Content uploaded by Robert Evert Cimera
Author content
All content in this area was uploaded by Robert Evert Cimera on Mar 14, 2014
Content may be subject to copyright.
http://aut.sagepub.com/
Autism
http://aut.sagepub.com/content/16/1/87
The online version of this article can be found at:
DOI: 10.1177/1362361311408129
2012 16: 87 originally published online 24 May 2011Autism
Robert Evert Cimera, Paul Wehman, Michael West and Sloane Burgess
spectrum disorder?
Do sheltered workshops enhance employment outcomes for adults with autism
Published by:
http://www.sagepublications.com
On behalf of:
The National Autistic Society
can be found at:AutismAdditional services and information for
http://aut.sagepub.com/cgi/alertsEmail Alerts:
http://aut.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://aut.sagepub.com/content/16/1/87.refs.htmlCitations:
What is This?
- May 24, 2011 OnlineFirst Version of Record
- Feb 21, 2012Version of Record >>
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
Autism
16(1) 87 –94
© The Author(s) 2012
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1362361311408129
aut.sagepub.com
Do sheltered workshops
enhance employment
outcomes for adults with
autism spectrum disorder?
Robert Evert Cimera
Kent State University, Ohio, USA
Paul Wehman
Virginia Commonwealth University, Richmond, Virginia, USA
Michael West
Virginia Commonwealth University, Richmond, Virginia, USA
Sloane Burgess
Kent State University, Ohio, USA
Abstract
This study investigated whether sheltered workshops help prepare individuals with autism spectrum
disorder (ASD) for competitive employment within the community. Two groups of individuals
were compared: (a) 215 supported employees who were in sheltered workshops prior to entering
supported employment and (b) 215 supported employees who were not in sheltered workshops.
Individuals from both groups were matched based on their primary diagnosis, secondary diagnosis
(if present), and gender. Results showed that there were no differences in rates of employment
between these two groups. However, individuals who participated in sheltered workshops
earned significantly less (US$129.36 versus US$191.42 per week), and cost signi ficantly more
to serve (US$6,065.08 versus US$2,440.60), than their non-sheltered workshop peers. Results
presented here suggest that individuals with ASD achieve better vocational outcomes if they do
not participate in sheltered workshops prior to enrolling in supported employment.
Keywords
autism spectrum disorder, sheltered workshops, supported employment
Corresponding author:
Robert Evert Cimera, 405 White Hall, Kent, Ohio 44242, USA.
Email: rcimera@kent.edu
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
88 Autism 16(1)
In the United States, approximately 7000 facility-based programs (e.g. sheltered workshops) serve
542,127 adults with mental, physical, and emotional disabilities (Braddock et al., 2008). These
programs offer skill training, special certificate subminimum wage work, prevocational services,
group work placements, and recreation and leisure activities. Each year, many young people with
autism, intellectual disabilities, and psychiatric conditions are referred to sheltered workshops as
the first step in their vocational rehabilitation process.
The underlying premise of sheltered workshops is that jobseekers with disabilities need cer-
tain skills prior to becoming competitively employed within the community. Further, sheltered
workshops and other facility-based programs are thought to teach jobseekers these skills and
‘prepare’ them for working in the community. Although much has been written about sheltered
workshops over the years, most of this discussion has been either on the merits of the philoso-
phies guiding sheltered workshops or their costs compared with those generated by supported
employment (cf. Bellamy et al., 1986; Mallas, 1976; Parent et al., 1989; Rosen et al., 1993;
Schuster, 1990; West et al., 1998; Whitehead, 1979, 1986). To date, very little attention has
focused upon whether sheltered workshops actually provide beneficial skills to jobseekers with
disabilities. In other words, it is unclear as to whether sheltered workshops are value-added
programs.
In economics, ‘value-added’ is the difference between the sale price of an item and the cost of
all of the materials and services utilized to create it (Levin and McEwan, 2000). Value-added,
therefore, is a term for the measurement of the enhancement that the company gives to its raw
materials when creating its products. Such a concept is often used when people are contemplating
selling their homes. For instance, sellers may evaluate the needs of updating their home (e.g.
remodeling a kitchen or adding another bathroom) in the hopes of improving the property’s mar-
ketability and sale price. In these evaluations, sellers may try to determine whether the costs of
completing the updates will be less than the eventual sale price of the home (i.e. the updates will
produce more benefits than costs or will be value-added).
