Background. The purpose of this study was to determine which factors influenced performance on the Thoracic Surgery In-Training Examination (TSITE) and whether the addition of a cardiac-specific didactic study course improved scores. Methods. Between 1989 and 2002, 59 TSITE scores (overall, cardiac [C-TSITE], and thoracic [T-TSITE]) were collected from 33 residents (23 cardiac track, 10 thoracic). Factors assessed with univariate and multivariate analysis included calendar year, year of training (31 year I, 28 year II), standardized test-taking history (average National Board and American Board of Surgery in-training scores), subjective faculty assessment of cardiothoracic knowledge, months on cardiac versus thoracic service, clinical performance, and participation in a cardiac-specific didactic series with faculty lectures and board question reviews (12 residents). Results. Cardiac-track residents had higher C-TSITE percentile scores (53% ± 27% versus 38% ± 27%, p < 0.05), whereas thoracic-track residents had higher T-TSITE scores (70% ± 24% versus 51% ± 25%, p < 0.01). Multivariate analysis identified 3 factors associated with higher overall TSITE scores: standardized test-taking history (p < 0.001), subjective faculty assessment of knowledge (p < 0.001), and year of training (p < 0.007). Inclusion in the cardiac-specific didactic series did not affect C-TSITE scores (50% ± 30% versus 48% ± 27%, p > 0.82) or overall TSITE scores (p > 0.23). Conclusions. Standardized test-taking history and subjective faculty assessment of knowledge were associated with higher TSITE scores, but implementation of a cardiac-specific didactic series had no influence. These findings suggest that independent study and reading may be the best way to improve TSITE scores.