Experimenter Expectation Affects Results

By William Broad and Nicholas Wade

From Betrayers of the Truth: Fraud and Deceit in the Halls of Science (Simon and Schuster, 1982)

From the chapter SELF-DECEPTION AND GULLIBILITY pp. 109-110


...But the habit of "blinding" the experimenter has not become as universal in science as perhaps it should. A dramatic demonstration of experimenter expectancy has been provided in a series of studies by Harvard psychologist Robert Rosenthal. In one of his experiments he gave psychology students two groups of rats to study. The "maze-bright" group of rats, the students were told, had been specially bred for its intelligence in running mazes. The "maze-dull" group were genetically stupid rats. The students were told to test the maze-running abilities of the two groups. Sure enough, they found that the maze-bright rats did significantly better than the maze-dull rats. In fact there was no difference between the maze-bright and maze-dull animals: all were the standard strain of laboratory rats. The difference lay only in the students' expectancies of each group. Yet the students translated this difference in their expectancies into the data they reported. Perhaps some of the students consciously invented data to ac- cord with the results they thought they should be getting. With others, the manipulation was unconscious and much more subtle. just how it was done is rather hard to explain. Perhaps the stu- dents handled more gently the rats they expected to perform better, and the treatment enhanced the rats' performance. Per- haps in timing the run through the maze the students would unconsciously press the button on the stopwatch a fraction too early for the maze-bright rats and a fraction too late for the maze-dull animals. Whatever the exact mechanism, the re- searchers' expectations had shaped the result of the experiment without their knowledge. The phenomenon is not just a pitfall for laboratory scientists. Consider the situation of a teacher administering IQ tests to a class. If he has prior expectations about the children's intelli- gence, are these likely to shape the results he gets? The answer is yes they do. In an experiment similar to that performed on the Psychology students, Rosenthal told teachers at an elementary school that he had identified certain children with a test that Predicted academic blooming. Unknown to the teachers, the test was just a standard IQ test and the children identified as "bloomers" were chosen at random. At the end of the school year, the children were retested, by the teachers this time, with the same test. In the first grade, those who had been identified to the teachers as academic bloomers gained fifteen IQ points more than did the other children. The "bloomers" in the second grade gained ten points more than the controls. Teachers' expectancies made no or little difference in the upper grades. In the lower grades, comments Rosenthal, "the children have not yet acquired those reputations that become so difficult to change in the later grades and which give teachers in subsequent grades the ex- pectancies for the pupil's performance. With every successive grade it would be more difficult to change the child's reputa- tion.



Please contribute today - buy our books - and spread the word to all your friends!

* * * Back to the Home Page of John "Birdman" Bryant, the World's Most Controversial Author * * *