Procedures to Eliminate Bias Stanford Achievement Test Series
All Panel members were brought to a meeting at Harcourt Brace Educational Measurement to discuss and define the problem of bias. At that time, the purpose, plan, and procedures for item review were established in a workshop setting. Examples of test questions that could be considered biased or stereotyped were examined, and solutions were considered. Following this meeting, Panel members were assigned to review particular content areas and test levels based upon their stated areas of interest, their educational backgrounds, and the special needs of a particular group representation. More reviewers were assigned to the Social Science, Reading Comprehension, and Language subtests than any other subtests, due to the potential for bias that is inherent in Reading and Language passages and in Social Science items that assess history and culture. An effort was made to divide test materials equally among Panel members and to assign a particular subtest across all upper or lower levels to the same individual. The Panel members took the materials with them from the meeting in order to review them over a sufficient period of time. All materials and written comments were returned to Harcourt Brace Educational Measurement. Following this review by the Advisory Panel, items that were found to be objectionable were either eliminated from the battery or revised to eliminate the objectionable content.