PEER EFFECTS AND DETERMINANTS OF ACADEMIC ACHIEVEMENT IN TURKEY: EVIDENCE FROM PISA DATA USING A MULTILEVEL APPROACH.

This paper investigates the determinants of academic performance in Turkey using data from the 2018 PISA survey, relying on a multilevel micro-econometric modeling framework. The analysis combines fixed effects, random effects, and the Mundlak model to address endogeneity arising from unobserved specific effects and to disentangle individual and contextual influences. The findings highlight the importance of individual characteristics such as gender, age, native student status, grade repetition, and parental cultural and socioeconomic capital. Moreover, school-level factors—such as disciplinary climate, classroom socioeconomic composition, school type, and access to technological resources—play a critical role in shaping student achievement. These results are discussed in light of recent educational reforms in Turkey, including the extension of compulsory schooling, the FATIH initiative for ICT integration, and the National Plan for Inclusive Education. While the study acknowledges notable progress, it also underscores persistent challenges related to social, regional, and gender inequalities. It offers insights for enhancing both equity and quality in the Turkish education system through targeted policies and more effective resource allocation.