Abstract
This paper reports the effect that background knowledge has on the item discriminability of reading comprehension items; item discriminability is an index of how well a test item discriminates bet ween weak and strong examinees in the ability being tested.
The questions from three reading comprehension tests were placed into one of the following categories: (1) textually explicit; (2) textually implicit; and (3) scriptally implicit. An item discriminability index was computed for each item. The results indicated that there were significant differences in the three categories of itèms with respect to item discriminability and that reading comprehension items which are heavily dependent on background knowledge do not ex hibit good item discriminability. The findings indicate that researchers must control for background knowledge in a reading test when they use test products as the basis for making inferences about the pro cesses underlying reading comprehension.
Get full access to this article
View all access options for this article.
