The earnings and employment outcomes of the 2005 cohort of Canadian postsecondary graduates with disabilities.
Zarifa, David ; Walters, David ; Seward, Brad 等
INTRODUCTION
CANADA'S FEAR OF FUTURE SKILL and labor shortages has brought
youth with disabilities to the forefront of public policy. Despite the
Canadian Charter of Rights and Freedom, Canadian Human Rights Act,
provincial accessibility standards acts (e.g., Ministry of Community and
Social Services [MCSS] 2005), and employer incentives by the provinces
to subsidize the workplace (e.g., MCSS 2012), individuals with
disabilities continue to face barriers in the workforce.
Youth with disabilities constitute an increasingly educated, yet
underrepresented social group in today's knowledge-based economy,
and have become an increasingly important at risk group for educators
and policymakers (CAF 2004; HREOC 2005; LDAC 2005; NCD 2007; Roeher
Institute 2004). Issues with secondary academic preparation and
inaccessible support and instruction opportunities can lead youth with
disabilities to drop out of high school, and can subsequently become a
barrier to postsecondary education (Kirby 2009). Yet, educational
achievement-based programs designed to accommodate students' needs
are growing across campuses. Moreover, many universities are now
reporting that nearly 10 percent of their graduating students have a
disability (CUSC 2009). As access to postsecondary education continues
to improve and many traditionally marginalized groups attend in greater
proportions, it remains increasingly important to keep a close eye on
groups that experience difficulties making the transition to the
workforce.
Drawing on Statistics Canada's 2005 National Graduates Survey
(NGS), we explore three major research questions. First, despite
increased access among youth with disabilities at the postsecondary
level, how do persons with a disability, who have obtained postsecondary
credentials, fare in the labor market relative to counterparts without a
disability? Second, what types of credentials appear to moderate the
effects of disability on workforce outcomes? That is, how do the
transition outcomes of disadvantaged groups compare across fields,
faculties, and types of programs? Do these relationships vary across
levels of education (college, trades, undergraduate degrees, and
graduate degrees)? Finally, do graduates with a disability experience
similar inequalities across multiple transition outcomes (i.e.,
earnings, employment status)?
A growing body of research is examining youth with disabilities in
their transitional outcomes from high school to postsecondary education
(Fleming and Fairweather 2012; Kirby 2009; Robson et al. 2014; Trainor
2008), persistence through postsecondary programs (Boyko and Chaplin
2012; Mamiseishvili and Koch 2011), transition to adulthood (Janus 2009;
Wells, Sandefur, and Hogan 2003), and transition from high school to
work (Sanford et al. 2011). However, only a handful of quantitative
studies exist that have examined the employment outcomes of
postsecondary graduates with disabilities (e.g., Blackorby and Wagner
1996; Dickinson and Verbeek 2002; Janus 2009; Kirchner and Smith 2005;
Madaus, Zhao, and Ruban 2008; Madaus et al. 2001; Rylance 1998; Sweet et
al. 2014). In Canada, more recent studies have typically been
qualitative or descriptive in nature and have concentrated on only a few
cities or regions (e.g., Bennett and Gallagher 2013; Fichten et al.
2012; Gillies 2012; Holmes and Silvestri 2011). Further, to our
knowledge, no existing studies have examined how disability status may
interact with postsecondary degree or program type in its impact on
early employment outcomes.
Our results will be of particular importance in informing
government policymakers, education officials, and key stakeholders
concerned with educational access, labor market transitions, and
workplace accessibility on how to tailor existing policies and programs
to eliminate systemic barriers to the labor market successes of youth
with disabilities.
REVIEW OF THE LITERATURE
Labor Market Outcomes of Postsecondary Graduates with Disabilities
In today's knowledge-based economy, workers are increasingly
expected to obtain the necessary skills to enjoy labor market successes.
Postsecondary education has consistently been shown to have positive
effects on the employment status, occupation, earnings, and job
satisfaction of recent graduates (Allen, Harris, and Butlin 2001; Frank
and Walters 2012; Walters, 2004a, 2004b; Walters, White, and Maxim
2004). Those lacking the knowledge and competencies to satisfy the
increasingly technical demands of the new economy, however, may find
themselves in part-time, less lucrative employment, or trapped in longer
bouts of unemployment.
Human capital theory has long explained higher education as a means
for individuals to increase their productivity, marketability, and skill
acquisition in preparation for the workforce (Becker 1993). While many
individuals choose to invest in education and training and hold the
skills and abilities necessary to make successful transitions to the
workforce, this relationship is not perfect. For instance, the skill
sets possessed by new graduates and those required in the job are often
mismatched, leading to over- and under skilling (Frank and Walters 2012;
Frenette 2004; Quintini 2011). The loose coupling between skill
development and labor market utilization may be in part attributable to
the signaling power of education (e.g., Arrow 1973; Spence 1974) or
credentialism processes (Collins 1979) rather than reflect increased job
demands for higher levels of skill (Handel 2011; Quintini 2011). In
addition, researchers have shown that youth from disadvantaged
socioeconomic backgrounds may be more prone to experience skill mismatch
and difficulties in their school-work transitions (Andres et al. 1999;
Krahn 2009; Krahn and Bowlby 1999). But, what do the early workforce
outcomes look like for recent postsecondary graduates with disabilities?
In the general working population, lower educational attainment among
people with disabilities is strongly associated with lower incomes, and
having a post-high school qualification significantly reduced the
likelihood of being unemployed (e.g., Cleland and Smith 2010). But, can
we expect to find the same inequalities among recent graduates that
exist in the entire working population?
Some researchers suggest that even individuals with disabilities
who complete postsecondary education earn significantly less and have a
harder time securing stable employment (NCD 2007; Roeher Institute 2004;
Shier, Graham, and Jones 2009). Wannell and Caron's (1994)
examination of graduates with disabilities from the 1990 NGS suggests
that this may be the case. In terms of access, about 3.9 percent of
graduates from university identified as having a disability compared to
approximately 6.5 percent among community college and trades graduates.
In terms of employment outcomes, the authors found clear disparities
across several measures. For example, labor force participation was
significantly lower (89.3 percent vs. 92.7 percent) for graduates with
disabilities across all levels of education (Wannell and Caron 1994:45).
For earnings, the gap was about 7 percent among university graduates and
about 1.5 percent among college and trades graduates employed full-time
in 1992 (Wannell and Caron 1994:24-25). The employment rate was also
significantly lower for graduates with disabilities, six percentage
points lower among university graduates and about 12 percent lower among
college and trades graduates (Wannell and Caron 1994:44-45).
A more recent study by Holmes and Silvestri (2011) examined the
employment experiences of Ontario postsecondary graduates (between
2004/05 and 2007/08) with learning disabilities. In terms of employment
status, descriptive analyses revealed only about 69 percent of college
and university graduates with learning disabilities were employed
(part-time or full-time) and about 16 percent were unemployed and about
11 percent had returned to school. Over two-thirds of graduates with
disabilities indicated that they were satisfied in their jobs (Holmes
and Silvestri 2011). Similarly, Sweet et al. (2014:55) found that only
half of the immigrants with disabilities in their survey were able to
find work (compared to 74% of immigrants without a disability)--of
those, only half had found full-time, full-year employment.
Yet, it remains uncertain whether or not these gaps have changed in
recent years. Existing studies have focused predominantly on outcomes in
the entire working populations rather than focus on recent postsecondary
graduates. Moreover, we were unable to locate previous studies in Canada
or elsewhere that have assessed earnings and employment inequalities
across disability status from varying levels of education and fields of
study. Existing studies have long shown that field of study and level of
education have significant effects on the early earnings and employment
outcomes of new graduates (e.g., Finnie, 2001; Walters 2004a), yet it is
unclear whether or not these educational choices interact with
disability status.
Educational and Workforce Factors Contributing to Unequal Outcomes
Researchers have pointed to a number of reasons to explain unequal
workforce outcomes. Oftentimes, youth with disabilities face
inaccessible workplaces, inadequate supports, inappropriate
expectations, and hiring discrimination (Devlin and Pothier 2006;
Gooding 1995; Holloway 2001; Oliver 1996; Wallace and Fenwick 2010).
Others suggest students with learning and hearing disabilities encounter
difficulties (e.g., quick pace of lectures, physically inaccessible
buildings, in-class discussions) in postsecondary education that limit
their abilities to acquire important skills for the workforce (Fuller,
Bradley, and Healey 2004; Holloway 2001). Further, workers with
disabilities often feel marginalized and socially and physically
isolated, are ignored by colleagues, and experience stereotyping and
harassment in the workforce (Robert and Harlan 2006).
