The study was an experimental comparison of the structured versus unstructed training of semiskilled production workers. The experiment was implemented using the following procedures, which are presented in detail: a representative semiskilled production job was selected, the two training programs were characterized and developed, trainees representative of a production worker profile were selected and hired, evaluation methods (for product quality, worker competence, cost effectiveness, and worker attitudes) were developed, and the experiment was executed comparing the two training programs. The conclusions, discussed according to six hypotheses, demonstrated that structured training as compared to unstructured training: took less time, produced a higher level of competence by time intervals, and involved lower production losses and a higher percentage of production problem solving. Training costs and trainee attitudes for the two groups were not significantly different. The appendixes (126 pages) contain: working papers (on cost effectiveness, a job analysis, and program characteristics), information on the operation of equipment used, foreman job and trainer characterization, training decisions for structured training, trainer and trainee manuals, material on selection of trainee subjects, sample items from the Bennett Mechanical Comprehension Test, Worker Attitude Inventory, statistical test results, program operational procedures, and summary of raw data. (Author/MS)