Content uploaded by Xin Wei
Author content
All content in this area was uploaded by Xin Wei on Jul 25, 2016
Content may be subject to copyright.
Remedial and Special Education
1 –12
© Hammill Institute on Disabilities 2015
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0741932515581495
rase.sagepub.com
Article
The prevalence of autism spectrum disorders (ASDs) has
steadily increased over the past few decades. Although esti-
mates vary, recent studies cite prevalence rates between 20
and 100 cases per 10,000 (Matson & Kozlowski, 2011).
Recent estimates indicate that 1 in 68 school-age children in
the United States has an ASD (Centers for Disease Control
and Prevention, 2014). Given their increasing numbers, it is
critical that children and youth with ASDs are provided the
services and supports that will promote positive post–high
school outcomes, including employment and postsecondary
education. Yet nationally, the combined 2-year and 4-year
college enrollment rates for youth with ASDs was 32%, the
third lowest among youth in 12 special education disability
categories and much lower than that of youth in the general
population (70%; Shattuck et al., 2012; Wei, Yu, Shattuck,
McCracken, & Blackorby, 2013).
This low rate of college participation has significant eco-
nomic and personal costs for youth with ASDs, their families,
and society. When multiplied by national high school gradu-
ation estimates (Hussar & Bailey, 2013), this prevalence rate
suggests that approximately 49,000 youth with ASDs will
graduate from high school in 2014 to 2015. At current rates,
almost 33,300 of them could fail to pursue any kind of post-
secondary education in the first several years after leaving
high school; fewer still are likely to pursue a college educa-
tion rather than a vocational course of study (Newman,
2005). The U.S. Bureau of Labor Statistics reports that work-
ers with a Bachelor’s degree earn an average of US$381 per
week more than workers with some college experience but
no degree or with no college education at all—a difference of
almost US$600,000 over a 30-year working life.
The high school years are the obvious time to intervene
to improve college enrollment for youth with ASDs.
However, little is known about the types of high school
policies, interventions, and services that can increase enroll-
ment rates and how high schools can use the transition plan-
ning process to do so. There is a growing need to identify
evidence-based interventions in this field.
The History of Transition Planning
The evolution of federal special education legislation since
the 1975 passage of the Education for All Handicapped
Children Act (EHA, PL 94-142) has seen a steady strength-
ening of the intent that youth with disabilities meaningfully
participate in planning their own post–high school transi-
tion and that their goals and interests guide the planning
process. That landmark legislation made the individualized
education program (IEP) and accompanying transition plan
the cornerstones of special education needs identification,
581495RSEXXX10.1177/0741932515581495Remedial and Special Education XX(X)Wei et al.
research-article2015
1SRI International, Menlo Park, CA, USA
Corresponding Author:
Xin Wei, SRI International, 333 Ravenswood Ave., Menlo Park, CA
94025, USA.
Email: xin.wei@sri.com
The Effect of Transition Planning
Participation and Goal-Setting on
College Enrollment Among Youth With
Autism Spectrum Disorders
Xin Wei, PhD1, Mary Wagner, PhD1, Laura Hudson, MSW1,
Jennifer W. Yu, ScD1, and Harold Javitz, PhD1
Abstract
This study used propensity score techniques to assess the relationship between transition planning participation and goal-
setting and college enrollment among youth with Autism Spectrum Disorders. Using data from Waves 1 through 5 of the
National Longitudinal Transition Study–2, this study found that 2- or 4-year college enrollment rates were significantly
higher among youth with ASDs who participated in transition planning and those who had a primary transition goal of
college enrollment. Educational implications are discussed.
Keywords
transition planning, autism, college enrollment, propensity score weighting
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
2 Remedial and Special Education
goal-setting, service and setting definition, and student
assessment. EHA also specified that students could partici-
pate in their transition plan meetings, as appropriate
(Gillespie & Turnbull, 1983).
With the 1997 reauthorization of what became the
Individuals With Disabilities Education Act (IDEA), stu-
dents ages 14 and older were not just to be allowed, but
actively invited to attend their transition planning meetings,
and all students were mandated to have a transition plan in
place by age 16. Students’ interests and preferences were to
shape transition decisions, and their post–high school aspi-
rations were to guide their high school course of study and
transition services (Grigal, Test, Beattie, & Wood, 1997;
Martin, Marshall, & Bale, 2004). The 2004 reauthorization
of IDEA further refined the process for developing a transi-
tion plan and mandated that a student be invited to any IEP
meeting that includes “consideration of postsecondary
goals” (U.S. Department of Education, 2007).
Student Transition Planning
Participation
Studies of students’ educational goal-setting have focused
largely on their participation in transition planning
(Heatherington et al., 2010; Williams-Diehm, Wehmeyer,
Palmer, Soukup, & Garner, 2008). Transition planning is an
opportunity for students to learn and demonstrate self-deter-
mination skills, which are broadly conceived as “the ability
to take primary control of one’s own life and to do so in per-
sonally meaningful ways” (Schuler & Baldwin, 1981, p. 115)
and are demonstrated in a process whereby students “make
choices, act on those choices, experience the results, and then
make new choices” (Agran & Hughes, 2008, p. 69).
Research suggests that students are increasingly attend-
ing IEP and transition planning meetings (Test et al., 2004),
but they participate relatively little without direct instruc-
tion regarding the purposes and procedures of those meet-
ings (Griffin, Taylor, Urbano, & Hodapp, 2014; Martin et
al., 2006; Mason, McGahee-Kovac, Johnson, & Stillerman,
2002; Wehmeyer, Palmer, Soukup, Garner, & Lawrence,
2007). Nonetheless, reviews of research on transition-
related best practices are united in asserting that student
involvement is an important element of effective transition
plans and programs (Greene, 2003; Hendricks & Wehman,
2009; Kohler, 1993; Landmark, Ju, & Zhang, 2010).
In analyses of data from a nationally representative study
of students receiving special education services, students in
the autism category reported high expectations for their
own postsecondary education. Overall, 84.4% of students
with autism reported that they “definitely” or “probably”
would get some form of postsecondary education, and
61.7% and 54.2% reported that they “definitely” or “prob-
ably” would complete a 2-year or 4-year college degree,
respectively (Wagner, Newman, Cameto, Levine, & Marder,
2007). Yet, only 22.9% of students with autism had a goal in
their transition plan of attending a 2- or 4-year college
(Cameto, Levine, & Wagner, 2004). This, coupled with evi-
dence that students with autism are less likely to attend or to
lead or actively participate in transition planning (Shogren
& Plotner, 2012; Wagner, Newman, Cameto, Javitz, &
Valdes, 2012), indicates a gap between students’ personal
expectations and plans laid out with the assistance of par-
ents, educators, and other professionals.