Shier et al. (2009) interviewed 56 people with disabilities
enrolled in employment training programs in Calgary and Regina. The
authors found that discrimination by both employers and colleagues as
well as labeling were key factors impeding workplace successes (Shier et
al. 2009). Interestingly, their study found that perceptions of
disability had stronger effects on workforce outcomes than lack of
accommodations in the workplace. Similar labeling effects have been
witnessed throughout the course of students' with disabilities
education, leading to a higher risk of dropout among persons with
disabilities at the secondary and postsecondary levels of education
(Trainor 2008).
Research from the United States suggests that new graduates with
learning disabilities need accommodations but often are reluctant to
disclose their disability. Individuals with learning disabilities are
often forced to weigh the costs and benefits to disclosing their
disability for fear of stigmatization and potential damage to their
public image (see Madaus 2008). Holmes and Silvestri (2011) found
similar results among Ontario postsecondary graduates. The great
majority of graduates did not disclose their disability for fear of
being judged and embarrassed, while some did not believe their
disability impacted their job duties. In terms of disclosure, some
groups more than others were willing to come forward, as females, older
students, college graduates, and those with lower job satisfaction were
more likely to disclose their disability (Holmes and Silvestri 2011).
Even prior to the workforce entry, individuals with disabilities
are experiencing barriers to educational attainment and achievement
(Michalko and Titchkosky 2010; Robson et al. 2014; Wallace and Fenwick
2010). For example, Robson et al.'s (2014) study on special needs
high school students from the Toronto District School Board found that
disability impacted postsecondary education enrollment rates, noting
that students with disabilities were far more represented at the college
level, while being less likely to pursue a university degree. The
authors attributed this to barriers inherent to social, cultural, and
economic capital differences (Robson et al. 2014:211). Similar studies
have looked at students with disabilities' success rates while
engaged in postsecondary education, noting that secondary school
academic preparation, teacher-student instruction accessibility, and the
costs associated with individualized services can have the greatest
impact on postsecondary education success (Kirby 2009).
As a result, researchers, governments, and policymakers in many
countries are becoming increasingly interested in the higher education
experiences and outcomes of young adults with disabilities (e.g., CAF
2004; Fuller et al. 2004; HREOC 2005; LDAC 2005; Michalko and
Titchkovsky 2010; NCD 2007; Roeher Institute 2004; Sweet et al. 2011;
Wallace and Fenwick 2010). While labor market outcomes for workers with
disabilities are typically less favorable in comparison to workers
without disabilities, improving access to postsecondary education and
removing barriers within the educational system may help to narrow the
gaps in labor force outcomes between youth with and without disabilities
(Roeher Institute 2004).
A recent report by Tsagris and Muirhead (2012) assessed the extent
to which postsecondary institutions offer support for students with
disabilities at Durham College and the University of Ontario Institute
of Technology. The authors examined the extent to which a Summer
Transition Program for students with learning disabilities taken prior
to postsecondary entry enhanced their education success through student
engagement and academic performance. Indeed, students who attended the
summer program were more familiar with the campus, more likely to access
student support and disability-specific support such as assistive
technology, professor accommodations, and enhanced counseling, all
contributing to higher academic performance (Tsagris and Muirhead
2012:8). Still, many students prefer not to disclose their disabilities
and request accommodations for fear of being stigmatized by their peers
and professors.
But, several questions remain unanswered. First, to what extent
does level and type of education play a role in securing favorable labor
market outcomes for graduates with disabilities? That is, do the
labor market outcomes between youth with and without disabilities vary
across levels of education and fields of study? To address this
question, we turn to interaction effects between disability status and
both level of education and field of study. While classical econometric
analyses of earnings disparities typically focus on workers who are
employed full-time, we believe that such analyses ignore much of the
marginalization within the labor market experienced by groups that may
also face greater barriers to securing full-time employment. Thus, we
examine which dimensions/types of workforce outcomes reveal the greatest
inequalities between graduates with and without disabilities (i.e.,
earnings or employment status).
METHODS
Data
This study employs data from Statistics Canada's 2005 NGS. The
survey contains information on 31,025 postsecondary graduates of various
programs across all provinces and territories, 1,604 of whom
self-identified as having a disability. Computer-Assisted Telephone
Interviews were employed to survey the educational history and
employment profiles of respondents in 2007, two years after completing
their degree. The survey population consisted of all Canadian graduates
who had completed the requirements of their postsecondary degrees,
diplomas, or certificates during the 2005 calendar year. The NGS serves
as a valuable resource for determining the labor market outcomes of
postsecondary graduates and past cycles have been used extensively in
the research literature (Allen et al. 2001; Finnie 2001; Finnie and
Frenette 2003; Frank and Walters 2012; Krahn and Bowlby 1999; Walters
and Zarifa 2008; Walters 2004a, 2006; Wannell and Charon 1994; Zarifa
and Walters 2008). However, few existing studies have focused on the
transitions of graduates with disabilities, or used the most recent
survey data on the early workforce experiences of graduates in 2005.
Moreover, by using the 2005 NGS, we are establishing a baseline measure
for the labor market experiences of graduates with disabilities, one
that can be further compared to future NGS waves to determine potential
changes to this group's experiences as they are released.
It is important to note that there are a number of limitations to
the NGS data. First, despite the benefits of the NGS in studying
postsecondary graduates' labor market outcomes, the cancellation of
the 2010 follow-up survey has restricted the 2005/2007 wave to only
allow for the study of short-term labor market outcomes. With the
absence of the 2010 follow-up survey, only the very early labor market
outcomes for postsecondary graduates from 2005 can be represented in the
data. Second, due to the coding of the disability identifier within the
data, this study can only assess the experiences of individuals living
with a self-identified disability as one homogenous group. Thus, it is
not possible to disentangle the impact of physical and learning
disabilities on employment outcomes. While an alternative Canadian
survey, the Participation and Activity Limitation Survey (PALS) provides
information on type, it is a postcensus survey of the entire working
population and fails to account for the early workforce experiences of
an entire cohort of postsecondary graduates from various fields of study
and levels of education. As a result, the NGS provides the most robust,
longitudinal sample of young Canadians with disabilities that have
completed postsecondary education available to understand the early
workforce experiences of this group. Finally, the NGS data only allows
for analysis among postsecondary graduates. Thus, we cannot compare the
labor market outcomes of those with and without postsecondary
credentials.
Variables
The key explanatory variables used in this study are disability,
level of education, and field of study. Disability status was coded as a
dichotomous "Yes" or "No" response to the question
that asked respondents to indicate whether she or he had a lasting
disability or handicap that had persisted at least six months. The level
of education variable distinguishes between the various postsecondary
programs that graduates surveyed in the NGS reported completing. These
levels are broken down into trades and vocational certification, college
diploma, university undergraduate degree programs, and university
advanced degree programs, which include graduates of masters, PhD, and
professional programs (i.e., law, medicine, dentistry, etc.). Finally,
the survey respondents were also asked to report their field of study.
Their responses were originally coded by Statistics Canada using the
Classification of Instructional Program system to match university,
college, and trade-vocational institutions' field of study
categories. Due to sample size concerns for this study, fields are
grouped into the following six categories: 1
1. Liberal arts (e.g., fine arts, humanities, and social sciences).
2. Commerce, management, and business administration.
3. Physical, agricultural, and biological sciences.
4. Engineering, computer sciences, and math.
5. Health and related fields.
6. Other.
The category classified as "other" includes areas that
could not be adequately captured in the above fields such as education,
recreational and counseling services, interdisciplinary studies,
unknown, or unclassified fields not specified and undeclared. These
categories are consistent with previous work with the NGS on fields of
study and workforce outcomes and thus serve as a point of comparison
with past research (Allen et al. 2001; Finnie and Frenette 2003; Frank
and Walters 2012; Zarifa and Walters 2008).
The response variables used in the study are the respondents'
annual earnings and their employment status. The earnings variable is
based on the following survey question: "Working your usual hours,
approximately what would be your annual earnings before taxes and
deductions at that job?" While the measure is an approximation of
what graduates would earn on an annual basis if the job were to last a
full year (adjusting for irregular work patterns), the variable has been
considered to be a well-defined measure in past versions of the NGS (see
Finnie 2000:201). Due to the skewed nature of the earnings variable, we
use the natural log of earnings as the dependent variable in our
ordinary least squares (OLS) regression models. The employment status
variable consists of three categories: (1) employed, full-time; (2)
employed, part-time (less than 30 hours per week); and (3) those who
were unemployed at the time of interview.
