An introductory geology class that satisfies a liberal arts distribution requirement was used to investigate the benefits of allowing discussion during assessments. For three term examinations, students completed short- to medium-length essay tests individually (individual examination) and then again as part of an assigned group of four to five students (group examination). The comprehensive final examination for the course was multiple-choice and true-false questions, with 75% of the questions covering material on the term examinations and 25% of the questions covering material not tested previously. Students generally favored the group examinations, both midway through the course and at the end of the course, but final examination results were mixed. Those whose scores increased the most in the group examinations tended to have higher percentage correct on both previously tested and new material on the final examination. Those whose group-examination scores were not much better than their individual scores performed at a level similar to or slightly worse than their performance on the term examinations. This suggests that low-achieving students benefit the most from the group examinations. Using the group-examination format in a large class will require a clear rubric for grading by multiple graders, but is expected to result in higher success for low achievers.