Model Persamaan Diskriminan untuk Prediksi Gaya Belajar Selmes Berdasarkan Kemampuan Metakognisi

Abstract

This study aims to determine how the discriminant equation model can be used to predict the type of Selmes learning style based on students' metacognitive abilities in Junior High Schools. This study is an associational research type with a quantitative approach, involving a sample of 79 students selected using the cluster random sampling technique. The instruments used consisted of a metacognitive ability test in the form of three descriptive questions that measured indicators of understanding, strategy and connection, as well as a closed questionnaire to identify four groups of Selmes learning styles, namely surface, deep, hard work and organized. The data analysis technique used is Linear Discriminant Analysis (LDA), the results of the study showed four LDA equation functions, each representing each learning style. This model has an accuracy rate of 83.33% in predicting learning styles with the results of the multivariate significance test using Wilks' lambda producing a  or  of 0.0000<0.05 (alpha), so it can be concluded that the discriminant function coefficient is significant for predicting Selmes' learning styles based on metacognitive abilities.