The variables age, sex, marital status, presence of dependent
children, visible and immigrant minority statuses, parental education,
region, and whether the respondent received funding from various sources
are available in the survey and have been included in the models as
control variables. These variables have been shown to influence labor
market outcomes in prior research conducted using the 2000 NGS (Walters
and Zarifa 2008; Zarifa 2012a, 2012b; Zarifa and Walters 2008).
Table A1 in the Appendix provides the descriptive statistics for
all focal variables from the NGS. (1) In terms of disability, the great
majority of respondents in the NGS did not self-identify as having a
disability; they represent 4 percent of the respondents in the earnings
models and 3 percent of the employment status models. Among fields of
study, the sampled graduates are most likely to hail from business,
liberal arts, and engineering and computer sciences (roughly 20
percent), followed by health-related fields and "other" fields
(15 percent each). Respondents are least likely to graduate from fields
in physical and agricultural sciences (only 5 percent of the
population). The majority of respondents received bachelor degrees (38
percent) followed by a college diploma (35 percent). The proportion of
respondents with trades-related certification was similar to those with
advanced university degrees (approximately 14 percent). In terms of
workforce outcomes, the majority of respondents held full-time
employment at the time of interview (85 percent), while the proportion
of those employed part-time and unemployed are 9 and 6 percent,
respectively. Finally, the average earnings of recent graduates in 2007
dollars is $45,900.
Analyses
Our statistical analyses include OLS regressions and multinomial
logistic regressions. The bivariate results for earnings and employment
models are shown in Tables A2 and A3 in the Appendix. For both the OLS
and multinomial regressions, we enter key variables into the models in
several stages. The first set of models regress earnings and employment
status separately on only the disability variable. Model 2 includes the
sociodemographic variables age, gender, marital status, presence of
dependent children, visible and immigrant minority statuses, parental
education, and the region of respondent as control variables. The third
set of models introduces whether the respondents received grants,
scholarships, or student loans to support their schooling as well as
graduates' field of study and level of postsecondary education. (2)
Finally, Models 4 and 5 include interactions between disability and
field of study and level of education to assess whether or not there are
differences in earnings and employment statuses between graduates with
and without a self-disclosed disability across educational types. With
the exception of age and age squared, all of the explanatory variables
used in the analysis are treated as categorical. (3)
RESULTS
OLS Regressions: Earnings
Do graduates with disabilities earn less than their counterparts
without disabilities? If so, does the relative gap vary across fields
and level of education? Table 1 shows the results of OLS regressions
addressing these questions. (4) As outlined above, Models 1 through 3
(available upon request) included key variables of interest in several
stages: (1) disability status; (2) sociodemographic variables; and (3)
education-related variables. As expected, individuals with a disability
do earn significantly less than their counterparts without a disability
(p < .001). This finding remains statistically significant (p <
.001), even when controlling for both sociodemographic and educational
characteristics. The exponentiated log-earnings estimates from Model 3
(not shown) reveal that there is approximately a $4,000 earnings gap
between postsecondary graduates with and without disabilities that is
not attributable to the other variables in the model.
[FIGURE 1 OMITTED]
To explore the extent to which this earnings gap varies by
educational characteristics, Models 4 and 5 include interaction terms
between disability and field of study and level of education. Indeed,
the results from Model 4 indicate that the interaction between
disability and field of study is statistically significant (p <
.001). That is, the effect of disability on earnings varies when related
to the field of study of the graduate. Figure 1A reveals that the
greatest earnings inequality between postsecondary graduates with and
without disabilities exists for those with postsecondary credentials in
a liberal arts and business program (approximately $6,000). (5) The
earnings difference between the two groups in health related fields is
also statistically significant (p < .05). The point estimates and
corresponding confidence intervals in Figure 1A also show that the
effect of disability on earnings is not statistically significant for
graduates of the sciences, engineering, and fields classified as other.
Finally, the results from Model 5 indicate that the interaction
between disability and level of schooling is also statistically
significant (p < .001). This finding suggests that the effect of
having a disability on earnings also is related to the level of
postsecondary schooling obtained by the graduates surveyed in this
study. (6) To further understand the relationship, the earnings
estimates and corresponding 95 percent confidence intervals for
graduates with and without disabilities are plotted separately by level
of schooling in Figure IB. (7) The figure reveals that there is a
significant difference in earnings for graduates of trades, college, and
advanced university degree programs. However, the estimate for
university graduates without disabilities just barely encompasses the 95
percent confidence interval for their counterparts with disabilities.
Thus, the earnings gap between these two groups of postsecondary
graduates (approximately $1,000), is not statistically significant.
Multinomial Logistic Regressions: Employment Status
To what extent is employment status related to disability status?
Second, do the employment outcomes of graduates with disabilities vary
by their field of study and their level of education? The second series
of regression models in Table 2 seeks to answer these questions by
distinguishing among graduates who are employed full-time, part-time
(< 30 hours per week) or who are unemployed at the time of the
survey. Since the response variable consists of three discrete
categories, we estimate a series of multinomial logistic regression
models, fit via maximum likelihood. As with the earnings models above,
the disability variable is entered first, followed by sociodemographic
characteristics, and eventually education-related variables.
In Models 1 through 3 (available upon request), the effect of
disability on employment status is statistically significant (p <
.001). In all three models, graduates without disabilities are more
likely to be employed fulltime than their counterparts who reported
having a disability. In fact, the predicted probabilities indicate that
graduates without disabilities are approximately .10 times more likely
(approximately .87 compared to .77) to be employed full-time (with and
without controls). Second, the results indicate that graduates with
disabilities are more likely to be employed part-time; however, the gap
is modest. The probability of part-time employment for graduates who
reported having a disability is approximately .11 in comparison to .07
for graduates without disabilities. Finally, the predicted probability
of being unemployed for graduates without disabilities is approximately
.06. By comparison, the predicted probability of being unemployed for
graduates who reported having a disability is more than twice as high
(.12). Thus, the substantially lower full-time employment levels of
postsecondary graduates with disabilities is attributable to their
higher levels of both part-time employment and unemployment.
As with the regression models for earnings, in Models 4 and 5, we
include a set of dummy variables to capture the interaction effects
between disability and field of study and level of education. As shown
in Table 2, the likelihood ratio chi-square test for the interaction in
Model 4 is statistically significant (p < .001), indicating that the
effect of disability on employment status is related to field of study.
(8) To provide a more meaningful interpretation of this result, we
created effect displays for each multinomial regression model (see Fox
and Andersen 2006) to reveal the predicted probabilities of being
unemployed, employed full-time, and employed part-time, separately for
graduates with and without disabilities and across fields of study.
Figure 2A displays the predicted probability of being employed full-time
for postsecondary graduates with and without disabilities in each of the
six fields of study. The results reveal that the greatest gap in
full-time employment between respondents with and without disabilities
is observed for graduates of the liberal arts (.61 vs. .81), followed by
business (.77 vs. .90) and then by engineering (.80 vs. .88). The gap in
full-time employment status among the two groups of graduates is also
statistically significant for graduates of engineering and fields
classified as other, but it is not the case for graduates of
health-related fields.
Figure 2B displays the predicted probability of being employed
part-time for postsecondary graduates with and without disabilities
across various fields of study. The greatest disparity between the two
groups of graduates in terms of part-time employment status can be found
among liberal arts graduates. The probability of being employed
part-time is .23 for graduates who identified themselves as having a
long-term disability and .11 for graduates who did not report having a
disability. The 95 percent confidence intervals for the estimates for
business and science graduates also suggest that graduates with
disabilities are significantly more likely to be employed part-time than
graduates without disabilities from the same field; however, these
disparities are not as large as those observed for liberal arts
graduates. At the same time, the confidence intervals in Figure 2B
reveal that graduates who reported having a disability and received
their certification in engineering, health, and other fields of study
are no more or less likely to be employed part-time than graduates of
those fields who did not report having a lasting impairment.
[FIGURE 2 OMITTED]
In terms of unemployment gaps across fields of study, Figure 2C
shows that the greatest disparities between graduates with and without
disabilities are found in business (.15 vs. .05), engineering (.16 vs.
.07), and the liberal arts (.15 vs. .08). The 95 percent confidence
intervals reveal that graduates who reported having a disability and
obtain certification in science, health, and the fields classified as
"other" are not more likely to be unemployed.
