Journal of Personality and Social Psychology 1981. Vol. 40, No. 3. 553-561 Copyright 1981 by the American Psychological Association, inc. 0022-3514/81 /4003-0553S00.75 Performance and Personality Correlates of Teachers' Susceptibility to Biasing Information Elisha Y. Babad and Jacinto Inbar Hebrew University of Jerusalem, Jerusalem, Israel Using Draw-A-Person Test scores attributed to a "high-status" and "low-status" child, groups of teachers of high and low susceptibility to stereotypically biasing information were identified and studied. In paper-and-pencil tasks, the groups differed from each other in responding to (a) some scales of a self-report per- sonality inventory and (b) open-ended educational events pertaining to teachers' failure. The groups did not differ in responses to the dogmatism scale and ed- ucational events pertaining to problems with individual children and to educa- tional ideology. Classroom observations revealed systematic differences in teacher and student behavior between the high-bias and no-bias groups. These differ- ences, indicative of authoritarianism, were validated by supervisors' evaluations, which correlated strongly with the classroom observations. Recently, Rosenthal and Rubin (1978) summarized "the first" 345 studies of ex- perimenter bias. They discussed the exis- tence and magnitude of the bias phenome- non, the major areas in which bias effects were reported, and the conditions that fa- cilitate or inhibit bias effects (see also Ro- senthal, 1971). Although the majority of studies in this area were designed to demonstrate overall bias effects, some researchers have been in- terested in differential effects of bias. We focus here on two differential designs ten- tatively labeled "the behavioral design" and "the personality design." The behavioral de- sign deals with mediation and transmission of bias effects, tracing how experimenters or teachers behave differently toward high- expectancy and low-expectancy subjects or students (see, for example, Brophy & Good, 1970; Friedman, 1967; Rubovitz & Maehr, 1973). The personality design deals with the identification of types of subjects who differ in their susceptibility to biasing information and the personality correlates of such sus- ceptibility (see e.g., Babad, 1979; Laszlo The authors thank Atara Sherman and the staff of the Wingate Teachers Seminar for making this project possible and Robert Rosenthal for his helpful comments. Requests for reprints should be sent to Elisha Y. Ba- bad, School of Education, Hebrew University of Jeru- salem, Jerusalem 24100, Israel. & Rosenthai, 1970; McFall & Schenkein, 1970). The two designs differ from each other as follows: The personality design deals with "suggestibility" to biasing information, whereas the behavioral design deals with "communicability" of bias (see Babad, 1979; Finn, 1972); one focuses on mediating be- haviors, and the other focuses on personality correlates. The behavioral design differen- tiates between the clients of the receivers of biasing information (how teachers in general differ in their behaviors toward high-expec- tancy and low-expectancy students), whereas, the personality design differentiates between the receivers themselves (how high-bias and no-bias teachers differ from each other in general behavioral or personality patterns). Investigators using the personality design typically employ self-report inventories and questionnaires (measuring dogmatism, anx- iety, task orientation, need for social ap- proval, repression, defensiveness, social de- sirability, locus of control, etc.) as correlates of susceptibility to bias. As Laszlo and Ro- senthal (1970) and Johnson (1978) con- clude, the results of such studies are rather weak, inconsistent, and sometimes even con- tradictory. Babad (1979) joins other writers in doubting the ecological validity of self- report questionnaires, arguing that most questionnaires are quite transparent and 553