Postsecondary Expectations of High-School Students With Autism Spectrum Disorders

Sample and Interview Procedure

In this cross-sectional descriptive study, 31 students with ASD from Wisconsin (n=22) and Tennessee (n=9) were drawn from two larger, ongoing longitudinal studies of adolescents with ASD (Smith, Greenberg, & Mailick, 2014; Taylor & Henninger, 2015). All study participants from both sites received an independent diagnosis of autism, Asperger Disorder, or PDD-NOS from a medical, psychological, or educational professional. Both samples were recruited through local autism groups, clinics and participant registries. All study participants included in our analytic sample had an intellectual quotient (IQ) of 70 or greater and were transition-aged youth, meaning they were enrolled in high school at the time of the interview. Further, all participants in our study were verbally fluent (defined as the ability to use sentences with two clauses). The Wisconsin participants completed the Social Communication Questionnaire (SCQ) and met criteria for verbal fluency (Rutter, Bailey & Lord, 2003). Because we did not have comparable data from the Tennessee sample, we assessed for verbal fluency through transcript reviews, in which nine cases were eligible. Our final combined sample was 79% male and 93% Caucasian (Table 1). The average age of the students was 16 years and the majority were in the 11^th or 12^th grade. IQs ranged from 70 to 128 as measured by either the Wechsler (Wechsler, 1999) or Stanford Binet V (Roid, 2003) for the Wisconsin and Tennessee samples, respectively. There were no significant differences in race, sex, or household income between the Wisconsin and Tennessee samples.

In both studies, students participated in an in-person interview which was part of a larger study protocol that included questions about friendships, activities, and the transition to adulthood. For the Wisconsin sample, interviews were taken as part of the Time 1 interview, collected before adolescents and parents participated in the Transitioning Together intervention. For the Tennessee sample, interviews were conducted as part of the Time 1 assessment, before the students left high school. The questions were constructed by the investigators and were reviewed and tested by teams at both sites in order to reach consensus regarding final wording and ordering. The questions were administered identically in both locations. In each case, questions were given at the end of a larger assessment battery which included cognitive testing and an administration of an ADOS (Lord, Rutter, DiLavore, & Risi, 2001). Both sites maintained a two probe rule for each question wherein the interviewer was able to ask two follow-up questions for clarification or to encourage an expansion of a response. Sites were in regular communication during the course of data collection and monitored transcripts to maintain fidelity of administration. Our analyses focused on five open-ended questions taken from these interviews. The first set of questions asked the respondent what they thought life would be like after they leave high school in terms of work, postsecondary schooling, friendships and living arrangement. The interviewers asked each question separately to ensure answers were collected within each domain. The final asked “what do you think someone has to do to be considered an adult?” Adolescent interviews were videotaped, with consent, and transcribed verbatim to written text for analysis. All transcriptions were imported into NVivo v. 10, a computer assisted data analysis software (CAQDAS) that assists in the process of organizing and coding raw data

Analysis

Arguably, qualitative methods are optimal for those who “wish to capture the lived experience from the perspectives of those who live it and create meaning from it” (Padgett, 2008, p. 15). Our goal for this study was not to draw inferences, but to describe this population; therefore, we chose fundamental qualitative description as our mode of inquiry. Unlike other forms of qualitative description such as grounded theory and ethnographic studies, fundamental qualitative description produces a complete end-product within itself and is not a prelude to further research studies (Sandelowski, 2000, p.335). Coders also seek descriptive validity rather than interpretive validity during analysis. We used content analysis, and more specifically directed content analysis, as our analytical approach because content analysis is often used to describe and quantify phenomena (Elo & Kyngäs, 2007). In directed content analysis, the codes are determined before and during data analysis; however, these pre-defined codes do not preclude the possibility of new or emerging themes (Hsieh and Shannon, 2005; Humble, 2009). The approach is sometimes referred to as theory-led thematic analysis (Hayes, 1997; Hendry & Kloep, 2010). This method is best suited for our study because, in addition to inductively coding participant responses, we had pre-defined codes that were informed by past literature. While different codes were applied, the same methods were used to answer both research questions. All researchers took part in the data analysis, independently crosschecked each other’s coding schemes and participated in debriefing sessions to discuss emerging themes (Krefting, 1990).

