COMPARING EQUATING ERRORS ON VARIOUS FACTORS FOR SUBTESTS WHICH HAVE ADDED VALUE

In this study dichotomous data which was according with two parameter logistic model (2PLM) was produced for form X and form Y. R 3.1.1. programming language was used to produce data. Each test form had two subtest. Ancor test had two subtest, too. Its’ subtest length was 40% of total test form X (Y). For both forms, corelation between subtests were altered in two level (0.70, 0.80 and 0.90). Moreover, average difficulty difference between subtest of form X and Y were altered in three levels (0.0, 0.4 and 0.7). Simulated forms were equated by using identity, chained linear, Braun/Holland and circle-arc methods for six different sample size (20, 25, 50, 100, 200 and 500) with 100 replications. The results obtained from this simulation study were evaluated based on equating error criterions. The findings indicated in the case when sample size was 100 and more, subtest length was 10, 15 and 30 and the level of average difficulty difference between form 0.0, it was concluded that equating forms would give better results than not equating. Furthermore circle-arc method was found to less equating error than other equating methods under most of the conditions studied.