Exploring the potentials of ChatGPT for instructional assessment: Lecturers' attitude and perception

Abstract

Lecturers play a crucial role in the educational process, offering unique insights and perspectives within the classroom. The issue of cred­ibility in educational assessment often rests on the shoulders of lecturers, who are responsible for eval­uating students' progress. The present study aimed to investigate lecturers' attitudes and perceptions re­garding the potential of ChatGPT for instructional assessment. A correlational research design was adopted, and purposive sampling was used to select 102 lecturers from Nigerian universities who had uti­lised ChatGPT for instructional assessment. Data was collected through an online structured question­naire. The normality and homogeneity of variance assumptions were met, as evidenced by kurtosis and skewness values falling within acceptable thresh­olds. The lecturers employed the instructional assessment questionnaire utilising ChatGPT to gather and analyse the data, employing t-tests and ANOVA. The findings revealed a statistically significant difference between perception (F (3, 98) =7.168, p=0.001 <0.05) and lecturers' years of experience in using ChatGPT for instructional assessment. The study indicated that lecturers held low attitudes and had poor perception levels when it came to exploring the potential of ChatGPT. However, it is recommended that training be provided to enhance lecturers' attitudes and perception levels to fully exploit the potential of ChatGPT for instructional assessment.