Differential effects of teacher comments.

Abstract

An analysis of the effects of three kinds of teacher comment (social-comparison, subject-matter, and intraindividuaUy oriented) accompanying grades in mathematics is presented. Subjects wen; 385 students from Grades 6 through 10 who were randomly assigned to one of three groups, each receiving one of the three types of comment, or to a control group. During the first half of the school year, teachers wrote comments concerning students' performance on mathematics examinations; during the second half, comments were not made so as to test the duration of effects resulting from the earlier comments. Students were observed at the beginning of the experiment, at the end of the treatment period, after the no-comment phase, and every time received a grade in mathematics. Dependent variables (besides later performance in mathematics) were cognitive-motivational variables and some school-related personality characteristics, which were selected with reference to a differentiated expectancy-value theory of achievement motivation. In analyses of variance, an interaction between type of comment and the student's prior performance level was revealed for almost all dependent variables. Effects of comments persisted after the experimental period for school-related personality variables only (some aspects of scopespecific locus of control and test anxiety). Implications for teaching strategies as well as for developmental and educational psychology are suggested. The results of the effects of written teacher comments about exam grades on student performance are inconsistent. Page (1958) reported positive effects (see, e.g., Mathis. Cotton, & Sechrest, 1970; Schmidt, 1978). Stewart and White (1976), however, surveyed 12 attempts to replicate Page's findings and found little agreement between Page's major finding and the replication studies. They concluded, somewhat cautiously, that (a) is no strong evidence to suggest that any type of comment retains its effectiveness over an extended period of time or if administered on more than one occasion (Stewart & White, 1976, p. 498); (b) there is no consistent evidence for comment effect at the elementary or secondary school level, although slight evidence is found at the college level; and (c) where comments were effective, they were encouraging and personalized in nature, rather than simple, standard statements (Stewart & White, 1976, p. 498). Recently, Elawar and Corno (1985) found strong main effects on achievement and attitude for individually oriented teacher comments on student homework (not exams). Most of the research has no theoretical base and is unrealistic in the sense of lacking external validity. These features of the research may be the reason for the uneven findings. In this study, 1 developed a theoretical perspective for research on teacher comments and student performance and then tested it in an empirical study.