The purpose of the study was to investigate the construct validity of the Jewish Employment and Vocational Service (JEVS) work samples. The specific objectives were: (1) to investigate the factor structure of the JEVS work samples; (2) to determine which work samples were good measures of general ability; and (3) to identify the work samples that possess trait specificity. One hundred twenty educationally handicapped students were administered 15 JEVS simulated work samples. Intercorrelations among raw scores of the work samples were factor analyzed utilizing principal axis factoring. Sixty-seven percent of the correlations were significant, indicating that work samples which emulate similar tasks share common abilities. A two factor rotated solution was found to have a simple structure and the most meaningful interpretation. Factor one was verbal comprehension and factor two was called perceptual organization. Six work samples showed 50 percent or higher of the total variance and the specific variance exceeded error variance. Two work samples had a moderate percentage of trait specificity. The other seven work samples did not meet the requirements for trait specificity. Results imply that caution should be used when using work samples for analyzing specific trait abilities. In addition sub-batteries of work samples might be used to assess the readiness skills or general ability and orientation toward vocational education instead of specific traits. (JAZ)