Advances in Consumer Research Volume 12, 1985 Pages 7-10
HEDONIC AND UTILITARIAN ASPECTS OF CONSUMER BEHAVIOR: AN ATTITUDINAL PERSPECTIVE
Olli T. Ahtola, University of Denver
Recent criticism of traditional attitude concept and measurement leveled by consumer researchers who emphasize symbolic, esthetic, and hedonic aspects of consumption is reviewed. Traditional attitude concept is reviewed and the validity of the criticism is partly accepted. A new conceptualization o f attitude is proposed that incorporates elements from both perspectives. Operationalization of the concept is discussed, and some preliminary empirical findings are referred to.
There has been of late an understanding that the traditional measures of consumer attitudes do not adequately reflect the various volitional, internally motivated, approach/avoidance response tendencies in different consumption and/or purchase situations.
A common criticism is that these attitude measures are based on conceptualizations that are too simplistic and/or too narrowly focused. One area of criticism has pointed out that the symbolic, hedonic, and esthetic aspects of consumption are not adequately reflected in these conceptualizations and consequently in the resultant operationalizations (e.g., Holbrook and Hirschman 1982).
The purpose of this paper is to respond to this criticism by proposing a modification-to the traditional attitude concept and measurement. This modification incorporates explicitly and as separate constructs both the utilitarian and the hedonic aspects of consumer attitudes. These two constructs determine the general aspect of attitudes. The three constructs and their relationships together form the overall attitude construct.
Hedonic, or experiential perspective, as it is also called (Hirschman and Holbrook 1982, Holbrook and Hirschman 1982, also Unger and Kernan 1983), does not limit its scope of interest only to attitudes, but many other aspects of consumer behavior are included. Of interest are various cognitive and behavioral responses in addition to affective ones. Also, much attention is focused on consumer and environmental 'input' variables. Perhaps, because this perspective has been discussed in the marketing literature in the form of rather short articles, the overall picture remains still very fuzzy with several more or less overlapping concepts and constructs that remain un- or vaguely defined. For example, Holbrook and Hirschman, in their various articles, discuss concepts such as esthetic products, esthetic criteria, and esthetic response without explaining these terms adequately. Is it, for example, proposed that evaluating esthetic products. esthetic criteria are used, and the resultant response is esthetic? Probably the main contribution of this literature to this date is that it has convincingly exposed the problem of the dominant view having been too limited, and has pointed out to some directions where eventually some solutions may be found.
As the title of this paper indicated, the hedonic perspective is examined here from an attitudinal perspective. More specifically, from the traditional attitudinal perspective (e.g., Likert 1932, Osgood, Suci, and Tannenbaum 1957).-Consequently, the focus is limited to those elements in the hedonic theories that are seen to provide inputs into accepting, rejecting, or adjusting the traditional attitude concept. The interest is thus almost exclusively on evaluative predispositions.
Hirschman and Holbrook (1982) and Hirschman (1983) make a point that we could classify products into utilitarian and esthetic. Traditional marketing research has concentrated on utilitarian products according to these authors and the measurement instruments reflect that fact. They argue that new methods and instruments need to be developed to study consumption of esthetic products (e.g., going to a concert). Furthermore, they argue that traditional attitude measures, such as expectancy-value multi-attribute models, measure only affect, which is only a subcategory of emotional response, which again is only a subcategory of hedonic response.
Hirschman and Holbrook (1982) make several valid points describing the conceptual differences between behaviors like "buying a washing machine" and "going to a concert". What remains unclear is whether also the attitudes toward various brands of washing machines are somehow utilitarian in nature as contrasted with emotional or esthetic. That is, is there a qualitative difference in attitudes toward utilitarian products vs. esthetic products? I cannot help feeling that the authors are implying that preferences among what they call utilitarian products are not hedonic or emotive in nature, even though they explicitly state that traditional attitude measures measure affect, which is a subconstruct of emotion. Perhaps most clearly this qualitative difference in evaluative responses is exposed by Holbrook (1983) referring to Holbrook and Zirlin (1983):
"Epistemic and emotional responses tend to correspond, respectively, to utilitarian and esthetic value judgements. Though the parallelism is not perfect, utilitarian (extrinsic) value tends to result from beliefs about the way product imagery serves consumption needs whereas esthetic (intrinsic) value tents to hinge on an emotional response to the sign or significate appreciated for its own sake."
