A modeling approach to identify academically resilient students: evidence from PIRLS 2016

Abstract

AbstractIn studies of academic resilience, the concept is typically operationalized by pre-defined cutoff values of students’ achievement level and their social background. A threat to the validity of such arbitrary operationalizations is that students around the cutoff values may be misclassified. The main objective of the current study is to apply a modeling approach to identify academically resilient students. Data come from the Progress in International Reading Literacy Study 2016. The primary method used was latent profile analysis. The study demonstrates that it is possible to identify a group of academically resilient students using a set of variables indicating achievement level and protective factors. Even though results suggest that academically resilient students have higher reading achievement compared to other less-advantaged students, it is evident that they have higher interest and confidence in reading than the non-resilient groups. A sensitivity analysis for other countries indicated similar results. Implications stemming from the results are discussed.