Abstract
BACKGROUND:
Parent expectations are known to be significant predictors of postsecondary outcomes among youth with disabilities. However, little is known about the relationship between parent and youth expectations and their differential prediction of postsecondary outcomes among youth with disabilities.
OBJECTIVE:
To determine the relationship between parent and youth expectations among students with disabilities in the United States.
METHODS:
Through analysis of the National Longitudinal Transition Study-2 (NLTS2), we compared wave two parent and youth expectations across disability categories (N = 1,940) and explored the extent to which they predicted youth postsecondary outcomes.
RESULTS:
We identified that parent and youth expectations are related but significantly different, with parents consistently holding less independent expectations than youth. Both parent and youth expectations were identified as significant predictors of postsecondary outcomes, with youth expectations more strongly predicting education and independent living outcomes, and parent expectations more strongly predicting employment and Social Security benefit outcomes.
CONCLUSIONS:
Transition providers should work with parents and youth to develop collaborative goals and expectations to best support youth with disabilities for independent activities in adulthood.
Introduction
Research has consistently revealed that young adults with disabilities experience poorer outcomes after graduating from high school than their peers without disabilities. In fact, Newman et al. (2011) report that, up to eight years after leaving high school, youth with disabilities were less likely than their typically-developing peers to live independently (45% vs. 59%), enroll in any type of postsecondary education program (60% vs. 67%), or complete a postsecondary program (41% vs. 52%). Individuals with disabilities also made less money than their peers on average ($10.40 per hour vs. $11.40 per hour) and were less likely to have a checking account (59% vs. 74%) or a credit card (41% vs. 61%). As would be expected, these outcomes vary widely based on disability category (Newman et al., 2011). Nonetheless, it is important to understand factors that contribute to these outcomes in order to improve transition assessment, intervention, and outcomes for individuals with disabilities.
Previous research has shown a strong relationship between parent expectations and student outcomes for individuals with and without disabilities (Zhang et al., 2011). In the disability literature, many studies have used NLTS2 data to examine the role of parent expectations about independent living, postsecondary education, and employment in predicting outcomes for students with disabilities. For example, parent expectations were identified as a significant mediator of postsecondary outcomes for students with autism spectrum disorder (ASD; Kirby, 2016). Parent expectations of students with hearing impairments have also been shown to correlate positively with independent living, employment, and postsecondary education attendance (Cawthon et al., 2015) and function as an important antecedent of autonomy development in this same population (Garberoglio, Schoffstall, Cawthon, Bond, & Caemmerer, 2016). In students with severe disabilities, parent expectations about employment predicted the likelihood that an individual would gain employment following graduation (Carter et al., 2012). In students with intellectual disability, Papay and Bambara (2014) found that parent expectations for employment and postsecondary education were among the strongest predictors of those outcomes. Parent expectations have also been shown to be significant predictors of high school graduation rates and employment after high school across students from all disability categories (Doren, Gau, & Lindstrom, 2012). In fact, Wagner, Newman, and Javitz (2014) found that parent expectations about high school graduation predicted a student’s likelihood of graduating, even when controlling for demographic and disability factors.
Despite the evidence of the relevance of parent expectations, it is important to also consider youth’s own perspectives given the important role of self-determination for youth with disabilities. Self-determination is described as a “combination of skills, knowledge, and beliefs that enable a person to engage in goal-directed, self-regulated, autonomous behavior” (Field, Martin, Miller, Ward, & Wehmeyer, 1998, p. 2). The self-determination of youth with disabilities has been shown to be associated with positive postsecondary outcomes (Weymeyer, 2014). Thus, their own expectations for their futures, and their advancement toward those expectations, should be important considerations during the transition process.
In the general adolescent development literature, youth’s own expectations have been consistently shown to predict future achievement through empirical research informed by the Expectancy-Value Theory of Achievement Motivation (Wigfield & Eccles, 2000) as well as Social Cognitive Career Theory (Lent, Brown, & Hackett, 1994). Both theories outline ways in which the expectations that individuals have for themselves influence their efforts, persistence, and choices about performance, leading to eventual impacts on outcomes. These theories have been empirically supported in non-disability samples, suggesting expectations are important considerations during the transition to adulthood process. Yet, much less is known about the predictive power of youths’ own expectations for their futures in disability samples. In a study of 735 young adults from the Netherlands with mild intellectual disability, the individuals’ own expectations about future work level predicted their likelihood of obtaining a job and maintaining it for at least six months (Holwerda, van der Klink, de Boer, Groothoff, & Brouwer, 2013).