Linking Student Transition Planning
Participation, Postsecondary Goals,
and Postsecondary Outcomes for
Youth With ASDs
Existing research has illuminated factors that are positively
associated with enrollment in postsecondary education
among youth with ASDs, including greater functional inde-
pendence, fewer limitations in functional areas (e.g., con-
versation, vision, and hearing; Cameto, 2005; Carter,
Austin, & Trainor, 2012; Newman, 2005), better high
school academic performance (Chiang, Cheung, Hickson,
Xiang, & Tsai, 2012; Shattuck et al., 2012; Taylor & Seltzer,
2011), higher income, non-Hispanic/non-African American
racial/ethnic status (Chiang et al., 2012; Shattuck et al.,
2012), and a longer time out of high school (Chiang et al.,
2012; Shattuck et al., 2012). However, few studies have
focused specifically on the connection between participa-
tion in transition planning and college enrollment or high-
lighted the ways in which transition goal-setting might
affect college enrollment for youth with ASDs.
Studies investigating postsecondary participation in any
field of study among students with ASDs have found that
high school experiences play a significant role in a student’s
successful enrollment and participation in postsecondary
education. Participation in transition planning during high
school is associated with participation in postsecondary
education for students with disabilities in general (Halpern,
Yovanoff, Doren, & Benz, 1995) as well as students with
ASDs (Chiang et al., 2012). Several studies have found that
having strong self-determination skills, a characteristic that
supports active transition planning participation, is also
associated with improved postsecondary outcomes for stu-
dents with disabilities (Getzel & Thoma, 2008; Morningstar
et al., 2010; Test, Mazzotti, Mustian, & Fowler, 2009).
Fundamental to transition planning is the responsibility
to establish goals for a student’s trajectory into his or her
early post–high school years and to incorporate those goals
into the transition plan and the activities that flow from it.
To date, very few studies have linked goals with post–high
school outcomes among individuals with ASDs. One cor-
relational study found that a transition goal of pursuing
postsecondary education increased the odds of enrollment
by 330% (Chiang et al., 2012). The National Longitudinal
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
Wei et al. 3
Transition Study–2 (NLTS2), the national data set used in
this study, provides a unique opportunity to investigate the
association between transition goals and attendance at post-
secondary institutions for youth with ASDs. However, it is
important to note that transition planning participation and
the outcomes that result, including a student’s transition
goals, are influenced by a variety of factors (Wagner et al.,
2012) that should be controlled for when examining the
relationship between transition planning and postsecondary
outcomes.
The Study Purpose
Although there is existing research on student involvement
in transition planning, studies of small, convenience sam-
ples predominate (e.g., Agran & Hughes, 2008; Fish, 2008;
Landmark, Zhang, & Montoya, 2007; Spann, Kohler, &
Soenksen, 2003), and even studies of larger samples have
limited generalizability (e.g., Virginia Department of
Education, 1998). Furthermore, these studies are often
descriptive and correlational in nature, and none used a
quasi-experimental design to adjust for confounding. In
addition, there is almost no research linking transition goal-
setting with post–high school outcomes except Chiang et al.
(2012). Because it is unethical in the context of an educa-
tional system to randomly assign students to participate in
transition planning or to set a primary goal to enroll in col-
lege or not to do so, a larger scale quasi-experimental study
is needed to explore the connection between transition plan-
ning, goal-setting, and college enrollment. Findings will be
useful for parents, advocates, and educators looking for
ways to improve postsecondary outcomes for the growing
ASD population.
We addressed this research need by focusing on two
research questions and applying propensity score modeling
methods to data from the NLTS2. Specifically, our research
questions examined whether the college enrollment rates
for students with ASDs were associated with (a) their par-
ticipation in transition planning and (b) having a primary
transition goal of college enrollment specified in their tran-
sition plans.
Method
Study Database
The NLTS2 is the largest, most comprehensive data set
available that generalizes to the experiences of youth with
disabilities nationally as they transitioned out of high
school. Conducted by SRI International for the U.S.
Department of Education, data were collected from parents
and/or youth in five waves, 2 years apart, from 2001 to
2009. The initial sample included more than 11,000 high
school students who were ages 13 through 16 and receiving
special education services on December 1, 2000, with about
1,100 of them receiving special education services in the
autism category. Each student’s eligibility for special edu-
cation services was determined by the school district or spe-
cial school from which the student roster was sampled
(special schools are those serving only students with dis-
abilities). It is important to note that the criteria for a special
education determination of autism may differ from state to
state and may differ from the criteria for ASD specified in
the Diagnostic and Statistical Manual of Mental Disorders
(4th ed.; DSM-IV; American Psychiatric Association, 1994).
However, more than 95% of children with a school designa-
tion of autism also met DSM-IV-based case criteria in pub-
lic health surveillance studies (Bertrand et al., 2001;
Yeargin-Allsopp et al., 2003).
The NLTS2 two-stage sampling approach first randomly
sampled local educational agencies (LEAs) and state-sup-
ported special schools stratified by region, district enroll-
ment, and wealth. Students with IEPs for special education
services were then randomly selected from rosters of LEAs
or special schools and weighted to yield nationally repre-
sentative estimates that generalize to all students in the
NLTS2 age range receiving special education services and
to those in each special education disability category
(Wagner, Kutash, Duchnowski, & Epstein, 2005).
Participants
This article includes data on a sample of approximately 920
youth with ASDs whose parents responded to a phone or
mail survey at Wave 1; approximately 660 of them remained
in the study at Wave 5. The study used data from parent/
youth telephone surveys or mail questionnaires across all
five waves as well as students’ high school transcripts and
responses to surveys of school staff who were familiar with
youth’s high school programs. The estimates reported here
used cross-wave weights that were suitable for analyzing
multiple waves of NLTS2 data (Valdes et al., 2013).
Unweighted sample sizes were rounded to the nearest 10, as
required by the U.S. Department of Education.
Intervention Variables
The intervention variables were extracted from Student’s
School Program Survey items relating to transition plan-
ning at Wave 1 in 2002 for older high school students and at
Wave 1 or 2 in 2004 for younger high school students.
Tran sit ion p la nni ng pa r ti ci pat ion . Transition planning partici-
pation was coded 1 if a student was reported by a school staff
member who was familiar with his or her school program to
have “provided some input into transition planning as a
moderately active participant” or “taken a leadership role in
the transition planning process, helping set the direction of
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
4 Remedial and Special Education
discussion, goals, and programs or service needs identified.”
Participation was coded 0 if a student had “not attended
planning meetings or participated in the transition planning
process” or had been “present in discussions of transition
planning but participated very little or not at all.”
Had a primary goal of college enrollment in their transition
plan. The Student’s School Program Survey asked respon-
dents to identify “the primary goal of this student’s educa-
tional program” and provided eight structured responses,
including attending a 2- or 4-year college; attending a post-
secondary vocational training program; finding competitive,
sheltered, or supported employment; living independently;
maximizing functional independence; and enhancing social/
interpersonal relationships and satisfaction. Typically, more
than one primary goal was indicated. If the goals included
attending a 2- or 4-year college, this variable was coded 1;
otherwise, it was coded 0.
