Level of aspiration as a case of judgmental activity in which ego-involvements operate as factors:[1] An experimental study of interpersonal relations.[2]
O. J. Harvey and Muzafer Sherif
University of Oklahoma
Introduction
When research areas developing more or less separately are brought together in a functional way, the findings and theory of both gain. In the present study, the level of aspiration is conceived as but one instance of judgmental activity in which affective factors enter into the determination of response. Since judgment, ego-involvement, level of aspiration are more frequently found apart than together, a brief statement of the general conceptual scheme may be in order.
The experimental investigation of judgmental activity, or discrimination, is one of the more elaborated areas in psychology. In reviewing the work on psychophysical, affective, social, and personal judgments, McGarvey (12) concluded "that certain fundamental processes and principles of judgment may be assumed to be independent of the type of stimulus-material and of the dimension in which the judgment is rendered." (p. 79).
One of these general findings concerning judgmental activities is their referential nature. Chapman and Volkmann (1) called attention to this "general fact that all judgmental activities take place within . . . referential frameworks" (p. 225), in harmony with Sherif's earlier conclusion (19, pp. 34-35). In judging physical stimuli, such as weights or lines, judgment is usually made 'in reference to certain anchoring points or reference points in the stimulus series (21). For example, these reference points may be introduced by the experimenter as a "standard" or comparison stimulus; or they may be the end points within (and, within limits, outside) the physical scale itself, etc. In judgments of social stimuli, the situation is complicated by a multiplicity of factors. Many of these factors are
( 121) internal or internalized, i.e., they are intervening variables. Physiological needs of the organism, emotional states, social attitudes are examples of such internal or, as in the case of social attitudes or motives, internalized factors. (Of course, even in judging weights, the standards established by past experience as, say, a weight lifter or college professor must also be taken into account in the total picture.)
In judgmental and perceptual situations in which factors both internal and external to the individual enter in a complicated way to determine response, analysis is more adequate in terms of an entire frame of reference in which various factors, external and internal, carry differing weights in the resultant reaction (19). Within this frame of reference, certain factors will have more weight or be more salient than others in different situations.
The concept of frame of reference, which is used simply to denote this functional interrelatedness of factors internal and external to the individual determining reaction at a given time, is not specific to judgmental activities. The bulk of evidence in the study of perception, judgment, memory, attitudes, ego-involvements and a variety of social behavior clearly indicates its general utility in the study of human behavior (19, 20) .
In different situations, the relative contribution of the external factors and internal factors varies. It is known that the relative prominence of external and internal factors varies with the degree of structure of the external stimulus situation. For example, in the autokinetic situation used by Sherif (18) to study these and related problems, the individual faces a tiny pinpoint of light in a completely dark room. In such unstructured situations, factors coming from the individual dominate the frame of reference within limits. The studies of Luchins (8, 9), Murphy and his students (6, 15) are among the pioneer studies in this area. The various "projective" tests, such as the Rorschach ink blot test; take advantage of such unstructured situations to study the effects of internal factors.
On the other hand, a graded scale of weights, or lines, or tones has more definite objective structure. When the individual faces such structured situations, internal factors carry relatively less weight in the reference frame, while the physical characteristics of the stimulus situation predominate. A more or less normal adult cannot ignore the compelling structure of the Empire State Building or Rockefeller Center.
One of the psychologist's principal tasks is to envision and conduct experiments, varying the stimulus situation and this factor then, that one later, to determine the relative weights of internal and external factors which are functionally interrelated at a given time in the structuring of experience
(123) and hence, response. The present investigation was conceived as another unit carried out within this conceptual scheme.
In daily social life, it is well known that our judgments are influenced by our relationships with other persons, groups, institutions and the like. It is a common human experience that we tend to overestimate the potentialities and actual accomplishments of those high in our esteem, and tend to minimize the potentialities and accomplishments of our rivals, competitors, and enemies.The effects of ego-involvements as dominant factors in the frame of reference may be illustrated by investigations of judgment. In an early study (1915), Cogan, Conklin and Hollingworth (2) found overestimates and underestimates in self-ratings according to the desirability of the "trait" in question. Marks' well-known study of judgments of skin color by Negro subjects (10) revealed that the subject's relation to the desired norm of light brown more or less determined the scale he established in judging other people's coloring.
Chapman and Volkmann (1) explicitly brought research on level of aspiration into functional relation with judgment: "The conditions which govern the setting of a level of aspiration (Anspruchs-niveau), in the sense of an estimate of one's future performance in a given task, may be regarded as a special case of the effect upon a judgment of the frame of reference within which it is executed." (p. 225) In their study, experimentally introduced anchorage points (scores on the task purportedly made by literary critics and W. P. A. workers) effectively lowered or raised the aspiration level. Several investigators confirmed their findings concerning the effect of "superior" and "inferior" standards upon estimate of future performance. (For example, Festinger (3), Preston and Bayton (14), and MacIntosh (11).)
This same conception of level of aspiration as an instance of the effect of the frame of reference within which judgment is executed guided Sears' study of academically successful and unsuccessful school children. In this case, the children's estimates of their future performance were made in terms of internalized "cultural pressure to excell and to keep the performance improving" and the child's awareness "of the position of the self relative to social norms" for performance (16, p. 528).
In a more recent study by Himmelweit in England (5), using neurotic patients in a hospital, it was noted that "In the case of the anxious and the depressed patient, no outside group norm is imposed---it is rather an interiorized one, based upon the conception that the patient has of his ability in
(124) relation to those of the group. Since he considers himself inferior . . . he behaves as if his performance had been. compared with that of a superior group." (p.57)
The general fact stated above that all judgments, including estimate of one's own performance, are made within referential frameworks was clearly substantiated in a review of the "level of aspiration" studies in 1944 by Lewin, Dembo, Festinger and Sears (7) . In accord with the earlier formulations of Chapman and Volkmann and of Sears, the authors conclude that such influences as temporary situational factors, standards on one's own and other groups, etc., upon the setting of an aspiration level "may be conceived of as frames, involving a scale of values, within which the individual makes his decision as to a goal." (p. 57)
The studies summarized here, representative of a body of experiments, indicate that ego-involvements may enter the total frame, of reference as dominant factors in judgmental activity, whether it involves a specific stimulus situation with social value or a future situation, as in the aspiration level studies. When judgmental activity is so viewed as taking place within a frame of reference involving both external and internal factors, demarcations between "fact" judgments and "value" judgments (or between the affective and cognitive) fade away. We may speak rather, of various degrees and kinds of internal factors (e.g., needs, ego-involvements, etc.) and of external factors (the concrete. stimulus situation, presence of other persons, etc.) operating in the judgmental situation.
