The Effects of Reusing Written Test Items: A Study Using the Rasch Model
Figure 1
Symbolized data matrices for estimating item difficulty parameters for Hypotheses 1 and 2. Items are in the columns and persons in the lines; shaded areas contain persons’ answers coded as 1 correct, 0 not correct; white areas contain no data, indicating that the person has not worked on the respective items. It is possible to estimate parameters using data from all four item/person groups simultaneously because there is a set of common items.