A Systematic Review of AI Adoption in Higher Education: Analyzing UTAUT Determinants Through Confidence Interval, Prediction Interval, and Pearson Correlation Assessments

In this investigation, we have reviewed 50 papers on the impact of artificial intelligence (AI) on higher education using the unified theory of acceptance and uses of technology (UTAUT) framework, and utilizing these data, we computed the central tendency, coefficient of variation, differential entropy, cumulative distribution function, and—more intriguingly—the confidence interval of the population mean, the prediction interval of future work of each parameter in the UTAUT framework, and the Pearson correlation between the parameters for the first time. Performance expectancy exhibits central tendency with a mean, median, and mode of 0.83759 and a variance of 0.07176, with excellent data consistency examined by the coefficient of variation and differential entropy. The confidence and prediction intervals of performance expectancy are 0.8041 to 0.871 and 0.5992 to 1.000, respectively, with a 99.9% confidence level. Similar to performance expectation, additional measures such as effort expectancy, social influence, fascinating conditions, and moderators (male and female) in the UTAUT framework have central tendency and variation of 0.83022 ±0.06128, 0.83793 ±0.06584, 0.81153 ±0.0595, 0.4793 ±0.12249, and 0.51817 ±0.12451, respectively. Additionally, at a 99.9% confidence level, these parameters demonstrate confidence and prediction intervals of (0.8041-0.871, 0.5992-1.0), (0.8017-0.8587, 0.6265-1.0), (0.8073-0.8685, 0.6191-1.0), (0.7838-0.8392, 0.6141-1.0), (0.4224-0.5362, 0.0723-0.863), and (0.4602-0.5762, 0.1052-0.9312). The Pearson correlation coefficient exhibits a strong correlation between the PE and EE (0.80), PE and FC (0.60), PE and EE (0.80), EE and SI (0.61), SI and Teachers (0.88), SI and Students (0.65), and Teachers and Male (0.56) pairs. Moreover, bibliometric keyword mapping shows 5 main research themes focusing on regularly studied issues, their linkages, and developing areas in the research field. After all, this work is crucial for future research on the effects of AI on higher education and similar studies for other sectors of scientific domains.