Drug users in or out of treatment face many barriers to employment when entering the labor market. Such barriers may consist of job-specific “hard skills,” such as the ability to work on a personal computer. In addition to deficiencies in hard skills, unemployed drug users may lack general “soft skills,” such as punctuality and interpersonal communication skills. To address this issue we design two studies to examine the existence and extent of specific barriers to employment in a sample of participants enrolled in a workplace therapeutic intervention for substance abuse. In Study 1, we characterized and examined predictors of participants' scores on a staff-related interpersonal skills scale. As predictors, we used three demographic variables: education, gender, and age. We collected data on demographic variables from Addiction Severity Index-Lite – Clinical Factors, Clinical/Training Version. We conducted a job termination to determine the length of time since the last job held and the type of termination. We also created a Work Behavior Inventory (WBI) composed of five subscales (social skills, cooperation, work habits, quality of work, personal presentation). Participants' age, education, gender, time since last job, and type of layoff were used as predictors of WBI scores. We calculated intercorrelations for the WBI scale scores and found that the social skills scale was highly correlated with both cooperation (r (77) = 0.85, p < 0.001) and personal presentation (r (77) = 0.78, p<0.001), and that the Personal Presentation scale was also highly correlated with cooperation (r(77) = 084, p<0.001). Based on these results, we combined the scales into a mean score that was used as criterion variables in a multiple linear regression analysis. The result indicated that participants in our sample did not consistently demonstrate high levels of professional interpersonal skills. Especially higher age was associated with lower WBI scores, while education, gender, and time since last job were not related to WBI scores. In Study 2, we examined whether participants had a lower level of IT knowledge than job seekers in the general population, and investigated possible predictors of IT knowledge in the sample. Participants in Study 2 (N=29) were enrolled in the same therapeutic workplace. Demographic variables were collected in the same manner as Study 1. Participants then completed the Prove It! Version 4.0 Basic Computer Literacy Test. The test consists of 34 multiple-choice and true-false questions and covers five thematic areas: hardware, Internet, security, software and the Microsoft Windows operating system. In order to determine whether Study 2 participants' scores differed from the community test takers' test scores, we compared the participants' mean score to the mean test score using the one-sample t-test (test value = 27.