The above mentioned criticism that the traditional attitude models are oriented toward utilitarian behaviors does not accurately reflect at least the intent of the researchers who developed these techniques. The various attitude measurement instruments have almost exclusively been developed by social psychologists, who have been interested in behaviors such as allowing communists to speak in public (Rosenberg 1956, cf. Hirschman 1983) and sexual behavior (Fishbein 1966).
In summary, it remains unclear whether the authors looking at the attitudes from the hedonic/experiential perspective argue that the traditional attitude work is focusing only on one, perhaps minor, aspect of hedonic response, or that this work focuses on responses that are not hedonic at all. Also, it remains unclear whether a rather comprehensive theory that explains both utilitarian and hedonic behaviors is suggested. Or is it a theory that relates only to esthetic/hedonic behaviors? However, a convincing case is made that we should take a closer look at distinctions between utilitarian value judgements and esthetic/ emotional response.
TRADITIONAL ATTITUDINAL PERSPECTIVE
The distinction between esthetic and utilitarian products or value judgements does not appear to have been a concern to those attitude theorists who have had major influence on consumer researchers. Those who consider attitude to be separate from cognitions and conations argue that it can be measured by a procedure which locates the subject on a bipolar affective or evaluative dimension vis-a-vis a given object. (For a review, see Fishbein 1975, Chapter 1.) Consequently, the attitudes are seen as unidimensional, and the same dimension applies to any subject or object. Furthermore, attitude researchers in psychology seem to equate attitude with feeling more than with preference (cf. Holbrook and Hirschman 1982). Many do not seem to make any difference, a notable example being Fishbein (1975, p. 11), who appears to have influenced the consumer attitude research the most. He uses terms "evaluation" and "affect" interchangeably. Also, in his measurement instruments, scales like pleasant/unpleasant and valuable/worthless are averaged to derive a measure of attitude.
It is also interesting that Fishbein in his earlier theoretical writings (Fishbein 1967) conceptualized three basic determinants of behavioral intentions (behavioral intention is considered by him to be the main determinant of subsequent behavior). These three determinants are 1. attitude (affect) toward the behavior, 2. social norms governing that behavior (referent influences), 3. personal norms governing that behavior. He later dropped the third component, i.e., personal norms, because it correlated too high with the behavioral intention, but at the same time, did not add to the predictive power of the model that included only the two other components (Fishbein 1975, pp. 305-6).
Fishbein may have dropped something theoretically meaningful too early based on empirical evidence that was equivocal. My hypothesis is that the personal norms did not help in predictions because these influences are confounded in his attitude (affect) measures and because of how they were conceptualized. He measured personal norms in terms of what the person believed s/he should/should not do with respect to the action s/he was considering. I believe this is too ambiguous.
Furthermore, I believe that his measure of attitude (affect) includes personal norms through the use of the scales like 'wise/foolish'. Even the scale like 'good/bad' that he always uses, is not independent of normative influences.
I should make it clear that my discussion here is focused on the concept of attitude alone, not on the relationship between, say, beliefs and attitudes. The so called expectancy-value models attempt to explain this relationship, that is, they try to measure attitudes through their determinants, and not directly. The direct measures of attitudes, utilized, for example, by Fishbein include techniques such as the evaluative dimension of the semantic differential (Osgood, Suci, and Tannenbaum 1957) and Likert's method of summated ratings (Likert 1932). These measures are used as criteria when the predictions of expectancy-value models are evaluated.
In summary, traditional attitude concept and measurement, especially as defined by Fishbein, was not developed for the purpose of predicting behaviors toward utilitarian objects. In fact, the distinction between esthetic and utilitarian objects does not seem to appear in this literature even implicitly. However, separately from attitudes, Fishbein (1967) considered but later discarded (Fishbein and Ajzen 1975) a possibility that people may have personal norms (in addition to social norms) which influence their behavior in addition to attitudinal predispositions. I suggest that these personal norms are rather closely related to the utilitarian motives of behaviors.
