The concept of the teacher expectancy effect continues to have a major impact on educational and psychological thinking. This article reviews the literature and concludes that the teacher expectancy effect, like the experimenter expectancy effect, is more difficult to demonstrate and less pervasive than has been claimed. It suggests that the most relevant issue is determining the relevant child characteristics that significantly cause a particular teacher to form an expectation about a particular child.L'effet produit par l'espoir que place un professeur en son 61~ve continue d'exercer une influence capitale sur la pens6e et la psychologie 6ducative. Cet article passe en revue les 6tudes consacr6es ~t ce sujet et conclut que l'effet en question, tout comme dans le cas d'un exp6rimentateur, est plus difficile ~ prouver et moins g6n6ralis6 qu'on ne le pr6tend. L'auteur estime que le probl~me capital consiste ~t d6terminer chez l'enfant les caract6ristiques valables qui incitent un professeur donn6 ~ fonder particuli~rement certains espoirs sur un enfant donn6.Editor's Note: The next issue of Interchange will contain comments on Dr. Grieger's article by Robert Rosenthal and a reply by Grieger and Saavedra.It is a fact of life that people behave in predictable ways. Societal norms, group pressures, and inner expectations of success or failure, among other things, serve to order and control a person's behavior. In addition, evidence suggests that one person's expectations concerning another's behavior may serve to increase the probability of the other's behavior, in other words, to act as a self-fulfilling prophecy (Merton, 1948).Most of the evidence concerning the self-fulfilling prophecy comes from laboratory situations where researchers have claimed to demonstrate that the expectancies of a behavioral scientist can significantly influence the outcome of his experimen,ts. These instances of the self-fulfilling prophecy have been termed the experimenter expectancy effect (Rosenthal, 1966). As illustration, Rosenthal and Fode (1963) studied the maze running of rats. Using an experimental psychology class, the authors told one-half of the experimenters that their rats were "maze bright" and the other half that their rats were "maze dull." Actually, all rats were ordinary laboratory rats. Results showed that rats believed to be "maze bright" performed significantly better ,than the other rats (p<.01). Further, those experimenters who believed they had bright rats viewed their rats as more likable and pleasant, and described their behavior toward the rats as more friendly and enthusiastic. From this and similar experiments, researchers claimed the experimenter expectancy phenomenon to be both powerful and general. Barber and Silver (1968), however, argued that this conclusion is totally unwarranted. They analyzed the design and data of 31 experimenter expectancy studies and concluded that only 12 unequivocally demonstrated the phenomenon. For the other 19, they were particularly critical of (1) failures to perform overal...