Problem In the present study, response style was defined as any tendency or habit which results in a score different from the one that the testee would earn when the items presented in a different form" Among the various response styles reported in the literature, two particular styles, social desirability and acquiescence, were selected for an examination in this study since most of the previous studies in the field have given much attentions to these particular styles and recognized as most relevant variables in personality testing. The major purpose of the present investigation was to find some possible answers to the following questions: (1) To what extent personality test scores are contaminated by the effects of response styles such as social desirability and acquiescence (2) If acquiescence response style is operated in the process of response choice, does it exist regardless of social desirability content of the items? (3) Are response styles determined by subject variable or item variables? Hypotheses The following experimental hypotheses were identified for testing in the present study. The directions of the intercorrelations among 10 experimental scales will be predicted as indicated in Table A. (Table A 참조) In Table A, the constructs of 10 experimental scales were as follows: Scale 1: consisted of 30 items of socially desirable contents and the direction of scoring is "True." Scale 2: consisted of 30 items of socially desirable contents and "False" was taken as the scoring direction. Scale 3: consisted of socially undesirable contents and "True" was taken as the scoring direction. Scale 4: consisted of 30 items of socially undesirable contents and "False" was taken as the scoring direction. Scale 5: consisted of 15 items of socially desirable contents in which "False" was' taken as ehe scoring direction and 15 items of socially undesirable contents in which "True" was taken as the scoring direction. Scale 6: consisted of 15 items of socially desirable contents in which "True" was taken as the scoring direction and 15 items of socially undesirable contents in which. "False" was taken as the scoring direction. Scale 7: consisted of 15 items of socially desirable contents and 15 items of socially undesirable contents and all items were' scored as "True." Scale 8: consisted of 15 items of socially desirable contents and 15 items of socially' undesirable contents and all items were-scored as "False." Scale 9: consisted of 30 neutral items with social, desirability content and "True" was taken as the scoring directions. Scale10: consisted of 30 neutral items with social desirability conten tand "False" was taken as the scoring direction. Method The sample for the study consisted of 67 male and 93 female students drawn from the College of Education, Seoul National University and the College of Education, Ewha Women's University respectively. The instrument used in the study was the: personality inventory consisted of 10 experimental scales which was specifically for the study constructed using selected items from MMPI on the basis of social desirabilty scale value. In order to construct 10 experimental scales, 10 judges rated each item in view of social desirability content and classified them thus to constitute 10 different scales. The nature of each experimental scales is presented along with the hypotheses of this study. The measure was administered under the standard instruction in group situation. Results The obtained intercorrelations among 10 experimental scale scores show that the effect of social desirability response style is present in the cases of positive or negative correlations predicted by social desirability content, while the effect of acquiescense response style is in the cases of zero or chance' correlations predicted. It is shown that the testees have a tendency to respond "True" in socially desirable items and "False" in socially undesirable items. It is noticed, however, that the tendency to reject socially undesirable items is higher than tendency to endorse socially desirable items. It is also noticed that acquiescence tendency is present in the neutral scaes (scale 9 and 10) and the balanced scales with social desirability content (scale 7 and 8). On the basis of the findings, it can be concluded that the scores assessed by personality inventories are heavily contaminated by the effects of response styles such as social desirability and acquiescence tendency. At the same time, it seems reasonable to conclude that social desirability style is operated especially in the items which disclose the social desirability content, while acquiescence tendency in the neutral items with social desirability content. The implications of the present study and the need for farther study in this field were discussed.