Item statistics derived from three-option versions of multiple-choice questions are usually as robust as four- or five-option versions: implications for exam design
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
Different versions of multiple-choice exams were administered to an undergraduate class in human physiology as part of normal testing in the classroom. The goal was to evaluate whether the number of options (possible answers) per question influenced the effectiveness of this assessment. Three exams (each with three versions) were given to each of two sections during an academic quarter. All versions were equally long, with 30 questions: 10 questions with 3 options, 10 questions with 4, and 10 questions with 5 (always one correct answer plus distractors). Each question appeared in all three versions of an exam, with a different number of options in each version (three, four, or five). Discrimination (point biserial and upper-lower discrimination indexes) and difficulty were evaluated for each question. There was a small increase in difficulty (a lower average score on a question) when more options were provided. The upper-lower discrimination index indicated a small improvement in assessment of student learning with more options, although the point biserial did not. The total length of a question (number of words) was associated with a small increase in discrimination and difficulty, independent of the number of options. Quantitative questions were more likely to show an increase in discrimination with more options than nonquantitative questions, but this effect was very small. Therefore, for these testing conditions, there appears to be little advantage in providing more than three options per multiple-choice question, and there are disadvantages, such as needing more time for an exam.
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 | 0.001 | 0.013 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| 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