Örneklem Büyüklüğünün Test Eşitlemeye Etkisi

Abstract

In this study, equating the scores belonging to SBS 8th Grade Math Subtests, which are done in 2011 and 2012, with linear and equipercentile equating methods under the condition of different sample size and suggesting the most appropriate method and also showing the effect of sample size on test equating were aimed.Population of this study was 517.382 people who taken 2011 and 2012 SBS 8th Grade Math Subtests examinations. Sample of the study was determined by computer randomly and it was composed of 1% of the population. Sample size was 6000. In this study, in order to equate two test forms the common-item nonequivalent groups design was used.According to the findings, it was concluded that with the increase in sample size less faulty equating could be made. In addition, in line with the findings obtained in the study, when using the all sample sizes used in the study the equipercentile equating method should be preferred instead of linear equating method.