant at p < .05.
To further understand the empirical relationship between VR services and case outcomes for clients with ASD, t tests were conducted. As shown in Table 10, clients with ASD who achieved an employment outcome, consistently across the decade, had utilized a significantly greater number of VR services (4.5 to 4.9) than clients who did not achieve an employment outcome (2.1 to 2.3). Even after correcting for unequal variances (i.e., Levene’s test) between the two groups, the differences between them were statistically significant (p < .001).
Service Total Difference in Employed (E) and Non-Employed (NE) Outcomes.
Adjusted with Levene’s test for unequal variances
Discussion
The purpose of this study was to examine characteristics of VR clients with ASD and the services they received to better understand their employment outcomes. Significance tests were conducted only on select variables to determine bivariate statistical differences to facilitate description of the population, not to establish a relationship between a dependent variable and independent variables (see Shadish, Cook, & Campbell, 2002). No causal inferences can be drawn. Results of the present study are important given the sparseness of published studies on employment outcomes of VR clients with ASD by analyzing multiple years of VR data. By examining 10 years of RSA-911, this study provides a more comprehensive picture of employment outcomes and services used by VR clients with an ASD. The discussion is organized by research question, followed by recommendations for future research and implications for practitioners.
What are the Demographic and Service Characteristics of VR Clients With ASD?
The number of VR clients with ASD increased by 571% across the decade, whereas the total number of all VR clients decreased by 10%. There were small increases in the percentages of male clients and White clients with ASD and small or no increases in female clients and non-White clients with ASD. These demographic findings are consistent with what is known about the connection between ASD identification, ethnicity, and gender. The male-to-female ratio found in this data set is similar to other reported findings (Chakrabarti & Fombonne, 2001; Werling & Geschwind, 2013). For years 2003 and 2004, the ratio in this RSA data set mirrors this proportion at 4:1 with an increase to 5:1 in the remaining years. Increase in the VR ASD population is also consistent with the Center for Disease Control’s noted increase in the prevalence of ASD in the general population (Baio, 2014). Reasons for this increase vary, from improved screening to possible over-identification or misidentification (Miller et al., 2013).
There was an increase in the percentage of VR clients with ASD who received some postsecondary education or training, which corresponds to a decrease in the percentage of ASD clients with less than a secondary education. Among young adults with ASD in the general population, the rates of postsecondary education did not improve statistically from late 1990s to late 2000s (Newman, Wagner, Cameto, Knokey, & Shaver, 2010). Unfortunately, a national estimate of postsecondary education attainment for older adults with ASD is unknown (Haley, 2011).
Relative to patterns of services, an average of 57% of the VR clients with ASD received assessment and/or VR counseling, and about 40% received diagnosis/treatment across the decade. These were the three most frequently used services, which were not surprising given an assessment is required to determine eligibility for services. In addition, VR counseling—defined as discrete therapeutic counseling and guidance services needed to achieve employment—is necessary to assist VR clients in finding employment options, and diagnosis and treatment directly addresses a client’s barriers to employment. However, none of the remaining 19 services were used by more than 33% of clients. Since ASD manifests so uniquely across the population (Attwood, 2007), a higher percentage of these other services utilized by clients might have been expected. Difficulties such as social skills, communication, executive functioning, and anxiety can pose significant barriers for individuals with ASD requiring specific supports to address them (Alverson, Lindstrom, & Hirano, 2015). It is possible that challenges with communication, social skills, and executive functioning were barriers preventing these VR clients from taking advantage of the full range of services available to them. Muller, Schuler, Burton, and Yates (2003) found individuals with ASD experienced several major obstacles to successful employment “(a) mastering the job application process, (b) acclimating to new job routines, (c) communication, and (d) navigating social interactions with supervisors and co-workers” (p 167). Within each major theme, these researchers identified minor themes affecting employment, such as “difficulties contacting potential employers by phone, difficulty reading between the lines, and taking too long to learn new tasks” (Muller et al., 2003, p. 167). These themes appear to describe the services available to clients from VR (e.g., job readiness training focused on building communication skills).
What are the Outcomes of VR Clients With ASD?
To understand employment outcomes in this context, one must first understand the various ways in which the RSA defines an employment outcome which include several categories such as “homemaker” and “unpaid family worker” (RSA, 2008, p. 32). This differs from the RSA definition of competitive employment, which is more restrictive—employment in an integrated setting, self-employment, and state-managed Business Enterprise Programs (RSA, 2008). Of the 49,623 cases with ASD, 36% (n = 17,922) achieved a competitive employment outcome. Research has consistently documented unemployment or underemployment of individuals with ASD (Frith, 2004; Hurlbutt & Chalmers, 2004). Recent reports estimate only 6% to 11% of adults with ASD are competitively employed (Cedurland, Hagberg, Billstedt, Gillberg, & Gillberg, 2008; Taylor & Seltzer, 2011). Shattuck et al. (2012) found 55.1% of individuals with ASD, ages 19 to 23 at data collection, had engaged in any paid work since leaving high school. Findings from this study are consistent with the wide range of employment findings.
Whereas the total number of VR clients with ASD increased from 2003 to 2012, the rates of closure with an employment outcome did not. Year-to-year fluctuations in the VR employment rates can be attributed to two other types of case closures that represented the largest groups of clients each year, those who exited without an employment outcome after (a) VR services and (b) being found eligible for services, but before development of the individual plan for employment (IPE). The consistent, substantial number of clients with these types of case closures can be accounted for, in part, by the reasons clients did not achieve an employment outcome. The four most frequently cited reasons that clients with ASD exited VR without an employment outcome accounted for 76% to 88% of cases (a) refused services/further services, (b) all other reasons, (c) unable to locate or contact, and (d) failure to cooperate. These reasons could be directly related to the common cluster of ASD challenges related to communication, social skills, and executive functioning, which could potentially be mitigated through expanded utilization of other VR services not commonly used.
