Abstract

The early work experiences of a nationally representative sample of youth with severe disabilities (i.e., intellectual disabilities, autism, multiple disabilities) were examined. Using data from the National Longitudinal Transition Study-2, we explored the extent to which various student-, family-, school-, and community-level factors were associated with paid work experiences during high school. Findings highlight the elusiveness of early work experiences for many youth with severe disabilities and call attention to malleable factors that may play a role in shaping employment success during high school. Recommendations for research and practice are highlighted.

An enduring emphasis of special education and transition services has been its focus on equipping youth with the skills, supports, and opportunities that prepare them for the world of work (D'Alonzo, 1978; Wehman & Revell, 1997). Early follow-up and longitudinal studies documenting the disappointing postschool employment experiences of young people with disabilities were driving factors in the initial establishment and subsequent expansion of transition services (Blackorby & Wagner, 1996; Hasazi, Gordon, & Roe, 1985). Within the Individuals with Disabilities Education Improvement Act (2004), transition services are now defined as a coordinated set of activities that are:

designed to be a results-oriented process, that is focused on improving the academic and functional achievement of the child with a disability to facilitate the child's movement from school to post-school activities, including post-secondary education, vocational education, integrated employment (including support employment), continuing and adult education, adult services, independent living, or community participation [italics added]. (§300.43)

Indeed, the extent to which young people with disabilities find and maintain meaningful employment after high school constitutes a key marker of the quality and impact of secondary education and transition services.

For young people with severe intellectual and developmental disabilities, however, meaningful employment after high school remains particularly elusive (Johnson, McGrew, Bloomberg, Bruininks, & Lin, 1997; Test et al., 2009). For example, Rusch and Braddock (2004) reported that less than one quarter of adults with severe disabilities participated in supported employment in 2002; the remainder received services within segregated day programs and sheltered workshops. Similarly, findings from studies conducted by the National Organization on Disability (2004) and National Longitudinal Transition Study-2 (Newman, Wagner, Cameto, & Knokey, 2009) further reveal the pervasiveness of dismal employment outcomes for individuals with severe disabilities. Collectively, these studies challenge secondary schools to consider anew the quality and impact of the career development and vocational experiences they promote among youth.

Both transition policy and recommended practice emphasizes the necessity of providing youth who have severe disabilities with a strong foundation of compelling career development experiences while they are still in middle and high school. Specifically, paid work experience within community-based job sites is advocated as an essential component of a well-rounded transition education (Rusch, Hughes, Agran, Martin, & Johnson, 2009). Moreover, connecting youth to these experiences earlier than the final year or two of high school not only mirrors typical adolescent work patterns (Zimmer-Gembeck & Mortimer, 2006), but it potentially provides more opportunities for students to develop a range of important vocational skills, knowledge, and aspirations (Carter et al., 2010). It is not surprising that every current transition framework stresses the contributions that early work experiences can make toward improving postschool employment outcomes (Certo et al., 2008; Kohler & Field, 2003; National Alliance for Secondary Education and Transition, 2005).

Despite this emphasis, relatively few researchers have conducted empirical studies to examine the work experiences of youth with severe disabilities during secondary school. Additional research is needed to increase the knowledge base in several ways. First, the extent to which youth with severe disabilities actually access early work experiences in their communities is not well-documented. Prior descriptive studies involving youth with intellectual disability, autism, and/or multiple disabilities suggest that school-sponsored, after-school, and/or summer work experiences may be particularly limited (Burbidge, Minnes, Buell, & Ouellette-Kuntz, 2008; Carter et al., 2010; Kraemer & Blacher, 2001). However, these prior studies—consistent with most studies involving students with severe disabilities—typically are constrained by relatively small sample sizes and geographically specific samples (e.g., a single school, district, state). Data documenting the ways in which students are accessing early work experiences at a national level could spur more focused and sustained efforts to promote career development for these youth.

Second, little is known about factors that may influence—directly or indirectly—the community-based work experiences of youth with severe disabilities. For example, research involving special education students broadly suggests that demographic characteristics (e.g., disability category, gender, race/ethnicity, or age) and student competencies (e.g., expressive and receptive communication; interpersonal, self-care, or mobility skills) may impact employment opportunities and outcomes (Test et al., 2009). The resources parents have (e.g., financial, transportation) and the expectations they hold for their children (e.g., household responsibilities, views regarding the future) also may shape the types of employment experiences youth with severe disabilities are offered throughout adolescence (Bianco, Garrison-Wade, Tobin, & Lehman, 2009; Chambers, Hughes, & Carter, 2004). In addition to student and family factors, the career and vocational programs schools provide can equip students with the critical skills, attitudes, and connections needed to find and maintain paid work in the community (Kramer & Blacher, 2001; Shandra & Hogan, 2008). Finally, the job opportunities youth encounter may vary depending on the type of community (i.e., rural, suburban, urban) within which they live and the transportation available within it. Understanding whether and how these student-, school-, family-, and community-related factors are associated with paid employment outcomes during school could provide guidance to practitioners and policymakers as they design and deliver transition services for this segment of the student population. Although these clusters of variables have been explored in analyses of postschool outcomes for heterogeneous groups of students with disabilities (Test et al., 2009), the specific roles these factors play in shaping the experiences of students with severe disabilities remains unexplored.

