The Social Judgment-Involvement Approach to Attitude and Attitude Change [Foundations of Communication Theory]

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting






I.

The study of speechmaking is necessarily audience centered. Rhetorical theory has always acknowledged the importance of the audience in its formulations. This interest in the audience has inevitably focused attention on the ways audiences felt about subject matters, and the ways these feelings affected their responses to speeches. Whether so named or not, such concern is concern with what we now call attitude.

A review of what rhetorical theory has had to say about attitudes through the years would be no simple matter. The length and complexity of such a review puts it beyond the scope of this paper. In some form or another, though, critical review takes place continually as rhetoricians attempt to develop and improve the theory with which they deal. Along with such development, certain ways of thinking about attitudes are discarded, while others are retained and refined.

Roger E. Nebergall: From Roger E. Nebergall, "The Social Judgment-Involvement Approach to Attitude and Attitude Change," Western Speech, 1966, 30, 203-215. Reproduced with permission of the author and publisher.

How and why do rhetoricians abandon one way of looking at the concept of attitudes, while retaining and refining another? Such selectivity implies certain criteria, not necessarily explicit, by which such views are evaluated. If the criteria are not explicit, of course, they may be vague and idiosyncratic. Explicit criteria, on the other hand, make evaluation easier and more consistent. Also, if it is clear how we evaluate, we may have a clearer notion of how to improve our understanding of a concept such as attitude.

This paper will attempt to develop two ideas or, more precisely, two parts of a single idea. First, an attempt will be made to develop explicit criteria which may be employed (and probably often are employed) in evaluating concepts in rhetorical theory. Second, the social judgment-involvement approach to the study of attitudes will be discussed, particularly as it relates to such criteria. The discussion will inevitably involve consideration of the fruitfulness of the approach for rhetorical theory.

In the long and richly varied history of rhetorical theory, many different, even contradictory, viewpoints have been advanced, criticized, and defended.

Certain consistencies have, however, been apparent. For one thing, rhetorical theory has always been intensely practical, and ultimately has appealed to effect as an evaluative criterion. The rhetorician may ask whether discourse is beautiful, or honest, or productive of philosophic insight. However, he first inquires as to the effects of that discourse upon audiences. Criteria by which to evaluate rhetorical concepts, then, necessarily involve determinants of those effects.

The first, most necessary criterion ought to be accuracy of description. How well does the theoretical concept (in this case attitude) describe the phenomenon to which it refers? Using this criterion to evaluate a concept demands that accuracy of description take precedence over convenience, elegance, and simplicity.

Theorists are often tempted to formulate concepts which are simple and elegant, and which seem intuitively complete and satisfying. Such criteria are not irrelevant, of course. Following accuracy of description they can be quite important.

But useful description of behavior has to be phenomenological. Considerations of how people actually behave have to take precedence over considerations of theoretical elegance, the application of handy mathematical procedures, or even the ease with which experimental subjects can be located and their responses noted.

It may not be inappropriate to suggest here that the proliferation of different ideas concerning attitudes in rhetoric, and the relatively short terms of popularity of each different idea might be in part explained by lack of attention to such considerations.

The second criterion must be that useful descriptions are those which permit predictions about future attitudes and future behaviors. The rhetorician, bound by his commitment to the results of communication, necessarily demands this. He wants to know what attitudes are like. But he also has to know how likely it is that a given attitude will change, and how. He wants to know how such changes are related to discourse. He is interested in discourse in the past that has shaped such attitudes, as well as future discourse which may modify them. Thus, the application of this criterion requires the development not only of a theory of attitude, but a theory of attitude change as well.

In summary then, these criteria require that theorizing about attitude give an accurate picture of the phenomenon in the real world, and that it do this at the expense of all other considerations. Further, such theorizing should include considerations of attitude stability and attitude change. Finally, and subordinate to the other two considerations, should be considered elegance, parsimony, simplicity, relationships to existing mathematical models, and other such considerations.

II.

The social judgment-involvement approach to the study of attitudes grows out of more than thirty years of research into psychological judgment processes.

This study of judgment process began with the study of the psychophysical judgment situation, and proceeded to the more complex kind of psychosocial scaling involved in attitude measurement. A complete review of this literature and, for that matter, an exhaustive review of the social judgment-involvement approach are beyond the scope of this paper. Both have been developed elsewhere. In this paper some of the basic concepts in the social judgment-involvement approach will be traced out briefly so that their relationships to rhetoric and communication may be explored at greater length.

