Data From: Using Large Language Models to Generate Script Concordance Test in Medical Education: ChatGPT and Claude

To access the paper (open-access): https://revistas.um.es/edumed/article/view/636331Dataset InfoExcel File:# The first row of the dataset identifies the columns.# The first column (Scenario) includes the scenarios that the script concordance test items were generated by large language models based on. (Left lower quadrant pain, right lower quadrant pain, right upper quadrant pain, epigastric pain, and acute onset flank pain)# The second column (Statement) includes 12 statements (criteria) that the expert panel used in the process for evaluation of script concordance test items.# The third column (Label) includes the labels used for representing 12 statements.# The fourth column (Model) includes the name of the large language model.# The fifth column (Response) includes the responses by the expert panel to 12 statements. (Yes, Uncertain, No)# The sixth column (Count) includes the frequency of responses. Word File:It includes the prompt template to generate script concordance test (SCT) items, and the generated SCT items using ChatGPT-4 and Claude 3 (Sonnet).