Year: 2024
Type: article
Abstract: The main objective of secondary education institutions is to provide quality education to its students. One way to achieve this is by introducing various teaching methods, one of which is tandem learning, which is a small-group cooperative learning method. Not everyone responds well to a one-size-fits-all method, and therefore, uncovering insights for predictive model selection tailored to individual students or classrooms becomes imperative for teaching institutions. The knowledge is embedded in the educational data set and is extractable through data mining techniques. The primary objective of the study was to identify the key factors that significantly influence student outcomes (including both emotional well-being and knowledge improvement) in tandem learning using machine learning algorithms. The study was conducted in a mathematics class during the course of one week of tandem learning implementation in the school year 2023/24 with a sample of 89 high school students from a selected Slovene high school and 13 predictor variables (gender, class, teacher, recent mathematics grade, MBTI variables, mathematical anxiety, motivation, qualitative interaction, quantitative interaction, and whether the student outperformed their partner). The outcome of interest was a three-state dependent variable indicating whether the student responded well to the implementation of tandem learning into the education environment. The present study tested which predictor variables were most important using mutual information and recursive feature elimination for all variables. The most important factors according to mutual information for predicting student response were outperforming the partner, class, and qualitative interaction within the tandem and according to recursive feature analysis qualitative interaction, outperforming partner and gender.
Authors: Bor Bregant, Daniel Doz, Sanela Hudovernik
Institution: University of Primorska
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W4404949412