19 4Examining the relationship between online distribution channels and tourist satisfaction and loyalty 
Home Page  

  • SciELO

  • SciELO


Tourism & Management Studies

 ISSN 2182-8458 ISSN 2182-8466

ULKU, Abdullah. Artificial intelligence-based large language models and integrity of exams and assignments in higher education: the case of tourism courses. []. , 19, 4, pp.21-34.   31--2023. ISSN 2182-8458.  https://doi.org/10.18089/tms.2023.190402.

^a

There is an increasing concern regarding the potential misuse of ChatGPT-4 in compromising the integrity of examinations and assignments. This study aims to examine the capabilities of ChatGPT-4 in critical thinking abilities, whether it poses a threat to examinations and assignments in higher education, and create a discussion agenda on this issue. ChatGPT-4 was asked to generate, answer, and criticize questions in tourism marketing, tourism management, tourism economics, tourist guidance, and gastronomy. The answers were evaluated according to universal critical thinking standards. The findings obtained from this study showed that ChatGPT-4 had commendable competence in several critical thinking standards and could produce human-like texts. However, there are certain domains that might be improved to comply more effectively with the expectations and norms of academia. Educators were recommended to use comprehensive approaches that combine technological and educational techniques to address the issue of cheating enabled by tools, such as ChatGPT-4, during assessment and exam processes.

^len^a

Há uma crescente preocupação em relação ao possível uso indevido do ChatGPT-4 comprometendo a integridade de exames e trabalhos académicos. Este estudo visa examinar as capacidades do ChatGPT-4 nas capacidades de pensamento crítico, avaliar se ele representa uma ameaça para exames e trabalhos no ensino superior e criar uma agenda de discussão sobre esse problema. Solicitou-se ao ChatGPT-4 que gerasse, respondesse e criticasse perguntas nas áreas de marketing turístico, gestão do turismo, economia do turismo, informação turística e gastronomia. As respostas foram avaliadas de acordo com padrões universais de pensamento crítico. Os resultados deste estudo mostraram que o ChatGPT-4 apresentou competência admirável em vários padrões de pensamento crítico e conseguiu produzir textos semelhantes aos humanos. No entanto, há certos domínios que podem ser aperfeiçoados para atender mais efetivamente às expetativas e normas académicas. Recomenda-se que os educadores adotem abordagens abrangentes que combinem técnicas tecnológicas e educacionais para lidar com o problema da fraude, facilitada por ferramentas como o ChatGPT-4, durante processos de avaliação e exames.

^lpt

: .

        · | |     ·     · ( pdf )