중학생들의 수학흥미와 수학가치인식이 수학학업성취도에 미치는 영향력에 대한 유형탐색

Abstract

This study detects heterogeneous latent classes and their models applying REBUS-PLS analysis based on the path model of math interest, math perception and math achievement using TIMSS 2015 8th grade data of Korea, under the context that the influence of related factors on the achievement of mathematics is different. As post-hoc analysis, the distribution for latent classes as dependent variables is analyzed by using math self-confidence and engaged math-learning as predictors. According to the analysis results, two latent classes were captured, and the math-interest oriented class and the math-perception oriented class were named. As a result of post-hoc analysis, CHAID analysis showed that students with high self-confidence in mathematics were distributed in the math-perception oriented class at a high proportion, while those with low self-confidence in mathematics were distributed in the math-interest oriented class at a high proportion. Next, under the classification by math self-confidence, students with more engaged math-learning in both high confidence and low confidence group were more distributed in math interest oriented class, and the less engaged math-learning, in math perception oriented class. This study presented a new method of research on school mathematics by applying the REBUS-PLS analysis method to detecting unit segments based on the path model.