Akademik Veri Yönetim Sistemi
JOURNAL OF MEASUREMENT AND EVALUATION IN EDUCATION AND PSYCHOLOGY-EPOD, cilt.3, sa.2, ss.291-296, 2012 (ESCI)
In this research has been used mixed design in Generalizability Theory in performance assessment applications' scores and has been made Generalizability and Decision Study for analyzing reliability of the scores. Research data has collected the 132 second tier students randomly selected 90 studens at a primary school in Kutahya in 2011-2012 spring term. According to the results of the study, person which is subject of measurement has the highest the percentage of total variance description, has seen insignificant as a source of variability in the level of class, as well as the scores of different classes can change partly seen from scorer to scorer. As a result of the Decision Study has been 0.911 G coefficient, phi coefficient that the 0.898, and increase or decrease the number of raters, increasing the number of tasks to reduce the effect was more than the reliability of the measurement.