What Are the Differences in Employment Service and Case Outcomes Among VR Clients With ASD Over Time?
In each year, VR clients with ASD who achieved an employment outcome had utilized a statistically significantly (p < .001) greater number of services, twice as many as clients who did not achieve an employment outcome. Common services utilized by the clients who achieved an employment outcome were (a) diagnosis and treatment of impairment, (b) job placement assistance, and (c) on the job supports. To further understand the empirical relationship between individual factors and VR case closure outcomes, client characteristics (i.e., predictor variables) were tested for statistical differences across two case outcomes—achieving or not achieving employment outcome at closure (i.e., criterion variable). Of the four predictors (gender, ethnicity, IEP status, and improvement in educational attainment from application to closure) across the decade, gender consistently had the strongest effect. Being male was the best predictor of whether a client with ASD would achieve an employment outcome at case closure, followed by ethnicity and improvement in educational attainment. This finding is not surprising given the economic and employment data in the general population with regard to the disparities between males and females (United States Department of Labor, 2014) combined with the fact that many more males are identified with ASD than are females (Attwood, 2007).
Recommendations for Future Research
We conducted a descriptive study to begin to understand the characteristics of VR clients with ASD as a population, the VR services they received, and their employment outcomes. From this beginning point, the field can pose research questions to discover and explain contributing factors, test hypotheses, and craft interventions that are likely to result in positive outcomes.
A multitude of important research questions can be addressed with further analyses of the RSA data. For example, how do the VR outcomes and services of other subpopulations compare with the outcomes and services of clients with ASD over multiple years? What are the relationships between the reasons clients with ASD did not achieve employment at case closure (e.g., refused services, failure to cooperate) and the types of services they received (and did not receive) while also accounting for differences across clients’ state economic conditions (e.g., unemployment rates of individuals with disabilities, predominant state industries or employers). Answering these research questions may also require further contextualization, such as surveying VR counselors, clients, clients’ coworkers, work supervisors, and others to more fully understand the individualized reasons for clients not achieving an employment outcome, such as inadequacy of workplace supports or mismatch between client skill-set and job requirements. Research could also directly address the financial costs and impact of VR services on case outcomes of clients with ASD, asking, what is the relationship of the changing national and state policies and fiscal expenditures on VR services and case outcomes of clients with ASD? This would entail analyzing the dollar cost of VR services, public assistance (e.g., Social Security Disability Insurance [SSDI], Temporary Assistance for Needy Families [TANF]), and number of hours worked and weekly average salary at case closure.
Implications for Practice
In light of the findings from this analysis, there are two important implications for practice. First, given that, across the 10 years examined in this analysis, 62% of the cases of clients with ASD did not achieve an employment outcome, it is imperative that those working with individuals with ASD (e.g., educators of transition age youth and adult services providers) maximize the individuals’ preparation for employment. Although VR is charged with helping individuals overcome barriers to employment, Individuals With Disabilities Education Act of 2004 (IDEA) charges educators with preparing young adults with disabilities for employment (C.F.R. §300.1(a)). Previous research (Benz, Lindstrom, & Yovanoff, 2000; Rabren, Dunn, & Chambers, 2002) has established the importance of having paid work experiences while in high school. For practitioners working with individuals who experience ASD and who do not receive IDEA services in school, the Workforce Innovation Opportunities Act of 2014 provides for the provision of pre-employment transition services to youth ages 14 to 24, including job exploration counseling, work-based learning experiences, and workplace readiness training.
Second, with the 571% increase in the number of individuals with ASD seeking services through VR, it is critical that those tasked with helping individuals get a job (e.g., VR counselors, job developers), as well as employers and natural supports understand the characteristics of ASD and how they manifest in individuals (Alverson et al., 2015). Valle et al. (2014) found specialized VR case loads to be an effective strategy for supporting individuals with ASD. Although VR counselors may not have a single-focused caseload, developing expertise in ASD may be beneficial. To help VR counselors understand the unique needs of their clients with ASD, they need training beyond describing of stereotypical characteristics of ASD. Furthermore, individuals with ASD should be comfortable explaining and describing how ASD affects them. Educators, families, and support personnel can help individuals with ASD understand and talk about their strengths and support needs.
Limitations
These findings should be considered within the following limitations. A limitation of using an extant data set is the lack of control over variables definitions, data collection, and entry. This limitation is evident in two ways. First, the RSA data set we received lacked an age variable, which did not allow for disaggregating a variety of key questions (e.g., average age of clients, closure rates—there may be different employment rates between older and younger clients). Second, the level of education variable for those who received special education services combined three distinct categories, individuals who (a) were still in school, (b) earned a certificate of completion or diploma, or (c) exited high school without receiving a certificate or diploma. It is important to note, this analysis represents only those individuals with ASD who sought VR services; these individuals may be significantly different from individuals with ASD who did not seek services.
Conclusion
By examining the characteristics of VR clients with ASD and the services they received, we have a better understanding of the employment outcomes of this fast-growing population. The RSA’s annual collection of these data from every state provides a snapshot of one segment of this population—those who sought VR services. Analyzing these data over a long period of time provides insights to avenues for further research related to improving service provision through interventions targeted to this population, improving the overall system, and developing service policies. This research provides a starting point to understanding the employment outcomes for individuals with ASD who seek services through VR.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