Aim 1

The first set of questions asked adolescents what they expected life to be like after high school. Our findings fell into one of three categories: (1) the student expected to fulfill a traditional transition role, (2) the student did not expect to fulfill a traditional transition role, or (3) the student did not know what to expect after high school. Overall, we found that the majority of students expected to attain traditional markers of adulthood in the years following high school; approximately 90% of respondents expected to attend more schooling, 62% planned to work and 77% had some plan around friendships (Table 3). Only thirteen students (42%) planned to leave the parental home for independent or semi-independent living arrangements.

Aim 2

Similar to the general population, our sample included both traditional and nontraditional adult roles in their definitions of adulthood (Molgat, 2007). Individualistic criteria, qualities of character, and transition roles were all categories in the data, as well as an unanticipated category of chronological age (shown in Table 4). A study conducted by Arnett (1997, 2001, 2004), found that individualistic criteria and qualities of character were highly endorsed characteristics of adulthood; this was also true of our sample. Individualistic criteria (e.g., financial independence, independent living skills, independent living and not depending on others) were cited by all but one student as important (Refer to Table 4). Character qualities (e.g., responsibility, maturity, acting like an adult and abiding by the rules) were cited by two thirds of students. Responsibility was the most frequently cited quality of character. Adolescents’ definitions of responsibility were vague, however; that is, many students stated the importance of ‘being responsible’ but did not explain what responsibility meant to them. Features of maturity were also loosely defined among the sample although the recognition of others’ feelings and ethical concerns were raised in reference to the meaning of maturity. Interestingly, students linked household skill sets (such as washing dishes, learning to do laundry and making meals) to independent living, personal survival and taking care of your own self. In other words, the ability to acquire “higher order” individualistic criteria was contingent on the attainment of independent living skills first.

In contrast to the frequently-mentioned themes of individualistic criteria and qualities of character, few transition roles (e.g., achievement of traditional markers needed to be an adult) emerged in our data with the exception of establishing a romantic relationship (cited by two students; 6.5%) and finding a job (cited by seven students; 22.6%). An unexpected standardized marker of adulthood, chronological age, also emerged and was cited by 13 students (41.9%). Eight of the 13 students (61.5%) who identified chronological age as an indicator of adulthood linked age with other markers of adulthood such as: buying a house, not depending on others for help, act like an adult and taking responsibility.

Nontraditional Adult Roles Should Be Incorporated in Transition Plans and Services

Nontraditional roles were considered to be more important to the adult identity than traditional markers in our sample and should be included in transition plans. Even though students with ASD have difficulties identifying prosocial behavior, the concept of maintaining a moral code was an important adult trait in our sample (Moran et al., 2011; Senland & Higgins-D’Alessandro, 2014). Two students believed that abiding by rules or laws is necessary for adulthood and nine adults reported personal responsibility was important. Becoming independent was another common theme in our sample. Adolescents linked passage into a particular adult role to being independent. Among the students who mentioned independence as a criterion, 30% also endorsed a transition marker (e.g., having a job or a romantic relationship). While the interplay between transition roles and individualistic criteria on adult identity is supported in prior literature, little is known of how this hierarchal sequencing of such criteria impacts the formation of postsecondary expectations (Shanahan, Porfeli & Mortimer, 2005). The interdependency between independence and traditional adult roles highlights the need for interventions that facilitate independence across all transition domains. Interventions designed to promote the independence of adolescents with ASD, such as self-monitoring, management video modeling and work systems, have shown to be effective (Hume et al., 2014).

As noted above, independent living is one of the three goal areas for individualized education plans. Independent living is associated with the acquisition of various domains of independence; individuals who live independently are more likely to run a household, form relationships, and have stronger interactions with parents (Kins & Beyers, 2010) and higher levels of discrete daily living skills also are associated with better outcomes during adulthood (Farley et al., 2009). Unfortunately, deficits in daily living skills are common for individuals with ASD, including those without co-occurring intellectual disability; however, significant growth in these skills is possible during adolescence and young adulthood (Smith, Maenner, & Mailick, 2012). Our data suggest that adolescents with ASD recognize the importance of daily living skills as indicators of adulthood and would benefit, if not appreciate, instruction in the high school environment before leaving the parental home. The difficulty that students on the autism spectrum face in completing activities of daily living combined with their awareness of the importance of these activities for adult life, supports the value of continued application of skills training.

