Statistical Analysis of Student Data and Machine Learning Models for Performance Prediction

In this paper, we conducted an analysis of student data to discern patterns in academic performance, utilizing the user-friendly statistical software, Jeffreys Amazing Statistics Program (JASP). The dataset was sourced from Kaggle. The study aimed to identify gender-based performance differences in various exams through a one-tailed t-test. To enhance the validity of this analysis, we employed the Brown-Forsythe Test to assess homogeneity of variance between male and female groups. To quantify the impact of different predictors on academic performance, Linear Regression was employed. The overall significance of the regression model was determined through the Analysis of Variance (ANOVA) statistical test. The explanatory capacity of the Linear Regression model was evaluated to ascertain its suitability for prediction. In assessing various machine learning models, it was observed that the best performing model for predicting student performance is Support Vector Machine (SVM). This project provides insights into the real-time application of Artificial Intelligence and Machine Learning algorithm, enabling educators to make informed decisions to enhance student performance.