PREDICTING ACADEMIC-PERFORMANCE OF MEDICAL-STUDENTS - THE FIRST 3 YEARS

Objective: The purpose of this exploratory study was to identify a clu ster of variables that would most economically explain variations in t he grade point averages of medical students during the first 3 years o f study. Method: Data were derived from a study of 92 students admitte d to the 3rd Faculty of Medicine in 1992-1993 academic year and who we re still in the medical school at the end of the sixth semester (third year). Stepwise regression analysis was used to build models for pred icting log-transformed changes in grade point average after six semest ers of study-at the end of the first, second, and third years. Predict or variables were chosen from four domains: 1) high school grade point averages in physics, mathematics, and the Czech language over 4 years of study, 2) results of admission tests in biology, chemistry, and ph ysics, 3) admission committee's assessment of the applicant's ability to reproduce a text, motivation to study medicine, and social maturity , and 4) scores on the sentimentality and attachment scales of the Tri dimensional Personality Questionnaire. Results: The regression model, which included performance in high school physics, results of the admi ssion test in physics, assessment of the applicant's motivation to stu dy medicine, and attachment scale score, accounted for 32% of the chan ge in grade point average over six semesters of study. The regression models using the first-, second-, and third-year grade point averages as the dependent variables showed slightly decreasing amounts of expla ined variance toward the end of the third year of study and within dom ains, changing the structure of predictor variables. Conclusions: The results suggest that variables chosen from the assessment domains of h igh school performance, written entrance examination, admission interv iew and personality traits may be significant predictors of academic s uccess during the first 3 years of medical study.