We examine the accuracy of various estimates of firms' repurchases of common stock used in earlier studies, and find high error rates in the most commonly used estimators. We also find that the procedure used to estimate open market share repurchases can significantly impact results. The Compustat-based measure, which is the most accurate, deviates from the actual number of shares repurchased by more than 30% in about 16% of the cases. We conclude that many studies should be revisited now that the SEC mandates disclosure of precise information about share repurchases in Forms 10-Q and 10-K.