The concept of value-added may appear foreign when put in the context of vocational services
for adults with disabilities; however, it lies at the heart of what all human service programs strive
to obtain. That is to say, we hope that our students, clients, or customers leave our programs better
off than when they first enter them. The specific outcomes depend on the program’s unique mis-
sion, but regardless of the program analyzed, it is hope that what is added to participants will
benefit them well into their future. Within this discussion, the question arises: ‘Do individuals who
participate in sheltered workshops benefit from the experience?’
To investigate this issue, a recent study (Cimera, in press) examined two groups of supported
employees – 4904 individuals with cognitive disabilities who were in sheltered workshops at the
time they enrolled in supported employment and 4904 individuals with cognitive disabilities who
were not in sheltered workshops prior to enrolling in supported employment. Individuals in both
cohorts were matched by their disability, the presence of a secondary disability, and their gender.
Cimera found that although both groups were equally likely to be employed (59.6% versus 60.4%,
respectively), individuals from sheltered workshops worked signi ficantly fewer hours, earned sub-
stantially less wages, and cost 74.8% more to serve than individuals who were not transitioning
from sheltered workshops. The author’s conclusion was that, for adults with cognitive disabilities,
sheltered workshops were ‘negative value-added’. That is, participating in sheltered workshops
diminished the future outcomes achieved once individuals became competitively employed, per-
haps because the skills and behaviors individuals learned in sheltered workshops had to be
‘unlearned’ in order for the workers to be successful in the community. It may be, however, that
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
Cimera et al. 89
sheltered workshops are more beneficial for certain populations than others, such as adults with
autism spectrum disorders (ASD).
As the numbers of individuals diagnosed with ASD continues to increase, so too has the interest
in the support services that they require to transition to adult life and succeed in the community
(cf. Chappel and Somers, 2010; Wehman, 2011). Therefore, it is critical to gain more evidence on
the vocational outcomes and benefits associated with participation in sheltered workshops since
many young people with ASD will present highly challenging communication and social behaviors
that could potentially route them into such programs. In other words, we need to know if sheltered
work is a ‘value-added’ service for these individuals.
The purpose of the present study was to extend the research conducted by Cimera (in press),
which involved only individuals with cognitive dis abilities, by comparing the outcomes of 215
adults with ASD who participated in sheltered workshops prior to applying for vocational rehabili-
tation services with 215 adults with ASD who did not participate in sheltered workshops. Individuals
in both groups were matched by their diagnosis of ASD, any other diagnoses they may have had,
and their gender. Outcomes investigated included: (a) rates of employment, (b) wages earned, (c)
hours worked, and (d) the cost of services received.
The tested hypothesis was that individuals who participated in sheltered workshops prior to
enrolling in supported employment programs would achieve significantly better vocational out-
comes than individuals who had not received pre-supported employment services. In other words,
this study attempted to discern whether individuals with ASD benefited from being in sheltered
workshops (i.e. are sheltered workshops value-added) or whether sheltered workshops actually
impair the vocational outcomes achieved by these persons as was found by Cimera (in press) for
indi viduals with cognitive impairments.
Methods
Source of data
The source of data was the Rehabilitation Services Administration’s (RSA) 911-database. This
national database contains detailed records on all persons who apply for services through voca-
tional rehabilitation. Data are entered by certified rehabilitation counselors and then crosschecked
by two computer programs for potential errors or duplicity (RSA, 2004).
Selection of participants
From 2002 to 2006, vocational rehabilitation counselors closed the cases (i.e. stopped receiving
services) of 3,182,126 individuals. Of these people, 14,378 had diagnoses of ‘autism’ (i.e. ASD).
Approximately 1.5% of these individuals (n = 215) were employed in sheltered workshops at the
time of their application for vocational rehabilitative services.
From the 14,163 individuals with ASD who were not employed in sheltered workshops when they
applied for services, a sample of 215 persons was randomly selected using SPSS’s random select
feature. Individuals selected were matched in pairs to participants from the sheltered workshop cohort
based on their disability (i.e. ‘autism’), their exact secondary dis ability (if present), and their gender.
These variables were identified as selection criteria because previous research has identified these
variables as significantly influencing employment outcomes and costs of services received (Cimera
et al., unpublished data). The demographics of these two cohorts can be found in Table 1.
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
90 Autism 16(1)
Variables
Disabilities. When an individual applied for services through vocational rehabilitation, they were
assessed by certified rehabilitation counselors. Based on these assessments, their disabling condi-
tions were then classified into 19 different ‘impairment codes’ (e.g. intellectual, physical, sensory
impairments) and 34 ‘cause codes’ (e.g. autism, cerebral palsy, traumatic brain injury). This cod-
ing was completed for both their ‘primary’ and ‘secondary disability’ (if present). Secondary dis-
abilities were noted in 74.8% of the participants, ‘mental retardation’ (i.e., cognitive impairments)
accounting for 46.1% of these. About a third of the sample (33.6%) had secondary disabilities
involving mental health (e.g., depression, anxiety dis orders, mental illness not otherwise
specified).