Our purpose in this study was to examine the early work experiences of a nationally representative sample of youth with severe disabilities (i.e., intellectual disability, autism, multiple disabilities). Specifically, we addressed the following research questions: To what extent are youth with severe disabilities accessing paid employment experiences during secondary school? What are the characteristics of those employment experiences? How are student-, family-, school-, and community-level factors each associated with paid work experiences during high school? A secondary focus was to provide a clearer description of the demographics and characteristics of secondary-age students with severe disabilities.

Method

National Longitudinal Transition Study-2

In the National Longitudinal Transition Study-2 (NLTS2), data were gathered over a 10-year period (2000–2010) from parents, youth, teachers, and schools to provide nationally representative information about students receiving special education services as they transition through secondary school into adulthood. The NLTS2 was commissioned by the U.S. Department of Education as a follow-up to the original National Longitudinal Transition Study (1985–1993). Only an overview of the design of the NLTS2 is provided here; more detailed information about the methodology and findings are available through the project website (http://www.nlts2.org/index.html) and in other publications (Wagner, Kutash, Duchnowski, & Epstein, 2005). The NLTS2 includes information for more than 11,000 youth receiving special education services who were ages 13 to 16 as of December 1, 2000 (SRI International, 2000).

Youth were selected in a two-stage sampling process. First, a stratified random sample of local education agencies was selected, with stratification based on geographic region (i.e., Northeast, Southeast, Midwest, West), district size (enrollment in grades 7 to 12), and community wealth (i.e., proportion of district living below poverty level). Second, students were randomly selected from each of 12 federally designated special education disability categories. Sampled youth were weighted to create a nationally representative sample by disability category and type of school district. We focused our analyses on data from Wave 1 because all youth were still enrolled in school at that particular time point.

Data from Wave 1 were collected from multiple sources using several different instruments (see www.nlts2.org for a description of data-collection tools). Our analyses combine data from three sources. The Parent Interview was conducted with a parent or guardian (hereafter referred to as parent); when a parent could not be reached by telephone, a mail survey with selected questions was sent. The response rate for combined parent data sources (hereafter referred to as Parent Interview) was 82.1%. The Student's School Program Survey—also sent by mail—was completed by school personnel who were most familiar with the student's school program (response rate  =  53.1%). The School Characteristics Survey was completed by the school principal or other administrator regarding the school as a whole (response rate  =  56.6%).

Students With Severe Disabilities

Because the population of students having “severe disabilities” is not clearly defined in the literature, we used the following criteria to identify our sample from this extant data set. Adolescents from within the primary disability categories of intellectual disability (we have chosen to use the term intellectual disability when referring to students with mental retardation, which reflects changes in terminology advocated since the launch of the NLTS2 study), autism, or multiple disabilities. We considered them for inclusion if they were enrolled in school or receiving instruction in a nonschool setting and had employment status (i.e., whether or not the student had a paid job) data available (n  =  2,620 students). All sample sizes were rounded to the nearest 10. Sums may not add to total due to rounding. From among this sample, students were considered to have severe disabilities if the Student School Program Survey revealed that they were eligible for the alternative assessment. If alternate assessment eligibility was not indicated, a student was only included in the sample if a parent reported that he or she had functional cognitive skill deficits (i.e., performs not at all well or not very well) in two or more of the following areas: reading and understanding common signs, telling time on a clock with hands, counting change, or looking up phone numbers and using the telephone. These criteria resulted in a sample of 1,510 of students with severe disabilities, including 390 students with a primary disability category of intellectual disability, 520 with autism, and 600 with multiple disabilities.

Employment Outcome

We considered several employment outcomes, including current employment status, paid community (i.e., not school-sponsored) job in the previous 12 months, and paid work–study. We classified a student as having paid work if he or she had a paid community job in the past 12 months and/or held a paid work–study job (see Table 1). Parents of students in all grades (7–12) were asked about paid community jobs, but only parents of students in high school were asked about work–study jobs. We classified students reported to have both community and work–study jobs as having a community job.

Table 1

Work Status by Primary Disability Category

Work Status by Primary Disability Category
Work Status by Primary Disability Category

In addition to examining paid work as an outcome variable, we also explored the types of jobs youth held and the work time frame (i.e., summer, school year, or both; available for both community and work–study jobs in the Parent Interview). When a student held both community and work–study jobs, we reported the type of community job they held. Data regarding pay, hours worked per week, and transportation mode(s) were only available for those with community jobs.

Predictor Variables

We examined groups of variables previously associated with employment outcomes for young people with disabilities, including student demographics and primary disability, student skills, family characteristics and expectations, school program characteristics, and community factors (Carter et al., 2010; Test et al., 2009). For questions with more than two response levels, we combined adjacent levels if fewer than 10% of respondents chose a response or if there were few or no students who had paid work experience at that response level.

Student demographic factors

Demographic variables included student age (in years, at the time of interview), gender, and race/ethnicity (i.e., White, African American, Hispanic, and other). We used terms drawn directly from the NLTS2 study. Although the other races/ethnicities group was relatively small (n  =  60), combining it with the next largest group (i.e., Hispanic students) would have substantially changed the association of Hispanic ethnicity with paid work. Demographics for the entire sample are displayed in Table 2.