Finally, Model 5 includes the interaction between the disability
variable and level of schooling. According to the results in Table 2,
the effect of the interaction is statistically significant (p <
.001), indicating that the effect of disability on employment status is
related to the level of schooling obtained by the postsecondary
graduates. (9)
To further explain this finding, Figure 3A displays the predicted
probability of being employed full-time for graduates with and without
disabilities who completed trades, community college, university
undergraduate, and advanced university programs. The difference in
full-time employment between graduates who reported having a disability
and those who did not is largest among university advanced degree
programs (.91 vs. .76), closely followed by graduates of trades programs
(.89 vs. .75), then by respondents with community college diplomas (.83
vs. .74). University graduates with undergraduate degrees who reported
having a disability have a predicted probability of being employed
full-time of .75, in comparison to .81 for their counterparts without
disabilities. (10)
Similarly, in Figure 3B, the point estimates and 95 percent
confidence intervals suggest that the difference in the probability of
being employed part-time for graduates with and without disabilities is
statistically significant at all levels of postsecondary schooling. The
predicted probability of being employed part-time for trade's
graduates with disabilities is . 14, which is much higher than it is for
trade's graduates without disabilities (.06). Among community
college graduates, the predicted probability of being employed part-time
for those with disabilities is . 16, which is .06 higher than their
counterparts without disabilities (.10), controlling for the other
variables in the model. The part-time employment gap is smallest among
respondents with undergraduate degrees, where graduates with
disabilities are only slightly less likely to be employed part-time than
graduates who did not report having a disability (.07 vs. .09). Finally,
graduates who report having a disability and have obtained advanced
university degrees are considerably more likely (.16) than counterparts
without a disability (.05) to be employed part-time.
[FIGURE 3 OMITTED]
Figure 3C displays the fitted probabilities of being unemployed for
each set of postsecondary graduates. While graduates who reported having
a disability with trades certification are almost twice as likely to be
unemployed (.11) as graduates who did not report having a disability
(.06), respondents with and without disabilities holding a community
college diploma are no more or less likely to experience unemployment
two years following graduation. Among university graduates with a
disability, the probability of being unemployed is .14 for those with an
undergraduate degree in comparison with .06 for counterparts without a
disability. Further, due to the large confidence interval for graduates
without a disability with advanced university degrees, the difference in
the predicted probability of being unemployed between the two groups of
graduates with advanced level university degrees is not statistically
significant. Thus, the large gap in full-time employment among graduates
with and without disabilities who have obtained advanced university
degrees is attributable to differences in part-time employment rather
than unemployment.
DISCUSSION
This study draws attention to the early employment difficulties
faced by new postsecondary graduates with disabilities. Existing studies
have placed increasing importance on the type and highest level of
education obtained on one's earnings and employment status in
today's knowledge-based economy (Allen et al. 2001; Frank and
Walters 2012; Walters 2004a, 2004b; Walters et al. 2004). Yet, until
this point, very little research has explored the early workforce
outcomes of postsecondary graduates with disabilities, or have existing
studies sought to examine if earnings and employment differences vary by
education level and field of study for youth with and without
disabilities.
Overall, our findings indicate that the presence of a disability
has a negative impact on graduates' earnings and employment status.
Moreover, our findings remain salient even after controlling for key
sociodemographic, educational, and financial factors. Despite holding
postsecondary credentials constant, across all models, postsecondary
graduates with disabilities experienced significant labor market
inequalities. These graduates earned significantly less (approx. $4,000)
than their counterparts without disabilities. It is important to note
that this gap exists only two years after the completion of their
postsecondary education. Theories of cumulative advantage suggest that
these inequalities will further entrench and that the earnings gap will
grow over the life course (for a review, see DiPrete and Eirich 2006).
Studies on the general working population with disabilities suggest that
these inequalities do not subside. Workers with disabilities remain much
more likely to encounter job insecurity, poor employment status,
underemployment, and are overrepresented in entry-level and part-time
jobs (Kaye 2009; Konrad et al. 2013; Schur et al. 2009). Moreover, the
compounding effect of disability over the life course can be seen when
examining the retirement ages of those reporting a disability condition.
That is, a substantial proportion of individuals with disabilities often
permanently leave the workforce involuntarily at around 48 years of age
(Sweet et al. 2011). By this time, however, many have not fully advanced
their careers, nor have they been employed long enough to be eligible
for retirement benefits (Sweet et al. 2014).
It is important to note that this is the earnings gap among
graduates who were employed full-time. As we anticipated, the earnings
analyses only reveal part of the story of how graduates with a
self-disclosed disability are marginalized within the Canadian labor
market. The most noteworthy findings of our analyses are obtained from
the regression models predicting employment status. Our employment
status analyses revealed even greater inequalities. Looking specifically
at youth with disabilities, these individuals were significantly less
likely to hold full-time employment, and were significantly more likely
to be employed part-time or be unemployed. What is most alarming is that
graduates who reported having a disability had a probability of. 12 of
being unemployed two years after graduation--twice as high as their
counterparts without a disability (.06). While one possible explanation
could be that work limitations resulting from graduates'
disabilities prevent these individuals from securing stable employment
(whether by choice or otherwise), a number of studies have observed
inaccessible and discriminatory workplaces suggesting that there may be
systemic employment barriers for persons with disabilities in the labor
market (Devlin and Pothier 2006; Gooding 1995; Holloway 2001; Holmes and
Silvestri 2011; McCloy and DeClou 2013; Oliver 1996; Robert and Harlan
2006; Shier et. al 2009).
The results also indicated that earnings and employment gaps vary
significantly by an individual's educational credentials. By
including field of study and level of education as predictors of labor
market outcomes, these analyses qualify the assumption that
postsecondary education translates to equally successful transition
rates into the labor market. For earnings and employment status,
graduates who reported having a disability and had obtained liberal arts
and business certification experienced the greatest relative
inequalities across fields of study. Engineering graduates with
disabilities also experienced difficulties, as they were significantly
more likely than engineers without a disability to enter unemployment
and part-time work. Level of education also changes the relationship
between disability status and workforce outcomes. While university
graduates with disabilities were significantly less likely to enter
part-time employment than those who did not report having a disability,
the results revealed larger relative gaps in terms of unemployment.
For scholars, the results of our study presents challenges for
human capital assumptions that suggest higher levels of education
translate into greater labor market rewards for all social groups. While
social inequality research has typically focused on difficulties across
class, race/ethnicity, gender, and age dimensions, our results
underscore the importance of examining disability as an increasingly
important dimension of social inequality. Despite holding similar
credentials both in terms of education level as well as field of study,
youth with disabilities did not receive the same returns to their
educational investments as their counterparts without disabilities.
First, the inability to disaggregate disability and examine
workforce outcomes by disability type is a limitation of our study, as
authors have found high variation among educational outcomes. Research
in the United States points to the importance of disentangling physical
and learning disabilities as well as disability severity (Garlaneau and
Radulescu 2009; Janus 2009; Robert and Harlan 2006; Sanford et al. 2011;
Trainor 2008). Trainor (2008), for example, finds 53% of American youth
with emotional and behavioral disabilities left high school without a
diploma or certification, compared to 27% of youth with learning
disabilities who also dropped out. It is possible that similar
variations across disability types may exist in the labor market. Should
data become available, an important consideration for future research is
to break down disability status by type or severity, providing further
insight into the early labor market inequalities observed here.
Second, and related to disaggregating disability, the lack of
consistent measures of disability poses challenges. As McCloy and DeClou
(2013:7) identify, the lack of a robust survey disaggregating disability
limits the ability to distinguish among disabilities, which can result
in inconsistent measurements for disability as an identifier within the
wider literature. While the NGS, for instance, allows only the
identification of whether an individual has a disability or not, the
PALS, due to sample size limitations when filtering, can only
disaggregate between "physical disabilities" and
"other" in many cases (McCloy and DeClou 2013:7). This can
lead to an inconsistent consensus within the literature, as comparisons
may be incompatible due to methodological differences.
The last limitation is that our study is only able to capture the
labor market experiences of persons with disabilities who graduated from
postsecondary education institutions and self-identified as having a
disability. That is, the NGS is unable to account for those who choose
not to pursue, or were unable to confirm postsecondary enrollment (see
Robson et al. 2014), as well as those who did not disclose a disability.
As this survey is a profile of graduates who have obtained their
certification from postsecondary institutions in Canada, those who also
did not persist to graduation are not included in the sample. This is a
similar problem for other data sources that include information on
persons with disabilities, such as the PALS (McCloy and DeClou 2013:7).
