The educational performance of 690 14 and 15 year old dropout prone students given limited labor market experience was evaluated in a mationwide study. The students worked 28 hours per week during the 1971-72 school year, which was found to be excessive. Working increased grade point averages up to a point, after which favorable impact declined or became negative. The extent of career exploration experienced and its implications for longrun labor market prospects could not be discerned from the results of the study. The control group did not come from the same population as the program students. Nonresponse bias also makes the impact of the program unclear. A broader extension among the population of students to be served is recommended. Approximately one half of the document is devoted to appendixes covering data forms, questionnaires, child labor regulations, and survey analysis tables. (MS)