Rate of employment. Rate of employment was calculated by dividing the number of individuals
who had their cases officially closed due to obtaining ‘an employment outcome’ (i.e. competitive
employment within an integrated setting earning at or above minimum wage) by the total number
of jobseekers in that cohort.
Wages earned. If participants had their cases closed because they obtained an employment
outcome, vocational rehabilitation counselors documented the participant’s average wages earned
Table 1. Demographics of adults with autism from the sheltered workshop and non-sheltered workshop
cohorts
Sheltered employees Non-sheltered employees
Sample 215 215
Average (SD) age 31.12 (9.07) years 37.75 (8.90) years
Percent female 20% 20%
Percent male 80% 80%
Percent with secondary conditions 74.8% 74.8%
Ethnicity
White 78.5% 83.3%
African American 16.4% 12.1%
Native American 1.9% 0.9%
Asian 4.2% 3.7%
Pacific islander 0.9% 0.5%
Hispanic 5.6% 1.9%
Source of referral to vocational rehabilitation
Educational institution (secondary) 8.4% 39.7%
Educational institution (post-secondary) 1.4% 1.4%
Medical personnel 5.1% 5.1%
Welfare agency 2.8% 0.9%
Community rehabilitation program 46.0% 9.3%
Social security administration 0.0% 1.4%
One-stop employment/training center 0.9% 0.9%
Self-referral 0.6% 17.3%
Other sources 29.3% 23.8%
Note: Participants were able to identify themselves as members of multiple ethnic groups.
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
Cimera et al. 91
in a week. Wages earned were gross wages, prior to the deduction of any taxes or other
withholdings.
Hours worked. As with wages earned, vocational rehabilitation counselors also documented the
average number of hours that successfully employed participants worked in a week.
Cost of services. In addition to wages earned and hours worked, vocational rehabilitation counse-
lors indicated in the 911-database the services that each participant received. They also docu-
mented the cost of services that were contracted to outside providers, such as job development and
training. In other words, the costs of services presented here represent the services funded by
vocational rehabilitation, but furnished by someone other than the participant’s vocational rehabili-
tation counselor.
Research questions
This study investigated four research questions. The first sought to determine whether individuals
who received services in sheltered workshops were more likely to be employed than individuals
who had the same demographic backgrounds, but did not receive services in sheltered workshops.
This study also investigated whether previous participation in sheltered workshops increased the
number of hours worked and wages earned in the community. Finally, this study explored whether
there was a significant difference in the cost of services received by these two groups. Differences
in the rates of employment were analyzed using a Pearson chi-square test. A two-tailed t-test for
paired samples was utilized for all other analyses.
Results
Question 1: Do former sheltered workers have a higher rate of employment than individuals
who were not from sheltered workshops?
As shown in Table 2, 98 (45.6%) of the 215 former sheltered workers were employed when their
cases were officially closed by their vocational re habilitation counselors, compared with 85 (39.5%)
of the 215 non- sheltered workers (p = .214).
Table 2. Employment outcomes achieved by adults with autism from the sheltered and non-sheltered
workshop cohorts
Sheltered employees Non-sheltered employees
Sample size 215 215
Employment rate 45.6% 39.5%
Hours worked per week 23.49 (11.40) 24.97 (12.33)
Wages earned per weeka US$129.36 ($89.66) US$191.42 ($118.83)
Cost of services (entire cohort)b US$6,065.08 ($9,879.33) US$2,440.60 ($4,585.63)
Cost of services (employed)c US$8,364.39 ($11,420.70) US$4,212.24 ($5,088.11)
Note: Standard deviations presented in parentheses.
at = 3.60; p = .001.
bt = 4.93; p = .001.
ct = 2.88; p = .001.
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
92 Autism 16(1)
Question 2: Do former sheltered employees work more hours than individuals who were not
from sheltered workshops?
Former sheltered employees with ASD who became competitively employed in the community
worked an average of 23.5 hours (SD = 11.4) per week; their matched peers from non-sheltered
workshops worked an average of 25.0 (SD = 12.3). As with rates of employment, the differences
between these groups were not statistically significant.
Question 3: Do former sheltered employees earn more than individuals who were not from
sheltered workshops?
