Despite numerous empirical studies, there is surprisingly little agreement about whether the Social Security earnings test affects male labor supply. In this paper, we provide a comprehensive analysis of the labor supply effects of the earnings test using longitudinal administrative earnings data and more commonly used survey data. We find that the response to the earnings test in survey data is obfuscated by measurement error and labor market rigidities. Accounting for these factors, our results suggest a consistent and substantial response to the earnings test, especially for younger men. (Contains 9 tables, 8 figures, and 27 footnotes.)