The quality of medical students' confidence judgments when using external information resources: The effects of different media formats, source of questions, and question formats.

摘要

This dissertation reports on the appropriateness of medical students' confidence in their answers to questions as part of a course where they used external information resources. Previous studies have investigated only subjects' confidence judgments using personal memory. The ability to judge appropriately one's skill in using medical information resources is important for medical students, as students who are overconfident rush to decisions without investigating all relevant information, while students who are underconfident waste important time rehashing information even after they have mastered it. Ideally, medical students should be appropriately confident to maintain the most efficient and effective use of their time.; Over 150 students taking a second year medical school class in Reproductive Biology were graded on their answers to three problem sets...given as part of the course. All of the questions required students to use an external information resource: either lecture notes and textbook, or research articles in a variety of formats: print, online, downloaded form, or a combination of the above. Students' scores were based upon the quality of their indicated confidence (in percentage form) in the right answer using Brier scores, resolution and calibration. The effects of media format, occasion, source of question, and question type were investigated.; Results indicate that medical students are uniformly very appropriately confident in their answers to questions about subject material in a course for which they are studying. The students with the best aptitude were the best judges of their performance. There was no effect of media format on quality of confidence. There was a marginal effect of occasion on quality of confidence (Brier scores and resolution), with poorer scores on the third of three problem sets than on the second. There was also no effect of question source on Brier scores or calibration, but a marginal effect of question source on resolution, with better scores on research questions than lecture material. Students' quality of judgment was better for all three measures of confidence on the multiple-choice than on true/false questions. The implications of this study are that medical students are already studying very effectively.