The purpose of this study was to examine the process of conformity change in interpersonal evaluation. It was assumed that perceiving one's own rating as deviating from the group atandard brings a sort of group pressure that tends to make one's re-rating conform to the perceived standard. 5 classes of 11th graders were asked to rate their teachers who took charge of their homeroom and taught them one or more subjects. Three months after ; (1) One of the classes, the control group was asked to rate their teacher again so taht the natural changes in rating scores between the two sets of rating may by be examined. (2) Four small groups, each selected out of the rest of the classes, were respectively exposed to 4 kinds of experimental situations and then asked to rate again. The rating scale comprised two parts, the one of personality and the one of efficiency, each consisting of 10 items, which were selected from some of the classical researches on teachers' competence and the preliminary research on ideal teacher-image entertained by teachers and pupils. The experimental situation was set by ; (1) Suggesting to each subject the simulated group standard of rating. (2) Demonstrating to each subject the simulated group mean score of rating as to some of the 20 items. (3) Demonstrating to each subject the simulated group mean score of rating as to the two parts of the scale. (4) Making 3 pupils who represent favorable, neutral and unfavorable attitudes in rating talk about the personality and the efficiency of their teacher. main results were as follows : (1) Interpersonal evaluation is amenable to group pressure and inclined to toward change conformity when perceived as deviating from the group standard. (2) Both suggesting and demonstrating the simulated group standard may bring about conformity changes in rating scores, as far as the gaps between the subject's rating and the simulated standard are maintained to some degree. The larger the gaps, the more effective is process. (3) It is easier to manipulate interperson 1 evaluation as a whole than partially. (4) A small group of 3 members different in evaluation to a specific person produces a group standard of its own as a natural effect of talking about the personal and efficient value of the person. difference among members decreases in rating scores quite remarkably at the second rating.