The effects of self-confrontation on the self-concept

After the Semantic Differential scores of 408 students enrolled in basic speech course at The University of Houston in the fall, 1971, were factor analyzed, two factors emerged as the dimensions on which the target concept, self-concept, could be measured. The same instrument was used to measure the 80 subjects who participated in this study. To determine whether a delay of one month after self-confrontation by videotape before measurement, the presence of a supportive companion during and after self-confrontation the sex of a subject, or an interaction between any of these variables affect the self-concept after self-confrontation, 80 subjects were stratified by sex and randomly assigned to one of four treatments. All subjects were videotaped under constant conditions while making a five-minute expository speech before an audience. Treatment thereafter was governed by the group assignment of the subjects. Subjects in the first group were shown their videotape by the female counselor, and left alone for ten minutes to complete an unrelated paper and pencil task used to equate the time spend with other subjects, before measurement with the Semantic Differential. Subjects in the second group were given supportive companionship. As soon as the videotape began, the counselor took the seat near the subject. She engaged in the following behaviors that constitute supportive companionship: (a) sitting beside the subject during the videotape replay and listening attentively to the speech, (b) inquiring about the feelings of the subject during the replay, (c) listening attentively and empathetically to the subject's responses, and (d) interspersing positive comments about the subject's behavior, voice quality, gestures, appearance, speech organization, topic choice, or whatever was commendable for ten minutes after the videotape finished playing. Then the measurement was made with the same instrument. Subjects in the fourth group were treated as those in the third, except that they returned for measurement in a month. The scores of subjects were multiplied by the factor weights assigned by the factor analysis, and then summed to produce two scores for each subject, one on the dynamism and one on the evaluative dimension. Two three-way analyses of variance were computed, one for each factor. Tests for homogeneity of variance and normality of distribution were made. The results, at the .05 level of significance, indicated that null hypotheses stating that the timing of measurement. the presence of supportive companionship, the sex of subjects or an interaction of any of these variables would make no significant difference in the self-concept scores were not rejected with one exception. The self-concept scores of. undergraduate males and females were significantly different on the dynamism factor of self-concept. Female subjects rated themselves significantly less dynamic than male subjects rated themselves