Table 2

Sample Demographics and Logistic Regression Model Predicting Paid Work

Sample Demographics and Logistic Regression Model Predicting Paid Work
Sample Demographics and Logistic Regression Model Predicting Paid Work

Student skill factors

We obtained ratings of student skills and abilities from the Parent Interview and Student School Program Survey. Parents rated students' communication skills, ability to understand others, self-care skills (i.e., feeding and dressing independently), ability to get to places outside the home independently, and social skills. As reflected in Table 3, some response levels were combined. Parents rated their child's social skills on 11 items (e.g., makes friends easily, seems self confident in social situation, starts conversations rather than waiting for others to start, receives criticism well), 9 of which were from the standardized Social Skills Rating System (Gresham & Elliott, 1990). Parents indicated the frequency with which children demonstrated these behaviors (never  =  0, sometimes  =  1, always  =  2). We reversed scores for negatively worded items and summed ratings to form a social skills scale. Consistent with Wagner, Kutash, Duchnowski, Epstein, and Sumi (2005), we considered scores one standard deviation (SD) below the mean (0–10) of the NLTS2 sample as low social skills, within one SD of the mean (11–16) as medium social skills, and greater than one SD (17–22) as high social skills.

Table 3

Student Skills and Logistic Regression Models Predicting Paid Work

Student Skills and Logistic Regression Models Predicting Paid Work
Student Skills and Logistic Regression Models Predicting Paid Work

We obtained teachers' ratings of students' classroom behaviors and skills from the Student School Program Survey. Ratings from special education and vocational classes were combined, with the special education classroom rating selected if both were available. We derived a classroom social scale (3 items) from teacher ratings of the student's ability to get along with peers, follow directions, and act appropriately in class (1  =  not at all well, 2  =  not very well, 3  =  pretty well, 4  =  very well). If a single variable was missing, we imputed its value using the mean of the other two items. A sum of 3 to 7 was considered low ability; over 7 to 9, as medium ability; and over 9 to 12, as high ability. We derived a classroom behavior scale (4 items) from teacher responses to how frequently (1  =  rarely, 2  =  sometimes, 3  =  usually, 4  =  almost always) the student completes homework on time, participates in class discussions, stays focused on his/her work, and withdraws from social contacts or class activities (with scores reversed for the negatively worded item). Scores of 4 to 9 were considered low; over 9 to 13, medium; and over 13 to 16, high. Self-advocacy was based on teacher ratings of how well the student asked for what he or she needed to do his or her best in class (1  =  not at all well, 2  =  not very well, 3  =  pretty well, 4  =  very well).

Family factors

Although several variables reflecting family income were available in the Parent Interview, poverty status was most closely correlated with paid work experiences. We also examined head of household education and employment status, as well as difficulty with transportation (see Table 4). Because parent expectations may impact youth employment, we examined parent ratings of whether they expect that the student will be self-supporting (definitely will not, probably will not, probably will, definitely will). A household responsibilities scale was created based on the sum of ratings of how frequently (1  =  never, 2  =  sometimes, 3  =  usually, 4  =  always) the student fixes his or her own breakfast, does laundry, cleans his or her room, and picks up a few things at the store. We considered scores of 4 to 8 to be low and compared them to moderate to high scores of 9 to 16. Because parental expectations regarding whether the student would ever have a paying job were highly correlated with current/recent work experience, we did not consider this question in our modeling.

Table 4

Family Factors and Logistic Regression Models Predicting Paid Work

Family Factors and Logistic Regression Models Predicting Paid Work
Family Factors and Logistic Regression Models Predicting Paid Work

School program

The Student School Program Survey included information on school programs that could prepare students for work and future careers (e.g., students' participation in prevocational or vocational classes, in-school work experience, Individualized Education Program (IEP) prevocational or vocational goals, and a variety of job-related and career experiences and supports (see Table 5). Three programs (i.e., internship, tech prep, and entrepreneurship programs) were uncommon and, thus, were combined for this analysis.

Table 5

School Programs and Logistic Regression Models Predicting Paid Work

School Programs and Logistic Regression Models Predicting Paid Work
School Programs and Logistic Regression Models Predicting Paid Work

Community characteristics

We used the School Characteristics Survey to determine the type of community (i.e., rural, suburban, or urban) in which the school was located and the availability of public transportation and/or transportation for individuals with disabilities (see Table 6). Information on whether the student attended his or her neighborhood school was obtained from the Parent Interview.

Table 6

Community Factors and Logistic Regression Models Predicting Paid Work

Community Factors and Logistic Regression Models Predicting Paid Work
Community Factors and Logistic Regression Models Predicting Paid Work

Data Analysis

Although our primary aim was to examine the relationship of predictors to paid work, our secondary interest was in describing the demographic and skill characteristics of secondary-age students with severe disabilities. We calculated descriptive statistics for demographic variables, paid work status, and each group of predictor variables. For both descriptive and modeling estimates (except for work characteristics data), we employed sampling weights for the appropriate instrument (i.e., Parent Interview, Student School Program Survey, or School Characteristics Survey). As recommended in the NLTS2 training materials, if multiple instruments were employed, weights for the instrument with the smallest sample size were used. Balanced repeated replication using 32 replicate weights provided by SRI were used to calculate standard errors and 95% confidence intervals.

