Keywords
Artificial intelligence
traditional methods
Abstract
The Study was on comparative analysis of Multiple-Choice Mathematics Test Items generated through artificial intelligence and traditional Methods in IkotAbasi LGA, AkwaIbom State. Two objectives, research questions and two null hypotheses guided the study. The researchers adopted descriptive survey research design. 1,260 Senior Secondary Students of 2022/2023 academic session formed the population. Simple random sampling technique was employed for this study. The sample size adopted was 250 Students. Two instruments titled: Artificial Intelligence Mathematics Performance Test on Probability (AIMPTP) and Traditional Mathematics Performance Test on Probability (TMPTP) were used. The two instruments were validated by three experts in measurement and evaluation for both face and content validity. A single test administration method was used for test reliability determination. The reliability of the instruments was determined using Kuder-Richardson 21 formula giving the reliability indices of 0.89 and 0.72 respectively. This showed that the instruments were reliable for the study. Research questions were answered using mean and standard deviationwhile the null hypotheses were tested using paired sample t-test and independent t-test statistics at 0.05 alpha level of significance respectively. The findings of the study revealed that a significant difference existed between the academic performance of Mathematics students exposed to artificial intelligence test and traditional methods. It also revealed that there is no significant difference between the academic performance of male and female Mathematics students exposed to artificial intelligence test. Based on the findings, it was recommended that Mathematics teachers should adopt artificial intelligence method during test development in order to enhance students’ performance in Mathematics.