Lecture Elaboration: Rosenthal's Work on Expectancy Effects
One of the criticisms of the use of IQ tests is that they can lead to labeling, which in turn can have a strong effect on behavior. Also, when applying the nature-nurture issue to IQ scores, it must be noted that environmental experiences do not occur independently of inherited factors. For example, if you're labeled as "slow", the environment is not likely to offer as many "intellectually nurturing experiences", relative to someone labeled as "smart." There is no doubt that labeling can have a tremendous effect on the way a person is perceived and treated by others. How "smart" a person is certainly one of this society's more prominent labels. Robert Rosenthal has done a great deal of research on these so-called expectancy effects.
Purpose of the study
Rosenthal is well known for his research on experimenter expectancy effects, the influence that a researcher can exert on the outcome of a research investigation. Perhaps the most famous case of an experimenter expectancy effect is the case of Clever Hans (as discussed earlier in class). Rosenthal and Fode (1963) attempted to demonstrate this phenomenon in the context of controlled laboratory research. In one of his early experiments, he tested the effects of experimenter expectancy on maze-running performance. He had two groups of students test rats, wrongly informing them either that the rats were specially bred to be "maze dull" or "maze bright." In reality, all rats were standard lab rats, and were randomly assigned to the "dull" and "bright" conditions. The results showed that the rats labeled as "bright" learned the mazes more quickly than those labeled as "dull." Apparently, students had unconsciously influenced the performance of their rats, depending on what they had been told. Rosenthal reasoned that a similar effect might occur with teachers' expectations of student performance.
Rosenthal and Jacobson tested children at Oak School with an IQ test, the Tests of General Ability (TOGA) at the beginning of the school year. This test was used because teachers were likely to be unfamiliar with it, and because it is primarily non-verbal, and not dependent on skills learned in school (i.e., reading and writing). In order to create an expectancy, the teachers were informed that the test was the "Harvard Test of Inflected Acquisition," which served as a measure of academic "blooming." Therefore, teachers were led to believe that certain students were entering a year of high achievement, and other students were not. In reality, the test had no such predictive validity.
Eighteen teachers at the school were informed of the students in their classes who had obtained scores in the top 20% of this test. These students were ready to realize their potential, according to their test scores. What the teachers didn't know is that students were placed on these lists completely at random. There was no difference between these students and other students whose names were not on the lists. At the end of the school year, all students were once again tested with the same test (the TOGA). In this way, the change in IQ could be estimated. Differences in the size of the changes for experimental and control group children could serve as an index of any expectancy effect.
Results and Discussion
Rosenthal and Jacobson's results demonstrated expectancy effects. There was a marked difference in IQ test score gains. Students who had been labeled as "ready to bloom" showed greater gains than those who had not been labeled in this way. One interesting qualification to these results was that they occurred only for the youngest children (1st and 2nd graders). No consistent difference in IQ scores was observed in older children. The authors offer a number of reasons for this age difference in expectancy effects. Perhaps younger children are more changeable because of their tender age, or were perceived as more malleable by their teachers. Another possible reason for the age difference was that perhaps younger children are more susceptible to the subtle influences that are characteristic of expectation effects.
Rosenthal and Jacobson's results demonstrate a powerful self-fulfilling
prophecy. Students believed to be on the verge of great academic success
performed in accordance with these expectations; students not labeled this
way did not. Later research has supported Rosenthal's original conclusion,
that teacher expectations can have a substantial effect on students' scholastic
Rosenthal, R., & Fode, K. (1963). The effect of experimenter bias on performance of the albino rat. Behavioral Science, 8, 183-189.
Rosenthal, R., &. Jacobson, L. (1963). Teachers' expectancies: Determinants
of pupils' IQ gains. Psychological Reports, 19, 115-118.