For most variables, frequencies for all response levels are reported. For school program and community data, many variables had only two response levels (yes or no), so only the percentage of affirmative responses is reported. Sample sizes for each group of variables vary because different surveys produced different response rates and because, in some cases, questions were not asked regarding all students. For example, data about vocational education and career-related services were only available for students in high school. Imputation for missing data was generally not used, except for a few variables described previously. To have a consistent data set for modeling each group of variables, we used a subset with no missing values for all variables in a group for descriptive and model statistics, as reported in Tables 1 through 6.

We used weighted logistic regression modeling to examine the relationship of predictor variables to the primary outcome (i.e., paid work). Unweighted models were also examined for the whole sample and stratified by primary disability. In stratified models, the associations between paid work and predictor variables in the autism group frequently differed from associations in the other two groups. Therefore, weighted models were chosen as more reflective of relationships between predictors and paid work in this sample of students with severe disabilities. Given these differences in associations, it was important to check interactions of predictor variables with primary disability in weighted models. No two-way interactions were significant at the .05 level in weighted models.

For each variable of interest, modeling proceeded in two stages. First, a logistic model was fit for each variable in a group with adjustment for demographic variables (age, sex, ethnicity, disability group). We examined two-way interactions of the variable of interest and all demographic variables. Second, we combined all variables that had a significant association, p < .05, with paid work in Stage 1 with demographic variables to produce a multivariate model for each group of predictors. Variables not included in this model are noted in the tables.

Results

What Are the Work Experiences of Adolescents With Severe Disabilities?

Overall, only 27.9% of students with severe disabilities in this sample had accrued paid work experience (see Table 1). The weighted percentage of students with paid work differed by disability group, with 31.3% of students with intellectual disability having paid work compared to 17.4% of students with multiple disabilities, and 11.0% of students with autism, Rao-Scott χ2(2, N  =  1510, p < .001. (The percentages and means in this section are unweighted estimates for students with paid work or community jobs.) Across all 270 students who had paid work, the top four types of jobs were categorized as maintenance (32.7%), food service (16.9%), clerk (12.4%), and personal care (10.5%); 45.7% worked only during the school year; 33.8% worked during both summer and school year; and 20.5% worked only during the summer. The 180 students who had community jobs (vs. school-sponsored jobs) worked an average of 8 hours/week (interquartile range  =  3 to 20) during the summer and 3 hours/week (interquartile range  =  2 to 7) during the school year. Average hourly pay for working students was $5.00 (interquartile range  =  $3.33 to $6.00). The most common form of transportation to a community job was getting a ride from a family member or friend/coworker (39.6%), but many students walked or rode a bike (34.1%). Public transportation (7.1%) and agency transportation (2.2%) were less frequently used. For students involved in paid work–study jobs, 75.7% also received school credit for this experience.

What Student Demographic Factors Are Associated With Paid Work During School?

Using weights that produce nationally representative estimates, we found that the prevalence of the three primary disability groups in the study sample was 78.3% for intellectual disability, 5.4% for autism, and 16.2% for multiple disabilities. As shown in Table 2, the average age was 15.8 years and the sample was predominantly White (58.2%). The proportion of boys was 62.9% in the entire sample, with a higher proportion of boys (81.3%, 95% CI 76.8–85.9) evident among students with autism. As shown in Table 2, the odds that students had paid work increased significantly with age. Hispanic students were significantly less likely to have paid work experience than were White students. Students with an intellectual disability had an almost four-fold odds of working than students with autism.

What Skill Factors Are Associated With Paid Work During High School?

Information regarding evaluations of students' skills is displayed in Table 3. We examined the association of each of these skills to paid work using a series of logistic regression models. In the first set of logistic models, we calculated odds ratios for each variable of interest adjusted for demographics (i.e., age, sex, race/ethnicity, and primary disability; see Odds Ratio 1 column). Students who communicated with others well and were independent in self-care had approximately three times the odds of having paid work than students who had difficulty with communication or self-care. The ability to get to places outside the home independently was strongly associated with paid work, with a linear trend of increasing work experience with increased ability to navigate. The ability to understand others and social skills scores were not significantly associated with paid work.

In the second stage of modeling, we combined variables that had significant relationships to paid work and demographics in a single multivariate logistic model. In this combined model, only the ability to get places outside the home remained significant, with increasing ability associated with increased odds of paid work when compared to students who could not get to places well (see Odds Ratio 2 column of Table 3).

