Collaborative Testing Improves Performance on Long Answer Questions, and Maintains Long‐Term Retention of Course Material
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Collaborative testing (CT) is an assessment strategy whereby students first write a test as an individual and then immediately following, write the same test again as a group, with the opportunity to discuss their thought process in reaching their answers. This strategy has been shown to increase performance on multiple‐choice (MC) tests, and to improve short‐term retention of material. However, this format has not been evaluated for tests using long answer (LA) questions, and measures of improved long‐term retention are inconsistent. The purpose of this study was to determine if CT improves performance on LA questions and whether CT could improve long‐term retention of course material compared to traditional teaching and testing methods. Two courses, (3 rd year Exercise Physiology, n=102 and 2 nd year Biochemistry, n=64) administered identical protocols which included an in‐class collaborative midterm (half MC and half LA questions), an unannounced, individual retention test 1 week later (short‐term) followed by a brief instructor‐led, in‐class review of the test, and finally another unannounced individual retention test 6 weeks later (long‐term). Performance was calculated as the difference in grades between collaborative and individual midterm. Retention was calculated as the difference in grades between a given retention test and the individual midterm. CT improved performance on both MC and LA questions, but the degree of improvement was greater on LA questions (16.5% ± 1.29%, 20.6% ± 1.41%, p<0.05). As expected, short‐term retention was better on questions that had been tested collaboratively compared to questions that were only seen individually (+3.7% ± 1.53% vs. −7.9% ± 1.50%, p<0.05). Surprisingly, long‐term retention was similarly maintained for both collaborative and individual questions (+0.69% ± 1.92% vs. −0.16% ± 1.89%), indicating that retention of the questions that had been tested individually had improved since the short‐term retention test. Our results show that CT can improve performance on LA questions and help students retain this information over several weeks, but also suggests that taking the time to review a test in class may be an equal strategy to improve long‐term retention of material.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it