3 Festinger (1954)
A Theory of Social Comparison Processes
Hochbaum (18) reports an experiment concerning the effect of knowledge of others’ opinions on
one’s own opinion which corroborates
Corollary II B. Half of the subjects in this experiment were
persuaded by the experimenter that they were extremely good at being able to make correct judgments
concerning things like the issue they were to discuss. The other half of the subjects were made to feel
that they were extremely poor in making such judgments. They were then asked to write their opinions
down and were handed back a slip of paper presumably reporting to them the opinions of each other
person in the group. In this way the subjects were made to feel that most of the others in the group
disagreed with them. Those subjects who were given an objective basis for feeling that their opinion was
likely to be correct did not change their opinions very often in spite of the disagreement with others in
the group. Those who had an objective basis for feeling their judgments were likely to be poor changed
their opinion very frequently upon discovering that others disagreed with them.
Hypothesis III: The tendency to compare oneself with some other specific person decreases as the difference between
his opinion or ability and one s own increases.
A person does not tend to evaluate his opinions or his abilities by comparison with, others who
are too divergent from himself. If some other person’s ability is too far from his own, either above or
below, it is not possible to evaluate his own ability accurately by comparison with this other person.
There is then a tendency not to make the comparison. Thus, a college student, for example, does not
compare himself to inmates of an institution for the feeble minded to evaluate his own intelligence. Nor
does a person who is just beginning to learn the game of chess compare himself to the recognized
masters of the game.
The situation is identical with respect to the evaluation of opinions. One does not evaluate the
correctness or incorrectness of an opinion by comparison with others whose opinions are extremely
divergent from one’s own. Thus, a person who believes that Negroes are the intellectual equals of whites
does not evaluate his opinion by comparison with the opinion of a person who belongs to some very
anti-Negro group. In other words, there is a self-imposed restriction in the range of opinion or ability
with which a person compares himself:
Corollary III A: Given a range of possible persons for comparison, someone close to one’s own ability or opinion will
be chosen for comparison.
There is some evidence relevant to this corollary from an experiment by Whittemore (24). The
purpose of the study was to examine the relation between performance and competition. Subjects were
seated around a table and given tasks to work on. There was ample opportunity to observe how the
others were progressing. After the experimental session, in introspective reports, the subjects stated that
they had almost always spontaneously selected someone whose performance was close to their own to
compete against.
Corollary III B: If the only comparison available is a very divergent one, the person will not be able to make a
subjectively precise evaluation of his opinion or ability.
There is evidence supporting this corollary with respect to abilities but no relevant evidence in
connection with opinions has been found.
Hoppe (20) in his experiment on level of aspiration reports that when subjects made a score very
far above or very far below their level of aspiration they did not experience success or failure
respectively. In other words, this extremely divergent score presented no grounds for self evaluation.
Dreyer (5) performed an experiment in which high school children were made to score either: very far
above the reported average for boys like themselves; at the reported average; or very far below the
reported average. After a series of trials they were asked, “How well do you feel you did on the test?”
There were five possible categories of response. The top two were good or very good; the bottom two
were poor or very poor. In the middle was a noncommittal response of fair. Both those who scored very
far below and those who scored very far above the reported group average gave the response “fair”
A Theory of Social Comparison Processes
4
significantly more often than did those who scored at the reported group average. Also, on the average,
the persons who had scored at the reported average felt they had done better than did those scoring far
above the group. Again the data support the hypothesis.
We may then conclude that there is selectivity in comparison on abilities and opinions and that
one major factor governing the selectivity is simply the discrepancy between the person’s own opinion
or ability and that of another person. Phenomenologically, the appearance of this process is different for
opinions and for abilities but conceptually it is exactly the same process. In dealing with opinions one
feels that those with whom one does not compare oneself are different kinds of people or members of
different groups or people with different backgrounds. Frequently this allegation of difference, to
support the non-comparability, is made together with some derogation. In the case of abilities, the
phenomenal process is that of designation of status inferior or superior to those persons who are
noncomparable to oneself. We will elaborate on this later.
Derivation A (from I, II, III): Subjective evaluations of opinions or of abilities are stable when comparison is available
with others who are judged to be close to one’s opinions or abilities.
Derivation B (from I, II, III): The availability of comparison with others whose opinions or abilities are somewhat
different from one’s own will produce tendencies to change one’s evaluation of the opinion or ability in question.
There are also data to show the effect which knowledge of group opinions or group abilities have
on the person’s evaluations which were initially formed privately. If the evaluation of an opinion or an
ability formed in the absence of the possibility of comparison with others is indeed unstable, as we have
presumed, then we would expect that, given an opportunity to make a comparison with others, the
opportunity would be taken and the comparison would have a considerable impact on the self
evaluation. This is found to be true for both abilities and opinions. “Level of aspiration” experiments
have been performed where, after a series of trials in which the person is unable to compare his
performance with others, there occurs a series of trials in which the person has available to him the
knowledge of how others like himself performed on each trial (1, 4, 6, 17). When the “others like
himself” have scores different from his own, his stated “level of aspiration” (his statement of what he
considers is good performance) almost always moves close to the level of the performance of others. It
is also found that under these conditions the level of aspiration changes less with fluctuations in
performance, in other words, is more stable. When the reported performance of others is about equal to
his own score, the stability of his evaluation of his ability is increased and, thus, his level of aspiration
shows very little variability. Dreyer, in an experiment specifically designed to test part of this theory (5).
showed clearly that the variance of the level of aspiration was smaller when the subject scored close to
the group than when he scored far above or far below them. In short, comparison with the performance
of others specifies what his ability should be and gives stability to the evaluation.
Festinger,
Gerard, et al. (10) find a similar situation with respect to opinions. When a person is
asked to form an opinion privately and then has made available to him the consensus of opinion in the
group of which he is a member, those who discover that most others in the group disagree with them
become relatively less confident that their opinion is correct and a goodly proportion change their
opinion. Those who discover that most others in the group agree with them become highly confident in
their opinion and it is extremely rare to find one of them changing his opinion. Again, comparison with
others has tended to define what is a correct opinion and has given stability to the evaluation. This result
is also obtained by Hochbaum (18).
We may then conclude that Derivations A and B tend to be supported by the available data.
Derivation C (from I, III B): A person will be less attracted to situations where others are very divergent from him
than to situations where others are close to him for both abilities and opinions.
This follows from a consideration of Hypothesis I and Corollary III B. If there is a drive toward
evaluation of abilities and opinions, and if this evaluation is possible only with others who are close