Teachers' ratings of student behaviors in the classroom are as follows. On the classroom social scale, 21% (95% CI 15.3–26.6) scored in the low range, 46.8% (95% CI: 39.6–54.1) in the medium range, and 32.2% (95% CI: 25.1–39.3) in the high range. On the classroom behavior scale, 22.3% (95% CI: 16.8–27.7) scored in the low range, 51.0% (95% CI: 42.5–59.5) in the medium range, and 26.7% (95% CI: 19.7–33.7) in the high range. For self-advocacy (i.e., how well the student asked for what he or she needed in the classroom), 10.7% (95% CI: 6.3–15.2) performed not at all well, 32.3% (95% CI: 26.6–38.1) not very well, 44.1% (95% CI: 35.2–53.0) well, and 12.8% (95% CI: 7.8–17.9) very well. Teacher rankings of classroom social skills, behavior, and self-advocacy were not significantly related to paid work (data not shown). Because these variables were not strongly related to paid work and because including data from the Student School Program Survey substantially decreased the sample size (from 1,410 to 720), we chose not to combine the two sources of data.

What Family Factors Are Associated With Paid Work During School?

Demographic information about the families of students with severe disabilities are presented in Table 4. In the logistic models, a parent having some college (compared to a college degree) and students having regular household chores were both associated with increased odds of paid work. Parental expectations were the family factors most strongly associated with paid work experience. Students whose parents expected them to eventually become self-supporting had a significantly increased odds of having paid work compared to students whose parents did not expect them to become self-supporting. There was no association of head of household employment, income, or transportation difficulty with paid work status. When combined in a single model (see Odds Ratio 2 column in Table 4), household responsibilities and parental expectations were both significantly associated with paid work.

What School Program Factors Are Associated With Paid Work During School

As shown in Table 5, most students were taking prevocational or vocational classes and received some type of career-related services. Among such services, career skills assessment and job skills training were significantly associated with paid work. Participation in internship, tech prep, or entrepreneurship programs had the strongest association with paid work, OR  =  4.34, but the total number of students participating in these programs was quite small. The two types of IEP goals, prevocational and vocational, had opposite relationships with paid work, with prevocational goals negatively associated with paid work and vocational goals positively associated. Because a large proportion of students had both types of goals, we decided to examine their interaction. There was a significant interaction of goal type, such that students who had only vocational goals had increased odds of paid work whereas those with prevocational goals (with or without vocational goals) were not significantly different from students who had no prevocational or vocational goals. When significant variables were combined (IEP goal type, career skills assessment, job skills training, and internship programs), only IEP goal type (including the interaction term) was significant.

What Community Factors Are Associated With Paid Work During School?

Information about the communities in which students with severe disabilities live is provided in Table 6. In logistic regression models, only the availability of transportation for people with disabilities was significantly associated with increased odds of paid work.

Discussion

The Individuals with Disabilities Education Improvement Act (2004) clearly identifies meaningful employment as a valued postschool outcome for students with disabilities. The work-related experiences that young people with disabilities have during middle and high school can provide both a strong foundation for and a critical bridge to their future careers (Rusch et al., 2009; Test et al., 2009). For adolescents with severe disabilities, whose outcomes into and throughout adulthood remain disappointing, these types of early transition experiences take on increased importance. In this study we provide (a) a national portrait of youth with severe disabilities and their career-related transition experiences and (b) exploration of student-, family-, school-, and community-level factors associated with paid work experience during secondary school. Our findings extend the transition literature in several important ways.

First, we found that a fairly small proportion of youth with severe disabilities was accessing paid work experiences during secondary school. Despite clear and longstanding consensus surrounding the importance of hands-on vocational experiences for these adolescents, nearly three quarters of students had not held a paid job at any point during the prior year. Moreover, work experiences were further restricted for youth served under the special education categories of multiple disabilities or autism as well as for youth who were Hispanic (cf. Hasnain & Balcazar, 2009). Such outcomes contrast sharply with the work experiences of adolescents without disabilities who commonly hold after-school, summer, and/or work–study jobs at some point during high school (Zimmer-Gembeck & Mortimer, 2006). Although the NLTS2 dataset offers few indicators of the overall quality and relevance of these work experiences, most positions do not reflect viable long-term positions for students. For example, the majority of jobs were time-limited (i.e., occurring either during the school year or during the summer) and provided limited monetary compensation and few working hours. Thus, although these jobs could provide a valuable context for accruing preliminary work experience, they should not represent the permanent postschool work destinations of students (Carter, Trainor, Ditchman, Swedeen, & Owens, in press).

Second, we identified several skill domains that were strongly associated with greater access to paid employment and may reflect potential foci of instructional efforts during middle and high school. Youth with severe disabilities who were perceived to have less difficulty related to communication and self-care skills were significantly more likely to report having paid employment. Although such skills are often cited as important to finding and maintaining employment, we are unable to disentangle whether these perceived skill deficits constitute direct barriers to employment, whether they instead influenced others' perceptions of students' capacities to work, or both. Given our focus on paid work experiences in the community, it is not surprising that mobility skills were so strongly associated with employment outcomes. The mobility difficulties many youth with severe disabilities encounter can introduce additional barriers to traveling to and from a job as well as navigating the workplace (Darrah, Magill-Evans, & Galambos, 2010).

In light of the emphasis such skills have received within the literature (Benz, Yovanoff, & Doren, 1997; Martorell, Guiterrez-Recacha, Pereda, & Ayuso-Mateos, 2008), we were initially surprised that teachers' ratings of social skills, behavior, and self-advocacy were not strongly associated with paid work during secondary school. Because teachers' ratings of these skill domains were anchored to classroom contexts, rather than to the workplace, it may be that different combinations of social-related and self-determination skills are more directly linked to job success (Carter & Wehby, 2003). Given the more prominent role that adults must often play in helping youth with severe disabilities identify and connect to community jobs, it may also be that such skills play a more central role in the work outcomes of students with high-incidence disabilities (Carter, Trainor, Ditchman, Swedeen, & Owens, 2009).

