We report the results of a questionnaire survey of manufacturing workplaces related to the lost-time frequency rates (LTFR) for Workers' Compensation claims. Six types of industry were chosen.' metal articles, plastic articles, grain products, textile manufacturing, printing, and automobile manufacturing. LTFR were standardized by type of industry. Stratifying simultaneously by number of employees and LTFR category, we sampled 718 workplaces. A mail questionnaire to labor and management representatives provided at least some information on 58%. Response rates were similar across LTFR categories, and telephone interviews of non-responders showed little difference in their replies from those obtained in completed questionnaires. A large number of variables were examined. Apart from statistical significance, we looked for consistency in trends across LTFR categories and in patterns for similar questions. Significant associations grouped into several areas. Lower LTFR were associated with: concrete demonstration by management of its concern for the workforce; greater involvement of workers in general decision-making; greater willingness of the Joint Health and Safety Committee to solve problems internally; and greater experience of the workforce. Variables that were not significant included profitability and financial performance. A final stepwise multiple regression explained 19% of the variance in LTFR, although this analysis suffered from several limitations.