The effect of prior contact with the examiner on the Wechsler Intelligence Scale for Children scores of third-grade children

This study is an experimental investigation of 'rapport' which is required in all individual intelligence testing. It was designed to answer the question of whether the experimenter effect is present and can be shown to affect individual intelligence test scores of children who are tested by different examiners. This experimenter effect is defined as that part of the results of experiments which can be attributed to the experimenter, the subject, or the interaction between them. Situational variables are also considered part of that effect. Subjects were tested under three experimental conditions: I, no prior contact; H, cold pretest contact; and III, warm pretest contact. It was hypothesized that (a) there would be a difference in results obtained by different examiners; (b) subjects with prior contact with the examiner would score higher than those with no prior contact; (c) subjects who had had a cold pretest interview would have significantly higher scores than those who had had no contact, but not as high as those with warm contact; and (d) subjects who had had a warm pretest interview would have the best test scores. Sixty third-grade subjects, thirty of each sex, were divided into three experimental groups. The two groups having pretest interviews in either the warm or cold condition spent about fifteen minutes with the examiner several days before the test. Both warm and cold behaviors in this pretest interview were specifically defined in terms of what was to be worn and what verbal remarks were to be made. Behavior was considered warm when the examiner wore a bright dress, and when the child was asked about his pets, his friends, and his favorite games. Careful attention was paid to anything the child said, and whatever he did was liberally praised. Primary reinforcement was offered in the form of an assortment of cookies. Behavior was considered cold when the examiner wore a drab dress or smock, when she did not look at the child, when she asked questions without allowing enough time for answers, when she frowned from time to time, and when things were mentioned which might be considered threatening to the child, such as the PRINCIPAL, the necessity of getting up earlier than usual, questions which he probably could not answer, etc. Verbal comments were deliberately designed to be slightly above his level of comprehension. The testing was done in the Galena Park Independent School District, Galena Park, Texas. The five examiners were females who were experienced in WISC testing. During the actual testing they were instructed to maintain their usual testing 'rapport.' The experimental design used was a 3x2x5 analysis of variance, with three conditions, two sexes, and five examiners; the actual number entered in the cells was the scaled score of the WISC. None of these hypotheses was confirmed. Conclusions to be drawn from this experiment were discussed, as were suggestions for future research. The 'bandwagon effect' was an incidental observation of reaction to 'reported' results of an experiment. This effect may be defined as the tendency of anyone even remotely connected with any research to insist that the results as reported were the results he knew would be found, regardless of the truth of these 'reported' results or whether or not his own part of the data confirms or disproves the reported results when it is analyzed objectively.Psychology, Department o