Şans başarısı için düzeltme formülü kullanılacağına ilişkin yönergenin testin psikometrik özelliklerine etkisinin araştırılması
Abstract
This study is a quasi-experimental research which aims to find out if there is a statistically significant difference on the test and item statistics when a multiple choice mathematics achievement test applied with or without a test instruction that indicate change formula scoring will be used. The research has been carried on 452 first year students of 14 vocational health schools. The measurement instrument developed by the researcher, constitutes 40 five-choice questions in a mathematics achievement test for first year classes of high school. Test has been applied on the last 3 weeks of the 2000- 2001 academic year and the academic scores of the study group for mathematics subject have been obtained from the school administrations. Findings point out that giving formula scoring instruction does not influence mean test scores, test validity and item difficulty indices but increases the KR-20 internal consistency reliability and item discrimination power values. High correlation like 0.99 between corrected (chance success has been cleared) and number-right test scores has been found.