A hozzáadottérték-modellek alkalmazása a tanulói teljesítménymérésben

Absztrakt

The aim of this paper is to review the international literature in the rapidly developing field of value-added analysis of schools. The evaluation of the results from schools seems to be a crucial part of improving the quality of a country’s school system,according to a report by McKinsey et al. (Mourshed, Chijioke & Barber, 2010). In addition, value-added analysis of schools results seems to be a better, fairer way to evaluate schools’ efforts in increasing students’ knowledge and competencies as comparedto analysing the results of a test unconditionally, given that schools’ intake varies considerably. Thus, value-added analysis of schools has become an important issue in many countries lately. However, value-added evaluation of schools is a complex and data-intensive area with many difficult policy and technical questions. This paper lists the most common types of test-based evaluation models and then describes a commonly accepted definition of value-added models and added value of schools. It summarizes the most important issues associated with the introduction of a value-added evaluation system, from the aims and consequences of the system through the requirements on quality and quantity of data to the choice of statistical models used in the analysis.The second part of the paper describes the largest Hungarian longitudinal student and school evaluation programmes, suitable for value-added analysis. The HELP programme in the Centre for Research on Learning and Instruction at the University of Szeged, the Hungarian Life Course Survey conducted by the Economics of Education Research Group at theInstitute of Economics, and the National Assessment of Basic Competencies at the Department of Assessment and Evaluation within the Educational Authority all track students throughout their educational careers with two or more data collection events to ascertain their cognitive skills, demographics and other background variables. Based on these, the National Assessment of Basic Competencies reports three differentvalue-added analyses for every school every year based on the reading and mathematics test booklets used in the study. The paper briefly introduces the models used to compute added value, a simple regression model at school level, a two-level (students within schools) and three-level (students within classes within schools) hierarchical linear model without background variables and a complex three-level hierarchical linear model considering various student-, class-and school-level variables in addition tostudents’ previous results. Together with reports on students’ ability scores and reports on ability scores compared to students’ socio-economic background, these value-added analyses provide schools with a complex and useful description of their students’ proficiencies in reading and mathematics.