Dataset of approximately 18,000 multiple-choice questions from a high-stakes medical exam¶
Product or Programme · All products and programmes
Evidence · 3 claims (3 for) · 1 study (1 associational),q2· 0 of 1 report an effect size · 3 claims rest on one study
Description¶
A research dataset assembled by Kang Xue et al. (2020) containing approximately 18,000 high-stakes medical-exam multiple-choice questions with item difficulty and response-time parameters for modeling and transfer-learning research.
Components¶
Claims¶
- All parts of the item were important for predicting response time [+W]
- Item difficulty was best predicted using signal from the item stem [+W]
- Transfer learning improves prediction of item difficulty when response time is used as an auxiliary task, but not the other way around [+W]
Related Products and Programmes¶
-
Key Sources¶
- Kang Xue, Victoria Yaneva, Christopher Runyon, Peter Baldwin. (2020). Predicting the difficulty and response time of multiple choice questions using transfer learning. 15th Workshop on Innovative Use of NLP for Building Educational Applications. https://www.nwea.org/research/publication/predicting-the-difficulty-and-response-time-of-multiple-choice-questions-using-transfer-learning/
Page metadata
| Field | Value |
|---|---|
type |
product |
id |
dataset-of-approximately-18000-multiple-choice-questions-from-a-high-stakes-medical-exam |
title |
Dataset of approximately 18,000 multiple-choice questions from a high-stakes medical exam |
description |
A research dataset assembled by Kang Xue et al. |
status |
review |
moved_from |
element/medical-exam-mcq-difficulty-response-time-dataset |
product_kind |
dataset |
generated |
by process:sweep-kinds on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — kang-xue-2020 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.