A medical school conducted a summer pre-matriculation program. The program provided basic sciences content comparable to first year medical student instruction along with clinical and other learning experiences. The study purpose was to examine self-confidence levels and reasoning skills of a single cohort of students. We examined the association between students' confidence, reasoning skills and performance measures. We also examined performance differences based on student gender, ethnicity and socioeconomic backgrounds. 32 students were given two confidential surveys in a pre/post-test format. Student confidence was measured using a modified version of the Motivated Strategies for Learning Questionnaire (MSLQ). Student reasoning skills were measured using the Health Sciences Reasoning Test (HSRT). Student performance was measured via written exams. Survey item mean differences and inter-item correlations were examined using the T-test and Spearman procedures. Data were analyzed using SPSS software, with significance levels for all tests set at p = 0.05. Completed pre- and post-test surveys were returned from 29 students (91%) for the MSLQ and 18 students (56%) for the HSRT. A single item on the MMSLQ was significantly different on the post-test. Student race, gender and socioeconomic background were poor predictors of overall performance but were associated with selected measures of student confidence and reasoning skills. Students enrolled in medical school preparatory classes may be overly confident prior to experiencing medical school. Demographic variables deserve attention by medical school staff, particularly when the school's mission focuses on successful performance by students from under-represented groups.