Developing and validating an Artificial Intelligence Dependence Questionnaire (AIDQ) for undergraduate students
Abstract
The excessive usage and dependency on artificial intelligence (AI) tools for overall academic performance have been a serious issue for the faculty in the last few years. Students have often overlooked generating original ideas and instead relied solely on machine learning/AI outputs for academic engagement and resources. Hence, this study develops and validates an Artificial Intelligence Dependence Questionnaire (AIDQ), a reference for assessing AI dependency among undergraduate students at a selected university in Lagos, Nigeria. It employed a quantitative approach and a cross-sectional research design. 812 undergraduate students were selected for the validation phase, with participants drawn from the University of Lagos, Nigeria. The students were selected based on convenience sampling, and they filled out the newly developed AIDQ for the validation phase. The data were analysed using split-half reliability, Cronbach's alpha reliability, factorial analysis, and scree plots. The results revealed that AIDQ’s internal consistency was 0.85, while the component analysis loaded and confirmed the factors responsible for AI dependency among the participants. The study concluded that the development and validation of AIDQ is a reliable and valid instrument for measuring AI dependency among undergraduate students. The research offers valuable insights for developing an assessment tool to measure and foster balanced AI use while upholding academic integrity.
