Contributions of ChatGPT Adoption and Academic Integrity to Learning Engagement among Undergraduates
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: This research work examined how the adoption of ChatGPT relates to learning engagement among undergraduates in public universities in Cross River State, Nigeria, namely, the University of Calabar, the University of Cross River, and the University of Education and Entrepreneurship, and whether self-regulated learning and academic integrity serially mediate this relationship. The study further compared the structural relationships across gender using multigroup analysis (MGA); a companion analysis of the same dataset examining age-group differences is reported in a separate paper. The study is grounded in theories of technology acceptance and self-regulated learning. Literature was reviewed based on the variables under study. The study adopted a cross-sectional survey design. The population of the study comprised 1,507 final-year undergraduate students. The sampling technique utilized for this research was proportionate stratified sampling. A serial mediation model was specified and tested on survey data from 452 undergraduates. Partial least squares structural equation modelling (PLS-SEM) was estimated in Smart PLS. The instrument for data collection was a questionnaire titled “ChatGPT Adoption, Academic Integrity, and Learning Engagement” (CAAILEQ). The measurement model satisfied reliability, convergent, and discriminant validity standards. Analyses were conducted using SmartPLS 4. Results show that ChatGPT adoption positively predicted self-regulated learning, academic integrity, and learning engagement; self-regulated learning predicted academic integrity; and academic integrity was the strongest predictor of learning engagement. The hypothesized four-step serial chain (ChatGPT adoption, self-regulated learning, academic integrity, and learning engagement) was statistically significant, and the model explained 38.7% of the variance in learning engagement. The multigroup analysis indicated that one structural path (adoption → integrity) differed significantly by gender, while all other paths were invariant across the gender groups compared. The findings inform the responsible integration of generative AI in Nigerian university education and in Cross River State specifically.
Source: Research Square
Authors: Achi Ndifon Bekomson, Catherine Kaning Agbongiasede, Ettah Ettah Omini, Dorn Cklaimz Enamhe, Matilda Ernest Eteng, Emeka Samuel Nnaji, Mary Iyaji Ngaji, Ajim Ferdinard Mpuah, Edmond Asu Odok, Roseline Asuquo Okon, Samson Akinwumi Aderibigbe, Ovat Egbe Okpa, Victoria Simon Akpama, Joy Anthony Ukam, Samuel Eburu Odey, Samson Chidubem Ugwu, John Edwin Effiom
Institution: University of Calabar
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7225423352