The Problem
In everyday life relationships, people are seldom impartial in their judgments and expectations of the activities of others, especially if the individuals stand in some more or less established relationship (positive or negative) to one another. This common observation has been neatly demonstrated in a laboratory situation by Zeaman (22). In the autokinetic situation, a naive subject made judgments of the distance of apparent movement, first with a person for whom he felt a good deal of affection and later with a person toward whom he tended to be antagonistic. The finding was that the naive subject's judgments tended to converge toward those of the person of whom he was fond and to shift away from those of the person toward whom lie was antagonistic. Thus, the direction of ego-involvement (positive or negative) determined the course of the judgments.
McGehee (13) compared individual's judgment of their own future performance and their judgments of another person's future performance in
(125) a simple dart throwing situation. Subjects were paired, no point being made of the relationship between subjects of the pairs. In this situation, he concluded that the subjects were more ego-involved in estimating their own performance than in estimating the other person's. This differential ego-involvement was reflected in the two judging situations.
On the other hand, when individuals are positively involved with one another, no such differential trend in judgments is found. In C. Sherif's study (17) [3] , subjects in a similar dart throwing situation were pairs of husbands and wives or of parents and children. Positively related to one another as relatives and loved ones, the subjects tended to be as ego-involved in their partner's performance as in their own. This ego-involvement was reflected in characteristically similar judgments of their own future performance and of their partner's (mate, parent, or child's) future performance. Discrepancy of judgment of future performance from actual performance scores and variability of judgments were highly similar (with no significant differences) for judgments of own performance and of the partner's performance. In some cases, the degree of involvement manifested was actually greater for the other person than for one's self. These findings were supported by the spontaneous remarks of encouragement and support made to the partners. In some cases these remarks were more tense and emotionally toned than remarks concerning own performance. Although no systematic attempt was made to determine the degree of positive involvement between the subjects, and hence differences from pair to pair were not individually analyzed, certain uniformities in the behavior of the various pairs were noted. In various degrees the husbands tended to be protective toward their wives, encouraging and comforting them. The wives, on the whole, tended to maintain an admiring role in relation to their husband's performance. Parents, paired with their children, manifested keen desire for their children's performance to be high and to improve; while the children seemed to expect good performance from parents and to be disappointed if they showed any weakness, even in dart throwing.
Thus, the discrepancy between judgment of another's future performance and erection of the aspiration level (judgment of own future performance) obtained by McGehee tended to disappear when positive ego-involvement between partners in the dart throwing situation was introduced. Commenting on this and other studies, Sherif (20) suggested that "further variations would be obtained if the subjects were involved with each other
(126) in a negative way, e.g., as bitter personal rivals, competitors, or members of antagonistic groups". (p. 292)
The present study, emanating from this suggestion, aims at extending the scope by investigating the effect of negative relationships as well as positive relationships in the estimation of own future performance and of another's future performance. In addition, different degrees of positive involvement were introduced. In brief, pairs of individuals who are definitely involved with one another in a negative way (antagonistic or competitive), pairs strongly involved in a positive way (sweethearts), and pairs who were merely friends were placed in a dart throwing situation, ostensibly a "test of eye-hand coordination." Each subject estimated his own future performance and also that of his partner in the task.The hypotheses to be tested were:
1. When individuals are strongly ego-involved with one another in a positive way, the discrepancies between performances and their estimates of each other's future performances will not differ significantly. Further, the discrepancies between performances and estimates of their own and of their partner's future performance will not differ significantly.
2. When individuals are strongly ego-involved with one another in a negative way, the discrepancies between performances and their estimates of each other's future performance will differ significantly. Further, the discrepancies between performances and estimates of their own and of their partner's future performance will differ significantly.
3. When individuals are mildly ego-involved with one another (such as friends who are not strongly committed to one another) differences will be found in the discrepancies between performances and estimates of each other's future performance. Further, differences will be found in discrepancies between performances and estimates of their own and of their partner's future performance. These differences should not be as great or as significant as those found for negatively involved pairs of subjects.
Subjects
Four groups of subjects, totalling 140, were used. The subjects were high school and college students. All subjects were paid at the rate of 60 cents an hour for participation in the experiment.
The four groups of subjects were:1. Coll. ps: pairs of college students positively (p) ego-involved with one another as sweethearts (S). There were 20 pairs of male-female combinations. The criteria of selection were that the subjects be engaged to be married or be married for less than 8 months.
2. H.S. ps: pairs of high school students positively (p) ego-involved with one another as "steadies" (S). Twenty pairs of male-female combinations were selected who were known by teachers, counselors and friends to be "going steady."
3. H.S. pf : pairs of high school students positively involved (p) as friends (f). Ten pairs of subjects (7 male and 3 female) were selected as friends on the basis of their constant companionship in and out of school and their special attempts to have classes together in a large high school.
4. H.S. n: pairs of high school students who were strongly and negatively (n) ego-involved. Twenty pairs of subjects, (17 male and 3 female) were selected who had recently fought with each other, attempted to fight but were restrained, or who were competing for the same boy or girl friend.
All cases from the high school were ones known to their teachers and counselors.It would have been desirable from the point of view of precision, to obtain subjects who were negatively involved along the same dimension as the positively involved groups, but in a different direction. This was not possible. Nor was it found possible, in spite of extensive efforts, to secure a sufficient number of college subjects who could meet the criteria defining a strongly negative relationship.
Apparatus
The apparatus consisted of a specially constructed dart board, S feathered darts, mimeographed score sheets and questionnaires.
The dart board was 36 by 44 inches surrounded by an 8 inch border of gray plywood. A piece of unbleached muslin containing 10 concentric circles ranging in score value from 2 (the outside circle) through 4, 6, 8, etc., to 20 (the bull's eye) was tightly stretched on a detachable frame. This target was removable. Behind it, was a second target of unbleached muslin which was blank with the exception of a dot in the center. This blank target was stretched across a beaverboard back. Darts were actually thrown at this blank target in an attempt to provide a stimulus situation with as little objective structure as possible. (See discussion of results on this point.) On the back side of the dart board, visible only to the experimenter, was a target containing concentric circles and appropriate score values so that as the dart pierced the beaverboard, the experimenter could record the scores out of the subject's sight. In addition, this hidden target shielded the experimenter as he recorded the relevant spontaneous remarks of the subjects in the experimental situation.
(128)
The use of the removable target with scored concentric circles will be specified below (Procedure).
Procedure
All subjects were tested in pairs with only the experimenter present. High school subjects were tested at Capitol Hill Senior High School, Oklahoma City and college subjects in an experimental room on South Campus, the University of Oklahoma. Each pair participated 'in one experimental session lasting approximately an hour. They were instructed that the study concerned eye-hand coordination. The validity of this explanation was not questioned. Its acceptance is indicated by statements on the questionnaires and oral remarks, such as "My hand and eye coordination is very poor," "My hand and eye coordination is better than I expected."
