Based on John Rawls' theory of justice as fairness, a nine-item rating scale was developed to serve as a criterion in studies of test item bias. Two principles underlie the scale: (1) Within a defined usage, test items should not affect students so that they are unable to do as well as their abilities would indicate; and (2) within the domain of a defined usage for any given test item, all students should have equality of opportunity to respond. A group of 316 twelfth-grade students rated a subset of 38 items of a reading test plus five contrived (biased) items. From this group, 247 students identified four or more of the five contrived items as biased. For the reading test, 8 of the 38 items were rated as biased by 30 to 50 percent of the 247 students. Explanations are given for the ratings of these eight items. (Author/CTM)