Only a few studies have examined the relationship between parent and youth expectations among youth with disabilities. One study of 341 young adults with intellectual disability in the Netherlands compared youth, parent, and teacher expectations about future employment (Holwerda, Brouwer, de Boer, Groothoff, & van der Klink, 2015). This study found that only teachers’ expectations significantly predicted employment after high school, with a complementary effect of parent expectations. However, this represents only a small group of people from one disability category. In a broader study, Powers and colleagues (2009) surveyed 279 parents and 242 youths with disabilities about their expectations for the future, finding that youth and parents identified the same three transition goals as top priorities: completing high school, having health insurance, and having a good doctor. However, youth were more likely than parents to report barriers to transition, placed more importance on getting married and having children, and reported higher levels of self-esteem than parents believed they had. Parents valued teacher support more highly than did youth. This study focused mainly on youth and parents’ desires for the future as opposed to their realistic expectations. Desires and realistic expectations have been shown to differ (Kraemer & Blacher, 2001), and desires likely do not predict future behavior as well as expectations. Powers and colleagues’ (2009) study did not follow up with students to collect data on outcomes after leaving high school, so it is unclear whether these desires were related to outcomes.
Youth’s own expectations for their futures have been shown to significantly predict outcomes in non-disability samples, whereas most research in disability groups has focused on the expectations of parents and how they predict postsecondary outcomes. Therefore, the purpose of the current study was to examine the relationship between youth and parent expectations, and how each relate to postsecondary outcomes in a nationally representative population of youth with disabilities. Accordingly, the current study utilized data from the National Longitudinal Transition Study-2 (NLTS2) to address the following aims: (1) Determine the consistency of youth and parent postsecondary expectations; (2) Describe differences between youth and parent postsecondary expectations with the entire sample and by disability category; and (3) Determine how youth and parent expectations differentially predict youth postsecondary outcomes. A prior NLTS2 report began the examination of the relationship between youth and parent expectations, using Wave 1 parent expectations and Wave 2 youth expectations. This report showed that overall parents held lower expectations than youth, but that their expectations were related (Wagner, Newman, Cameto, Levine, & Marder, 2007). However, this report did not use the longitudinal nature of the NLTS2 dataset to examine how these expectations differentially predict outcomes.
Methods
NLTS2 dataset
The NLTS2 is a longitudinal, nationally-representative study of students enrolled in special education services (Institute of Education Sciences [IES], n.d.). Data on the transition process were collected from individuals with disabilities as well as their parents, teachers, and schools. Data collection began in the year 2000 and occurred in five waves, each wave approximately two years apart. Youth were ages 13–16 when the study began and 21–25 when it was complete. School districts and students were randomly sampled and stratified based on geography, district/school size, demographic characteristics, age, and disability classification (IES, n.d.). The final sample included over 11,000 youth. The authors obtained a Restricted-Use Dataset Agreement, and all security procedures were followed. Sample sizes reported in this paper have been rounded to the nearest 10, in accordance with the agreement.
Sample
Sample Description
Sample Description
Notes. SSI, Supplemental Security Income. Dataset included the sample from Wave 2 that had expectation data for parent and youth; unweighted sample sizes and weighted percentages provided (using Wave 2 weights for all but the final four variables, for which we used “any wave” weights); percentages based on the available data for each variable. Unweighted sample sizes were rounded to the nearest 10 in accordance with the data-use agreement with the Institute for Education Sciences. *Final listed variable was created by combining multiple variables to separate those living on their own (including with roommate, partner/spouse, and in college dorm) from those reported to live with parents or in other supervised settings.
NLTS2 variables relevant to the research aims were examined statistically prior to analysis (e.g., normality, means, standard deviations, frequency, skewness, kurtosis). Table 1 contains NLTS2 variable names to allow for replication. The primary variables of interest in this analysis were from questions about youth and parent expectations. Youth were asked about their expectations at Waves 2 and 3, and parents were asked about their expectations at Waves 1, 2, and 3; individual expectation questions were skipped if the youth had already accomplished the outcome. Each time, respondents were asked how likely they felt it would be for the youth to attain different outcomes (e.g., graduate high school, attend postsecondary school, live away from home without supervision, get a paid job, earn enough to be financially self-sufficient; response options: definitely will, probably will, probably will not, definitely will not). Three expectation questions (i.e., attend postsecondary school, live away from home without supervision, and earn enough to be financially self-sufficient) were selected for this analysis because they addressed expectations about three key postsecondary outcome areas of interest. Expectation variables were reverse-scored for the purposes of this analysis (see Table 1). Additional variables were calculated to facilitate analysis of our research aims; specifically, we calculated a variable to capture the absolute value of individual differences between parents and youth on each expectation, and also created dichotomized variables indicating whether or not parents and youth agreed or disagreed in their expectations about whether the youth would or would not achieve each postsecondary outcome.
