Files
Abstract
Assessments are used to measure test takers knowledge in a particular area. In this study, the Credibility Interval Approach using a Rasch model was used to identify differential item functioning (DIF) on an undergraduate chemistry achievement test. Results from this test can have impacts such as course placement and grade distribution. DIF, a psychometric characteristic that occurs when people from different groups who are matched on some latent trait have a significantly different probability of giving a certain response to a test item, is present for a portion of the items on this test. Item difficulties and 95% posterior credibility intervals of the items were analyzed to determine which items exhibited DIF. Items that were shown to have DIF were further explored to assess whether males or females were advantaged. In addition, it was determined whether DIF was consistent by group, test form, or concept grouping.