In order to insure interest in the task, the subjects were told that the results revealed certain characteristics of their personalities, as well. This remark was introduced after pre-tests indicated that some subjects, especially on the college level, were not sufficiently challenged by the dart throwing itself.Subjects were shown the removable target marked with score values to acquaint them with the basis of scoring. Darts were actually thrown at the blank target, and estimates made on this basis. One thrown dart constituted a trial. This was chosen as a more sensitive method of scoring than the total of several darts used by previous experimenters.
Subjects were instructed as follows:This is an experiment aimed at testing your hand and eye coordination, that is, it will attempt to measure how closely you can coordinate the motor skill of dart throwing with visual skill in two situations: in estimating the performance of yourself and in estimating the performance of another. The results of this test will reveal certain aspects of your personality so it is important that you do your best on it.
You will notice at the front of the room a target containing ten concentric circles ranging in numerical score value in series of two's from two to 20. The worst score you can make, of course, is zero and the best possible score is 20. Look carefully at the target containing these concentric circles and score values and remember their order because you will not throw at this target. It will be removed and you will throw at a target of the same size but without the circles and score values. (The removable target was detached, revealing the blank target.)
After 10 practice trials, I want each of you to have 50 trials at the blank target. First, one of you will take your 60 trials, and then the other. The one of you who is throwing will record before each trial
(129)
the score you actually expect to make on that trial and also announce it aloud. After you have thrown you will record and call aloud the score you think you actually made on the dart just thrown.
The one of you who is not throwing will record before each trial the score you expect the other will make on that particular trial. You will also record after each trial the score you think your partner actually made on the dart just thrown. You are not to call your estimates aloud. You are to make your estimates on this sheet before the one who is throwing announces his aloud. You may signify this by "O. K." (Blanks on the score sheet for recording estimates were indicated.)
I will not be able to tell you your score. However, I will remind you that 20 is the best possible score. If there are no further questions, I shall toss a coin to see which one of you throws first.
The subject thus selected to throw first was given 5 darts and the target with concentric circles removed from sight. Each subject threw from a line 15 feet from the target. The experimenter offered no suggestions on how to throw the darts. He stood behind the dart board, recording the scores and spontaneous remarks of the subjects without comment.
After 10 practice trials and 50 test trials, the subject throwing (Bidder) became the observer (Estimator) while the second subject, who had been observing, began to throw. Each subject (as Bidder) thus estimated his own future score on each trial and the score he believed he actually made. As Estimator, each subject estimated his partner's future score on each trial and the score he believed his partner actually made on each trial. Consequently, the data for each subject consisted of 100 judgments related to his own performance (before and after each trial) and 100 judgments related to his partner's performance (before and after each trial). The procedure followed for both subjects of the pair was the same.
In this dart throwing situation, the task was such that actual performance is readily perceived visually by the subject. The formal announcement of performance to the subject by the experimenter amounts, in this situation, to repeating to the subject what is already obvious to him. Announcement of the obvious to the subject in this case would serve merely to clutter the experimental situation unnecessarily. Therefore, in the present procedure, it was deemed unnecessary and unnatural to make an announcement of performance to the subjects. In preliminary work, comparison of subjects' own estimates of performance which they recorded after each trial with the experimenter's record of actual score revealed inappreciable differences (indicating that the removal of the structured target with scored concentric circles did not produce a sufficient degree of unstructuredness).
As supplementary material, a questionnaire was administered after the
(130) session was completed. The questions concerned their degree of certainty as to their estimates, the difficulty of estimating, their feelings when they and their partner made scores higher and lower than estimated, and general impressions of the session.
Results
To test our hypotheses, the main task in the analysis of data becomes comparison of judgments made by the subjects, standing in the different relationships specified before, in the Bidder and Estimator roles of the experimental situation. The data were analyzed by comparison of discrepancy scores (D scores). The D score is the difference between estimate and performance. Since in the procedure used the actual performance scores were not announced to the subjects for reasons mentioned previously, their estimates of actual performance after each trial were used rather than the score recorded by the experimenter. Thus, performance scores used in the computation of D scores were those perceived and recorded by the subjects themselves. As mentioned before, preliminary work indicated that in this simple situation with one dart constituting a trial, relatively inappreciable differences were found between actual performance and perception of actual performance.
The first comparison of discrepancy scores concerns the relative accuracy of estimation. D scores-the difference between estimates of future performance and subsequent estimates of the actual performance-were computed for every trial. In this case, the comparison made was between D scores of the two paired subjects on the same series of trials-one subject serving as Bidder and estimating his own performance while his partner served as Estimator, estimating the Bidder's performance. The means of these D scores for the Bidder and for the Estimator were obtained. The difference between the average D scores of the two paired subjects, Bidder and Estimator, was computed as follows: Mean DB1 - Mean DE2. Here DB refers to the Bidder's discrepancy scores, DE to the Estimator's discrepancy scores, the subscripts 1 and 2 merely indicating that in this analysis the Bidder and Estimator were the two different subjects in each pair, both estimating at the same time the future performance and actual performance of the Bidder.Table I gives frequency of distributions of the differences in accuracy (Mean DB1-Mean DE2) between the Bidders and Estimators. Negative values in the table mean that the Estimator overestimated the Bidder's future performance to a greater extent than did the Bidder himself. Data are presented for subjects in the 4 experimental groups-college sweethearts
(131) (Coll. ps), high school "steadies" (H.S. ps), high school friends (H.S. pf), and high school students negatively involved as rivals (H.S. n) (See Subjects). It is seen in Table I that for the most positively involved groups of subject-pairs (Coll. ps and H.S. ps), the differences between Bidder and Estimator in mean D scores cluster around zero, about 66% of cases being within 1.49 on either side of the zero point. In contrast, most of the cases (85 %) in the group of rivals (H.S. n) pile up between 1.5-4.49 on the positive side of zero. These findings are perhaps all the more significant when it is considered that some competition enters into this experimental situation almost unavoidably (See Discussion of Results).
| TABLE I | ||||
| Frequency Distribution of Differences in Mean D Scores (Accuracy) for Estimating One’s Own Future Performance and for Another Person’s Estimate of One’s Future Performance | ||||
| Mean DB1-Mean DE2 | ||||
| Difference | Group Coll. ps N = 40 |
Group H.S. ps N = 40 |
Group H.S. pf N = 20 |
Group H.S. n N = 40 |
| 6.00 - 7.49 | — | — | — | 1 |
| 4.50 - 5.99 | — | — | — | 2 |
| 3.00 - 4.49 | 1 | 3 | — | 18 |
| 1.50 - 2.99 | 8 | 8 | 5 | 16 |
| 0.00 - 1.49 | 16 | 13 | 12 | 3 |
| 0.00 - - 1.49 | 12 | 12 | 2 | — |
| -1.50 - -2.99 | 2 | 3 | 1 | — |
| -3.00 - -4.49 | 1 | — | — | — |
| -4.50 - -5.99 | — | 1 | — | — |
| -6.00 - -7.49 | — | — | — | — |
Table II presents the means of these differences in discrepancy scores for Bidder and Estimator and the statistical significance of this mean difference for each group of subjects.