Former sheltered employees who became competitively employed in their community earned
an average of US$129.36 (SD = US$89.66; median = US$106.50) per week, 32.4% less than
the US$191.42 (SD = US$118.83; median = US$152.50) the wages earned by the non-sheltered
workshop group (p = .001).
Question 4: Do former sheltered employees cost less to serve than non-former sheltered
employees?
Formerly sheltered workers received services costing vocational rehabilitation an average of
$6065.08 (SD $9879.33) per person. The non-sheltered workshop cohort, however, received ser-
vices costing 59.8% less (M = $2440.60; SD $4585.63) (p <.001). This difference persisted when
comparing only those in each cohort who achieved employment. Former sheltered employees who
became employed as a result of their participation in vocational rehabilitation programs received
services costing an average of $8364.39 (SD $11,420.70) compared with an average of $4212.24
(SD $5088.11) for the employed non-sheltered employees (p = .001).
Discussion
In order for human service programs to be beneficial, they must give participants skills, aptitudes,
or dispositions that will help them maximize functioning or increase their community participa-
tion. In other words, participants must leave programs better off than when they first entered them.
In economics, this term is often referred to as ‘value-added’. This study examined whether shel-
tered workshops are value-added for indi viduals with ASD.
The tested hypothesis was that individuals who received services in sheltered workshops would
achieve better vocational outcomes than individuals with similar demographics who had not
received such pre- vocational services. Specifically, it was assumed that if sheltered workshops
were beneficial (i.e. value-added programs), individuals from sheltered workshops would be
employed at higher rates, work more hours, earn more wages, and cost less to serve in the com-
munity than individuals who had not received services in sheltered workshops.
The findings here mirror those found by Cimera (in press) for vocational rehabilitation clients
with cognitive disabilities. Specifically, individuals with ASD who transitioned to supported
employment from sheltered workshops were employed in the community at comparable rates and
worked nearly identical hours per week once employed in the community as individuals who had
not transitioned to vocational rehabilitation from sheltered workshops.
When wages earned and cost of services received were examined, individuals from the non-
sheltered workshop group fared much better than their peers who were from sheltered workshops.
More precisely, individuals who did not receive pre-vocational services in sheltered workshops
earned significantly more per week than their peers who did. Further, individuals with ASD who
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
Cimera et al. 93
were not in sheltered workshops also generated only 40% of the costs of those who were in shel-
tered workshops.
A limitation of this study is that measures of severity in the RSA dataset are very limited. There
may be other characteristics of the sheltered and non-sheltered cohorts that contributed the dispari-
ties in employment outcomes from vocational rehabilitation services. For example, those with
sheltered work histories could have had more significant barriers to employment (i.e. more behav-
ioral issues than the non-sheltered group). Although individuals in both groups were matched by
primary and secondary conditions, it may be that individuals in sheltered workshops had more
severe impairments than individuals who were not in the sheltered workshop cohort.
Despite this limitation, there are important implications related to these findings. Debate regard-
ing the value of sheltered work for individuals with disabilities in comparison with community-
integrated employment has con tinued since the early 1970s. One of the often stated benefits of
sheltered workshops is that they serve as a stepping-stone to community- integrated employment by
providing essential employment training and work preparation (Inge et al., 2009). That assertion is
called into question by: (a) research findings indicating that only a small percentage of work shop
employees make the transition to integrated employment, even after many years of training and
preparation (Blanck et al., 2003), and (b) findings from this study and from Cimera (in press) indi-
cating that parti cipation in sheltered workshops did not significantly improve chances for eventual
competitive employment within the community.
Additionally, the findings presented here indicate that those who did not receive services in
facility-based programs earned significantly more and had significantly lower service costs than
those who did. Stated another way, this study found that sheltered workshops appeared not to be
‘value-added’. In fact, data presented here suggest that they generated negative value for their par-
ticipants in relation to vocational outcomes in the community (i.e. fewer dollars earned and higher
costs of services to taxpayers).
There are a number of possible explanations for these findings. For instance, motivations of the
non-sheltered clients and their family members to obtain and keep higher paying jobs may be greater
than for the sheltered employees, or that the sheltered employees and their families had more con-
cerns over loss of disability benefits, which were not included in cost calculations in this study. It
is also possible that the sheltered employees were more difficult to place and train as a result of
their workshop experi ences, such as due to learned helplessness or developing work behaviors that
might be acceptable in the sheltered setting but unacceptable in competitive positions.
