Exploratory Analysis for Stable Stanines of College Scholastic Ability Test

While current CSAT(College Scholastic Ability Test) score reporting includes percentiles and standardized scores as well as stanines, Minister of Education and Human Resource announced that only stanine score will be reported from 2008 school year CSAT. The purpose of this study is to evaluate several conditions for stable stanines. To complete this goal, we compared stanine proportions according to raw scores, linear standardized scores and normalized standard scores. Also we analyzed effects of item difficulties on stanine scores. We found that there were not much differences of stanine proportions in Korean Language, Mathematics and Foreign Language (English) whether we used raw scores, linear standardized scores or normalized standard scores. But raw scores made stanines more stable in Social Studies/Science/Vocational Education and Foreign Languages/Chinese Characters and Classics since raw scores had more score points compared with linear standardized scores and normalized standard scores. Also we found that stanines approached target proportions when a test included various item difficulty levels(i.e., difficult, moderately difficult and easy). In sum, CSAT stanine scales would be stable when a test consists of various levels of item difficulty and raw scores are used for the final scaling. But this suggestion may not be feasible when a test includes a relatively few items (i.e., 20 items for Social Studies). So it needs to be considered to increase the number of items by combining similar subjects in future.