Outcomes
College enrollment. We identified whether youth with
ASDs were ever enrolled in college by examining parents’
and/or youths’ answers to two survey items at Waves 2
through 5 for youth who were out of high school at those
waves: whether the youth had ever since leaving high
school attended (a) a 2-year or community college or (b) a
4-year college or university. A measure of any 2- or 4-year
college enrollment was coded 1 if either of the survey
items was yes in any wave and 0 if both items were no in
all waves. All outcome variables were measured after the
intervention variables. Specifically, older high school stu-
dents whose transition planning activities were measured
at Wave 1 were out of high school by Wave 2 and reported
their college enrollment status at Wave 2. Younger high
school students whose transition planning activities were
measured at Wave 2 were out of high school at Wave 3 or
later and reported their college enrollment status as early
as Wave 3.
Covariates
An important aspect of estimating the propensity score in
propensity score modeling is the selection of covariates.
Researchers suggest that covariates that affect both inter-
vention participation and outcomes should be included in
the estimation of the propensity score (Caliendo & Kopeinig,
2008; Heckman, Ichimura, Smith, & Todd, 1998; Lechner,
2002; Ravallion, 2001). Covariates included in this study
were derived from correlates of college enrollment and
transition planning participation that have been cited in the
literature described above: youth’s gender, age, race/ethnic-
ity, disability severity, high school achievement, family
income, mother’s education level, whether parents ever
enrolled in a postsecondary school/program, and parents’
expectation of youth attending college.
Covariates were measured in the Wave 1 Parent Phone
Interview and/or Mail Survey in 2001; thus, measurement
of all covariates preceded the interventions. Five factors
were used to assess disability severity: the presence of
Attention Deficit/Hyperactivity Disorder (ADD/ADHD)
and measures of youth’s social skills, conversational ability,
self-care skills, and functional cognitive skills. For ADD/
ADHD, parents reported whether youth had the disorder.
Youth’s social skills were measured by summing the
responses to 11 questions from the Social Skills Rating
Systems (SSRS) parent version (Gresham & Elliott, 1990),
which asked parents to rate how often (1 = never, 2 = some-
times, 3 = very often) their child was able to do the follow-
ing: join groups, make friends, end disagreements calmly,
seem confident in social settings, avoid trouble, start con-
versations, receive criticism well, control temper when
arguing, keep working until finished, speak in an appropri-
ate tone, and cooperate with family members. The social
skills score ranged from 11 to 33, with a reliability of α =
.79. Parents rated children’s conversational ability as 1 =
doesn’t converse at all, 2 = has a lot of trouble conversing,
3 = has a little trouble conversing, or 4 = converses as well
as other children his or her age. Self-care skills were mea-
sured by a summing scores ranging from 1 (low) to 4 (high)
on items indicating how well youth could dress and feed
themselves independently. Functional cognitive skills were
measured using a scale from 4 (low) to 16 (high) based on
parents’ reports of how well their children were able to do
the following four tasks without help: tell time on an analog
clock, read and understand common signs, count change,
and look up telephone numbers and use a telephone. Each
item had four response categories: 1 = not at all well, 2 = not
very well, 3 = pretty well, 4 = very well. A summation of the
scores on the four items measures students’ overall cogni-
tive functioning, with internal consistency reliability of .93.
Youths’ academic achievement was measured by their
high school grade point average (GPA), measured in
Carnegie units and extracted from their high school tran-
scripts. One Carnegie unit is associated with a student pass-
ing a course that meets for approximately 1 hr per day, 5
days per week for a total of 24 weeks. We also calculated a
dichotomous variable coded 1 for students who were able to
take standardized achievement assessments at either Wave
1 or 2. For the remaining youth, for whom an alternative
assessment was completed by an adult familiar with the stu-
dent due to a student’s physical, cognitive, or behavioral
limitations, the item was coded 0.
Propensity Score Methodology
Propensity score methods are quasi-experimental approaches
that were developed to approximate findings obtained from
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
Wei et al. 5
randomized control trials (RCTs; Becker & Ichino, 2002).
They have been increasingly used in analysis of observa-
tional data to reduce selection bias in estimating treatment,
policy, or intervention effects when RCTs are not feasible or
ethical. Propensity score methods enable quasi-experimen-
tal contrasts between students experiencing naturally occur-
ring treatments and comparison groups whose members are
similar on other factors included as covariates in the mod-
els. This study used propensity score methods to test the
effect of particular aspects of a special education policy—
transition planning and goal-setting—on college enrollment
rates. The propensity score is the predicted probability of
participating in an intervention based on a set of potentially
confounding covariates (e.g., student demographic and dis-
ability characteristics, student academic achievement, par-
ents’ expectations of students attending college) using
logistic regression. Propensity scoring attempts to equalize
the mean values of potentially confounding covariates in
the treatment and comparison groups, and assures that dif-
ferences in outcomes are not the result of differences in
mean values on those covariates. Although it aims to gener-
ate rigorous and unbiased estimates of the effects of a treat-
ment on the outcome of interest, propensity scoring cannot
account for unobserved confounders and requires a suffi-
ciently large sample that has overlap between treatment and
comparison groups.
Two types of analyses to estimate the average treatment
effect on the treated (ATT) were conducted. One analysis
estimated the ATT on the treated students in the sample
(SATT). The other analysis estimated the ATT on treated
students in the population (PATT) represented by NLTS2
students with ASDs.
Analyses of SATT adjusted for confounding were per-
formed using an inverse propensity score estimator, as rec-
ommended by Curtis, Hammill, Eisenstein, Kramer, and
Anstrom (2007); Hirano, Imbens, and Ridder (2003); and
Rosenbaum and Rubin (1983). Specifically, the weight for
treated students was their survey weight (or in the case
where the intent is not to project to a population, the weight
is 1.0), and the weight for comparison students was equal to
their survey weight multiplied by (pi / [1 − pi]), where pi is
the propensity score for the ith comparison student (Harder,
Stuart, & Anthony, 2010; Hirano et al., 2003). Analyses of
PATT adjusted for confounding used the approach recom-
mended by DuGoff, Schuler, and Stuart (2014). The weight
for treated students was their survey weight, and the weight
for comparison students was equal to their survey weight
times their propensity score transformed to an odds scale
(DuGoff et al., 2014).