As Table II indicates, the mean difference in accuracy (D scores) of the Bidder (estimating his own future performance) and of his partner (estimating the Bidder's future performance) is small and not significant statistically for each of the positively involved groups of subjects. For subjects most strongly involved in positive relationships, college sweethearts and high school "steadies", these mean differences are .304 and .459, respectively. For the less positively involved subjects, high school friends, the difference in means is .894. Since none of these differences is significantly
(132) different, we may say that the relative accuracy of the aspiration level (estimates of own future performance) and another person's estimate of that future performance for subjects who are positively ego-involved with one another is similar. It is noted that the least involved of these three groups of subjects with positive relationships (friends) had relatively a much larger mean difference in accuracy scores (.894).
| TABLE II | ||||
| Differences Between Mean Accuracy Scores For the Experimental Groups and Their Significance | ||||
| Differences in accuracy of estimating one’s future performance and of another’s estimate of this performance | ||||
| Mean DB1-Mean DE2 | ||||
| Mean diff. | Group Coll. ps .304 |
Group H.S. ps .459 |
Group H.S. pf .894 |
Group H.S. n 3.075 |
| t | 1.394 | 1.663 | 1.337 | 17.083 |
| P | >.10 | >.10 | >.10 | <.01 |
In contrast, the mean difference in accuracy for the negatively involved group of high school students (H. S. n) is 3.075 which is significant at less than the .01 level. This means that on the average the difference between estimate of future performance and estimate of actual performance was significantly greater for the Bidder than for the Estimator, when the two were negatively involved with each other.
This significant difference between accuracy of estimating one's own future performance and accuracy of a negatively involved partner's estimate of this future performance is one of the striking findings of this study, especially when compared with the mean difference in accuracy of Bidder and Estimator in the positively related groups. The obtained mean difference in D scores for negatively involved subjects (3.075) is almost seven times greater than the mean difference in D scores for Group H. S. ps (.459) and ten times greater than for Group Coll. ps (.304). The actual difference between the means of Coll. ps and H. S. n and between H. S. ps and H. S. n are 2.77 and 2.62, respectively.
Testing the significance of the above differences, we find the following: When Group Coll. ps is compared with Group H. S. n, the obtained difference is significant at less than .01 level, t = 9.75. Comparing Group H. S. ps and Group H. S. n, the obtained difference is likewise significant at less than .01 level, t = 7.76. Thus it can be concluded that the difference between
( 133) D scores of the Bidder and Estimator when the relationship between them is strongly positive is significantly less than the difference between D scores of the Bidder and Estimator when their relationship is negative.
The hypotheses to be tested require that comparison also be made between the accuracy of each subject's judgment as Bidder, estimating his own future performance, and as Estimator, estimating his partner's future performance. In this case, comparisons made between the mean D scores of each subject when he was Bidder and when he was Estimator. The difference between average D scores was computed as follows:
Mean DBI-Mean DEI
the subscript 1 indicating that in this analysis the same subject's D scores as Bidder (indicated by subscript B) and as Estimator (indicated by subscript E) of his partner's performance were compared.
| TABLE III | ||||
| Frequency Distribution of Differences in Mean D Scores (Accuracy) for Estimating One’s Own Future Performance and Estimating A Partners Future Performance | ||||
| Mean DB1-Mean DE1 | ||||
| Difference | Group Coll. ps N = 40 |
Group H.S. ps N = 40 |
Group H.S. pf N = 20 |
Group H.S. n N = 40 |
| 7.50 - 8.99 | — | — | — | 2 |
| 6.00 - 7.49 | — | — | — | 2 |
| 4.50 - 5.99 | — | 1 | — | 4 |
| 3.00 - 4.49 | 2 | 1 | 3 | 11 |
| 1.50 - 2.99 | 5 | 11 | 4 | 14 |
| 0.00 - 1.49 | 19 | 14 | 6 | 3 |
| 0.00 - - 1.49 | 8 | 6 | 6 | 3 |
| -1.50 - -2.99 | 5 | 5 | 1 | 1 |
| -3.00 - -4.49 | 1 | — | — | — |
| -4.50 - -5.99 | — | — | — | — |
| -6.00 - -7.49 | — | 2 | — | — |
Again, as shown in Table III, a frequency distribution of these mean D scores was made. In this comparison between one subject estimating his own future performance and estimating his partner's future performance, the majority of the subjects in the two most positively involved groups (Coll. ps and H. S. ps) again clusters around zero, while subjects of the rival pairs (H. S. n) pile up between 1.50--4.49 on the plus side. However, there are
(134) four subjects in the negatively involved group (H. S. n) constituting an exception to this tendency, inasmuch as the mean difference in D scores was negative (i.e., D score was larger when subject was in the Estimator role). These 4 subjects seemed characteristically to set low aspiration levels which coincided closely with actually obtained scores. A plausible explanation in this few atypical cases, in contrast to greater inaccuracy in judging the partner's future performance, seems to lie in the possibility of protecting the self in the presence of a hostile rival by hitting the nail on the head in relation to one's own capabilities as much as possible. While most subjects in these groups tended to keep the aspiration level high, these subjects for various reasons did not care to expose themselves to possible ridicule by allowing large discrepancies to appear between performance and aspiration level. The problem which this touches upon relates to individual techniques of maintaining self satisfaction or stability. This is not the topic of the present investigation. However, the specific relationship between the pair to which one of these four subjects belonged illustrates the point. The subject was an undergrown boy who had recently fought his partner. Investigation revealed that he was afraid of his larger partner and, at the time of the experiment, was making peaceful overtures to him in spite of their reciprocal antagonism and violent dislike for one another. As a result, he kept his own aspiration level within safe and easy range of his performance, while greatly overestimating the future performance of his stronger and feared partner in the experiment.
Table IV gives the mean of these differences in discrepancy scores for the same subjects serving as Bidder and as Estimator and the statistical significance of the mean differences for each group of subjects. Table IV shows that the difference in average D scores for the subjects in the two most positively involved groups in estimating their own future performance (aspiration level) and, when they changed to the Estimator role, estimating their partner's future performance are small. These differences for college sweethearts and high school "steadies" are .395 and .413, respectively. Neither difference is statistically significant. In this comparison, between each subject's estimate of his own future performance and of his partner's future performance, a larger and statistically significant difference between D scores for the less positively involved group of high school friends was found, namely .945. This is in harmony with the direction of difference found in Table II.Again, by far the largest and most significant difference was found between the D scores of subjects estimating their own future performance and
(135)
| TABLE IV | ||||
| Differences Between Mean Accuracy Scores For the Experimental Groups and Their Significance | ||||
| Differences in accuracy of estimating one’s future performance and of another’s estimate of this performance | ||||
| Mean DB1-Mean DE1 | ||||
| Mean diff. | Group Coll. ps .395 |
Group H.S. ps .413 |
Group H.S. pf .945 |
Group H.S. n 3.076 |
| t | 1.466 | 1.105 | 3.975 | 3.076 |
| P | >.10 | >.10 | <.10 | <.01 |
estimating the future performance of a partner toward whom they were antagonistic. For the negatively involved group of high school subjects, the mean difference was 3.076, significant at less than .01 level.