Secondary Institutions Need to Prepare Students for the Uncertainties of Adulthood

The unpredictability of adulthood can be challenging for persons with ASD as they are characteristically resistance to change and uncertainty. Although change in routines can be difficult, research shows that individuals with ASD often follow a nonlinear trajectory to adulthood. Nationally, a third of college students with an ASD start school at a two-year college and transfer to a four-year college (Wei et al., 2014a). Employment pathways also vary. In a national study of young adults with ASD, of youth with ASD primarily interested in employment, roughly 10% attended postsecondary schooling up to two years after high school and then later worked part time and 10.1% of this group worked part time and switched from a vocational training program to college (Wei et al., 2014b). Further, in a sample of individuals with ASD without intellectual disability, Taylor and Mailick (2014) found only 25% were consistently engaged in competitive employment or post-secondary education over a 10 year period, whereas 40% were sometimes engaged in these activities and sometimes not, suggesting that maintaining activities over time may be difficult. Taken together, these findings indicate that multiple transitions and changes in daily routines may be expected for individuals with ASD during young adulthood. However, the students in our sample did not anticipate that such dynamic changes would happen to them. In fact, most students had linear postsecondary plans. Educators should consider how to prepare students not only to achieve their post-secondary goals but also to have coping skills needed for multiple life course transitions.

Eight of the 11 students in our sample who believed chronological age is an indicator of adulthood also endorsed another aspect of adulthood such as: buying a house, not depending on others for help, act like an adult and taking responsibility. This suggests a broad view of the benchmarks of maturity and may reflect how engagement in transition planning (which emphasizes that secondary school services are tied to chronological age), influences thinking about markers of adulthood. It also may reflect literal thinking regarding definitions of adulthood, as many individuals on the spectrum rely on hierarchal thinking and rigid rule-based reasoning (Tarbox, Zuckerman, Bishop, Olive & O’Hora, 2011). Regardless of the possible reasons for the linkages between age and other markers in our sample, it is important to note that the transition to adulthood is taking longer in contemporary society and adult roles are uncommonly met before the age of 21 (Furstenberg, Kennedy, McLoyd, Rumbaut & Settersten, 2004). Transition service providers can help guide the formation of expectations in the context of these normative trends and support realistic goal setting as part of the transition planning process.

There is a Need for Community-based Services that Target Non-college Bound Students

The sequencing of transition events reported for many students in our sample began with the enrollment or completion of college. Although postsecondary schooling was not cited as an important indicator of adulthood, it was central to many students’ future plans and was linked with the attainment of other adult roles. Many students felt that they would to go to college and then make new friends, leave their parent’s home or find a job. The idea that human development is structured through a series of ordered stages (e.g., age-grading) is commonly found among developmental stage theorists (Arnett et al., 2011). Of notable importance is the sequential order in which the individual moves through these stages; one cannot graduate to the next stage until the previous stage is realized (Hogan & Astone, 1986). This expectation of ordered-stages was observed in the participants in our study, with 37% of respondents discussing events as being sequenced. Although this is not surprising, the expectation that postsecondary life will be linearly formed and sequenced could have negative consequences if it is difficult for continuity to be achieved. Based on past research we know that many individuals with ASD do not enroll in postsecondary education (Shattuck et al., 2011a). Similarly, 70% to 80% of individuals with ASD continue to show marked social impairment into adulthood (Farley et al., 2009; VanBergeijk, Klin & Volkmar, 2008) and many young adults continue to experience social isolation in college (Madriaga, 2010; Welkowitz & Baker, 2005). In hopes to increase the enrollment and retention of students with ASD, colleges are finding new ways to provide social supports to students with ASD by including peer mentors and support groups and introducing etiquette courses into university settings (Adreon & Durocher, 2007; Farrell, 2004). However, there are fewer social supports available to young adults not enrolled in school (Tobin, Drager & Richardson, 2014). The Aspirations program, a social skills groups designed for 18 to 30 year old adults with ASD, and the PEERS intervention, designed for persons 18–23, have shown to positively impact the social outcomes for this population (Gantman, Kapp, Orenski & Laugeson, 2012; Hillier, Fish, Cloppert & Beversdorf, 2007). Further efforts must be made to replicate and disseminate community-based interventions.