The SATT and PATT of transition planning participation
and transition goal-setting were estimated using a weighted
logistic regression model. The odds ratio (OR) from each
model can be interpreted as the measure of association
between transition planning participation or goal-setting and
college enrollment rates, adjusted for the estimated propen-
sity of participation or having a college enrollment transition
goal. This essentially weights the comparison group to create
balance with the treatment group on observed covariates and
thus estimates the effect of the two interventions for the indi-
viduals who actually participated in them. Weighting was
selected over other approaches, such as matching, because of
its good performance in this data set (details below), flexibil-
ity with the distribution of the data, capability to deal with
time-dependent covariates and censored data, and because it
retains all participants in the analysis. After propensity score
weighting for comparison students, we examined the stan-
dardized mean score (the difference in means for the treat-
ment and comparison groups, divided by a pooled standard
deviation) to assure that they were less than 0.25, a standard
recommended by the Institute of Education Science’s What
Works ClearingHouse (Institute of Education Sciences,
2014), thereby demonstrating covariate balance.
Handling of Missing Data
Missingness rates for covariates ranged from no missing
to 52%. Missing data on covariates were imputed 20 times
using Stata’s Imputation by Chained Equations (ICE) pro-
cedure (Royston, 2004, 2005, 2007, 2009; Royston,
Carlin, & White, 2009; White, Royston, & Wood, 2011).
Imputations were performed on all variables used in the
analyses to avoid bias associated with listwise deletion
and to capture the information contained in the correlation
between covariates and the outcome and treatment vari-
ables. However, we did not use imputed values for the out-
comes or treatments in the analyses, as recommended by
Little (1992), Little and Rubin (2002), White et al. (2011),
and Von Hippel (2007). Analyses conducted on imputed
data were aggregated using the Stata mim procedure, a
command for analyzing multiply imputed data sets that
combines regression results across imputations and adjusts
the standard error estimates to accurately reflect the uncer-
tainty due to missingness.
Results
Table 1 shows the characteristics of youth with ASDs
weighted to represent the population. Consistent with epi-
demiological estimates, 85.41% of youth were male. The
sample was diverse in terms of ethnicity, race, and family
socioeconomic position. Pertinent to the study’s outcome of
interest, parents of more than 7 in 10 youth reported having
themselves enrolled in some form of postsecondary educa-
tion, and more than one third definitely or probably thought
their student would also. Functionally, 54.91% were
reported to have either a lot of trouble or no ability to con-
verse, and 56.07% were not able to participate in the direct
assessment of their academic skills, emphasizing that this
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
6 Remedial and Special Education
was not a predominantly high-functioning group of youth
with ASDs. Four in 10 youth participated in transition plan-
ning and about one fourth had a primary transition goal of
college enrollment in their transition plan. Overall, 3 in 10
attended a 2- or 4-year college after high school.
To ensure that the propensity score method successfully
created balanced treatment and comparison groups, we com-
pared the standardized mean differences between the two
groups for each covariate before and after propensity score
weighting for both SATT and PATT. The balance on the
Table 1. Descriptive Analysis of Youth With ASDs.
Full sample
Variables used in this study Weighted % or M SE Unweighted n
Covariates
Malea85.41 2.00 960
African American 21.84 0.98 920
Hispanica10.20 0.85 920
Agea,c
13 7.44 1.53 960
14 27.76 3.00 960
15 23.07 2.49 960
16 25.44 2.83 960
17 16.31 2.13 960
Incomea
Low: ≤US$25,000 24.93 0.86 830
Medium: US$25,001–US$50,000 29.11 1.04 830
High: >US$50,000 45.96 0.84 830
Mother’s education levela
Less than high school 8.04 1.26 850
High school graduate or GED 25.33 2.62 850
Some college 34.32 3.05 850
BA or higher degree 32.32 2.74 850
Parent ever attended postsecondary educationa71.41 2.28 880
Parent expectation of youth attending collegea
Definitely will not 36.63 2.81 870
Probably will not 25.99 2.27 870
Probably will 23.19 2.38 870
Definitely will 14.19 1.56 870
Has ADD/ADHDa18.77 2.52 740
Social skills scale scoreb11.37 0.15 860
Cognitive functioning skillsb10.94 0.24 910
Conversation abilitya
Does not carry on a conversation at all 18.10 2.30 890
Has a lot of trouble carrying on a conversation 36.81 2.86 890
Has a little trouble carrying on a conversation 30.62 3.14 890
Converses as well as other children 14.47 1.91 890
Had a direct assessment scorea43.93 3.16 960
Self-care skills scale scoreb6.97 0.08 910
High school GPAb3.03 0.07 460
Intervention
Transition planning participationa40.29 2.67 630
Had a primary transition goal of college educationa24.20 2.90 700
Outcomes
Attended any 2- or 4-year college since high schoola29.63 2.63 710
Source. NLTS2, Waves 1 through 5.
Note. Means and percentages were weighted to population levels. Weighted means are presented for continuous variables and weighted percentages
for categorical variables. Unweighted ns were rounded to the nearest 10. ASDs = Autism Spectrum Disorders; GED = general education development;
ADD/ADHD = Attention Deficit/Hyperactivity Disorder; GPA = grade point average; NLTS2 = National Longitudinal Transition Study–2.
aCategorical variables. bContinuous variables. cAge as of July 15, 2001.
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
Wei et al. 7
covariates in the analysis was greatly improved after apply-
ing the propensity score weighting method for both SATT
and PATT. Due to space limitations, data presented in Table
2 show the balance for SATT. Before propensity score
weighting, the differences on covariates in the two models
ranged from −0.48 to 1.63 standard deviations; whereas
after propensity score weighting, the average differences on
covariates ranged from −0.14 to 0.23 standard deviations.
For example, before propensity score weighting, students
with ASDs who participated in transition planning were less
likely to be African American (column B = −0.33), less
likely to come from low-income families (−0.28), and more
likely to have taken the direct assessment (0.64) than their
peers who did not participate in transition planning. Students
with ASDs who participated in transition planning also had
higher cognitive functioning skills (1.11), conversation abil-
ity (0.86), self-care skills (0.45), and parental expectation of
youth attending college (0.87) than their peers who did not
participate in transition planning. After propensity score
weighting, the difference between students who participated
in transition planning and those who did not was reduced to
−0.10 to 0.08 (column C) standard deviations on all baseline
covariates. Following the same pattern, students with ASDs
who had a primary transition goal of college enrollment also
had a higher likelihood of being included in the direct assess-
ment (column E = 0.77), had higher cognitive functioning
skills (1.52), higher social skills (0.44), higher conversation
ability (0.84), and higher parent expectations of youth attend-
ing college (1.63). By conducting propensity score weight-
ing, these large differences were reduced from − 0.14 to 0.23
(column F) standard deviations across the covariates.
Therefore, after propensity score weighting, participants and
nonparticipants were very similar on all potentially con-
founding covariates that were included in the analyses.
Table 2. Treatment and Control Balance Statistics on Covariates After PSW for ATT Students in the Sample.