AN ATTITUDE WITH UTILITARIAN AND HEDONIC COMPONENTS
I suggest that it is useful to make a distinction between utilitarian and hedonic motives underlying consumption decisions. These motives form two of the three aspects of attitudes toward behavioral acts. Utilitarian aspect of an attitude toward a behavior relates to usefulness, value, and wiseness of the behavior as perceived by the consumer. Hedonic aspect relates to pleasure experienced or anticipated from the behavior. Hedonic aspect results from the esthetic/ emotional feelings listed by Holbrook and Hirschman (1982) and Hirschman and Solomon (1984), such as love, hate, fear, joy, boredom, etc., in addition to like/ dislike.
Both the utilitarian and hedonic aspects are bipolar in nature and vary in magnitude. Hedonic aspect includes painful and unpleasant feelings as well as pleasant. Utilitarian aspect includes judgements about foolishness and irrationality as well as wiseness and rationality.
I want to emphasize that I am talking about a past, present, or anticipated state, not about a process. Consequently, my utilitarian aspect differs from that of Hirschman and Solomon (1984). They seem to talk about rational processing of information, while my focus is on the affective state. These affective states, whether hedonic or utilitarian in nature, may be the result of rational or irrational, components or Gestalt processing (cf. Holbrook and Moore 1981). Consequently, clear distinction need to be drawn between processing of information and the nature of the resultant state.
I do not believe it to be conceptually sound to divide behaviors into either hedonic or utilitarian. I feel that practically every behavior has both utilitarian (positive and/or negative) and hedonic (positive and/ or negative) aspects. Consequently, it is not theoretically sound to classify behaviors to utilitarian and hedonic, even though it is clear that some behaviors are more hedonically motivated than some others.
Going back to Hirschman and Holbrook (1982) article, their definition of 'hedonic response' is much broader than what is being offered here as 'hedonic aspect of attitudes'. Here it refers only to anticipated or experienced pleasure reactions to an event or a group/ sequence of events. The multisensory or fantasy experiences most likely are powerful determinants of this emotional reaction, but they are not considered here to be part of it. Their place is earlier in the causal chain leading to attitudes. A parallel to the above distinction is the relationship between beliefs and attitudes. Rosenberg (1956) considers beliefs to be part of the attitude construct. Fishbein (1967), on the other hand, considers beliefs to be determinants of attitudes, but not part of it. The probable reason for this distinction is that Rosenberg predicts that beliefs can directly affect behaviors while Fishbein assumes that their effect on behaviors is totally mediated by attitudes.
If it is true that the essence of hedonic response can be measured by the pleasure reaction, we are probably much closer to developing an instrument to measure total affect, than if we have to include also multisensory and fantasy measures. Of course, if the purpose is to diagnose emotional reactions, and understand their determinants, we need to develop instruments and methods to study these phenomena. Furthermore, even though it is valuable to measure emotional reactions during the purchase and consumption behaviors for better diagnostics, for the purpose of measuring affective predispositions, it should be adequate to measure anticipated pleasure reactions. The hedonic aspect, however, does not incorporate all the motivational forces. We need to measure also the utilitarian aspect and the general aspect.
THE OVERALL ATTITUDE CONSTRUCT
The utilitarian aspect is considered to be separate from the hedonic aspect, not a subconstruct of it, as Hirschman and Holbrook (1982) seem to be implying. Again, for the purpose of measuring affective predisposition it should be adequate to measure the anticipated utilitarian value of the behavior. For this purpose it is not necessary to know how different pieces of information were processed to come up with the value.
It is anticipated that the utilitarian and hedonic aspects are normally positively correlated. For example, most cognitive balance theories would predict that.
