Performance of ChatGPT-3.5 and ChatGPT-4 in the Taiwan National Pharmacist Licensing Examination: Comparative Evaluation Study
Bibliographic record
Abstract
Background: OpenAI released versions ChatGPT-3.5 and GPT-4 between 2022 and 2023. GPT-3.5 has demonstrated proficiency in various examinations, particularly the United States Medical Licensing Examination. However, GPT-4 has more advanced capabilities. Objective: This study aims to examine the efficacy of GPT-3.5 and GPT-4 within the Taiwan National Pharmacist Licensing Examination and to ascertain their utility and potential application in clinical pharmacy and education. Methods: The pharmacist examination in Taiwan consists of 2 stages: basic subjects and clinical subjects. In this study, exam questions were manually fed into the GPT-3.5 and GPT-4 models, and their responses were recorded; graphic-based questions were excluded. This study encompassed three steps: (1) determining the answering accuracy of GPT-3.5 and GPT-4, (2) categorizing question types and observing differences in model performance across these categories, and (3) comparing model performance on calculation and situational questions. Microsoft Excel and R software were used for statistical analyses. Results: GPT-4 achieved an accuracy rate of 72.9%, overshadowing GPT-3.5, which achieved 59.1% (P<.001). In the basic subjects category, GPT-4 significantly outperformed GPT-3.5 (73.4% vs 53.2%; P<.001). However, in clinical subjects, only minor differences in accuracy were observed. Specifically, GPT-4 outperformed GPT-3.5 in the calculation and situational questions. Conclusions: This study demonstrates that GPT-4 outperforms GPT-3.5 in the Taiwan National Pharmacist Licensing Examination, particularly in basic subjects. While GPT-4 shows potential for use in clinical practice and pharmacy education, its limitations warrant caution. Future research should focus on refining prompts, improving model stability, integrating medical databases, and designing questions that better assess student competence and minimize guessing.
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 | 0.003 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| 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".