Two-stage collaborative testing is an assessment strategy that involves students initially writing a test individually and then immediately afterward writing the same (or similar) test again in groups. Current evidence shows that two-stage testing improves performance on multiple-choice tests as well as short-term retention of material, but little is known about the effect on long-answer questions and retention across a longer time frame. The purpose of this study was to determine (a) if two-stage testing improves performance on both multiple-choice and long-answer questions, (b) if two-stage testing improves short- and long-term knowledge retention, and (c) whether there are differences in knowledge retention based on question type. A two-stage midterm with both question types was administered in two undergraduate science courses, followed by a short-term and long-term retention test. Performance on both question types improved, with comparable improvement on both question types. Two-stage testing also maintained knowledge retention from the original midterm for both question types in the short term, although the learning gains for long-term retention were less apparent.