To begin with, the social judgment-involvement approach denies the adequacy of a single score as representative of an attitude. No doubt such a score can be related to an attitude, and the utilization of such a single score has proved useful in various kinds of research in the past. If we discover that the phenomenon attitude is too complex to be accurately represented by a single score, how ever, then adequate theorizing must undertake to describe that complexity.

Consider for a moment that attitudes are inferred from behavior. Consider also that the behavior is toward referents and grows out of evaluation. Such evaluation is judgment, and leads us necessarily to think of attitudes as involving a judgment process, a choosing between alternatives. The alternatives which are judged are, of course, the various possible positions on the issue at hand.

What sort of judgment is involved? What sorts of evaluations of alternative stands go into the specifications of an attitude? One such judgment would be, of course, the selection of the one stand that an individual regards as his own.

-------------

1. See particularly Muzafer Sherif and Carolyn W. Sherif, An Outline of Social Psychology, Rev. ed. (New York, 1956), Chaps. 2, 3, 15, 16; Muzafer Sherif and Carl Hovland, Social Judgment (New Haven, 1961).

---------------

Many people would regard this selection as the most important part of an attitude or perhaps as all that is involved in attitudes. If we were to ask someone about his attitude on a social question, and if he were to choose to attempt to answer, he would almost certainly respond by identifying the stand with which he agreed, his "own" position. It is possible, though, that people who would identify the same position or alternative as their own would judge differently regarding other possible stands. If they did, then the single position which they most preferred would not reveal their attitude fully. Further, if differences in these judgments about other stands were related to the kinds of characteristic behaviors associated with attitudes, then an assessment of attitudes which did not take them into account would not predict attitude related behavior with the same precision as an assessment which did.

One such difference, which has appeared in repeated research findings, involves the positions which persons who identify the same stand as their own will identify as positions with which they also agree. The number of alternative stands other than their own with which these persons will also agree will vary.

The specific stands chosen will also differ, some choosing and some not choosing to agree with a specific alternative.

Further, if persons are asked to designate the alternatives with which they disagree, they will certainly designate one or more of the alternative stands known to them as stands with which they disagree. These stands will vary, even among individuals who identify the same position as their own. Finally, these two judgment processes may not exhaust all the alternative stands. There are often other stands which an individual will not make either judgment about, but toward which he will prefer to remain noncommittal.

---------------

2. Much of this research is reported in Carolyn W. Sherif, Muzafer Sherif and Roger E. Nebergall, Attitudes and Attitude Change: The Social Judgment-Involvement Approach ( Philadelphia, 1965).

------------------

These alternative ways of responding contain much useful attitude information. These alternatives have been conceptualized as latitudes of acceptance, rejection, and non-commitment, respectively. It would take a far longer paper than this one to summarize the research which has established the importance of these concepts in attitude research. They have been found to be stable indicators of important dimensions of attitude. More important, they have been found to pro vide information about attitudes not available from information about the one position which an individual regards as his own, or the single score which may represent it. Persons who would, following some procedures of attitude assessment, have the same "score" have been found to differ along such dimensions as susceptibility to attitude change, direction of change, the way in which messages are understood, and so forth. Thus, to overlook these judgments which people do make about positions on a matter other than their own is to ignore much of what we hope to learn when we investigate attitudes in the first place, i.e., the degree to which attitudes as we study them affect behaviors in which we are interested.

In addition to the emphasis on the inadequacy of the single score to represent an attitude, the social judgment-involvement approach stresses the importance of the degree of involvement an individual has in the matter. The degree to which an individual is ego-involved in the process of judgment is a crucial factor in the kind of judgments he makes. Thus, ego-involvement is a vital difference between the process operative in psychophysical scaling (in which the anchor is a relatively neutral stimulus supplied by the experimenter) and psychosocial scaling (in which the anchor becomes the person's own position and is far, far from neutral for him). The approach also regards as crucial the relationships between the anchor and the position being evaluated, and the influences of these relations upon the judgments being made. An assimilation-contrast phenomenon has been a consistent research finding both in earlier psychophysical research and in later research involving psychosocial scales. Briefly stated, the consistent finding is that when an object to be judged (a stand in attitude research) is relatively close to the anchor (the individual's own position in attitude research) it will be assimilated.

