The Prediction of Academic Performances in the Subject of Probability and Statistics Subject, Royal Thai Naval Academy, Using a Designed Computer Program Based on the Principle of Multiple Linear Regression Analysis Model

Abstract

The objectives of this research are as follows: 1) to develop a computer program applying multiple linear regression models to predict academic performance as a tool for lecturers to raise their students’ learning efficiency. The prediction is calculated based on multiple linear regression models and classified by using distributed confidence level technique into three groups, which are high-performing, intermediate-performing, and low-performing groups; 2) to evaluate the accuracy of the prediction. The population for this research is 317 naval cadets (class of Probability and Statistics, RTNA, Samut Prakarn). Percentage, ratio, and multiple linear regression analysis are used along with naval cadet opinion questionnaire and a designed computer program. The tools used in the research are Visual Studio Code for coding, PHP for backend algorithm, HTML for user interfaces, along with Apache Web Server for serving the application to browsers.

The results show that 1) the software can predict academic achievement and 2) the error of the prediction is less than 5% in each academic performance group when the same academic year data are used.