A novel approach for calculating the item discrimination for Likert type of scales

Item analysis is performed by developers as an integral part of the scale development process. Thus, items are excluded from depending on item prior to factor analysis. Existing discrimination indices calculated based correlation, yet with different response patterns likely have a similar index. This study proposed new index that can be used in Likert type scales and examined its effect results. For this purpose, simulative datasets were generated, according .20, .30 .35 criteria, exploratory was for single factor. Accordingly, it found more variance could explained fewer compared other when .20 criterion slope coefficient suggested study. Similar findings obtained using indices. In context, recommended use additional calculation method