That is, the evaluator will overestimate the extent to which the alternative is like his own position and will therefore judge it closer to his view. Similarly, if the position under consideration is relatively far from the judge's own position it will be contrasted, seen as farther away and more violently in disagreement.

Such effects clearly affect latitudes of acceptance, rejection, and non-commitment, of course. They also affect the ways in which an individual with a certain attitude interprets and responds to messages he gets which are relevant to that attitude.

It would appear, then, that these considerations, to the extent that they have been and continue to be verified in a variety of research settings, provide a solid basis to argue that the social judgment-involvement approach is our best effort to date to satisfy the first criterion described above, i.e., accuracy of description.

People do make judgments about positions other than their own. They do agree with them, disagree with them, or withhold judgment on them. People are ego-involved in varying degrees concerning matters on which they hold social attitudes.

People do assimilate and contrast the positions to which they are exposed. It does not seem reasonable to develop theories of attitude which proceed as though they do not.

III.

The second suggested criterion of the adequacy of a theoretical formulation about attitudes is the ability of the formulation to generate verifiable predictions.

The social judgment-involvement approach should be continually tested as more research projects undertake to utilize its formulations for predictive purposes.

Several brief examples (not a full review) of research findings involving such predictions provide good early evidence of the usefulness of the approach in pre diction.

DIRECTION OF CHANGE OF POLITICAL ATTITUDES

By employing symmetry and asymmetry of latitudes of acceptance (whether the latitudes of acceptance included the same number of positions on each side of the subject's own position or not) and the location of the one position the subject disagreed with most, 246 of 342 (72 percent) changes in a subject's own position were predicted as to direction (either more or less favorable to one party or the other). The actual changes were obtained from data obtained in two testing sessions several days apart just before the 1960 election. Notice that the data on which these predictions are based are data which attitude measuring procedures eliciting a single attitude score do not gather. With only a single score available, only a random choice model, or perhaps a regression model could be employed to predict direction of change. If students of communication are interested in predicting direction of attitude change, this is an important matter. This is a preliminary finding, of course, but the identification of persons who identify the same position on a controversial matter as their own, but who are likely to change in opposite directions is rich in implications for the rhetorical theorist.

ATTITUDE STABILITY-SUSCEPTIBILITY TO CHANGE

The student of communication is also interested in the stability of attitudes. Re search employing a variety of attitude change measures has shown that such stability is related to extremity of stand. However, the social judgment-involvement approach can serve to specify extremity with more precision than by the location of a response somewhere in a distribution of test scores. Specifically, the size of the latitude of non-commitment gives promise of being a measure of susceptibility to change, with a person remaining non-committal on a range of alternatives being more likely to change than a person with a narrower non-commitment range. One important exception to the generalization that the more extreme one's own position the more stable his attitude, may involve those who identify the middle or neutral position as their own. Recent research findings have shown that those whose own position is neutrality tend to divide into those who are unconcerned and tolerant on the one hand, and those who oppose both extremes on the other. The first are characterized by the narrow latitude of rejection and wide latitude of non-commitment characteristic of unstable attitudes, while the other shows the narrow non-commitment latitude, and wide range of rejection characteristic of those with stable attitudes, typically partisans. This may suggest a most useful way to get at the distinction many have suspected between the un involved neutral whose neutrality shows lack of interest and the neutral whose neutrality combines indecision and preoccupation. The man who simply can't make up his mind on a matter of great importance to him may have the latter attitude structure. At the risk of belaboring the point, note that attitude measures which obtain a single score must view these types of neutrals as indistinguishable. Predictions which cannot include data as fundamental as these must continue to be unnecessarily imprecise.

---------------

3. Sherif, Sherif and Nebergall, pp. 178-180. 4 Ibid., 176-178.

----------------

SUMMARY

This paper has developed the argument that a theory of attitudes will be useful to rhetoric and communication if it is both accurately descriptive and able to make useful predictions concerning communication behavior. The social judgment-involvement approach has been briefly discussed with particular emphasis on latitudes of acceptance, rejection, and non-commitment; on ego-involvement; and on assimilation-contrast effects. The usefulness of these ideas for precise description has been developed. Finally, some specific early findings concerning prediction have been developed. Throughout the paper, the relative merit of the social judgment-involvement approach and those procedures concerned with a single attitude score has been explored.

5. Ibid., 58-59.


Also in Part 4:


Prev. | Next

Top of Page  | Index |  Home

Updated: Monday, 2021-08-30 12:29 PST