Random Forest Analysis of Factors Influencing the Students' Reading Literacy Levels: Using PISA 2018 Korea Data

Abstract

The purpose of this study is to explore the variables that have a major impact on the classifying of groups by reading literacy level. To research the study, the groups were divided to the general group and the below basic literacy level group by reading literacy level and the random forest was used to explore the main variables to classify the group by reading literacy level. And the subjects of this study were the 6,630 students in the 168 schools who participated in PISA 2018 and the 286 variables of student, parents, teacher, and school were used for analysis. As a result, the variables of student were found to play a major role in classifying students into the general group rather than the below basic literacy level group. Specifically, the group classification was influenced by the variables related to the various reading strategies, the experiences, frequency and cognition of reading activities, and the difficulty or effort of the PISA testing or survey. Also, other variables related to individual and family backgrounds affected the classification of groups.In addition, by using the random forest, this research confirmed the nonlinear relationship between group classification and affecting variables, taking into account complex relationships among variables. Based on these results, this study suggested the educational supports needed to improve the students' reading literacy.

Random Forest Analysis of Factors Predicting Science Achievement Groups: Focusing on Science Activities and Learning in School

“…Random Forest Since data mining and machine learning techniques have emerged, many researchers have introduced various methods using these techniques to do analyses in the educational field (Peña-Ayala, 2014;Kamath, 2016). In conventional statistical analysis, we follow the steps of building a new hypothesis based on knowledge from the theoretical background and verifying the hypothesis with the data analysis.…”

“…Among these data mining techniques, random forest has demonstrated superior performance, although the actual dataset contains both numerical and categorical predictors. Son et al (2020) explored the major factors according to reading literacy level group using random forest analysis with PISA 2018 Korean dataset. When there is imbalanced data, the results tend to be biased toward a group with a large number of cases, so a method of correcting this by adjusting the sampling rate was used.…”