Evaluating GPT-4 Turbo`s Ability to Design English Reading Test Items for Language Learners
Bibliographic record
Abstract
This study evaluates the Generative Pre-trained Transformer (GPT-4 turbo) to design English reading multiple-choice questions (MCQs) for intermediate learners by addressing a deficiency in studies that examine GPT-4 turbo`s capabilities to generate MCQs using various prompt engineering techniques and evaluate their psychometric properties. Utilizing a descriptive quantitative method, a cohort of eight-item writers and 150 preparatory students participated in the study. Both the questionnaire and the generated online test were used to collect data. The findings reveal that zero-shot prompting demonstrates the highest level of agreement compared to few-shot prompting across three of the six aspects: text coherence, question stem quality, and answer options quality. The study finds that although all generated MCQs using few-shot prompting exhibit significantly higher discrimination values than those generated using zero-shot prompting, all MCQs displayed a low level of difficulty across all three prompt engineering techniques (zero-shot prompting, few-shot prompting (two-shot & four-shot). Taken together, this study proposes some profound implications for language assessment developers by illustrating how prompt design influences both the perceived and measured quality of AI-generated items. It also contributes to the literature by providing meaningful insights into using large language models, especially GPT-4 turbo, for AIG.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.069 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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 source (direct Gemma or distilled Codex), 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".