Predicting employment outcomes of rehabilitation clients with orthopedic disabilities | Knowledge Translation for Employment Research Center

Authors:	Chapin, M. H. & Kewman, D. G.
Year Published	2006
Publication	Disability and Rehabilitation
Volume	28
Number	5
Pages	257-270
Publisher	Taylor & Francis
Background	Work is fundamental to persons with disabilities having some sort of independence and quality of life. Vocational rehabilitation services measure their effectiveness largely on employment outcomes of their clients. Employment outcomes of clients and thus what services should be offered to them can be predicted to some degree using data mining techniques such as the chi-squared automatic interaction detector (CHAID). Using CHAID to study data sets can yield a wealth of information about that data set.
Purpose	The purpose of this study was to use CHAID to mine the Rehabilitation Service Administration (RSA)-911 data set to determine what influences employment rates of rehabilitation clients with orthopedic disabilities.
Setting	This study included individuals with orthopedic disabilities served by multiple vocational rehabilitation agencies in various settings.
Sample	The study participants (N=74,861) had their data mined from the RSA-911 database. All of the cases were closed by public rehabilitation agencies in fiscal year 2001. The sample was 55% male. European Americans numbered 73.2% of the sample. The mean age of the participants was 41.4 years. Forty-three percent of the sample completed high school and 28% had some college education. Eighty-eight percent had a severe disability.
Data Collection	Variables that predicted employment outcomes were grouped into personal variables: gender, race, disability severity, age, education, and government benefits and rehabilitation variables. Rehabilitation variables consisted of: receiving an initial assessment, medical restoration, post-secondary education, business and vocational training, adjustment training, on the job training, miscellaneous training, substantial counseling and guidance, job finding services, job placement, transportation, maintenance, personal assistive services, rehab engineering, assistive technology, and other. Data was analyzed using CHAID.
Control	The employment outcome was the comparison condition.
Findings	Job placement services significantly enhanced employment outcomes but were very much underutilized (only 25% of the study sample used them). Clients that had work disincentives such as social security disability insurance had lower employment rates than clients without such disincentives.
Conclusions	The CHAID analysis proved to be an effective approach for examining the data set and interpreting relationships between variables. Vocational rehabilitation counselors should be aware of the predictors to employment that this study suggests.
URL	http://www.ncbi.nlm.nih.gov/pubmed/16492620
Disabilities	Orthopedic impairments
Populations	White / Caucasian
Outcomes	Employment acquisition \| Return to work
NIDILRR Funded	Yes
Research Design	Observational
Peer Reviewed	Yes

Year Published

2006

Publication

Disability and Rehabilitation

Volume

28

Number

5

Pages

257-270

Publisher

Taylor & Francis

Background

Work is fundamental to persons with disabilities having some sort of independence and quality of life. Vocational rehabilitation services measure their effectiveness largely on employment outcomes of their clients. Employment outcomes of clients and thus what services should be offered to them can be predicted to some degree using data mining techniques such as the chi-squared automatic interaction detector (CHAID). Using CHAID to study data sets can yield a wealth of information about that data set.

Purpose

The purpose of this study was to use CHAID to mine the Rehabilitation Service Administration (RSA)-911 data set to determine what influences employment rates of rehabilitation clients with orthopedic disabilities.

Setting

This study included individuals with orthopedic disabilities served by multiple vocational rehabilitation agencies in various settings.

Sample

The study participants (N=74,861) had their data mined from the RSA-911 database. All of the cases were closed by public rehabilitation agencies in fiscal year 2001. The sample was 55% male. European Americans numbered 73.2% of the sample. The mean age of the participants was 41.4 years. Forty-three percent of the sample completed high school and 28% had some college education. Eighty-eight percent had a severe disability.

Data Collection

Variables that predicted employment outcomes were grouped into personal variables: gender, race, disability severity, age, education, and government benefits and rehabilitation variables. Rehabilitation variables consisted of: receiving an initial assessment, medical restoration, post-secondary education, business and vocational training, adjustment training, on the job training, miscellaneous training, substantial counseling and guidance, job finding services, job placement, transportation, maintenance, personal assistive services, rehab engineering, assistive technology, and other. Data was analyzed using CHAID.

Control

The employment outcome was the comparison condition.

Findings

Job placement services significantly enhanced employment outcomes but were very much underutilized (only 25% of the study sample used them). Clients that had work disincentives such as social security disability insurance had lower employment rates than clients without such disincentives.

Conclusions

The CHAID analysis proved to be an effective approach for examining the data set and interpreting relationships between variables. Vocational rehabilitation counselors should be aware of the predictors to employment that this study suggests.

URL

http://www.ncbi.nlm.nih.gov/pubmed/16492620

Disabilities

Orthopedic impairments

Populations

White / Caucasian

Outcomes

Employment acquisition | Return to work

NIDILRR Funded

Yes

Research Design

Observational

Peer Reviewed

Yes