For adults with ASD and other developmental disabilities, segregated facility-based programs,
such as sheltered workshops, continue to be the primary model of service delivery. Across the
United States, over 88% of participants with intellectual disabilities are being served in segregated
services (Butterworth et al., 2010). Additionally, data collected from the National Longitudinal
Transition Study (NLTS-2) show that secondary students with ASD are more likely to have a post-
school goal of sheltered employment than any other group of students (Cameto et al., 2003). In
order to make informed decisions, individuals with ASD, their families, teachers, and transition
coordinators need to be aware that participation in sheltered workshops may be beneficial in tran-
sitioning jobseekers to competitive employment in the community.
Certainly, other arguments have been made regarding the value of sheltered employment, such as
greater safety, maintaining longstanding social relationships, and lack of available competitive jobs
or transportation in the community (Migliori et al., 2008). This study cannot address those argu-
ments. It can only address the value of sheltered work experiences in pro moting future competitive
employment and improved vocational outcomes, such as higher wages earned. Certainly, the data
presented here to not support an economic argument for the value of sheltered work experiences.
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
94 Autism 16(1)
References
Bellamy GT, Rhode LE, Bourbeau PE and Mank DM (1986) Mental retardation service in sheltered workshops
and day activity programs: Consumer benefits and policy alternatives. In: Rusch FR (ed.) Competitive
Employment Issues and Strategies. Baltimore, MD: Brookes, 257–72.
Blanck PD, Schartz HA and Schartz KM (2003) Labor force participation and income of individuals with
disabilities in sheltered and competitive employment: cross-sectional and longitudinal analyses of seven
states during the 1980s and 1990s. William and Mary Law Review 44: 1029–108.
Braddock D, Hemp R and Rizzolo MC (2008) The State of the States in Developmental Disabilities: 2008.
Washington, DC: American Association on Intellectual and Developmental Disabilities.
Butterworth J, Smith FA, Hall AC, Migliore A and Winsor J (2009) State Data: The National Report on
Employment Services and Outcomes. Boston, MA: Institute for Community Inclusion, University of
Massachusetts.
Cameto R, Marder C, Wagner M and Cardoso D (2003) Youth Employment, NLTS2 Data Brief, National
Center on Secondary Education and Transition. Minneapolis, MN: University of Minnesota.
Chappel SL and Somers BC (2010) Employing persons with autism spectrum disorders: A collaborative
effort. Journal of Vocational Rehabilitation 32: 117–24.
Cimera R (in press) Does being in sheltered workshops improve the employment outcomes of supported
employees with intellectual disabilities? Journal of Vocational Rehabilitation.
Inge KJ, Wehman P, Revell G, Erickson D, Butterworth J and Gilmore D (2009) Survey results from a
national survey of community rehabilitation providers holding special wage certificates. Journal of
Vocational Rehabilitation 30: 67–85.
Levin HM and McEwan PJ (2000) Cost-effectiveness Analysis: Methods and Applications. Thousand Oaks,
CA: Sage.
Mallas AA (1976) Current workshops strengths and weaknesses. Education and Training in Mental
Retardation 11: 334–48.
Migliori A, Grossi T, Mank D and Rogan P (2008) Why do adults with intellectual disabilities work in shel-
tered workshops? Journal of Vocational Rehabilitation 28: 29–40.
Parent W, Hill M and Wehman P (1989). From sheltered to supported employment. Journal of Rehabilitation
55(4): 51–7.
Rehabilitation Services Administration (2004) Reporting Manual for the Case Service Report (RSA-911).
(Report No. RSA-PD-04–04). Washington DC: RSA.
Rosen M, Bussone A, Dakunchak P and Cramp J (1993) Sheltered employment and the second generation
workshop. Journal of Rehabilitation 59(1): 30–4.
Schuster JW (1990) Sheltered workshops: Financial and philosophical liabilities. Mental Retardation 28(4):
233–9.
Wehman P (2011) Essentials of Transition Planning. Baltimore, MD: Paul Brookes Publishing Co.
West M, Revell G and Wehman P (1998) Conversion from segregated services to supported employment:
A continuing challenge to the VR service system. Education and Training in Mental Retardation and
Developmental Disabilities 33: 239–47.
Whitehead CW (1979) Sheltered workshops in the decade ahead: Better work and wages, or welfare. Journal
of Rehabilitation 45(2): 77–80, 90.
Whitehead CW (1986) The sheltered workshop dilemma: Reform or replacement. Remedial and Special
Education 7: 18–24.
at OhioLink on March 14, 2014aut.sagepub.comDownloaded from