Column
A B C D E F
Transition planning participation Had a primary transition goal of college education
Covariates
Treatment
% or mean Pre-PSW balance Post-PSW balance
Treatment
% or mean Pre-PSW balance Post-PSW balance
Malea85.10 0.06 −0.04 87.29 0.15 −0.13
African Americana17.70 −0.33 −0.001 15.18 −0.34 0.04
Hispanica5.84 −0.22 0.03 2.41 −0.35 −0.02
Ageb14.90 −0.14 0.03 14.90 −0.11 0.23
Income lowa14.37 −0.28 −0.02 12.01 −0.48 0.04
Income mediuma30.45 0.03 −0.002 29.53 0.02 0.02
Mother’s education levelb2.99 0.13 −0.07 3.26 0.53 −0.02
Parent ever attended
postsecondary
educationa
77.39 0.17 −0.09 88.70 0.52 0.04
Parent expectation of
youth attending collegeb
1.74 0.87 −0.03 3.24 1.63 −0.004
Has ADD/ADHDa17.72 −0.04 −0.10 17.50 −0.02 0.03
Social skills scale scoreb12.16 0.32 0.08 12.61 0.44 −0.13
Cognitive functioning
skillsb
12.89 1.11 −0.09 14.24 1.52 0.03
Conversation abilityb1.85 0.86 −0.07 1.96 0.84 −0.14
Had a direct assessment
scorea
67.49 0.64 0.01 76.40 0.77 0.22
Self-care skillsb7.41 0.45 −0.04 7.55 0.52 −0.02
High school GPAb3.04 0.00 0.03 2.97 −0.20 0.03
Source. NLTS2, Waves 1 through 5.
Note. To reduce the number of covariates in the model, mother’s education, parent expectation of youth attending college, and conversation ability were
treated as continuous variables with the lowest level being 1 and highest level being 4. Balance statistics are measured by the standardized mean difference,
which is the difference in means between the groups, divided by the pooled standard deviation of both the treatment and comparison groups. Treatment
and control balance statistics on covariates after propensity score weighted for ATT students in the population were similar to the balance statistics for ATT
students in the sample. Due to space limitation, they are not presented here. PSW = propensity score weighting; ATT = average treatment effect on the
treated; ADD/ADHD = Attention Deficit/Hyperactivity Disorder; GPA = grade point average; NLTS2 = National Longitudinal Transition Study–2.
aCategorical variables. bContinuous variables.
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
8 Remedial and Special Education
The unadjusted 2- or 4-year college enrollment rates dif-
fered significantly in both the sample and population esti-
mates between transition planning participants and
nonparticipants and between those who had a college
enrollment goal in their transition plans and those who did
not have such a goal (comparing “Treatment” column and
“Comparison” columns in Table 3). However, the last col-
umn in Table 3 shows that when models were weighted,
transition planning participants had significantly higher
odds of attending a 2- or 4-year college than nonparticipants
in the sample estimates but not in the population estimates
(see Note 1). Youth who had a primary transition goal of
college enrollment had significantly higher odds of attend-
ing 2- or 4-year college than those who did not have such a
goal in both the sample and population estimates.
Discussion
These results provide a national picture of the effect of high
school transition planning participation and goal-setting on
college enrollment rates among students with ASDs. About
40.29% of youth with ASDs actively participated in their
transition planning meetings, and 24.20% had a primary
transition goal of college enrollment in their transition plan.
We found that both transition planning participation and
having a primary transition goal of college enrollment dur-
ing secondary school were associated with higher odds of
attending a 2- or 4-year college among the sample of youth
with ASDs. Findings related to having a primary transition
goal of college enrollment also generalize to the population.
These findings add empirical evidence to the literature
(Greene, 2003; Hendricks & Wehman, 2009; Kohler, 1993;
Landmark et al., 2010) on the benefits of student involve-
ment in transition planning and goal-setting for students
receiving special education services.
Transition planning begun early in high school provides
the context within which students with disabilities can artic-
ulate their post–high school goals and work with parents,
school staff, and others to chart a course toward them. The
transition plan itself is required by law to specify the transi-
tion services needed to assist students in achieving their
goals (IDEA Partnership, 2004). This study suggests that
participation in transition planning is a valuable opportunity
to intervene to improve postsecondary education outcomes
for secondary school students with ASDs. However, there is
a marked contrast between the large percentage of youth
with ASDs who expect to attend a postsecondary institution
(84.40%) and the low percentage who have postsecondary
education goals included in the transition plan (24.20%;
Bhandari & Wagner, 2006; Wagner et al., 2007). This
emphasizes the urgent need to effectively engage youth in
the transition planning process so that their interests and
desires are reflected in their plans. This study finds that
specifying a primary goal related to college attendance in
transition plans also can effectively boost the odds of
attending college by 564% (OR = 6.64, 95% confidence
interval [CI] = [1.89, 29.16]) for youth with ASDs in the
population as shown in Table 3.
Postsecondary education benefits youth with disabilities
by increasing their potential to become self-reliant, tax-pay-
ing, and civically engaged citizens. Over the last decade,
there has been an expansion of opportunities in higher edu-
cation for individuals with disabilities and of their full
inclusion in the college classroom. There are reports that as
Table 3. ATT Effect of Transition Planning Participation on College Enrollment Rates for Youth With ASDs.
Any 2- or 4- year college enrollment rates
Intervention ATT estimates Treatment (%)aComparison (%)b
Propensity weighted
comparison (%)c
Propensity adjusted
OR [95% CI]d
Transition planning
participation
Sample 54.08 17.03*** 37.65* 1.95* [1.10, 3.45]
Population 50.28 18.83*** 34.98 1.88 [0.81, 4.33]
Had a primary
transition goal of
college education
Sample 76.09 14.49*** 41.80** 4.43** [1.51, 13.00]
Population 80.02 12.09*** 37.62** 6.64** [1.89, 29.16]
Source. NLTS2, Waves 1 through 5.
Note. ATT = average treatment effect on the treated; ASD = Autism Spectrum Disorder; OR = odds ratio; CI = confidence interval; NLTS2 = National
Longitudinal Transition Study–2; LOR = logged odds ratio.
aTreatment column indicates treatment group percentages rates. bComparison column indicates the control group percentage. The treatment-
comparison differences in college enrollment rates were tested by weighted chi-square tests. cPropensity weighted comparison is the estimated
propensity of the comparison group enrolling in college based on the weighted logistics regression model controlling for demographic, disability,
academic, and parent expectation. It is calculated as 100 × Pt / [OR (1 − Pt) + Pt], where Pt is the treatment group college enrollment rates and OR is
the propensity adjusted OR. The significance level indicates whether treatment and comparison groups are significantly different in the weighted logistic
regression model. dPropensity adjusted OR controlled for demographic, disability, academic achievement, and parent expectation covariates in the
weighted logistic regression model. The significance level indicates whether treatment and comparison groups are significantly different in the weighted
logistic regression model. Effect size for the ORs can be calculated using the Cox Index LORCox = ln(OR) / 1.65 (Cox, 1970).
*p < .05. **p < .01. ***p < .001.