Third, the important and influential role of families in the lives of transition-age youth is regularly emphasized but less frequently studied (Kim & Turnbull, 2004; Trainor, 2008). We explored several family-level factors associated with paid work outcomes for youth. Although parental education and income level have previously been addressed within postschool follow-up studies, neither was found to have a strong association with work during secondary school. Instead, two other factors emerged as particularly salient. Youth whose parents expected them to eventually be self-supporting had more than three times the odds of holding a paid job versus youth whose parents did not hold such expectations. Further, youth who were given more household responsibilities had more than twice the odds of holding a paid job. Although the nature of these associations are likely quite complex and myriad unexplored factors may shape parents' expectations, these findings do support the value of encouraging parents to communicate high expectations for their children with severe disabilities and to provide informal opportunities for youth to assume responsibilities at home that may enhance employment success. Blacher, Kraemer, and Howell (2010) found that parents held varied work-related expectations depending on the nature of their child's disability.

Fourth, although paid work was quite limited for students with severe disabilities during this first wave of the NLTS2, the large majority of students were reported to have received some type of vocational or career-related coursework, instruction, and/or support. For example, nearly three quarters of students with severe disabilities took at least one course that was categorized as prevocational or vocational. Moreover, nearly all of the school-sponsored programmatic efforts were associated with increased odds of paid work, although this association was significant for only three variables (i.e., career skills assessment; job skills training; and internship, tech-prep, or entrepreneurship programs). Although the introductory and preparatory career development experiences provided by schools are certainly important, it should not automatically be assumed that these experiences will seamlessly translate into actual work experiences during adolescence without additional efforts.

In addition to programmatic efforts, the focus of students' IEP goals also appears to be a relevant factor. The transition mandates within the Individuals With Disabilities Education Act Amendments—IDEA (2004) dictate that the IEPs of youth must include “appropriate measurable postsecondary goals based upon age-appropriate transition assessments related to training, education, employment, and, where appropriate, independent living” [italics added] (300.320(b)). Youth with vocational goals written in their IEPs had more than six times the odds of holding a paid job. Students with prevocational goals in their IEPs, however, were no more likely to report such employment than were those with no prevocational or vocational skills goals. Moreover, nearly one third of students did not have any type of vocational-related goal as a primary goal in their IEP. Previous research suggests that teachers' expectations can impact students' employment experiences (Carter et al., 2010; Trainor, Carter, Owens, & Swedeen, 2008). Although we cannot evaluate the specific nature of students' vocational goals or the contexts in which they were derived, it may be that the assessment, planning, collaboration, and accountability associated with goal setting in this domain facilitates the steps required to connect youth to community jobs.

Fifth, although relatively little research has been focused on community-level differences associated with the employment of adolescents with disabilities, we were surprised that a strong association did not exist between paid work and community type (i.e., rural, suburban, urban), enrollment in a neighborhood school (an indicator of a students' residential proximity to school programming), or the availability of public transportation. Only the availability of accessible transportation (i.e., transportation for individuals with disabilities) predicted work experiences. Given the mobility skill difficulties experienced by this segment of students, these findings reinforce the importance of addressing logistical considerations during planning and establishing community-level supports as part of comprehensive transition programming.

Limitations and Recommendations for Future Research

Several limitations to this study suggest areas for future researchers addressing early work experiences of youth with severe disabilities. First, as with any secondary data analysis, we were limited to those variables contained in the extant database. In an effort to address the breadth of experiences encountered by youth with disabilities, the NLTS2 necessarily sacrificed depth related to some domains. Additional research—perhaps drawing upon state- or district-level data—is needed to provide greater insight into the nature of students' work experiences (e.g., alignment with long-term interests, inclusiveness, benefits) and the additional factors that may influence involvement (e.g., job development efforts of educators or parents, availability of on-the-job support, local economic opportunities). For example, we know little about the quality of the jobs that students held (Brooke, Revell, & Wehman, 2009). Second, we focused narrowly on paid employment experiences. Such an emphasis reflects prevailing values and high aspirations for students with severe disabilities (Johnson, 2004), and there is ample evidence that such experiences are both attainable and sustainable (Brown, Shiraga, & Kessler, 2006; Certo et al., 2008). Yet, this emphasis is not intended to minimize the contributions that internships, volunteer experiences, and other informal work opportunities can make to a student's early career preparation. Third, in an effort to characterize the work experiences of students with severe disabilities, we neglected to examine whether and how identified associations might be similar for youth served under other special education categories and/or for youth without disabilities. Understanding which factors may be salient for all students could inform the design and delivery of inclusive career development programs that better address the needs of a broad range of students within a given school. Fourth, although correlational analyses such as ours can provide important insight into the associations between students' experiences and key factors, experimental studies remain sorely needed to elucidate how specific transition strategies—alone or in packaged interventions—influence the outcomes students with severe disabilities attain during high school (cf. Carter et al., 2009).

