The purpose of this study was to investigate relationships among customer demographics, case service variables, and successful closure in supported or competitive employment for Vocational Rehabilitation (VR) customers with autism aged 18-25. Utilizing the RSA-911 data file for fiscal year 2004, logistic regression was used to measure relationships among customer demographic and case service variables related to successful closure. T-tests were used to examine the means for earnings per week at closure, hours per week at closure, and case service costs between all participants in supported and competitive employment, and within African American and White participant groups closed in supported and competitive employment. Study results found that African American and White participants closed in supported employment were older than African American and White participants closed in competitive employment. African American participants were more likely than White participants to be closed in supported employment. African American participants were more likely to have a secondary disability of mental retardation or specific learning disability, and receive public assistance than White participants closed in supported or competitive employment. Mental illness was more commonly reported among White participants than among African American participants closed in supported or competitive employment. Results of the logistic regression indicated that a larger percentage of White participants closed in competitive employment received college training and job readiness training than among White participants closed in supported employment. African American participants closed in supported employment were more likely to receive on-the-job supports than African American participants closed in competitive employment. Job placement assistance was provided more frequently to African American and White participants closed in competitive employment than for participants closed in supported employment. Individuals closed in competitive employment, regardless of race, had a higher weekly wage and worked approximately five hours more per week than participants closed in supported employment. No statistically significant findings were noted between African American and White participants on wages, hours worked per week, or case service costs, regardless of successful closure status. Implications for practice and suggestions for future research are discussed.