Predictors of Korean Students’ Science Achievement PISA 2022: A Post-Selection Inference Approach

• In Event: AER Roundtable Session Friday 1:45 pm - Gold 4
  In Roundtable Session: Understanding Science Achievement Using Large-Scale and Longitudinal Data

Abstract

This study investigated predictors of Korean students’ science achievement in PISA 2022 using a Post-Selection Inference (PSI) approach. A LASSO model with 10-fold cross-validation identified candidate variables, and PSI was applied 100 times to test significance. In total, 38 predictors were selected, spanning attitudes toward science, school climate, digital resource use, parental support, and out-of-school activities. Expected results showed that high achievers enjoyed science, felt safe at school, and used fewer digital resources during lessons. Mathematics achievement was positively linked to science, but students who liked or found math easy performed worse. Unexpectedly, extracurricular activities, including science clubs, correlated negatively with achievement. Digital use had mixed effects, while parental encouragement promoted achievement and excessive monitoring predicted lower performance.

Authors

• Nam-Hwa Kang, Korea National University of Education

• Minjeong Rho, Korea National University of Education