Conclusion

In this article we provide new insight into the foundational work experiences of adolescents and call upon practitioners and policymakers to consider the ways in which such adolescent experiences might enhance the early career trajectories of all students. These findings highlight the elusiveness of early work experiences for youth with severe disabilities and call attention to malleable factors that may play a role in shaping success during high school. Although providing youth with these experiences is an essential endeavor, it is also critical that policies, service delivery models, and other programmatic issues be addressed concurrently to increase the opportunities, supports, and encouragement that young people with severe disabilities need in order to find and maintain meaningful work.

Acknowledgments

The research reported here was supported by the Institute of Education Sciences, U.S. Department of Education, through Grant R324S060023. The opinions expressed are those of the authors and do not represent views of the U.S. Department of Education.

References

References
Benz
,
M.
,
P.
Yovanoff
, and
B.
Doren
.
1997
.
School-to-work components that predict postschool success for students with and without disabilities.
Exceptional Children
63
:
151
165
.
Bianco
,
M.
,
D. F.
Garrison-Wade
,
R.
Tobin
, and
J. P.
Lehman
.
2009
.
Parents' perceptions of postschool years for young adults with developmental disabilities.
Intellectual and Developmental Disabilities
47
:
186
196
.
Blacher
,
J.
,
B. R.
Kraemer
, and
E. J.
Howell
.
2010
.
Family expectations and transition experiences for young adults with severe disabilities: Does syndrome matter?
Advances in Mental Health and Learning Disabilities
4
:
3
16
.
Blackorby
,
J.
and
M.
Wagner
.
1996
.
Longitudinal postschool outcomes of youth with disabilities: Findings from the National Longitudinal Transition Study.
Exceptional Children
62
:
399
413
.
Brooke
,
V. A.
,
G.
Revell
, and
P.
Wehman
.
2009
.
Quality indicators for competitive employment outcomes: What special education teachers need to know in transition planning.
Teaching Exceptional Children
41
(
4
):
58
66
.
Brown
,
L.
,
B.
Shiraga
, and
K.
Kessler
.
2006
.
The quest for ordinary lives: The integrated post-school functioning of 50 workers with significant disabilities.
Research and Practice for Persons with Severe Disabilities
31
:
93
121
.
Burbidge
,
J.
,
P.
Minnes
,
K.
Buell
, and
H.
Ouellette-Kuntz
.
2008
.
Preparing to leave school: Involvement of students with intellectual disabilities in productive activities.
Journal on Developmental Disabilities
14
:
19
26
.
Carter
,
E. W.
,
N.
Ditchman
,
Y.
Sun
,
A. A.
Trainor
,
B.
Swedeen
, and
L.
Owens
.
2010
.
Summer employment and community experiences of transition-age youth with severe disabilities.
Exceptional Children
76
:
194
212
.
Carter
,
E. W.
,
A. A.
Trainor
,
N.
Ditchman
,
B.
Swedeen
, and
L.
Owens
.
in press
.
Community-based work experiences of adolescents with high-incidence disabilities.
The Journal of Special Education
.
Carter
,
E. W.
,
A. A.
Trainor
,
N.
Ditchman
,
B.
Swedeen
, and
L.
Owens
.
2009
.
Evaluation of a multi-component intervention package to increase summer work experiences for transition-age youth with severe disabilities.
Research and Practice for Persons with Severe Disabilities
34
:
1
12
.
Carter
,
E. W.
and
J. H.
Wehby
.
2003
.
Job performance of transition-age youth with emotional and behavioral disorders.
Exceptional Children
69
:
449
465
.
Certo
,
N. J.
,
R. G.
Luecking
,
S.
Murphy
,
L.
Brown
,
S.
Courey
, and
D.
Belanger
.
2008
.
Seamless transition and long-term support for individuals with severe intellectual disabilities.
Research and Practice for Persons with Severe Disabilities
33
:
85
95
.
Chambers
,
C. R.
,
C.
Hughes
, and
E. W.
Carter
.
2004
.
Parent and sibling perspectives on the transition to adulthood.
Education and Training in Developmental Disabilities
39
:
79
94
.
D'Alonzo
,
B. J.
1978
.
Career education for handicapped youth and adults in the '70s.
Career Development for Exceptional Individuals
1
:
4
12
.
Darrah
,
J.
,
J.
McGill-Evans
, and
N. L.
Galambos
.
2010
.
Community services for young adults with motor difficulties—A paradox.
Disability and Rehabilitation
32
:
223
229
.
Gresham
,
F. M.
and
S. N.
Elliott
.
1990
.
Social Skills Rating System
.
Circle Pines, MN
American Guidance Service
.
Hasazi
,
S. B.
,
L. R.
Gordon
, and
C. A.
Roe
.
1985
.
Factors associated with the employment status of handicapped youth exiting high school from 1979 to 1983.
Exceptional Children
51
:
455
469
.
Hasnain
,
R.
and
F.
Balcazar
.
2009
.
Predicting community-versus facility-based employment for transition-aged young adults with disabilities: The role of race, ethnicity, and support systems.
Journal of Vocational Rehabilitation