This difference in accuracy of estimating one's own future performance and in estimating the future performance of a partner with whom one is negatively involved (3.076) is strikingly greater than the mean difference in D scores for Group Coll. ps (.395) and Group H. S. ps (.413). The actual difference between the means of Coll. ps and H. S. n is 2.681 and that between H. S. ps and H. S. n is 2.66.
The significance of the above differences was tested. When Group Coll. ps was compared to Group H. S. n, the difference was significant at less than .Ol level, t = 6.168. Comparing Group H. S. ps to Group H. S. n, the obtained difference was significant at less than .O1 level, t = 5.27. Thus, the difference between D scores for a subject as Bidder and for the same subject in the changed role of Estimator was found to be significantly less when the subject pairs were positively - involved than when they were negatively involved with one another.
The second set of comparisons of discrepancy scores related to the consistency of judgment. The problem in this analysis was the effect of actual performance, as estimated by the subjects, upon the subsequent judgment of the subject's own future performance and partner's judgment of his future performance. Frank (4) mentioned the "rigidity" of the aspiration level, i.e., its consistency in spite of variation in performance.
The consistency score was obtained for each trial by subtracting from each estimate of future performance the score the subject perceived as made on the immediately preceding trial. We can conveniently designate estimate of future performance (aspiration level) on the kth trial by AK and the
(136) estimate of actual performance on the kth trial by PK. Accordingly, the consistency score (C) is the difference between estimate of future performance on the kth trial and estimate of actual performance on the (k-1)th trial, i.e.,
C = AK-PK-1
These consistency scores were computed for every subject as Bidder and as Estimator. The mean consistency scores of the two paired subjects were compared as follows:
Mean CB1 -Mean CE2
C refers to consistency score, B to Bidder, E to Estimator, and the subscripts 1 and 2 to the fact that comparison here is made between the two different subjects of a pair, one serving as Bidder and one as Estimator.
| TABLE V | ||||
| Differences Between Mean Consistency Scores For the Experimental Groups and Their Significance | ||||
| Differences in consistency of estimating one’s future performance and of another’s estimate of this performance in terms of prior judgment of actual performance. | ||||
| Mean DB1-Mean DE2 | ||||
| Group Coll. ps |
Group H.S. ps |
Group H.S. pf |
Group H.S. n |
|
| Mean diff | .410 | .372 | .911 | 2.985 |
| t | 1.677 | 1.266 | 3.780 | 6.769 |
| P | >.10 | >.10 | <.10 | <.01 |
Table V gives the obtained mean differences in average consistency scores for Bidder and Estimator and the statistical significance of these differences. The differences of .410 and .372 found for college sweethearts and high school "steadies", which are not statistically significant, indicate that the effect of past performance on estimate of future performance for the most positively involved pairs of subjects tended to be similar in terms of this analysis.
For the least involved of the positively related group of pairs (high school friends) and especially for the negatively involved group of pairs (high school rivals), past performance lead a significantly greater effect on the Estimator's judgments of his partner's future performance than on the Bidder's own judgments of his future performance. These differences of .911 for friends and 2.985 for rivals are both significant at less than the .01 level.
Consistency scores were then computed for each subject serving
( 137) successively in the session as Bidder and as Estimator. This time the comparison made was between the means of the consistency score for each subject when judging his own future performance and when judging his partner's future performance: Mean CBI-Mean Ce1, the subscript 1 indicating that the comparison is between the same subject's judgments as Bidder (B) and as Estimator (E). The mean difference obtained in this way and the statistical significance of this difference is shown for each of the four experimental groups in Table VI.
The small mean differences obtained for the two most positively involved groups, .122 and .361, indicate that when an individual is strongly involved with his partner in a positive way, past performance exerts no significantly greater effect upon estimate of that person's future performance than upon the erection of an aspiration level for one's self.If on the other hand, two individuals are merely friends or especially if they are actual enemies, the checking effect of past performance is greater when judging the other person's future performance than when estimating own future performance. The mean differences for the group of friends and for the group of rivals are .956 and 3.145, respectively, both being statistically significant.
The findings summarized statistically above, indicating similarities in judgments of partners of strongly and positively involved pairs, are substantiated by the analysis of material obtained by the questionnaire administered to the S's after the session was completed. Perhaps the most telling substantiation came from four questions concerning the subjects' feelings at
| TABLE VI | ||||
| Differences Between Mean Consistency Scores For the Four Experimental Groups and Their Significance | ||||
| Differences in consistency of estimating one’s future performance and of another’s estimate of this performance in terms of prior judgment of actual performance. | ||||
| Mean DB1-Mean DE1 | ||||
| Group Coll. ps |
Group H.S. ps |
Group H.S. pf |
Group H.S. n |
|
| Mean diff. | .122 | .361 | .958 | 3.145 |
| t | .455 | .989 | 2.876 | 9.798 |
| P | >.10 | >.10 | <.01 | <.01 |
their own and their partner's successes and failures. These questions were as follows:
5. How did you feel when you made a score higher than you estimated?
( 138)
6.How did you feel when your partner made a score higher than you estimated?
7.How did you feel when you made a score lower than you estimated?
8.How did you feel when your partner made a score lower than you estimated?
Subjects responded to these questions by checking one category on a 7 point scale ranging from "very pleased" to "very displeased", with "indifferent" in the middle category.
