Key points are not available for this paper at this time.
Hypotheses regarding self-fulfilling prophecies, perceptual biases, and accuracy were tested using longitudinal data relating 98 6th-grade math teachers' expectations to 1,731 students' performance. Consistent with the self-fulfilling prophecy hypothesis, teacher expectations predicted changes in student achievement beyond effects accounted for by previous achievement and motivation. Consistent with the perceptual bias hypothesis, teacher expectations predicted their own evaluations of students' performance more strongly than they predicted standardized test scores
Jussim et al. (Tue,) studied this question.