It could be argued that those students who persist to graduation have
"made it," and therefore provide a skewed profile of students
with disabilities. Still, this exploratory study serves to frame the
systemic inequalities faced by new graduates with disabilities entering
the labor market in Canada. As our study shows, even those who are in
the best position to succeed in the labor market experience significant
difficulties. Therefore, the call for further policy initiatives is
exacerbated when considering that the individuals at risk are at the
best-possible position among this group.
The results of our study suggest several challenges remain for
policymakers. First, given the positive role postsecondary education
plays on workforce outcomes, education officials, and policymakers need
to continue to improve access and offer institutional supports within
postsecondary education for persons with disabilities. Studies such as
Tsagris and Muirhead (2012) suggest that summer transition programs
greatly benefit students with disabilities and contribute to higher
academic performance (see also Irving 2012; Winn and Hay 2009). In
another example, the Ontario Learning Opportunities Fund established 13
learning disabilities transition centers across postsecondary
institutions in Ontario (McCloskey et al. 2011:10). These centers
provided youth with disabilities coping strategies, self-advocacy, and
individual skill instruction (McCloskey et al. 2011:10). These
initiatives mirror a growing trend for special needs education--a
movement away from systems based on "deficit" approaches that
emphasize risk reduction to "strengths" based strategies that
mobilize protective factors such as positive self-image, and a voice in
decisions concerning academic futures (Sweet et al. 2011:42).
However, many students with disabilities are fearful to disclose
their needs. Thus, the challenge that remains is twofold. First,
continued investment in these and other successful programs is needed.
Second, policymakers and education officials need to continue to build
an inclusive learning environment in postsecondary institutions whereby
individuals feel comfortable disclosing their disabilities and level the
playing field through these services.
Second, despite holding postsecondary credentials, many recent
graduates with disabilities still experienced labor market difficulties,
particularly in securing full-time employment, perhaps suggesting that
more comprehensive employment policies are needed and should consider
alternative initiatives targeted toward new graduates with disabilities.
Canadian policymakers have already started launching programs to aid in
the participation of talented, underrepresented groups (for recent
policy initiatives, see HRSDC 2013), but a clearly defined policy for
youth with disabilities may be warranted. Further, while the government
offers several financial benefits to aid persons with disabilities
(e.g., Disability Tax Credit; Registered Disability Savings Plan), these
benefits are contingent upon the ability to first obtain employment
(Service Canada 2014), and would not serve to reduce the gaps in
full-time employment rates among new graduates.
Alternatively, it may be beneficial for employment policies to
account for disability as a whole, as overarching strategies that focus
on all demographics of persons with disability have been successfully
achieved through advocating and legal action (Canadian Council on
Learning 2009; Martin Prosperity Institute 2010; Wilton 2006). In this
way, disability policies assist graduates both entering the labor
market, as well as further along the course of their careers,
potentially resulting in more stable outcomes for this demographic
(Sweet et al. 2014). Indeed, much emphasis is being placed on
recognizing the social and emotional dimensions of disability, and on
the increased access to training meant to retain and sustain employees
with disabilities (Deloitte 2010; Public Service Commission of Canada
2011). While there are still issues particular to recent graduates with
disabilities that require attention, these overarching policies may
better serve graduates as they progress in their careers.
Finally, our results highlight a pressing need for building
stronger connections between student service offices on campus and local
organizations. From the employer side, Markel and Barclay (2009) suggest
an ethos of organizational social responsibility as a key ingredient in
reducing the underutilization of graduates with disabilities in the
labor market. The authors suggest that employers offer targeted training
for coworkers, accessible job postings, develop partnerships with campus
disability support service offices and local disability services, and
provide ongoing support for the continued employment of persons with
disabilities. Our results indicate that those who reported the most
difficulties transitioning to the workforce were trades and college
graduates and graduates from the liberal arts and business. Likewise,
among graduates with advanced university degrees, graduates with
disabilities are far more likely than their counterparts without
disabilities to be employed part-time. Thus, individuals with
disabilities who hold these educational credentials may be the most in
need of stronger school-work linkages.
DAVID ZARIFA
Nipissing University
DAVID WALTERS AND BRAD SEWARD
University of Guelph
The authors are very grateful to the editor and the anonymous
reviewers for their helpful comments and suggestions.
David Zarifa, Department of Sociology, Nipissing University, 100
College Drive, Box 5002, North Bay, ON PIB 8L7, Canada. E-mail:
[email protected]
References
Allen, M., S. Harris and G. Butlin. 2001. Finding Their Way: A
Profile of Young Canadian Graduates. Cat. No. 81-595-MIE2003003. Ottawa:
Statistics Canada.
Andres, L., P. Anisef, H. Krahn, D. Looker and V. Thiessen. 1999.
"The Persistence of Social Structure: Cohort, Class and Gender
Effects on the Occupational Aspirations and Expectations of Canadian
Youth." Journal of Youth Studies 2(3):261-82.
Arrow, K.J. 1973. "Higher Education as a Filter." Journal
of Public Economics 2:193-216.
Becker, G.S. 1993. Human Capital: A Theoretical and Empirical
Analysis, with Special Reference to Education. 3d ed. Chicago: The
University of Chicago Press.
Bennett, S.M. and T.L. Gallagher. 2013. "High School Students
with Intellectual Disabilities in School and Workplace: Multiple
Perspectives on Inclusion." Canadian Journal of Education
36(1):96-124.
Blackorby, J. and M. Wagner. 1996. "Longitudinal Postschool
Outcomes of Youth with Disabilities: Findings from the National
Longitudinal Transition Study." Exceptional Children 62:399-413.
Boyko, L. and E.K. Chaplin. 2012. "Students with Learning
Disabilities: Access to Higher Education in Canada--A Review of the
Literature." Retrieved October 12, 2014
(https://tspace.library.utoronto.ca/bitstream/1807/31997/1/
Literature%20Review_Boyko_Chaplin%20FINAL.pdf).
CAF [Canadian Abilities Foundation]. 2004. Neglected or Hidden:
Connecting Employers with People with Disabilities in Canada. Toronto.
Canadian Council on Learning (CCL). 2009. A Barrier to Learning:
Mental Health Disorders among Canadian Youth. Ottawa: Canadian Council
on Learning.
Cleland, G. and A. Smith. 2010. Journey to Work: Creating Pathways
for Young Disabled People in New Zealand. New Zealand: CCS Disability
Action.
Collins, R. 1979. The Credential Society. New York: Academic Press.
CUSC [Canadian University Survey Consortium]. 2009. Graduating
Students Survey. Retrieved September 15, 2013 (http://www.cusc-
ccreu.ca/publications/rpt_CUSC_Master_V3_20091.pdf).
Deloitte. 2010. "The Road to Inclusion: Integrating People
with Disabilities into the Workplace." Deloitte Roundtable on
Diversity. Retrieved September 15, 2013
(www.deloitte.com/assets/Dcom-Canada/Local%20Assets/Documents/About%20us/ Diversity/ca_en_dialogue_on_diversity_vs_080710.pdf).
Devlin, R. and D. Pothier. 2006. "Introduction: Toward a
Critical Theory of Dis- Citizenship." Pp. 1-22 in Critical
Disability Theory: Essays in Philosophy, Politics, Policy, and Law,
edited by D. Pothier and R. Devlin. Vancouver, BC: UBC Press.
Dickinson, D.L. and R.L. Verbeek. 2002. "Wage Differentials
between College Graduates with and without Learning Disabilities."
Journal of Learning Disabilities 35:175-84.
DiPrete, T.A. and G.M. Eirich. 2006. "Cumulative Advantage as
a Mechanism for Inequality: A Review of Theoretical and Empirical
Developments." An nual Review of Sociology 32:27-197.
Fichten, C.S., S. Jorgensen, A. Havel, M. Berile, V. Ferraro, M.
Landry, D. Fiset, J.C. Juhel, C. Chwojka, M.N. Nguyen, R. Amsel and J.
Asuncion. 2012. "What Happens after Graduation? Outcomes,
Employment, and Recommendations of Recent Junior/Community College
Graduates with and without Disabilities." Disability &
Rehabilitation 34(11):917-24.
Finnie, R. 2000. "From School to Work: The Evolution of Early
Labour Market Outcomes of Canadian Postsecondary Graduates."
Canadian Public Policy 26:197-224.
