Bulgaria - Piloting statistical models for estimation of schools’ value-added using the results from the national assessments : key results and findings (Vol. 1 of 2)

BULGARIA Piloting Statistical Models for Estimation of Schools’ Value-Added Using the Results from the National Assessments Key results and findings March 2015 WORLD BANK EUROPE AND CENTRAL ASIA REGION EDUCATION Acknowledgements This report was prepared under the Programmatic Education Sector Technical Assistance Program of the World Bank (WB) in response to the request of the Bulgarian Ministry of Education and Science (MES) to pilot and document the implementation of a set of statistical models for estimation of the value-added of Bulgarian schools using data from the national census-based student assessments. The pilot and the report were prepared by a team led by Plamen Danchev (Education Specialist, World Bank) and comprising of the following researchers from the Bulgarian Association for Educational Measurement and Evaluation (BAEME): prof. Kiril Bankov (Lead Consultant, BAEME), Vessela Stoimenova (Consultant, statistical analyses, BAEME), Dimitar Atanasov (Consultant, statistical analyses, BAEME). The scope of the pilot implementation, the key parameters for the activity and the required data for the analysis were jointly defined by the WB and a working group of MES experts led by Mr. Krassimir Valchev, Secretary General of the MES and comprising Evgenia Kostadinova (Director for Curriculum and School Programs, MES), Orlin Kouzov (Director for IT in Education, MES), Neda Kristanova (Director, CKOKUO – National Testing Agency), Penka Ivanova (Director of Policy Directorate, MES), Sashko Arabadjiev (Expert, CKOKUO), Rumiana Tomova (Expert, MES), Stela Mitsova (Expert, MES) and Radosveta Drakeva (Adminsoft, developer of the Education Management Information System of the MES). Summary At the request of the Ministry of Education and Science (MES), the World Bank (WB) has expanded the scope of the schools value added measurement pilot using test results from the national census based student assessments. The first pilot was implemented in 2013. The present report summarizes the results of the second phase of the pilot. The value added measures (VAM) of school performance try to isolate the contribution of the schools to the academic growth of the students from the factors that are beyond the control of schools teachers and management. As such, the VAMs are arguably among the most precise and equitable test-based measures of school performance, as long as their estimation is based on valid and reliable student assessments and appropriate statistical models. The results from the present pilot show that the three tested models (a fixed effect linear regression model, a random effects model and a random effects model benefitting from the inclusion of the language spoken at home by the observed students) could be used for the purposes of estimating the value added of Bulgarian schools, but the model with the most advantages is the random effect model using the “language spoken at home