Medical and dental students were administered two short confidence-scored tests on cellular and molecular biology. Increases in test reliability and predictive validity were found when test scores were adjusted for realism, but were not statistically significant. (Author/MH)

Presents a discussion of trying to best account for the variation resulting from guessing on a multiple-choice exam. A mathematical process is given to determine the amount of variance that could be added or subtracted from a student's score to account for guessing. Examples are provided. (MVL)