ABSTRACT
Very few instructors working in higher education, and, for that matter, instructors working at any level of education, receive much, if any, formal training in the construction of classroom tests and their uses. Item analysis is a process of calculating and interpreting statistics for individual items on a test. This chapter addressses the topic of practical item statistics for classroom item review. These analyses are often called studies of 'differential item functioning (DIF)'. Item difficulty is the average performance of students on an item. If the item is scored 0 (for an incorrect answer) and 1 (for a correct answer), one say that the item is 'binary-scored' or 'dichotomously scored'. Item discrimination indicates the capability of a test item to distinguish between high- and low-achieving students on the criterion; something one would like each test item to do. The polychoric correlation is in essence an extension of the tetrachoric correlation for polytomous data.