Finnie, R. 2001. "Fields of Plenty, Fields of Lean: The Early
Labour Market Outcomes of Canadian University Graduates by
Discipline." Canadian Journal of Higher Education 31(0:141-76.
Finnie, R. and M. Frenette. 2003. "Earning Differences by
Major Field of Study: Evidence from Three Cohorts of Recent Canadian
Graduates." Economics of Education Review 22(2): 17992.
Fleming, A.R. and J.S. Fairweather. 2012. "The Role of
Postsecondary Education in the Path from High School to Work for Youth
with Disabilities." Rehabilitation Counselling Bulletin
55(2):71-81.
Fox, J. and R. Andersen. 2006. "Effect Displays for
Multinomial and Proportional-Odds Logit Models." Sociological
Methodology 36(1):225-55.
Frank, K. and D. Walters. 2012. "Exploring the Alignment
between Postsecondary Education Programs and Labour Market
Outcomes." Canadian Journal of Higher Education 42(3):93-115.
Frenette, M. 2004. "The Overqualified Canadian Graduate: The
Role of the Academic Program in the Incidence, Persistence, and Economic
Returns to Overqualification." Economics of Education Review
23:29-45.
Fuller, M., A. Bradley and M. Healey. 2004. "Incorporating
Disabled Students within an Inclusive Higher Education
Environment." Disability & Society 19(5):455-68.
Garlaneau, D. and M. Radulescu. 2009. Employment among the
Disabled. Catalogue No. 75001-X. Ottawa: Statistics Canada.
Gillies, J. 2012. "University Graduates with a Disability: The
Transition to the Workforce." Disability Studies Quarterly 32(3).
Retrieved June 10, 2015 (http://dsqsds.org/article/view/3281/3115).
Gooding, C. 1995. "Employment and Disabled People: Equal
Rights or Positive Action." Pp. 6476 in Removing Disabling
Barriers, edited by G. Zarb. London: Policy Studies Institute.
Handel, M. 2012. "Trends in Job Skill Demands in OECD
Countries." Working Paper No. 143, OECD Social, Employment and
Migration. OECD Publishing. Retrieved September 15, 2013
(http://dx.doi.org/10.1787/5k8zk8pcq6td-en).
Holloway, S. 2001. "The Experience of Higher Education from
the Perspective of Disabled Students." Disability & Society 16:
597-615.
Holmes, A. and R. Silvestri. 2011. Employment Experience of Ontario
Postsecondary Graduates with Learning Disabilities. Toronto: Higher
Education Quality Council of Canada.
HREOC [Human Rights Equal Opportunity Commission]. 2005.
"WORKability II: Solutions, People with Disability in the Open
Workplace." Final Report of the National Inquiry into Employment
and Disability. Sydney, Australia: HREOC.
HRSDC. 2013. "Funding Programs." Human Resources and
Skills Development Canada. Ottawa: HRSDC Canada. Retrieved May 16, 2013
(http://www.hrsdc.gc.ca/eng/funding/index.shtml).
Irving, B.A. 2012. "Access, Opportunity, and Career:
Supporting the Aspirations of Dis/Abled Students with High-End Needs in
New Zealand." International Journal of Inclusive Education 17:1-13.
Janus, A.L. 2009. "Disability and the Transition to
Adulthood." Social Forces 88(1):99-120.
Kaye, H.S. 2009. "Stuck at the Bottom Rung: Occupational
Characteristics of Workers with Disabilities." Journal of
Occupational Rehabilitation 19:115-28.
Kirby, D. 2009. "Widening Access: Making the Transition from
Mass to Universal Post-Secondary Education in Canada." Journal of
Applied Research on Learning 2:1-17.
Kirchner, C. and B. Smith. 2005. "Transition to What?
Education and Employment Outcomes for Visually Impaired Youths after
High School." Journal of Visual Impairment & Blindness
99:495-504.
Konrad, A.M., M.E., Moore, E.S.W. Ng, A.J. Doherty and K. Breward.
2013. "Temporary Work, Underemployment and Workplace
Accommodations: Relationship to Well-being for Workers with
Disabilities." British Journal of Management 24:367-82.
Krahn, H. 2009. "Choose Your Parents Carefully: Social Class,
Post-Secondary Education, and Occupational Outcomes." Pp. 171-89 in
Social Inequality in Canada: Patterns, Problems, and Policies, edited by
E. Grabb and N. Guppy. Toronto, ON: Pearson Prentice Hall.
Krahn, H. and J.W. Bowlby. 1999. "Education-Job Skills Match:
An Analysis of the 1990 and 1995 National Graduate Surveys."
Applied Research Branch for Strategic Policy. Ottawa, ON: Human
Resources Development Canada.
LD AC [Learning Disabilities Association of Canada]. 2005. A
Literature Framework to Guide the Research Study: Putting a Canadian
Face on Learning Disabilities (PACFOLD). Ottawa: LDAC.
Madaus, J.W. 2008. "Employment Self-Disclosure Rates and
Rationales of University Graduates with Learning Disabilities."
Journal of Learning Disabilities 41:291-99.
Madaus, J.W., T.E. Foley, J.M. Mcguire and L.M. Ruban. 2001.
"A Follow-Up Investigation of University Graduates with Learning
Disabilities." Career Development for Exceptional Individuals
24(2):133-46.
Madaus, J.W., J. Zhao and L. Ruban. 2008. "Employment
Satisfaction of University Graduates with Learning Disabilities."
Remedial and Special Education 29:323-32.
Mamiseishvili, K. and L.C. Koch. 2011. "First-to-Second-Year
Persistence of Students with Disabilities in Postsecondary Institutions
in the United States." Rehabilitation Counselling Bulletin
54(2):93-105.
Markel, K.S. and L.A. Barclay. 2009. "Addressing the
Underemployment of Persons with Disabilities: Recommendations for
Expanding Organizational Social Responsibility." Employee
Responsibilities and Rights Journal 21:305-18.
Martin Prosperity Institute. 2010. Releasing Constraints:
Projecting the Economic Impacts of Increased Accessibility in Ontario.
Toronto, ON: Author.
McCloskey, L., K. Figura, K. Narraway and B. Vukovic. 2011.
Transitions Longitudinal Study: 7th Annual & Final Report to the
Ministry of Training, Colleges and Universities. Toronto, ON: Ontario
Ministry of Training, Colleges and Universities.
McCloy, U. and L. DeClou. 2013. Disability in Ontario:
Postsecondary Education Participation Rates, Student Experience and
Labour Market Outcomes. Toronto: Higher Education Quality Council of
Ontario.
Michalko, R. and Titchkovsky, T. 2010. "There and Not There:
Presence and Absence of Disability in the Transition from Education to
Work." Pp. 109-24 in Challenging Transitions in Learning and Work,
edited by P. Sawchuk and A. Taylor. Boston: Sense Publishers.
Ministry of Community and Social Services [MCSS]. 2005.
Accessibility for Ontarians with Disabilities Act. Government of
Ontario. Ottawa: MCSS. Retrieved May 16, 2013
(http://www.elaws.gov.on.ca/html/statutes/english/elaws_statutes_05all_e.htm).
Ministry of Community and Social Services [MCSS]. 2012.
"Government of Ontario." Retrieved May 16, 2013
(http://www.mcss.gov.on.ca/en/mcss/programs/accessibility/index.aspx).
NCD [National Council on Disability]. 2007. "Employment for
Americans with Disabilities: Breaking Barriers to Careers and Full
Employment." Washington, DC: NCD.
Oliver, M. 1996. Understanding Disability : From Theory to
Practice. New York, NY: St. Martin's Press.
Public Service Commission of Canada. 2011. Recruitment of Persons
with Disabilities: A Literature Review. Ottawa, ON: Equity and Diversity
Directorate.
Quintini, G. 2011. "Right for the Job: Over-Qualified or
Under-Skilled." Working Paper No. 120, OECD Social, Employment and
Migration.
Robert, P.M. and S.L. Harlan. 2006. "Mechanisms of Disability
Discrimination in Large Bureaucratic Organizations: Ascriptive
Inequalities in the Workplace." Sociological Quarterly
47(4):599-630.
Robson, K.L., Anisef, P., Brown, R.S. and Parekh, G. 2014.
"The Intersectionality of Postsecondary Pathways: The Case of High
School Students with Special Education Needs." Canadian Review of
Sociology / Revue Canadienne de Sociologie 51:193-215.
Roeher Institute. 2004. Improving the Odds: Employment, Disability
and Public Programs in Canada. North York, ON: Roeher Institute.
