Unemployment and health in a community sample

This paper reports the first results from a survey of unemployment in Southeastern Michigan. The study includes samples of currently unemployed, previously unemployed and stably employed respondents. Special efforts were made to obtain information about the context surrounding each job loss reported, thus allowing the explicit analysis of selection bias in the estimation of unemployment effects. The analysis documents the substantial influences that unemployment has on self-reported physical health, somatization, anxiety, and depression that cannot be explained by selection bias. (March 1987)