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
Wei et al. 9
many as 200 college and university programs across the
country actively support students with disabilities in their
academic programs, career development, and campus life
(Blalock, 2014; Grigal & Hart, 2010). One of the founding
concepts of “inclusive postsecondary education” is to
embed individuals with disabilities, particularly those with
intellectual disabilities, within normative pathways to the
maximum extent possible (Uditsky & Hughson, 2012).
Similar to students without disabilities, educators should
explore college as a viable option with youth and parents
and begin to prepare students with ASDs for college at the
start of the transition planning process. Based on a partner-
ship between educators and a youth’s family, student-
focused planning should enable student participation in
decision making and goal-setting, particularly if the student
expresses goals related to postsecondary education. The
process should support high school coursework based on
students’ goals and interests, self-evaluation of their prog-
ress in meeting their goals, and development of self-
determination and other skills to achieve goals (Kohler,
1993, 1996, 1998; Kohler & Field, 2003).
Although it makes a valuable contribution to the transi-
tion planning literature for youth with ASDs, this study has
several limitations. First, unobserved confounding is a con-
cern in propensity score modeling. There is a possibility
that an unmeasured factor might be correlated with both the
likelihood of transition planning participation and goal-set-
ting and the likelihood of college enrollment. Second, the
college enrollment data were collected via surveys, not col-
lege registration records, which may result in reporting
biases. Future research could validate the results of this
study by using other data sources, such as enrollment data
from a university disability support office. Third, future
studies should test the mechanism underlying the positive
association between transition planning and goal-setting
and college enrollment. For example, it is possible that
improved self-determination skills, more academically ori-
ented secondary school course-taking patterns, or the inter-
play of the two may mediate the relationship between
transition planning and goal-setting and college enrollment.
Finally, NLTS2 surveyed youth’s expectations of attending
college as well as parents’. However, this variable was col-
lected at Wave 2 and our analysis only included covariates
from Wave 1. In addition, only youth who were able to
respond to an interview or complete a survey and those who
were not attending or who had not previously attended a
postsecondary education institution were asked this ques-
tion; therefore, it has extensive missing data (57.38%).
Thus, this variable could not be included as a covariate in
the analyses, although it has been reported previously for
descriptive purposes only in Wagner et al. (2007).
Despite these limitations, this study breaks new ground
in the understanding of best practices in transition planning
for youth with ASDs. The national sampling frame and the
large size and diversity of the study sample increases the
external validity of the findings. The use of propensity score
methods is innovative and strengthens the case for the effect
of transition planning participation and goal-setting on col-
lege enrollment. The extensive list of covariates included in
both the propensity score weighting procedure and ATT
estimation not only ensures the participants and nonpar-
ticipants were similar on all included factors, but also
makes the estimation of the ATT effect of transition plan-
ning participation and goal-setting more robust. Last, the
measures of transition planning participation and goal-
setting were based on school records, which are more reli-
able than parent- or student-reported transition planning
participation rates.
In sum, the findings from this study lay the ground-
work for better understanding the association between
transition planning practice and goal-setting and college
enrollment among youth with ASDs. This study empow-
ers all parties (e.g., policymakers, students, parents,
teachers, etc.) who strive to expand postsecondary educa-
tion opportunities for youth with disabilities to start the
secondary school transition planning process as early as
possible so that students’ course taking and other high
school experiences can be aligned with and support
achievement of transition goals. To increase college
enrollment rates and the benefits of a college education
for youth with ASDs, high school personnel can ensure
that youth with ASDs are given the training and supports
needed to participate actively in their own transition plan-
ning and explore whether postsecondary education goals
can be set and met in their transition planning process.
Future studies in the area of ASD and postsecondary edu-
cation should continue identifying evidence-based prac-
tices and interventions that increase the likelihood of
postsecondary participation among the growing popula-
tion of young adults with ASDs and extend the analyses
to address college completion, through which the benefits
of postsecondary education can be realized.
Authors’ Note
Any opinions expressed are those of the authors and do not repre-
sent the positions or polices of the funding agency.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: This
research was supported by Grant R324A120012 from the U.S.
Department of Education, Institute of Education Sciences and
Grant HRD-1130088 from the National Science Foundation.
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
10 Remedial and Special Education
Note
1. It is likely that there are differential treatment effects across
different subgroups of Autism Spectrum Disorders (ASD).
When the treatment effect was weighted to the population,
the average treatment effect on the treated (ATT) population
estimates are different from the ATT sample estimates.
References
Agran, M., & Hughes, C. (2008). Asking student input: Students’
opinions regarding their individualized education pro-
gram involvement. Career Development for Exceptional
Individuals, 31, 69–76.
American Psychiatric Association. (1994). Diagnostic and statis-
tical manual of mental disorders (4th ed.). Washington, DC:
Author.
Becker, S. O., & Ichino, A. (2002). Estimation of average treat-
ment effects based on propensity scores. Stata Journal, 2,
358–377.
Bertrand, J., Mars, A., Boyle, C., Bove, F., Yeargin-Allsopp,
M., & Decoufle, P. (2001). Prevalence of autism in a United
States population: The Brick Township, New Jersey, investi-
gation. Pediatrics, 108, 1155–1161.
Bhandari, A., & Wagner, T. (2006). Self-reported utilization of
health care services: Improving measurement and accu-
racy. Medical Care Research and Review, 63, 217–235.
doi:10.1177/1077558705285298
Blalock, G. (2014). Expert update: Inclusive post-secondary
education for individuals with developmental disabilities.
Retrieved from http://www.gcdd.org/blogs/gcdd-blog/2639-
expert-update-inclusive-post-secondary-education-for-indi-
viduals-with-developmental-disabilities.html
Caliendo, M., & Kopeinig, S. (2008). Some practical guid-
ance for the implementation of propensity score matching.
Journal of Economic Surveys, 22, 31–72. doi:10.1111/j.1467-
6419.2007.00527.x
Cameto, R. (2005). Employment of youth with disabilities after
high school. In M. Wagner, L. Newman, R. Cameto, N. Garza
& P. Levine (Eds.), After high school: A first look at the post-
school experiences of youth with disabilities. A report from
the National Longitudinal Transition Study-2 (NLTS2) (pp.
5–1–5–21). Menlo Park, CA: SRI International.
Cameto, R., Levine, P., & Wagner, M. (2004). Transition plan-
ning for students with disabilities: A special topic report
from the National Longitudinal Transition Study-2 (NLTS2).
Menlo Park, CA: SRI International.
Carter, E. W., Austin, D., & Trainor, A. A. (2012). Predictors of
postschool employment outcomes for young adults with severe
disabilities. Journal of Disability Policy Studies, 23, 50–63.
