Question format is the best predictor of item discrimination: a multivariable analysis
Bibliographic record
Abstract
ABSTRACT Item discrimination, the point-biserial correlation between performance on a question and total exam score, was analyzed in relation to question format, difficulty, and Bloom’s taxonomy, which are rarely studied together in a joint framework. To estimate their independent contributions in undergraduate biology examinations, simultaneous multivariable linear and logistic models were conducted on an upper year biology item bank ( n = 754). Predictors were question format (multiple-choice question [MCQ], short answer [SA], true/false [TF]), difficulty (easy >80%, moderate 60%–80%, hard <60%), and Bloom (recall, understanding, application, analysis). Question format was the strongest independent predictor. Relative to MCQs, SA items showed higher discrimination and greater odds of meeting the ≥0.35 threshold (odds ratio [OR] = 4.18), while TF items were less discriminating (OR = 0.58). Moderate and hard items exceeded easy questions (ORs = 2.75 and 2.45, respectively). For Bloom, higher-order items outperformed recall questions overall; both understanding and analysis items discriminated better than recall items (ORs = 1.4 and 3.56, respectively). A post hoc grouping of moderate difficulty questions showed that SA + higher-order items (0.53) exceeded MCQ + recall items (0.35), with 89% vs 53% of these items, respectively, meeting a discrimination threshold of ≥0.35. In simultaneous models, question format was the strongest independent predictor of item discrimination, moderate difficulty optimized discrimination, and higher-order objectives exceeded recall. These results provide exploratory insights from a single-institution case study and suggest that adding some higher-order short-answer questions of moderate difficulty may represent a pragmatic strategy for improving assessment quality.
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.
How this classification was reachedexpand
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".