Descriptive variables used from Wave 1 included age, gender, race/ethnicity, disability category, and whether or not the youth was eligible for free or reduced meals in school (i.e., socioeconomic status). Data on relevant postsecondary outcomes (i.e., ever attended postsecondary institution, does not live with parents or other supervised situation, ever worked for pay outside the house, currently not receiving Supplemental Security Income [SSI]) were from Wave 5.
Data analysis
The Statistical Package for Social Sciences, Version 24, was used for data analysis. Cases were included in analyses where valid data was available for the statistical test (i.e., pairwise removal of missing data). Significance was set at p < 0.01. NLTS2-provided sampling weights were applied to all analyses in order to account for the sampling strategy and obtain estimates that correspond to the national population of students with disabilities at the start of the study. For within-Wave 2 analyses (aims 1 and 2), Wave 2 weights were used, and for cross-wave analyses (aim 3), weights designed to be used with any wave were applied. To determine consistency between youth and parent expectations (aim 1), we calculated Pearson correlation coefficients, examining general consistency in score patterns, and ran paired t-tests, examining if there were any significant changes between youth and parent. Summary and descriptive statistics were used to describe differences between youth and parent expectations, and groups were compared on frequency of parent-youth agreement using chi-square analyses with post-hoc z-tests and Bonferroni correction (aim 2). Finally, to determine how youth and parent expectations differentially predict youth postsecondary outcomes (aim 3), we conducted logistic regression analyses examining youth and parent expectations as predictors of four binary outcome variables from Wave 5 (yes or no: ever attended postsecondary education, lives independently or in an unsupervised setting, ever worked for pay outside the house since leaving high school, currently receives money from social security). In the logistic regression analyses, we controlled for gender, age, race/ethnicity, and receipt of free/reduced meals because these are well-measured variables in NLTS2 that have consistent evidence of relation to expectations. Each outcome was run in a unique model, with parent expectation (Wave 2), youth expectation (Wave 2), and covariates included as predictors.
Results
Aim 1: Consistency between youth and parent expectations
Youth and Parent Expectations Comparisons in Wave 2
Youth and Parent Expectations Comparisons in Wave 2
Note. Weighted analyses using Wave 2 sampling weights. ***p < 0.001
Percentages of Parent-Youth Expectation Agreement by Disability Category
Percentages of Parent-Youth Expectation Agreement by Disability Category
Notes. Agreement indicates both parent and youth rated either ‘probably/definitely will’ or ‘probably/definitely will not’ on expectation questions. Group comparison chi-squared analyses were weighted using Wave 2 sampling weights; values in each column not sharing the same subscript letter are significantly different at p < 0.01 in the two-sided test of equality using Bonferroni correction.
Logistic Regression Results Examining Youth and Parent Expectations and Predictors of Post-School Outcomes
Logistic Regression Results Examining Youth and Parent Expectations and Predictors of Post-School Outcomes
Notes. Four total logistic regression models were run, one for each postsecondary outcome. Analyses were weighted using sampling weights designed for any wave. Respective parent and youth expectations, along with covariates (gender, age, race/ethnicity, and receipt of free/reduced meals) were included in all models. SSI, Supplemental Security Income. OR, odds ratio; CI, confidence interval. ***p < 0.001.
This study aimed to determine the relationship between parent and youth expectations for the future among students with disabilities in the United States. Through analysis of the NLTS2 dataset, we identified that parent and youth expectations are related but significantly different. Parents consistently had lower expectations for their youth than the youth held for themselves. Although prior literature suggests that parent expectations may be more accurate (e.g., Finlay & Lyons, 2002), in this study, it was not always the parents’ expectations that best predicted postsecondary outcomes. Youth expectations contributed more strongly to postsecondary education and independent living outcomes while parent expectations contributed more strongly to employment and SSI benefit outcomes. This adds important findings to the literature that youth expectations also play an important role in prediction of post-school outcomes, especially expectations for postsecondary education and independent living. We also identified that youth in some disability categories have relatively more parent-youth expectation agreement (i.e., learning disability, speech impairment) while others are relatively less likely to agree (i.e., intellectual disability, emotional disturbance, multiple disabilities).