Centers for Disease Control and Prevention. (2014). Prevalence
of autism spectrum disorder among children aged 8 Years—
Autism and developmental disabilities monitoring network,
11 sites, United States, 2010. Retrieved from http://www.cdc.
gov/mmwr/preview/mmwrhtml/ss6302a1.htm
Chiang, H.-M., Cheung, Y., Hickson, L., Xiang, R., & Tsai, L.
(2012). Predictive factors of participation in postsecond-
ary education for high school leavers with autism. Journal
of Autism and Developmental Disorders, 42, 685–696.
doi:10.1007/s10803-011-1297-7
Cox, D. R. (1970). The analysis of binary data. London, England:
Methuen.
Curtis, L. H., Hammill, B. G., Eisenstein, E. L., Kramer, J. M.,
& Anstrom, K. J. (2007). Using inverse probability-weighted
estimators in comparative effectiveness analyses with obser-
vational databases. Medical Care, 45, 158–165.
DuGoff, E. H., Schuler, M., & Stuart, E. A. (2014). Generalizing
observational study results: Applying propensity score methods
to complex surveys. Health Services Research, 49, 284–303.
Fish, W. W. (2008). The IEP meeting: Perceptions of parents of
students who receive special education services. Preventing
School Failure, 53, 8–14.
Getzel, E. E., & Thoma, C. A. (2008). Experiences of college students
with disabilities and the importance of self-determination in
higher education settings. Career Development for Exceptional
Individuals, 31, 77–84. doi:10.1177/0885728808317658
Gillespie, E. B., & Turnbull, A. P. (1983). It’s my IEP: Involving
students in the planning process. Teaching Exceptional
Children, 16, 26–29.
Greene, G. (2003). Best practices in transition. In G. Greene & C.
A. Kochhar-Bryant (Eds.), Pathways to successful transition
for youth with disabilities (pp. 154–196). Upper Saddle River,
NJ: Pearson Education.
Gresham, F. M., & Elliott, S. N. (1990). Social skills rating system
manual. Circle Pines, MN: American Guidance Service.
Griffin, M., Taylor, J., Urbano, R., & Hodapp, R. M. (2014).
Involvement in transition planning meetings among high school
students with autism spectrum disorders. The Journal of Special
Education, 47, 256–264. doi:10.1177/0022466913475668
Grigal, M., & Hart, D. (2010). Think college: Postsecondary
education options for students with intellectual disabilities.
Baltimore, MD: Paul H. Brookes.
Grigal, M., Test, D., Beattie, J., & Wood, W. (1997). An evalu-
ation of transition components of individualized education
programs. Exceptional Children, 63, 357–372.
Halpern, A. S., Yovanoff, P., Doren, B., & Benz, M. (1995).
Predicting participation in postsecondary education for
school leavers with disabilities. Exceptional Children, 62,
151–164.
Harder, V. S., Stuart, E. A., & Anthony, J. C. (2010). Propensity
score techniques and the assessment of measured covariate
balance to test causal associations in psychological research.
Psychological Methods, 15, 234–249.
Heatherington, S. A., Dirant-Jones, L., Johnson, K., Nolan, K.,
Smith, E., Taylor-Brown, S., & Tuttle, J. (2010). The lived
experiences of adolescents with disabilities and their par-
ents in transition planning. Focus on Autism and other
Developmental Disabilities, 25, 163–172.
Heckman, J., Ichimura, H., Smith, J., & Todd, P. (1998).
Characterizing selection bias using experimental data.
Econometrica, 66, 1017–1098. doi:10.2307/2999630
Hendricks, D. R., & Wehman, P. (2009). Transition from school
to adulthood for youth with autism spectrum disorders:
Review and recommendations. Focus on Autism and other
Developmental Disabilities, 24, 77–88.
Hirano, K., Imbens, G. W., & Ridder, G. (2003). Efficient esti-
mation of average treatment effects using the estimated pro-
pensity score. Econometrica, 71, 1161–1189. doi:10.1023/
a:1020371312283
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
Wei et al. 11
Hussar, W. J., & Bailey, T. M. (2013). Projections of education
statistics to 2021. Washington, DC: Institute of Education
Sciences, National Center for Education Statistics, U.S.
Department of Education.
IDEA Partnership. (2004). IDEA regulations—Individualized
education program (IEP). Author. Retrieved from http://
www.ideapartnership.org/using-tools/topic-briefs/regulatory-
provisions/1679-individualized-education-program-regula-
tions.html
Institute of Education Sciences. (2014). What Works
Clearinghouse procedures and standards handbook (Version
3.0). Washington, DC: Author.
Kohler, P. D. (1993). Best practices in transition: Substantiated
or implied? Career Development for Exceptional Individuals,
16, 107–121.
Kohler, P. D. (1996). Taxonomy for transition programming:
Linking research and practice. Champaign: Transition
Research Institute, University of Illinois.
Kohler, P. D. (1998). Implementing a transition perspective of
education: A comprehensive approach to planning and deliv-
ering secondary education and transition services. In F. R.
Rusch & J. Chadsey-Rusch (Eds.), High school and beyond:
Transition from school to work (pp. 179–205). Belmont, CA:
Wadsworth.
Kohler, P. D., & Field, S. (2003). Transition-focused education:
Foundation for the future. The Journal of Special Education,
37, 174–183.
Landmark, L. J., Ju, S., & Zhang, D. (2010). Substantiated
best practices in transition: Fifteen plus years later. Career
Development for Exceptional Individuals, 33, 165–176.
Landmark, L. J., Zhang, D. D., & Montoya, L. (2007). Culturally
diverse parents’ experiences in their children’s transition:
Knowledge and involvement. Career Development for
Exceptional Individuals, 30, 68–79.
Lechner, M. (2002). Some practical issues in the evaluation of het-
erogeneous labour market programmes by matching methods.
Journal of the Royal Statistical Society, Series A: Statistics in
Society, 165, 59–82. doi:10.1111/1467-985X.0asp2
Little, R. J. A. (1992). Regression with missing X’s: A review.
Journal of the American Statistical Association, 87, 1227–
1237. doi:10.2307/2290664
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with
missing data (2nd ed.). Hoboken, NJ: Wiley.
Martin, J. E., Marshall, L. H., & Bale, P. (2004). A 3-year study
of middle, junior high, and high school IEP meetings.
Exceptional Children, 70, 285–297.
Martin, J. E., Van Dycke, J. L., Christensen, W. R., Greene, B.
A., Gardner, J. W., & Lovett, D. L. (2006). Increasing student
participation in their transition IEP meetings: Establishing the
self-directed IEP as an evidenced-based practice. Exceptional
Children, 72, 299–316.
Mason, C. Y., McGahee-Kovac, M., Johnson, L., & Stillerman, S.
(2002). Implementing student-led IEPs: Student participation
and teacher reactions. Career Development for Exceptional
Individuals, 25, 171–192.
Matson, J. L., & Kozlowski, A. M. (2011). The increasing prev-
alence of autism spectrum disorders. Research in Autism
Spectrum Disorders, 5, 418–425.
