Providing Validation Evidence for a Clinical-Science Module: Improving Testing Reliability with Quizzes
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
DESCRIPTION OF THE PROBLEM: High-stakes decision-making should have sound validation evidence; reliability is vital towards this. A short exam may not be very reliable on its own within didactic courses, and so supplementing it with quizzes might help. But how much? This study's objective was to understand how much reliability (for the overall module-grades) could be gained by adding quiz data to traditional exam data in a clinical-science module. THE INNOVATION: In didactic coursework, quizzes are a common instructional strategy. However, individual contexts/instructors can vary quiz use formatively and/or summatively. Second-year PharmD students took a clinical-science course, wherein a 5-week module focused on cardiovascular therapeutics. Generalizability Theory (G-Theory) combined seven quizzes leading to an exam into one module-level reliability, based on a model where students were crossed with items nested in eight fixed testing occasions (mGENOVA used). Furthermore, G-Theory decision-studies were planned to illustrate changes in module-grade reliability, where the number of quiz-items and relative-weighting of quizzes were altered. CRITICAL ANALYSIS: One-hundred students took seven quizzes and one exam. Individually, the exam had 32 multiple-choice questions (MCQ) (KR-20 reliability=0.67), while quizzes had a total of 50MCQ (5-9MCQ each) with most individual quiz KR-20s less than or equal to 0.54. After combining the quizzes and exam using G-Theory, estimated reliability of module-grades was 0.73; improved from the exam alone. Doubling the quiz-weight, from the syllabus' 18% quizzes and 82% exam, increased the composite-reliability of module-grades to 0.77. Reliability of 0.80 was achieved with equal-weight for quizzes and exam. NEXT STEPS: Expectedly, more items lent to higher reliability. However, using quizzes predominantly formatively had little impact on reliability, while using quizzes more summatively (i.e., increasing their relative-weight in module-grade) improved reliability further. Thus, depending on use, quizzes can add to a course's rigor.
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 | 0.036 | 0.684 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.022 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 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