31
:
175
188
.
Individuals with Disabilities Education Improvement Act of 2004, 20 U.S.C. § 1400 et seq
2004
.
Johnson
,
D. R.
2004
.
Supported employment trends: Implications for transition-age youth.
Research and Practice for Persons with Severe Disabilities
29
:
243
247
.
Johnson
,
D. R.
,
K.
McGrew
,
L.
Bloomberg
,
R. H.
Bruininks
, and
H. C.
Lin
.
1997
.
Results of a national follow-up study of young adults with severe disabilities.
Journal of Vocational Rehabilitation
8
:
119
133
.
Kohler
,
P. D.
and
S.
Field
.
2003
.
Transition-focused education: Foundation for the future.
The Journal of Special Education
37
:
174
183
.
Kraemer
,
B. R.
and
J.
Blacher
.
2001
.
Transition for young adults with severe mental retardation: School preparation, parent expectations, and family involvement.
Mental Retardation
39
:
432
455
.
Martorell
,
A.
,
P.
Gutierrez-Recacha
,
A.
Pereda
, and
J. L.
Ayuso-Mateos
.
2008
.
Identification of personal factors that determine work outcome for adults with intellectual disability.
Journal of Intellectual Disability Research
52
:
1091
1101
.
National Alliance for Secondary Education and Transition
2005
.
National standards and quality indicators: Transition toolkit for systems improvement
.
Minneapolis
University of Minnesota, National Center on Secondary Education and Transition
.
National Organization on Disability
2004
.
Survey of Americans with disabilities
.
Washington, DC
Author
.
Newman
,
L.
,
M.
Wagner
,
R.
Cameto
, and
A.
Knokey
.
2009
.
The post-high school outcomes of youth with disabilities up to 4 years after high school: A report from the National Longitudinal Transition Study–2 (NLTS2)
.
Menlo Park, CA
SRI International
.
Rusch
,
F. R.
and
D.
Braddock
.
2004
.
Adult day programs versus supported employment (1988–2002): Spending and service practices of mental retardation and developmental disabilities state agencies.
Research and Practice for Persons with Severe Disabilities
29
:
237
242
.
Rusch
,
F. R.
,
C.
Hughes
,
M.
Agran
,
J. E.
Martin
, and
J. R.
Johnson
.
2009
.
Toward self-directed learning, post-high school placement, and coordinated support constructing new transition bridges to adult life.
Career Development for Exceptional Individuals
32
:
53
59
.
Shandra
,
C. L.
and
D. P.
Hogan
.
2008
.
School-to-work program participation and the post-high school employment of young adults with disabilities.
Journal of Vocational Rehabilitation
29
:
117
130
.
SRI International
2000
.
1
20
The National Longitudinal Transition Study-2 (NLTS2): Conceptual framework and research questions
.
Menlo Park, CA
Author
.
Test
,
D. W.
,
V. L.
Mazzotti
,
A. L.
Mustian
,
C. H.
Fowler
,
L.
Kortering
, and
P.
Kohler
.
2009
.
Evidence-based secondary transition predictors for improving postschool outcomes for students with disabilities.
Career Development for Exceptional Individuals
32
:
160
181
.
Trainor
,
A. A.
2008
.
Using cultural and social capital to improve postsecondary outcomes and expand transition models for youth with disabilities.
The Journal of Special Education
42
:
148
162
.
Trainor
,
A. A.
,
E. W.
Carter
,
L.
Owens
, and
B.
Swedeen
.
2008
.
Special educators' perceptions of summer employment and community participation opportunities for youth with disabilities.
Career Development for Exceptional Individuals
31
:
144
153
.
Wagner
,
M.
,
K.
Kutash
,
A. J.
Duchnowski
, and
M. H.
Epstein
.
2005
.
The Special Education Elementary Longitudinal Study and the National Longitudinal Transition Study: Study designs and implications for children and youth with emotional disturbance.
Journal of Emotional and Behavioral Disorders
13
:
25
41
.
Wagner
,
M.
,
K.
Kutash
,
A. J.
Duchnowski
,
M. H.
Epstein
, and
W. C.
Sumi
.
2005
.
The children and youth we serve: A national picture of the characteristics of students with emotional disturbances receiving special education.
Journal of Emotional and Behavioral Disorders
13
:
79
96
.
Wehman
,
P.
and
W.
Revell
.
1997
.
Transition into supported employment for young adults with severe disabilities: Current practices and future directions.
Journal of Vocational Rehabilitation
8
:
65
74
.
Zimmer-Gembeck
,
M. J.
and
J. T.
Mortimer
.
2006
.
Adolescent work, vocational development, and education.
Review of Educational Research
76
:
537
566
.

Author notes

Erik W. Carter, PhD (e-mail: erik.carter@vanderbilt.edu), Associate Professor, and Audrey A. Trainor, PhD, Associate Professor, Department of Rehabilitation Psychology and Special Education, University of Wisconsin-Madison; 1000 Bascom Mall, Madison, WI 53705. Diane Austin, MS, Department of Occupational Therapy, University of Wisconsin-Madison, 3122 Center Ave., Madison, WI, 53704. The first author is now affiliated with the Department of Special Education, Peabody College, Vanderbilt University.