Table VII summarizes the answers to questions 5 and 6. About 93% of the most positively involved subjects (36 in Coll. ps and 38 in H. S. ps), indicated some degree of pleasure when either they or their partners made scores higher than estimated. In fact, slightly more subjects in both groups were more pleased at their partner's success than their own. The majority of friends (H. S. pf) were pleased at both their own and their partner's successes; but more subjects were "indifferent" toward their partner's successes than toward their own successes.In contrast, all but one of the 40 negatively involved subjects were pleased with their own successes; but 26 (65%) were indifferent or displeased in some degree at their partner's successes.
| TABLE VII | ||||||||
|
Comparison of Expressed Feelings Accompanying Ones Own and One’s Partner’s
Success: Performance Higher than Estimated |
||||||||
| Mean DB1-Mean DE1 | ||||||||
| Group Coll. ps N = 40 |
Group H.S. ps N = 40 |
Group H.S. pf N = 20 |
Group H.S. n N = 40 |
|||||
| Self | Partner | Self | Partner | Self | Partner | Self | Partner | |
| Very displeased | — | — | — | — | — | — | — | — |
| Displease | — | — | — | — | — | — | — | 4 |
| Slightly displeased | — | — | — | — | — | — | — | 5 |
| Indifferent | 4 | 4 | 2 | 3 | 1 | 7 | 1 | 17 |
| Slightly pleased | 8 | 6 | 10 | 3 | 4 | 4 | 13 | 10 |
| Pleased | 22 | 18 | 19 | 22 | 11 | 5 | 19 | 4 |
| Very Pleased | 6 | 12 | 9 | 12 | 4 | 4 | 7 | — |
These findings concerning the subject's feelings of success are borne out by the responses to questions 6 and 7 concerning response to failure (Table VIII). About 87% of the most positively involved subjects (32 in Coll. psi and 37 in H. S. ps), were displeased when they or their partners made scores lower than they estimated, with none being pleased. In the negatively involved group (H. S. n), 26 subjects (65%) were indifferent or pleased in
(139)
| TABLE VIII | ||||||||
|
Comparison of Expressed Feelings Accompanying One’s Own and One’s Partner’s
Failure: Performance Lower than Estimated |
||||||||
| Mean DB1-Mean DE1 | ||||||||
| Group Coll. ps N = 40 |
Group H.S. ps N = 40 |
Group H.S. pf N = 20 |
Group H.S. n N = 40 |
|||||
| Self | Partner | Self | Partner | Self | Partner | Self | Partner | |
| Very displeased | 2 | — | 3 | 1 | 2 | 1 | 6 | — |
| Displease | 11 | 9 | 19 | 22 | 10 | 3 | 17 | 4 |
| Slightly displeased | 19 | 18 | 15 | 12 | 6 | 8 | 13 | 10 |
| Indifferent | 8 | 13 | 3 | 5 | 2 | 7 | 4 | 16 |
| Slightly pleased | — | — | — | — | — | — | — | — |
| Indifferent | — | — | — | — | — | 1 | — | 4 |
| Slightly pleased | — | — | — | — | — | — | — | — |
some degree when their partners failed to achieve the score they estimated. But 36 of the 40 subjects in this group were displeased in some way at their own failures to achieve the estimated score.
Spontaneous Reactions: The spontaneous reactions of the subjects during the experimental sessions supply crucial verification of the statistical findings. In the most positively related groups, disappointment and elation at the partner's failures and successes were expressed frequently and exuberantly, as well at one's own successes and failures. Poor performance by a partner called for grunts, groans, shrieks, whistles, and a host of exclamations, such as "Goodness," "My heaven !" "Lordee!," etc.
Encouragement was frequently given to the subject throwing in these groups by their partners. For example, one girl told her boy friend: "Try to hit the jackpot for us this time." A boy said to his girl friend, who was throwing: "Get me a bull's eye." Success was greeted by such remarks as "That's fine!", "Good, good. You're getting good." Or, as one girl exclaimed when her boy friend made a consistently high score: "Goodness, you're as good as I think you are."
Subjects in these groups frequently instructed their partner who was throwing on how to improve his score. For example, a boy to his girl friend: "Throw from the wrist, honey. Don't hesitate to throw, now, for you make worse when you hesitate. And be sure to aim for where you call." Another boy: "AA , you're throwing too hard; not so hard." Still another male subject to his sweetheart after she made a series of low scores: "Honey, you're disappointing me. Now slow down; you can beat that."
Although subjects in all three of the positively related groups (Coll. ps, H. S. ps, and H. S. pf) expressed in varying degrees remarks of encouragement
(140) to their partners and disappointment or elation at performance, this was most marked among college students. The relatively less spontaneity among high school subjects was due to the fact that they perceived the experiment more as a formal test than did the college students. This was indicated by their whole demeanor in the experimental situation and by their respectful treatment of the experimenter, whom they usually addressed as "Sir," responding to his instructions and questions with "Yes, Sir" and "No, Sir." The college students, on the other hand, saw the experimenter as one like themselves, and were much more 'informal.
This general concern in the positively related groups that one's partner's, as well as one's own performance meet one's expectations of success, was conspicuously lacking in the negatively involved subject pairs (H. S. n). Not only did the negatively involved subjects keep comment on their partner's performance to a minimum, these same subjects voiced only a few expressions of concern over their own success and failure. In the presence of an individual toward whom one is antagonistic, overt concern over performance was suppressed on the whole.Discussion of Results
The analysis of data in this experiment reveals the differential effects of varying one important factor in the frame of reference determining behavior. In the present instance, as internal factor—ego-involvement—was varied in direction and degree by placing in the same stimulus situation pairs of subjects with objectively differing personal relationships.
The selection of strongly and positively involved pairs was made from sweethearts, newlyweds, and "steadies" so that there would be great likelihood that the degree and direction of ego-involvement would follow that expected of their relationship. The friends were selected on the basis of actual behavior, rather than mere pencil and paper preferences. The negatively involved pairs had actually fought or been restrained from fighting one another or, in a few cases, were competing for the same goal. Nevertheless, no assumptions were made that there would be the same intense degree of positive involvement between pairs of "steadies" or newlyweds, or the same degree of reciprocal antagonism between partners, in each of the negatively involved pairs. In fact, there might have been some cooling off or even friction between a few positively involved pairs, as results of a few atypical pairs would indicate. It would have been very desirable to make an intensive study of these few atypical positive pairs to ascertain whether or not, at the time of the experiment, some situation of friction existed between the members of
(141) the pair. Nor is the assumption made that the positive or negative involvement between members of a pair 'is reciprocal with the same degree of intensity. Even though no intensive study was made, on the basis of these data we should think that in the case of some pairs, one partner was the admiring and more devoted one, whose admiration or devotion was not reciprocated 'in the same degree by his or her partner. This point is strikingly illustrated by the results obtained from a pair of high school subjects engaged to be married. The results computed from these subjects' estimates following their participation in the study showed a high positive relationship. A few days after their participation in the experiment, however, conflict arose between the pair, and the engagement was broken on the girl's initiative. In order to compare the judgments of these subjects after this break in their positive relationship with the results already obtained while they were engaged to be married, they were brought back to participate in the experimental situation together once more. The results of this second testing were in keeping with expectation. In the case of the girl who broke the engagement with the boy, there was a large discrepancy between her estimates of his future performance and his own estimates of his future performance, as well as between her aspiration level and her estimates of his future performance. On the other hand, there was an even smaller discrepancy between the boy's estimates o f the girl's future performance and her estimates o f her own future performance than at the first testing while they were engaged. This increased affection on the boy's part at the second testing was confirmed by his behavior in the experimental situation on this second occasion. The boy made obvious attempts to regain the girl's affection, for example, rubbing her cheeks between his hands, caressing the back of her neck, standing close to her and peering directly into her eyes. The couple was later married.
