Introduction. It is possible to conduct electronic tests at Warsaw Medical University (WMU) in the academic year 2014-2015. The final test for the ‘Law in Medicine’ course was the first one in the history of WMU that was conducted on the e-exam platform. Aim. To analyse the quality of questions. Material and methods. 200 questions with three answer options, one correct answer. Analysis of reliability, easiness and differentiation power of particular questions by the round of exam and number of draws of a particular question. Analysis of the easiness of the question database, frequency of using questions in the database, assessment of interrater reliability assumed for the estimated question easiness, assessment of easiness by the frequency of using a particular question. STATISTICA 10.0, licensed to WMU, Kruskal-Wallis ANOVA test. Results. Seven rounds of exams, easiness of question database: medium: 0.715, median: 0.750, SD: 0.23, assessment of interrater reliability assumed for the estimated question easiness: significant correlation (p < 0.05), moderate agreement γ = 0.433. Parameters of the test questions were independent of the round of exam: Kruskal-Wallis ANOVA test, easiness of H = 2.946; p = 0.8156, differentiation power of H = 2.842; p = 0.8283. Estimated easiness of a particular question dropped with the increase of the number of drawings of this question: question’s easiness dropped by 0.002. Conclusions. 1. Owing to a small measuring scale of the questions, the study is only a pilot and needs to be continued. 2. Due to the fact that easiness of questions used in the test was higher than expected, modification of questions needs to be considered. 3. Repeatability of questions in the test decreased the easiness of questions. Concerns about the reuse of previously drawn questions seem to be unjustified. 4. It seems necessary to conduct a series of trainings for teachers covering broad issues concerning phrasing test questions and evaluating test exams.

Key words: computer-aided examinations, quality of education, students of health sciences, didactic measurement.