Development And Validation Of The Generative Artificial Intelligence Usage And Exposure Scale (Gaiues)

The purpose of this study was to develop and validate the Generative Artificial Intelligence Usage and Exposure Scale (GAIUES) to measure university students' usage and exposure to generative artificial intelligence. The present study included two phases. Phase I involved the development of the scale through theoretical generation of items, expert evaluation, pilot testing, and psychometric validation. First, a set of items was generated and evaluated by six psychologists using Content Validity Ratio (CVR), followed by selection of 40 items for pilot testing. Exploratory Factor Analysis (EFA) was performed on the data gathered from 1200 university students, yielding a six-factor solution. After excluding weak and cross-loading items, the final GAIUES scale included 13 items explaining 52.63% of variance in the data. Factor structure of GAIUES was also validated using Confirmatory Factor Analysis (CFA), showing good fit of the model. Furthermore, convergent and divergent validity of the scale proved to be satisfactory. In summary, results of this study suggest that GAIUES scale is a reliable and valid multidimensional measure of generative artificial intelligence usage and exposure.