This non-reciprocation of a strongly positive or negative involvement was also evident in the case of the negatively involved pair already mentioned (p. 16). It will be recalled that in this case, one subject greatly overestimated his partner's performance. This subject was overawed and fearful of his antagonist and was, at the time of the study, making overtures to him to establish friendly relations. The general tendency which came out in our results would have been further clarified had it been feasible to study intensively the degree of positive and negative relationship between members of each pair and the reciprocation or non-reciprocation of positive or negative involvements.The situation in which these subject pairs were placed was simple: dart throwing and estimating future and past performance. Most subjects
(142) found some challenge in the task, although a very few were indifferent to their performance (Table VII and VIII). Neither task nor instructions involved competition between pairs. However, one partner followed another at throwing; and in the cultural setting, competition in such activities is so generally taken for granted that some competitive elements were probably present in the situation.
In view of this, the findings seem all the more significant. In spite of such competitive tendencies, subjects of the most positively related groups (college sweethearts and high school "steadies") were on the average as ego=involved in their partners' performance, in their success and failure, as in their own. For these groups, the size of discrepancies between actual performance and estimates of future performance were similar for the two partners of a pair, one as Bidder and one as Estimator, and for the same subject setting his aspiration level (as Bidder) or estimating his partner's future performance (as Estimator) (See Tables II, IV). In all these instances, the subjects had positive expectations and goals related to the task for partner as well as for self. Estimates of self and of other were made in these terms. As Tables I and III indicate, this represents the modal tendency for these groups. In some cases, higher goals were set for the partner than he set for himself, or than one set for one's self. This over-estimation, characteristically associated with judgment relating to one's own performance (aspiration level), is thus found also in judgmental activity in which there is strong positive involvement with another person.
It should be noted, however, that the evaluation of results of the present study are not based merely on measures of over- or under-estimations of scores. The measures used in this analysis are all difference scores. While over-estimation may indeed be a characteristic indication of strong ego-involvement, under-estimation may, in some cases, also be such an indication. Some individuals, to protect themselves or to shelter a loved one, may keep their goals relatively low. Since this problem is not the topic of the present study, analysis was made simply in terms of discrepancies between judgment and performance.
For the most positively related subject pairs, past performance as perceived by the subjects characteristically exerted about the same effect on the estimation of future performance by the Bidder and by the Estimator, as well as on one's aspiration level and on his estimate of partner's future performance (Tables V and VI). In other words, the subject's goals for his own performance were as consistent as were his partner's goals for his performance. And, his goals for his own performance were as consistent as
( 143) were his goals for his partner. For these pairs, pleasure characteristically accompanied own success and partner's success; displeasure accompanied failures (Tables VII and VIII).
In contrast, the negatively related subjects estimated their own future performance in a characteristically different way than did their partners. This differential is also revealed in comparing their estimates of own future performance with their estimates of their partner's future performance. Discrepancies between one's estimate of own future performance and actual performance were significantly greater than the discrepancies for the partner's estimate of this performance, or than one's estimate of partner's performance (Tables I-IV). In the presence of an antagonist, goals are set with less regard for one's own actual performance than one's partner pays, judging his rival in action. Past performance exerts significantly less effect on the estimates of he who throws and must keep his chin up, than on his antagonist who watches and is, perhaps, not even ashamed to be pleased at his failures and a bit disgruntled at his successes (Tables V, VII, and VIII). And when this person who was throwing sits down to observe his antagonist, he too can afford to pay more heed to the realities of past performance in estimating his rival's performance than he could when he was on the spot (Table VI).
The smaller and less significant discrepancies in accuracy and consistency found in the cases of pairs of friends are in line With the hypotheses and clarify the results. When the pairs were strongly involved in a positive way, they brought relatively little competition and strong identification into this situation. Discrepancies in accuracy and consistency were small and insignificant. The negatively related subjects (H. S. n) brought not only competition, but hostile rivalry with one another into the objectively same situation. Discrepancies in accuracy and consistency scores were large, with Bidder's being the least accurate and most consistent. The friends were chosen as good friends in terms of their associations. While they were positively involved with one another, while identification was present so that there was, on the whole, pleasure at each other's successes and disappointment at each other's failures (Tables VII and VIII), they also competed as friends. This competition produced significant discrepancies, but smaller than for antagonistic rivals (H. S. n), in accuracy and consistency of Bidder and Estimator, with the Bidder being less accurate and more consistent than Estimator.
In addition to the competitive factors that entered into the production of these small, but significant discrepancies in accuracy and consistency for pairs mildly involved as friends, it is probable that the task was too structured
( 144) to reveal in any clear-cut way the involvements of these subjects. Internal factors are most fully revealed in situations with little objective structure. It will be recalled that this dart throwing situation even with the scored target removed, was relatively structured, as revealed by the close correspondence between actual performance and the subject's perception of that performance (p. 11). In a situation with less objective structure, we would expect discrepancies between judgments of 'individuals mildly involved in a positive way to be even smaller and, if little or no competition entered the situation, insignificant.
Thus, these paired subjects, with differing personal relationships and involvements, the validity of which is found in life situations, reacted in characteristically different fashions in this situation. The discrepancies between judgment and performance of Bidder and of Estimator were in line with those predicted in terms of the direction (positive and negative) and intensity of their involvements with each other.The practical possibility which these findings open is the use of such an indirect method as a predictive index of the relationship between individuals. The difficulties and faults of direct verbal or paper-and-pencil methods are too well known to warrant discussion here. The advantages of discovering the relationship between individuals without their being aware of this aim are obvious. The analysis of the measures used in the present study indicates that this design could serve as a model for such a method. It is probable that dart throwing would not be an appropriate task for all kinds of subjects. Subjects at the high school level seem to be challenged by the task more than subjects on the college level. From a theoretical point of view, if the task to be judged has less objective structure than the present one, greater discrepancies between groups should be found, inasmuch as internal factors (ego-involvements in this particular case) would assume even greater weight in the total situation. The procedures used in this study are effective in tapping strong involvements either in the positive or negative direction. However, they are not sufficiently sensitive to reveal in clear cut fashion milder involvements, as shown by the significant, though small discrepancies between friends. In order to do so, a less structured situation should be employed. But it must be remembered that a situation of little structure will also reveal competitive tendencies more clearly while individuals strongly involved in a positive way bring such strong identifications to the situation that rivalry is virtually impossible and the negatively involved pairs brought their intense rivalry into this situation as well as in daily life encounters, the friends competed in a friendly fashion, as friends do at such tasks. The situation to be developed
(145) to tap milder relationships, therefore, must not only be less structured but must contain no features which situationally impose competition, e.g., as does a task commonly associated with competitive sports.