Rylance, B.J. 1998. "Predictors of Post-High School Employment
for Youth Identified as Severely Emotionally Disturbed." The
Journal of Special Education 32:184-92.
Sanford, C., L. Newman, M. Wagner, R. Cameto, A.M. Knokey and D.
Shaver. 2011. The Post-High School Outcomes of Young Adults with
Disabilities up to 6 Years after High School. Key Findings from the
National Longitudinal Transition Study-2 (NLTS2) (NCSER 20113004). Menlo
Park, CA: SRI International.
Schur, L.A., D. Kruse, J. Blasi and P. Blanck. 2009. "Is
Disability Disabling in All Workplaces? 'Disability, Workplace
Disparities, and Corporate Culture.'" Industrial Relations
48:381410.
Service Canada. 2014. "Living with a Disability."
Government of Canada. Ottawa: Service Canada. Retrieved February 10,
2015 (http://www.servicecanada.gc.ca/eng/life events/disability.shtml).
Shier, M., J.R. Graham and M.E. Jones. 2009. "Barriers to
Employment as Experienced by Disabled People: A Qualitative Analysis in
Calgary and Regina, Canada." Disability & Society 24(l):63-75.
Spence, A.M. 1974. Market Signaling: Informational Transfer in
Hiring and Related Screening Processes. Cambridge: Harvard University
Press.
Sweet, R., P. Anisef, R. Brown, M., Adamuti-Trache and G., Parekh
2011. Special Needs Students and Transitions to Post-Secondary
Education. Final Report to Higher Education Quality Council of Ontario.
Toronto: HEQCO.
Sweet, R., P. Anisef, S.-D. Stone and M. Adamuti-Trache 2014.
"Toward Improving the Relationship between Educational Credentials
and Employment for Immigrants with Disabilities. Pp. 53-66 in Immigrant
Integration: Research Implications for Future Policy, edited by K.
Kilbride. Toronto: Canadian Scholars Press Inc.
Trainor, A. 2008. "Using Cultural and Social Capital to
Improve Postsecondary Outcomes and Expand Transition Models for Youth
with Disabilities." The Journal of Special Education 42(3): 148-62.
Tsagri, D. and B. Muirhead. 2012. Evaluating Postsecondary Supports
for Ontario Students with Learning Disabilities. Toronto: Higher
Education Quality Council of Ontario.
Wallace, J. and Fenwick, T. 2010. "Transitions in Working
Dis/Ability: Able-ing Environments and Disabling Policies." Pp.
309-24 in Challenging Transitions in Learning and Work edited by P.
Sawchuk and A. Taylor. Boston: Sense Publishers.
Walters, D. 2004a. "A Comparison of the Labour Market Outcomes
of Postsecondary Graduates of Various Levels and Fields over a
Four-Cohort Period." Canadian Journal of Sociology 29(l):l-27.
--. 20046. "The Relationship between Postsecondary Education
and Skill: Comparing Credentialism with Human Capital Theory." The
Canadian Journal of Higher Education 34(2):97-124.
--. 2006. "An Examination of the School-to-Work Transitions of
Male and Female College and University Graduates of Applied and Liberal
Arts Programs in Canada." Higher Education Policy 19:225-50.
Walters, D., J. White and P. Maxim. 2004. "Does Postsecondary
Education Benefit Aboriginal Canadians? An Examination of Earnings and
Employment Outcomes for Recent Aboriginal Graduates" Canadian
Public Policy 30(3):283-301.
Walters, D. and D. Zarifa. 2008. "Do Cooperative Programs
Improve the School-to- Work Transitions of Postsecondary Graduates in
Canada?" Journal of Vocational and Educational Training
60(4):377-400.
Wannell, T. and N. Charon. 1994. A Look at Employment-Equity Groups
among Recent Postsecondary Graduates: Visible Minorities, Aboriginal
Peoples and the Activity Limited. Cat. No. 11F0019MPE-69. Ottawa:
Statistics Canada.
Wells, T., G.D. Sandefur and D.P. Hogan. 2003. "What Happens
after the High School Years among Persons with Disabilities?"
Social Forces 82(2):803-32.
Wilton, R.D. 2006. "Working at the Margins: Disabled People
and the Growth of Precarious Employment." Pp. 129-50 in Critical
Disability Theory: Essays in Philosophy, Politics, Policy, and Law,
edited by D. Pothier and R. Devlin. Vancouver, BC: UBC Press.
Winn, S. and I. Hay. 2009. "Transition from School for Youths
with a Disability: Issues and Challenges." Disability & Society
24(1):103-15.
Zarifa, D. 2012a. "Choosing Fields in an Expansionary Era:
Comparing Canada and the United States." Research in Social
Stratification and Mobility 30(3):328-51.
--. 2012b. "Persistent Inequality or Liberation from Social
Origins? Determining Who Attends Graduate School in Canada's
Expanded Postsecondary System." Canadian Review of Sociology 49(2):
109-37.
Zarifa, D. and D. Walters. 2008. "Revisiting Canada's
Brain Drain: Evidence from the 2000 Cohort of Canadian University
Graduates." Canadian Public Policy 34(3):305-20.
Appendix
Table A1
Descriptive Statistics for Variables in National Graduates
Survey, 2005
Employment Earnings
status models models
Variable Proportion/mean Proportion/mean
Disability
Yes 0.04 0.03
No 0.96 0.97
Sex
Male 0.43 0.46
Female 0.57 0.54
Marital status
Not married 0.41 0.42
Married 0.59 0.58
Dependent children
Yes 0.23 0.22
No 0.77 0.78
Visible minority status
Visible minority 0.19 0.19
Nonminority 0.81 0.81
Immigrant status
Immigrant 0.14 0.14
Nonimmigrant 0.86 0.86
Parental education
No postsecondary 0.41 0.41
At least one parent with 0.59 0.59
postsecondary
Region
Atlantic provinces 0.06 0.06
Quebec 0.32 0.32
Ontario 0.36 0.35
Western provinces 0.26 0.27
Bursaries/grants
Yes 0.21 0.2
No 0.79 0.8
Government loans
Yes 0.44 0.44
No 0.56 0.56
Scholarships
Yes 0.3 0.31
No 0.7 0.69
Other loans
Yes 0.27 0.28
No 0.73 0.72
Field of study
Liberal arts 0.21 0.19
Business 0.24 0.26
Physical and agricultural 0.05 0.05
sciences
Engineering/Comp. Sci. 0.2 0.22
Health 0.15 0.14
Other 0.15 0.14
Level of schooling
University (undergrad.) 0.38 0.38
Trades 0.15 0.14
College 0.35 0.35
University (advanced) 0.12 0.13
Employment status
Full-time 0.85 --
Part-time 0.09 --
Unemployed 0.06 --
Age 30 29
Income -- 45,900
N 27,683 19,626
Table A2
Descriptive Statistics for Recent Graduates with and without
Disabilities, Earnings Models (2005 National Graduates Survey)
Respondents with Respondents
disability without disability
Variable Percentage/mean Percentage/mean
Income * 44,286 45,957
Sex
Male 48 46
Female 52 54
Marital status
Not married 45 42
Married 55 58
Dependent children **
No 30 22
Yes 70 78
Visible minority status ***
Nonminority 8 20
Visible minority 92 80
Immigrant status ***
Immigrant 5 15
Nonimmigrant 95 85
Parental education **
Parents did not attend 47 41
postsecondary
At least one parent with 53 59
postsecondary
Region ***
Ontario 36 35
Atlantic provinces 7 6
Quebec 25 32
Western provinces 32 27
Bursaries/grants *
Yes 22 20
No 78 80
Government loans
Yes 42 44
No 58 56
Scholarships
Yes 30 31
No 70 69
Other loans *
Yes 31 27
No 69 73
Field of study ***
Liberal arts 19 19
Business 21 26
Physical and agricultural 5 5
sciences
Engineering/Comp. Sci. 22 22
Health 12 14
Other 21 14
Level of schooling ***
University (undergrad.) 29 39
Trades 15 14
College 44 34
University (advanced) 12 13
Age *** 33 29
N 787 18,839
Notes: * p < .05; ** p < .01; *** p < .001. The asterisks
denote the results from chi-square tests used to compare
sample proportions and t-tests for differences in sample
means. Percentages are shown for all 51 variables except
for age and income. Age is measured in years and income
in dollars.
