Evaluating the quality of tourism research using ChatGPT

This article evaluates the quality of tourism research by comparing the results obtained from citation analysis with those obtained from ChatGPT quality scores based on the Research Excellence Framework criteria, applied to 50 of the largest tourism and leisure journals. Whilst there was an article-level weak correlation between normalized citation rates and normalized ChatGPT scores, there was a moderately strong journal-level correlation between the two measures. The latter supports the value of normalized ChatGPT scores for journal quality indicators. The results also suggest important nuances between and within qualitative and qualitative methods used in tourism. Articles using experiments, advanced statistical tests, and theories scored well on both indicators but the opposite for those based on surveys and convenience sampling.