Creating test materials that are fair to examinees is a critical part of the test
development process. Bias prevention is a central concern because tests are typically
used to gather important information that will support decisions about examinees.
Educational tests conform to professional standards that guide their development and implementation—the most central of which is called "validity." Validity is the property of a test that permits it to accomplish its intended purposes and that supports the meaningful interpretation of scores derived from the test. A valid test measures what it is supposed to measure and does not contain content or constructs that are irrelevant to the goals of the assessment.
Freedom from bias is a requirement for validity. A test is supposed to be a balanced representation, or sample, of the entirety of the knowledge and skills for which examinees are responsible. Tests that measure the domain of knowledge in an unbalanced or biased manner are likely to favor some examinees with particular knowledge or backgrounds.
When a lack of balance—or bias—produces a test that promotes the performance of some examinees and hinders that of others, test results cannot be interpreted clearly. For example, if a test that is supposed to measure knowledge of U.S. geography focuses heavily on one region (e.g., the Southwest), that test is biased because its content is likely to be more familiar to some examinees (e.g., residents of that region) than to others. As a result, scores derived from this test do not reflect examinees' knowledge of U.S. geography as a whole. In other words, the bias in the test prevents the test from measuring what it is supposed to measure, and this affects the validity of the test.
Top of Page