Perhaps one more methodological caution is needed. If the ego-involvements of two individuals with a certain relationship are to be tapped, not only must they be unaware of this aim, but also the situation must not strongly impose a new, temporary relationship, e.g., relatedness merely as two subjects being tested by a psychologist. For this reason, the atmosphere must be easy and casual, so that the actually established relationship between subjects is revealed. Even sweethearts may act and feel like strangers in an overwhelmingly formal situation.Summary and Conclusions
The present experiment was conceived as a specific case of differential behavior resulting from variation of a crucial factor within the frame of reference at a given time. In this study, the variable factor was internal—the direction and intensity of ego-involvements between subjects. The objective task—dart throwing and estimating one's own and one's partner's past and future performance—was the same for all subjects.
Four groups of subjects were selected on the basis of observed behavior which indicated the relationship and involvement in question.
Two groups of subject pairs, college sweethearts (N = 40) and high school "steadies" (N = 40) were selected as being strongly and positively ego-involved with one another.
One group of high school subject pairs (N = 20) was selected as friends.One group of high school subject pairs (N = 40) was selected as negatively ego-involved on the basis that they had fought or attempted to fight with one another or were competing for the same goal.
Each subject of a pair had a turn at throwing darts and estimating his own future and actual performance for 50 trials (as Bidder), and a turn at observing his partner throw and estimating his partner's future and actual performance for 50 trials (as Estimator).
Data for each S thus consisted of 100 judgments of future and actual performance on each of his own trial-, (as Bidder) and 100 judgments of future and actual performance on each of his partner's trials (as Estimator). After the session, each subject filled in a questionnaire.The results indicate:
(1) When two individuals are strongly ego-involved with one another in a positive way, there are no significant differences between one's estimate
( 146) of future performance and one's partner's estimate of that same future performance in terms of accuracy and of consistency. Nor are there significant differences between estimates of own future performance and estimates of one's partner's future performance in terms of accuracy and consistency. Pleasure accompanies success and displeasure accompanies failure both for self and for partner.
(2) When two individuals are less strongly involved in a positive direction as friends, some friendly competition enters the situation. Small discrepancies are found between estimating own future performance and another's estimate of this future performance. The Bidder is significantly more consistent in setting his goals than the Estimator is in estimating the Bidder's future performance. And estimates of own future performance are significantly less accurate and more consistent than estimates of one's partner's future performance.
(3) When two individuals are negatively involved as antagonists, the rivalry produces rather wide and significant discrepancies between setting of goals for his own performance by the Bidder and estimation of this future performance by his rival (viz. Estimator). The Bidder is significantly less accurate and more consistent than is the Estimator. And the estimates of subjects as Bidders are significantly less accurate and more consistent than estimates of these same subjects when their roles were changed to that of Estimators. The expressed feelings at one's own success and failure were strikingly different from feelings at the partner's success and failure.
From a theoretical point of view, these results clearly indicate that it is erroneous to draw demarcations between aspiration level and judgment. Differences in judgmental activity related to self and to others can properly be explained in terms of differences in direction and intensity of ego-involvements of the individuals in question which enter in functional relation with other internal and external factors in the frame of reference. This is but one example of differential reactions resulting from variation of crucial factors in the frame of reference within which reaction takes place. With such functional analysis, sharp delineations between various sorts of discriminatory activity, e.g., fact and value judgments, will become superfluous.
Since for every group, the similarities and discrepancies in estimating were in line with the predictions made on the basis of the human relationships between pairs, this experiment offers practical leads for the development of an indirect method for predicting the relationship (positive or negative and degrees thereof) between individuals.
(147)
REFERENCES
1. Chapman, D. W. and J. Volkmann, A social determinant of the level of aspiration. J. abnorm. soc. Psychol., 1939, 34, 225-238.
2. Cogan, L. S., A. M. Conklin, and H. L. Hollingsworth, An experimental study of self-analysis, estimates of associates and the results of tests. Sch. & Soc., 1915, 2, 171-179.
3. Festinger, L., Wish, expectation, and group standards as factors influencing level of aspiration. J. abnorm. soc. Psychol., 1942, 37, 184-200.
4. Frank, J. S., Individual differences in certain aspects of the level of aspiration. Amer. J. Psychol., 1935, 47, 119-129.
5. Himmelweit, H. T., A comparative study of the level of aspiration of normal and of neurotic persons. Brit. J. Psychol., 1947, 37, 5-59.
6. Levine, R., I. Chein, and G. Murphy, The relation of the intensity of a need to the amount of perceptual distortion. J. Psychol., 1942, 13, 283-293.
7. Lewin, K., T. Dembo, L. Festinger, and P. S. Sears, Level of Aspiration. Chapt. 10, 333-378, in J. McV. Hunt (edit.), Personality and the Behavior Disorders, Vol. 1, N. Y.: Ronald Press, 1944.
8. Luchins, A. S., On agreement with another's judgment. J. abnorm. soc. Psychol., 1944, 39, 97-111.9. Luchins, A. S., Social influences on perceptions of complex drawings. J. soc. Psychol., 1945, 21, 257-273.
10. Marks, E., Skin color judgments of Negro college students. J. abnorm. soc. Psychol., 1943, 38, 370-376.
11. MacIntosh, A., Differential effect of the status of the competing group upon the level of aspiration. Amer. J. Psychol., 1942, 55, 546-554.
12. McGarvey, H. R., Anchoring effects in the absolute judgment of verbal materials. Arch. Psychol., 1943, No. 281.
13. McGehee, W. Judgment and the level of aspiration. J. gen. Psychol., 1940, 22, 3-15.
14. Preston, M. G. and J. A. Bayton, Differential effect of a social variable upon three levels of aspiration. J. exp. Psychol., 1941, 29, 351-369.
15. Schafer, R. and G. Murphy, The role of autism in a visual figure-ground relationship. J. exper. Psychol., 1943, 32, 335-343.
16. Sears, P. S., Levels of aspiration in academically successful and unsuccessful children. J. abnorm. soc. Psychol., 1940, 35, 498-536.
17. Sherif, C., Ego-involvement as a factor in judgment. Reported in M. Sherif (20), 289-292.
18. Sherif, M., A study of some social factors in perception. Arch. Psychol., 1935, No. 187.
19. _____ , The Psychology of Social Norms, N. Y.: Harper, 1936.
20. _____ , An Outline o/ Social Psychology, N. Y.: Harper, 1948.
21. Volkmann, J. Scales of judgment and their implications for social psychology. 273-296 in J. Rohrer and M. Sherif (edits.), Social Psychology at the Crossroads, N. Y.: Harper, 1951.
22. Zeaman, D., Reported in M. Sherif (20), 293-296.
