Date

2019-1-24

Author

Cansız, Mehmet
Ozbaylanlı, Bilgehan
Çolakoğlu, Mustafa Hilmi

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

479
views

285
downloads

Cite This

This study aims to investigate the causal effects of school type on the student achievement. Schools type involves two categories: public schools and private schools; whereas student achievement is defined in terms of an overall measure named as Basis for Admission Score (BAS), which is a weighted score reflecting both the grades obtained from courses in each school and the points obtained at nation-wide centralized exams. Factors used as controls in the study include gender of the student, parental attributes (are they alive and live together; their levels of education and occupations), type of the house (own/rent/public housing), separate room of the student, city the family lives in and the level of development of the geographical region. This study utilizes a dataset comprising 3,752,374 secondary school students, which covers all of the student population of Turkish secondary schools within 2014-2016 period. This comprehensive dataset is utilized for the first time in such a study. At the first step, we present the literature on the effects of school type on academic achievement measured by test scores. This literature can be traced back to 1960's. On the other hand, a second line of literature, which focuses on the evaluation of causal effects of policies and of programs has progressed swiftly starting from 1980's and promises an important methodological framework for evaluation of policies and programs in education science. The methodology of this study is designed by bringing together the mentioned two lines of research. Methodology involved the application of regression adjustment, inverse probability weighting, and exact matching techniques in a complementary style in order to ensure the robustness of estimations. Under the assumptions discussed in detail in the study, we have found that school type has significant impact on student achievement in Turkey. Being a private school student instead of public school student leads to 87 points increase (29.6%) on average in BAS score. It is found that school type has a comparatively larger effect in Turkey compared to other country examples. Based on the findings of the study, a set of research topics are suggested with the objective of improving equality of opportunity in education and of identifying new policies to improve quality of education in public schools.

Subject Keywords

Education, School type, Public school, Private school, Student achievement, Causal inference, Evaluation, regression adjustment, Inverse probability weighting, Exact matching, Doubly robust estimation

URI

https://hdl.handle.net/11511/52177

Collections

Department of Economics, Article