Morningstar, M. E., Frey, B. B., Noonan, P. M., Ng, J., Clavenna-
Deane, B., Graves, P., . . .Williams-Diehm, K. (2010). A
preliminary investigation of transition preparation and self-
determination for students with disabilities in postsecond-
ary education settings. Career Development for Exceptional
Individuals, 33, 80–94.
Newman, L. (2005). Postsecondary education participation of
youth with disabilities. In M. Wagner, L. Newman, R. Cameto,
N. Garza & P. Levine (Eds.), After high school: A first look at
the postschool experiences of youth with disabilities. A report
from the National Longitudinal Transition Study-2 (NLTS2)
(pp. 4/1–4/17). Menlo Park, CA: SRI International.
Ravallion, M. (2001). The mystery of the vanishing benefits: An
introduction to impact evaluation. The World Bank Economic
Review, 15, 115–140. doi:10.1093/wber/15.1.115
Rosenbaum, P., & Rubin, D. B. (1983). The central role of the
propensity score in observational studies for causal effects.
Biometrika, 70, 41–55.
Royston, P. (2004). Multiple imputation of missing values. Stata
Journal, 4, 227–241.
Royston, P. (2005). Multiple imputation of missing values: Update
of ICE. Stata Journal, 5, 527–536.
Royston, P. (2007). Multiple imputation of missing values: Further
update of ice, with an emphasis on interval censoring. Stata
Journal, 7, 445–464.
Royston, P. (2009). Multiple imputation of missing values: Further
update of ice, with an emphasis on categorical variables. Stata
Journal, 9, 466–477.
Royston, P., Carlin, J. B., & White, I. R. (2009). Multiple imputa-
tion of missing values: New features for mim. Stata Journal,
9, 252–264.
Schuler, A. L., & Baldwin, M. (1981). Nonspeech communica-
tion and childhood autism. Nonspeech Communication and
Childhood Autism, 12, 246–257.
Shattuck, P. T., Narendorf, S. C., Cooper, B., Sterzing, P. R.,
Wagner, M., & Taylor, J. L. (2012). Postsecondary educa-
tion and employment among youth with an autism spectrum
disorder. Pediatrics, 126, 1042–1049. doi:10.1542/peds.
2011-2864
Shogren, K. A., & Plotner, A. J. (2012). Transition planning for
students with intellectual disability, autism, or other disabili-
ties: Data from the National Longitudinal Transition Study-2.
Intellectual and Developmental Disabilities, 50, 16–30.
Spann, S. J., Kohler, F. W., & Soenksen, D. (2003). Examining
parents’ involvement in and perceptions of special education
services: An interview with families in a parent support group.
Focus on Autism and other Developmental Disabilities, 18,
228–237.
Taylor, J. L., & Seltzer, M. M. (2011). Employment and post-
secondary educational activities for young adults with autism
spectrum disorders during the transition to adulthood. Journal
of Autism and Developmental Disorders, 41, 566–574.
doi:10.1007/s10803-010-1070-3
Test, D. W., Mason, C., Hughes, C., Konrad, M., Neale, M., &
Wood, W. (2004). Student involvement in individualized edu-
cation program meetings. Exceptional Children, 70, 391–412.
Test, D. W., Mazzotti, V. L., Mustian, A. L., & Fowler, C. H.
(2009). Evidence-based secondary transition predictors for
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
12 Remedial and Special Education
improving postschool outcomes for students with disabilities.
Career Development for Exceptional Individuals, 32, 160–
181. doi:10:1177/0885728809346960
Uditsky, B., & Hughson, E. (2012). Inclusive postsecondary edu-
cation—An evidence-based moral imperative. Journal of
Policy and Practice in Intellectual Disabilities, 9, 298–302.
doi:10.1111/jppi.12005
U.S. Department of Education. (2007). IDEA regulations:
Secondary transition. Retrieved from http://idea.ed.gov/
explore/view/p/,root,dynamic,TopicalBrief,17,
Valdes, K., Godard, P., Williamson, C., Van Campen, J.,
McCracken, M., & Jones, R. (2013). National Longitudinal
Transition Study-2 (NLTS2) Waves 1, 2, 3, 4 & 5 data docu-
mentation and dictionary. Menlo Park, CA: SRI International.
Virginia Department of Education. (1998). 1997 Project
UNITE Individual Transition Plan: Information summary
(Unpublished data). Richmond, VA: Author.
Von Hippel, P. T. (2007). Regression with missing y’s: An
improved strategy for analyzing multiple imputed data.
Sociological Methodology, 37, 83–117. doi:10.1111/j.1467-
9531.2007.00180.x
Wagner, M., Kutash, K., Duchnowski, A. J., & Epstein, M. H.
(2005). The Special Education Elementary Longitudinal
Study and the National Longitudinal Transition Study.
Journal of Emotional and Behavioral Disorders, 13, 25–41.
doi:10.1177/10634266050130010301
Wagner, M., Newman, L., Cameto, R., Javitz, H., & Valdes, K.
(2012). A national picture of parent and youth participation
in IEP and transition planning meetings. Journal of Disability
Policy Studies, 23, 140–155. doi:10.1177/1044207311425384
Wagner, M., Newman, L., Cameto, R., Levine, P., & Marder,
C. (2007). Perceptions and expectations of youth with dis-
abilities: A special topic report of findings from the National
Longitudinal Transition Study-2 (NLTS2). Menlo Park, CA:
SRI International.
Wehmeyer, M. L., Palmer, S. B., Soukup, J. H., Garner, N. W., &
Lawrence, M. (2007). Self-determination and student transi-
tion planning knowledge and skills: Predicting involvement.
Exceptionality: A Special Education Journal, 15, 31–44.
Wei, X., Yu, J. W., Shattuck, P. T., McCracken, M., &
Blackorby, J. (2013). Science, technology, engineering,
and mathematics (STEM) participation among college stu-
dents with autism spectrum disorder. Journal of Autism and
Developmental Disorders, 43, 1539–1546. doi:10.1007/
s10803-012-1700-z
White, I. R., Royston, P., & Wood, A. M. (2011). Multiple impu-
tation using chained equations: Issues and guidance for prac-
tice. Statistics in Medicine, 30, 377–399.
Williams-Diehm, K., Wehmeyer, M. L., Palmer, S. B., Soukup, J.
H., & Garner, N. W. (2008). Self-determination and student
involvement in transition planning: A multivariate analysis.
Journal on Developmental Disabilities, 14(1), 25–36.
Yeargin-Allsopp, M., Rice, C., Karapurkar, T., Doernberg, N.,
Boyle, C., & Murphy, C. (2003). Prevalence of autism in a
U.S. metropolitan area. The Journal of the American Medical
Association, 289, 49–55.
at SRI INTERNATIONAL LIBRARY on May 4, 2015rse.sagepub.comDownloaded from
