Binary Logistics Regression To Predict The Opportunity Of SNMPTN Graduation In Statistics Study Program Of Tanjungpura University

Abstract

The National Selection of State University Entrance or Seleksi Nasional Masuk Perguruan Tinggi Negeri (SNMPTN) is one of the selections for high school students seeking higher education. The Statistics Study Program as one of the study programs at Tanjungpura University has a capacity of 20 seats in the SNMPTN. This limited capacity causes prospective students to prepare the right strategy in order to be accepted through the SNMPTN. In this study, logistics regression was used to predict the probability of graduation status on the SNMPTN path in the Statistics Study Program of Untan. Binary logistic regression is a statistical analysis technique for representing the relationship between a response variable with two (binary) categories and one or more predictor variables on a continuous or categorical scale. Data for this study were primary data from a questionnaire that received 93 samples. The response variable used is graduation status (Y) through the SNMPTN in Statistics Study Program, Tanjungpura University classified as 1 (passed) and 0 (not passed). Based on the results of the study, it is known that the variables that have a significant effect on graduation status are the status of choice in Statistics Study Program (X1), national level achievement ownership (X3), the average value of Mathematics (X4), the average value of Chemistry (X6), Biology average score (X7), Indonesian average score (X8), and English average score (X9). Meanwhile, provincial level achievement (X2) and Physics average (X5) did not have a significant effect on graduation status. The binary logistic regression model obtained has an accuracy error of 15,05% with an accuracy rate of 84,95%, meaning that this model has a good criteria.