Table A3
Descriptive Statistics for Recent Graduates with and without
Disabilities, Employment Models (2005 National Graduates Survey)
Respondents
Respondents with without
disability disability
Variable Percentage/mean Percentage/mean
Employment Status ***
Full-Time (30 hours or more) 73 85
Part-Time (29 hours or less) 14 9
Not Employed 13 6
Sex
Male 42 42
Female 58 58
Marital Status **
Not Married 41 41
Married 59 59
Dependent Children
No 30 23
Yes 70 77
Visible Minority Status ***
Non-Minority 10 20
Visible Minority 90 80
Immigrant Status ***
Immigrant 7 14
Non-Immigrant 93 86
Parental Education ***
Parents Did Not Attend 47 41
Post-Secondary
At Least One Parent with 53 59
Post-Secondary
Region ***
Ontario 38 35
Atlantic Provinces 7 6
Quebec 23 33
Western Provinces 32 26
Bursaries/Grants **
Yes 24 21
No 76 79
Government Loans
Yes 43 44
No 57 56
Scholarships **
Yes 29 30
No 71 70
Other Loans
Yes 29 27
No 71 73
Field of Study ***
Liberal Arts 24 21
Business 21 24
Physical and Agricultural 5 5
Sciences
Engineering/Comp. Sci. 19 20
Health 11 15
Other 20 15
Level of Schooling ***
University (Undergrad.) 32 38
Trades 15 15
College 42 35
University (Advanced) 11 12
Age *** 34 29
N 1,311 26,372
Notes: * p < .05; p < .01; p < .001. The asterisks denote
the results from chi-square tests used to compare sample
proportions and t-tests for differences in sample means.
Percentages are shown for all variables except for age and
income. Age is measured in years and income in dollars.
(1.) The earnings analyses are restricted to only those who were
employed full- time and full-year.
(2.) Results from Models 1 through 3 are available upon request.
(3.) Indicator coding (0/1) was used for categorical variables, and
orthogonal polynomial contrasts were used for age and aged squared in
all regression models to account for the collinearity between these two
variables.
(4.) The models are estimated for respondents who reported
full-time and full- year employment at the time of survey--a standard
practice in the econometric literature predicting wages in Canada.
(5.) The predicted probabilities are calculated separately for
graduates with and without self-disclosed disabilities using the
following equation: [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII].
For respondents who reported having a disability, the estimate for this
variable is multiplied by 1. For respondents who did not report having a
disability, the estimate for this variable is multiplied by 0. For both
groups, the control variables are held constant at typical values, where
means are used for quantitative variables (i.e., age) and proportions
are used for categorical variables (i.e., gender). Thus, the regression
estimate in the above equation for age would be multiplied by the mean
age for the entire sample, while the dummy variable for gender would be
multiplied by the proportion of males in the entire sample.
(6.) Additional analyses (not shown) indicated that the three-way
interaction among the field of study, level of schooling, and
disability, however, is not statistically significant.
(7.) The parameter estimates for the other variables in the model
remained relatively unchanged when the interaction was included in the
model.
(8.) The coefficient of--1.410 in Model 4, for example, corresponds
to graduates from a health-related field who reported having a
disability.
(9.) Additional analyses (not shown) revealed that the likelihood
ratio chi- square test for the three-way interaction among disability,
field of study, and level of schooling, however, was not statistically
significant.
(10) The 95 percent confidence intervals reveal that all contrasts
are statistically significant.
Table 1
Ordinary Least Squares Regressions of the Log of Earnings
for the 2005 Cohort of Postsecondary Graduates in Canada
Model 4
b p
Disability
Yes -0.191 ***
No -- --
Field of study ***
Liberal arts -- --
Business 0.192 ***
Physical and applied sciences 0.101 ***
Engineering/Comp. Sci. 0.249 ***
Health 0.263 ***
Other 0.093 ***
Level of schooling ***
University (undergrad.) -- --
Trades -0.365 ***
College -0.207 ***
University (advanced) 0.211
Disability * Field of study
Disability * Liberal arts -- --
Disability * Business 0.025
Disability * Physical and applied sciences 0.130
Disability * Engineering/Comp. Sci. 0.168 ***
Disability * Health 0.095
Disability * Other 0.167 ***
Disability * Level of schooling
Disability * University (undergrad.)
Disability * Trades
Disability * College
Disability * University (advanced)
N 19,626
R-Square 0.3228
Adj. R-Square 0.3218
Model 5
b p
Disability
Yes -0.047
No -- --
Field of study ***
Liberal arts -- --
Business 0.193 ***
Physical and applied sciences 0.105 ***
Engineering/Comp. Sci. 0.255 ***
Health 0.266 ***
Other 0.100 ***
Level of schooling ***
University (undergrad.) -- --
Trades -0.365
College -0.205 ***
University (advanced) 0.212
Disability * Field of study
Disability * Liberal arts
Disability * Business
Disability * Physical and applied sciences
Disability * Engineering/Comp. Sci.
Disability * Health
Disability * Other
Disability * Level of schooling
Disability * University (undergrad.) -- --
Disability * Trades -0.040
Disability * College -0.082 *
Disability * University (advanced) -0.054
N 19,626
R-Square 0.3222
Adj. R-Square 0.3212
Notes: * p < .05; ** p < .01; *** p < .001. Asterisks
aligned with variable names indicate the results of multiple
parameter likelihood ratio chi/square tests. All models
include controls for sex, age, age squared, marital status,
visible minority status, immigrant status, parental
education, region, bursaries/grants, government student
loans, scholarships, and other loans. Results from Models 1
through 3 are available upon request.
Standard errors are available upon request.
Table 2
Multinomial Logistic Regressions of Employment Status
for the 2005 Cohort of Postsecondary Graduates in
Canada (Employed Full-Time is the Reference Category)
Model 4 Model 4
Part-time Unemployed
b p b p
Disabled
Yes 0.994 *** 0.989 ***
No -- -- -- --
Field of study ***
Liberal arts -- -- -- -
Business -0.964 *** -0.586 ***
Physical and -0.776 *** -0.234
agricultural sciences
Engineering/Comp. Sci. -1.201 *** -0.162
Health -0.014 -0.805 ***
Other -0.120 -0.237 **
Level of schooling *** ***
University (undergrad.) -- -- -- --
Trades 0.593 *** 0.317 ***
College 0.457 *** 0.187 **
University (advanced) -0.096 -0.204 *
Disability * Field of study *** ***
Disability * Liberal arts -- -- -- --
Disability * Business -0.172 0.352
Disability * Physical and -0.001 -1.014
agricultural sciences
Disability * -1.233 ** -0.004
Engineering/Comp. Sci.
Disability * Health -1.410 *** -1.220 *
Disability * Other -0.676 ** -0.577
Disability * Level of
education
Disability * University --
(undergrad)
Disability * Trades
Disability * College
Disability * University 0.239
(advanced)
LR (58)
= 1,783.54
p > [chi.sup.2]
=0.000
Pseudo [R.sup.2]
= 0.0607
N = 27,683
Model 5 Model 5
Part-time Unemployed
b p b p
Disabled
Yes -0.758 *** -0.190 ***
No -- -- -- --
Field of study *** ***
Liberal arts -- - -- --
Business -0.770 *** -0.288 *
Physical and -1.259 *** -0.160 *
agricultural sciences
Engineering/Comp. Sci. -0.076 -0.853 ***
Health -0.164 * -0.281 ***
Other 1.059 *** 0.903 ***
Level of schooling *** ***
University (undergrad.) -- -- -- --
Trades 0.622 *** 0.348 ***
College 0.525 *** 0.181 **
University (advanced) -0.125 -0.197 *
Disability * Field of study
Disability * Liberal arts
Disability * Business
Disability * Physical and
agricultural sciences
Disability *
Engineering/Comp. Sci.
Disability * Health
Disability * Other
Disability * Level of *** ***
education
Disability * University -- -- --
(undergrad)
Disability * Trades -0.475 -0.421
Disability * College -1.319 *** 0.022
Disability * University -0.095
(advanced)
LR (54)
= 1,786.65
p > [chi.sup.2]
=0.000
Pseudo [R.sup.2]
= 0.0608
N = 27,683
Notes: * p < .05; ** p < .01; *** p < .001. Asterisks
aligned with variable names indicate the results of multiple
parameter likelihood ratio chi/square tests. All models
include controls for sex, age, age squared, marital status,
visible minority status, immigrant status, parental
education, region, bursaries/grants, government student
loans, scholarships, and other loans. Results from Models 1
through 3 are available upon request. Standard errors are
available upon request.
