The Effect of Self-Interest on Scores Made on the Allport Test for Measuring Ascendance-Submission
- Author:
Charles W. Manzer
New York University
In any group of subjects to whom the Allport Test for Measuring Ascendance-Submission has been given there are frequently found individuals whose personality types seem to be in sharp contrast with the types indicated by the scores made on the Allport Test. Such discrepancies between estimates and scores seem to indicate a possible divergence between the possession of certain personality traits and acquaintance with the social implications of these traits. The question arises, To what extent, if at all, are scores on the Allport Test affected by self-interest ?
An answer to this question could be obtained from a comparison of scores on the Allport Test with two different kinds of measures of ascendance-submission: the first, a measure that would be independent of the subject’s knowledge of the social significance of ascendance-submission, and the second, a measure that would be a direct expression of the subject’s understanding of the social value of this personality trait. In the absence of any such measures it has been thought that an approximate answer to the question stated above could be secured by comparing scores on the Allport Test with (a) ratings by associates and with (&) self-ratings. Ratings by associates are independent of the self-interest of the individual rated, while self-ratings have been found to be considerably affected by the individual’s self-interest. Since in self-ratings subjects have been found to flatter themselves, it would be expected that selfratings for ascendance-submission would avoid the extremes of this trait (very submissive and very ascendant) and would consequently cluster about the mean as the most favorable rating.
The subjects of this study were sophomore and junior students in three sections of introductory psychology. The three groups numbered 37, 45 and 46 students, a total of 128 subjects. The agerange was from 18 to 24 years. There were 62 women and 66 men in the total group. Each student rated every other student in his group for ascendance-submission on a five-category scale. These catagories ranged from (1), most submissive to (5) most ascendant. Each student also rated himself. After the ratings were completed the Allport Scale for Measuring Ascendance-Submission was given to the subjects in each group.
Treatment of Results
For the purpose of making comparisons the group ratings, the self-ratings and the scores made on the Allport Test have been converted into standard scores. On the basis of the standard scores in the group ratings the subjects have been divided into five subgroups, ranging from the most submissive fifth to the most ascendant fifth. Table 1 contains the mean of the standard scores of group ratings, self-ratings and Test scores of each of the sub-groups. In Table 1 the mean standard score of the sub-group rated as most subTable I Means of Standard Scores of Group Ratings, of Self-Ratings and of Scores Made on the Allport Test for Each of the Sub-groups
I N = 25 II N = 25 III N = 28 IV N = 25 V N = 25 Range I to V Group Ratings., Self-Ratings… Test Scores…. -1.35 -0.43 -0.54 -0.61 -0.15 -0.28 -0.04 -0.21 -0.02 +0.58 +0.26 +0.26 + 1.42 +0.80 +0.86 2.77 1.23 1.40
missive (sub-group I) is 1.35 S.D. below the mean of the total group. The mean standard score of the self-ratings of this sub-group is 0.43 S.D. below the mean, while the mean standard score of this sub-group on the Allport Test is 0.54 S.D. below the mean. The mean standard score of the sub-group rated as most ascendant (subgroup Y) is 1.42 S.D. above the mean; the mean standard score of the self-ratings of this sub-group is 0.80 S.D. above the mean and their mean test score is 0.86 S.D. above the mean. Comparisons in the other sub-groups are similar to those in the two extreme groups but they are less marked. In general the foregoing comparisons show that self-ratings are most closely grouped about the mean, the Test scores are somewhat more widely scattered while the group ratings are the most widely scattered of the three measures. It will be seen (Table 1) that the range of the standard scores of the group ratings between the means of the two extreme sub-groups is from ?1.35 to -+ 1.42 S.D. or a total of 2.77 S.D. The range of the self-ratings is from ? 0.43 S.D. to + 0.80 S.D., a total of 1.23 S.D.; in the case of the test scores the range is from ? 0.54 S.D. to + 0.86 S.D., a total of 1.40 S.D. The widest range is that of the group ratings, the narrowest range is that of the self-ratings while the range of the Test scores is intermediate between these two ratings. The range of the Test scores is, however, nearer that of the self-ratings than that of the group ratings. This comparison of the three ranges shows that the Test scores more closely resemble the selfratings than the group ratings; that is, the Test scores more closely resemble the more “flattering” of the two kinds of rating with which they have been compared. The inference seems to be justified that scores made on the Allport Test for Measuring AscendanceSubmission are subject to the same tendency to be affected by the subject’s knowledge of the social value of ascendance-submission (self-interest) as are self-ratings for this personality trait.
Summary
Scores on the Allport Test for Measuring Ascendance-Submission have been compared with (a) group ratings and with (&) selfratings. The distribution of the Test scores is nearer that of the self-ratings than it is that of the group ratings. The inference is drawn that the tendency to make “flattering” scores, admittedly present in self-ratings, is also present to a considerable extent in the scores made on the Allport Test.
Disclaimer
The historical material in this project falls into one of three categories for clearances and permissions:
Material currently under copyright, made available with a Creative Commons license chosen by the publisher.
Material that is in the public domain
Material identified by the Welcome Trust as an Orphan Work, made available with a Creative Commons Attribution-NonCommercial 4.0 International License.
While we are in the process of adding metadata to the articles, please check the article at its original source for specific copyrights.