The results obtained by three test administration modes were examined with a focus upon systematic bias due to lack of standardization of test administration procedures. The My Class Inventory was administered to ninety classrooms of fifth grade students in a large suburban school district. The 45 items on the inventory are distributed over five factors: satisfaction, friction, competitiveness, difficulty, and cohesiveness. The test was administered in three ways: (1) by the teacher; (2) by outside test administrators; and (3) by children, using pre-recorded tape cassettes. Significant differences were found for factors of satisfaction, friction, and cohesiveness, but there were no significant differences among methods on competitiveness and difficulty. There were factor differences and overall differences between teacher administered and the other techniques. There were no significant differences between outside administrator and the student cassette tape method. Student tape cassettes were the preferred technique because they minimized teacher bias, were inexpensive, and standardized directions and procedures. Results also indicated that the pre-recorded taped directions administered by students more closely approximated the results obtained by an outside evaluator than did the results obtained by teacher administration of the same instrument. (Author/JAC)