Abstract
In a previous study the labeling effect of ability grouping was definitely found to exist. A second experiment, similar in detail to the first, failed to indicate the operation of the labeling effect. The primary difference between the two studies was that in the first experiment the teachers did not know that the sections were grouped randomly, but in the second the instructors knew that their students had not been grouped according to ability. This indicates that a primary factor in the production of the labeling effect is the teacher’s own reaction to his perception of pupil ability rather than the pupil's reaction to teacher expectation.