The behavioral intentions to integrate AI in teaching science among distance and contact education student teachers : a comparative study

Abstract

Integrating Artificial Intelligence could reduce educational challenges and improve learning outcomes in teaching science. However, the intention of teachers to use these technologies is poorly understood. Informed by the Theory of Planned Behavior, this comparative survey study examined the behavioral intentions of science student teachers from two South African universities to integrate AI into teaching science. An online questionnaire was used to collect data from purposively sampled final-year students from Central University (n = 97) and East Coast University (n = 85). Data analysis involved Exploratory Factor Analysis to identify the structure of constructs, ordinal logistic regression to identify predictors of behavioral intention, and the Mann-Whitney U test to compare behavioral intentions between the two samples. The results suggested that the constructs conformed to the Theory of Planned Behavior with high reliability (Cronbach’s Alpha: .846–.935). Seven factors explained 72.67% of the variance, with strong loadings. Both samples demonstrated a positive intention to integrate AI, with the Central University reporting higher control beliefs (p = .005). However, Ordinal Logistic Regression indicated the constructs did not significantly predict behavioral intentions (p > .05). The findings highlight the importance of providing context-specific training programs to support the integration of AI in science education. IMPACT STATEMENT : This study examined South African science student teachers’ intentions to integrate Artificial Intelligence (AI) into teaching, comparing distance and contact education contexts. The findings show strong readiness to embrace AI, highlighting opportunities for innovation in science education. However, differences in control beliefs reveal structural inequities in training, access, and support that shape confidence in AI adoption. By exposing these disparities, the study demonstrates the need for responsive, hands-on professional development and equitable resource provision. It advances educational practice and policy by offering evidence to strengthen AI integration, bridge systemic gaps, and promote sustainable, equitable improvements in science education.