Examining the Discrimination of Binary Scored Test Items with ROC Analysis

In this study, it was claimed that ROC analysis, which is used to determine what extent medical diagnosis tests can be differentiated between patients and non-patients, also examine the discrimination of binary scored items in cognitive tests. order obtain various evidence for claim, 2x2 contingency table analysis adapted accordance with logic item discrimination. It suggested article areas under curves (AUC) obtained by using sensitivity specificity values calculated considered as a measure The results statistical analyses made on simulation data showed AUC were positively highly correlated D, ????bis parameter items, from different sized samples consistent. Additionally, more stable against range narrowing than other methods. respect, concluded very large groups not needed proposed method.