Data collected by the Federal Trade Commission from the Educational Testing Service and Commercial Coaching Schools in the New York Metropolitan area are used in an observational study to test the hypothesis that commercial coaching can improve performance on the Scholastic Aptitude Test. Evidence is presented that many students who sought commercial coaching had scored below expectations on previous standardized exams. Their poor performance, however, was found not to be due to chance but rather to a systematic inability to do well on such tests. The data are consistent with the hypothesis that coaching can, at the very least, help such students to overcome their inability to perform as one would predict from their demographic and personal characteristics.