The Individuals with Disabilities Education Act (IDEA, 2004) mandates that students are required to participate in the transition planning process; however, the ways in which that happens can vary greatly (Cameto, Levine, & Wagner, 2004). Because youth, parents, and teachers often participate in the process, differing expectations between parents and youth, as was observed in this study, could potentially contribute to confusion during transition planning. This confusion may result in missed opportunities for students. In fact, a study by Riesen and colleagues (2014) described parent and youth expectations as barriers to successful employment, because they are often, “unclear, too specific, too high, or too low” (p. 36) according to providers. Thus, it is critical for all participants in the special education transition planning process to develop clear expectations and plans collaboratively.
Although disagreement between parents and adolescents is considered typical in the general population, agreement is believed to be a positive thing for adolescents and families. In non-disability samples, parent-youth agreement about expectations has been associated with better psychological adjustment, and is considered adaptive (P
Although much research has examined the role of parent expectations in predicting postsecondary outcomes for youth with disabilities, the current analysis of NLTS2 data expands knowledge in this area by revealing that youth expectations may play a stronger predictive role than parent expectations for outcomes related to postsecondary education and independent living. This should re-emphasize the importance of involving the student in transition planning, and suggests that parents should be aware and supportive of youth postsecondary goals, rather than relying only on their own goals and expectations for the youth. Parents may hold more restrictive expectations, especially around independent living, because of concern for their child’s safety. Given the current study’s findings that higher youth expectations are associated with greater independence after high school, parents should begin early to help youth work toward their individual goals. It is possible that activities promoting self-determination skills in youth with disabilities would assist with this process (Wehmeyer, 2014), although more work would need to be done to determine the relationship between self-determination and youth expectations.
Our results related to financial independence expectations are notable. In this category, we identified the weakest correlation between parent and youth expectations (r = 0.25). Financial independence is also the area in which youth expectations played a lesser role in predicting outcomes (i.e., paid employment and SSDI benefits) than parent expectations did. These results suggest that parents and youth with disabilities are not in agreement about their potential for financial independence, with youth more likely to expect that they will be financially independent, but parent expectations more accurately and strongly predicting eventual paid employment and receipt of SSI benefits. Existing literature postulates that the observed accuracy of parent expectations in broader populations may be mediated by student motivation or self-efficacy, parental involvement, or teacher expectations (Yamamoto & Holloway, 2010). Therefore, encouraging higher parent expectations, or directly addressing student motivation, student self-efficacy, parent involvement, and teacher expectations, could be an avenue for promoting more positive postsecondary outcomes related to financial independence.
Limitations and future directions
The NLTS2 provided a unique opportunity to explore our research questions about parent and youth expectations in connection with longitudinal outcomes. However, the use of this dataset presents some limitations worth noting. One such limitation is the age of the dataset (i.e., data collection was completed in 2010), which could potentially limit generalizability in the current day, as the landscape surrounding transition planning continues to change rapidly. This study was also limited by the availability of existing variables in the dataset. For example, there was no non-disability comparison group to include as a statistical control. Furthermore, only students deemed capable of answering questions were included, which likely excluded individuals with more significant support needs from the present study. NLTS2 outcome data is based on self-and parent-report rather than more objective measures of outcomes, potentially limiting reliability and accuracy. Additionally, while NLTS2 used a well-constructed, nationally representative sample, missing data at Wave 5 may have affected the representativeness of the sample used in the current analyses. Last, we chose not to examine disability category as part of the third aim due to the high number of disability categories, reduced sample sizes at wave 5, and the lack of a clear reference group for use in the logistic regressions (e.g., no non-disability comparison group). Understanding differential prediction of expectations to postsecondary outcomes by disability category may be important for future study.
Future research should further examine factors associated with agreement and disagreement among parents and youth with disabilities to further inform transition practices that can benefit the most students. Specifically, youth functioning is a potentially important variable that was not included in the present analysis due to the lack of good adaptive behavior measures available in the NLTS2 dataset. Additionally, understanding the role of self-determination in the relationship between youth and parent expectations and youth outcomes will be an important area for future study. Future research should also examine whether parent and youth expectations change over time to become more or less similar as the youth ages out of school, and whether direct intervention could improve agreement among parents and youth.
Conflict of interest
None to report.
Footnotes
Acknowledgments
The data used in this study was acquired through a data-use agreement with the Institute for Education Sciences. Anne V. Kirby’s research efforts were supported by the University of Utah Program in Personalized Health and the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number KL2TR001065. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
