The purpose of this study was to assess the degree to which teachers' expectations are related to children's social class characteristics. A sample of 96 elementary school teachers drawn from four schools serving lowerand middle-upper class neighborhoods was asked to judge the performance potential and related characteristics, including the SES background, of a set of photographs of both black and white grade school children. While expectations were found to be positively and significantly related to perceived SES of the children in the photos, variations by school SES and number of years teaching experience in the mean number of judgments made suggest that teachers' class bias operates in a more complex manner than is usually thought. In addition, teachers were found to respond more often when choosing children for success than for failure categories. Regardless of perceived SES, white children were more often expected to succeed in school than black children. Recently, Rosenthal (1973) and Rosenthal and Jacobson (1968) presented data supporting the hypothesis that a teacher's expectations of a child greatly influence the child's actual behavior. (Students of whom teachers were led to expect exceptional performance were found generally to live up to the expectations.) They speculated that the artificially created positive expectations the teachers had for certain children in their classes caused them to be more warmly encouraging; to give more feedback; to teach more effectively; and to ask more of these chosen few. Thus the teachers were transmitting high expectations to children who behaved accordingly. Though not directly tested in their study, there is the implied corollary that the teacher who expects poor performance of a child may also communicate this expectancy through her attitudes and interactions with the child, and the child will perform poorly. Since the initial publication, Rosenthal and Jacobson's study has been attacked on points of methodology, procedure, and analysis, and has been criticized as being unreproducible (cf. Barber and Silver, 1968; Elashoff and Snow, 1971; Fleming and Anttonen, 1971). Nonetheless, the hypothesized relationship between expectations and behavior remains convincing, so much so that in some circles it seems to have become a basic tenet in the social psychology of the classroom, no doubt owing in large part to the considerable popularization of the Pygmalion study. (See the book review by Snow, 1969, on the uncritical acceptance and popularization of the study; and Johnson, 1970, especially pp. 141-52.) Furthermore, it is widely assumed that teachers develop expectations about their students on the basis of perceived SES of
Read full abstract