Akademik Veri Yönetim Sistemi
PAMUKKALE UNIVERSITESI EGITIM FAKULTESI DERGISI-PAMUKKALE UNIVERSITY JOURNAL OF EDUCATION, cilt.0, sa.53, ss.110-149, 2021 (ESCI)
This study aims to determine the individual and socio-economic variables that affect the secondary school-age population to stay out of school by gender in Turkey. Microdata sets of the household budget surveys conducted by the State Statistical Institute (SIS) in 2003 and by the Turkish Statistical Agency [TUIK] in 2011 and 2016 years are used to study of the change in the historical process of the effect of the selected variables. Since the dependent variable is binary and discrete, Binary Logistic Regression Model is used to analyze the data. In the study, the goodness of fit of the model was examined through the so-called R-2 value. Cox-Snell and Nagelkerke R-2 statistics were used to measure the relationship between the dependent and independent variables. It turned out that independent variables are significant predictors, and their effect levels differ by year. While the variables of disability status, health assurance, and education level of parents are effective predictors for secondary school age boys to stay out of school in the data for every three years; it turned out that settlement unit in 2011 data, household size in 2016 data and annual income per capita in 2003 and 2016 data are the effective predictors. On the other hand, as the variables of disability status, settlement unit, education level of parents, and household size are significant predictors to secondary school age girls to stay out of school in the data for every three years; it turned out that health assurance in 2003 and 2011 data, annual income per capita in 2003 and 2016 data are the effective predictors.