It is also hypothesized that the utilitarian and hedonic aspects of attitudes together determine the third attitudinal construct that I call the general aspect of an attitude. This third aspect comes closest to the traditional notion of attitude. It relates to the general pro or con orientation toward the behavior. Is the behavior seen as good or bad, positive or negative? The reason I do not conceptualize this general aspect to be the sole or even the cent al part of the e- all attitude construct is because I hypothesize that the hedonic ant/or the utilitarian aspects have also a direct impact on behaviors. Their impact is not totally mediated by the general aspect (cf. Fishbein 1967). This is especially likely to be the case when the same behavior takes place under different circumstances. The demands of the situation and the mood of the individual affect the relative influences.
Figure 1 summarizes the proposed overall attitude construct. In the nutshell, what is being proposed here is that the general aspect of an attitude is a function of two basic components 1. hedonic, and 2. utilitarian. We can measure the general aspect by using scales like 'positive/negative', 'good/bad', 'favorable/unfavorable'. The hedonic component can be measured by using scales like 'pleasing/annoying', 'happy/sad', 'beautiful/ugly', and the utilitarian component by using scales like 'valuable/worthless', 'foolish/wise', 'rational/irrational'. There are probably significant individual, cultural, and situational differences that determine the relative weights for the three components as determinants of behaviors.
I have recently conducted studies where attitudes toward several different behaviors in different populations were measured using evaluative dimension of the semantic differential with a large number of scales. Preliminary findings indicate that when the scales are further factor analyzed, two factors emerge. One factor has the hedonic scales consistently loading on it, and the other factor has the utilitarian scales loading on it. General aspect scales often have high factor loadings on both factors. Sometimes one factor is dominant, sometimes the other. This pattern appears systematic. For example, 'eating candy' has the hedonic factor more dominant, while 'doing laundry' has the utilitarian factor more dominant. The less dominant factor, however, explains also a sizeable part of the total variance in the data. This data is still being analyzed using more sophisticated techniques, such as LISREL, and has not been reported in any published form yet.
SOME RELATED ISSUES
An interesting related issue is pointed out by Hirschman and Holbrook (1982). That is, it is well documented that people often choose to experience something they expect will cause them emotional pain. Masochism aside, this phenomenon can be explained by utilitarian reasons. As pointed out in the same article, the reason often seems to be a need to learn to teal better with past unhappy events. I would call this utilitarian. Another utilitarian reason might simply be a need to learn more to be better prepared for some future situation. An avid anticommunist may want to read Das Kapital to be more able to argue against communism.
Another related issue is how the proposed conceptualization relates to the work by Fishbein. My interpretation here is that even though Fishbein does not distinguish between hedonic and utilitarian aspects of attitudes, he did try to get to utilitarian aspects of behaviors through his "personal norms" component (even though the relationship between personal norms and utilitarian motives may not be quite straightforward). My recommendation differs significantly from the early and since discarded conceptualization by Fishbein in that it considers the utilitarian aspect to be part of the attitude, not a separate personal normative component like Fishbein conceptualized.
Furthermore, the proposed conceptualization requires a ---h more careful and restrictive use of scales that load on the evaluative dimension of the semantic differential. Some of those scales are general attitudinal, some are hedonic, and some utilitarian. It is hoped that when the different and unique motivational aspects of these components are considered, better behavioral predictions will also follow. It is, by no means, suggested here that attitudes are the sole determinant of behaviors, but they certainly seem to be one of the most important. Consequently, improvements in their measurement should have a significant effect.
Fishbein, Martin E. (1966) "Sexual behavior and propositional control," paper presented at the annual meeting of the Psychometric Society.
Fishbein, Martin E. (1967) Readings in attitude theory and measurement, Wiley & Sons, pp. 477-492.
Fishbein, Martin E. and Icek Ajzen (1975) Belief, attitude, intention, and behavior: An introduction to theory and research. Reading: Addison-Wesley.
Hirschman, Elizabeth C. (1983) "Aesthetics, ideologies and the limits of the marketing concept," Journal of Marketing, 47 (Summer), 45-55.
Hirschman, Elizabeth C. and Morris B. Holbrook (1982) "Hedonic consumption: Emerging concepts, methods and propositions," Journal of Marketing, 46 (Summer), 92-101.
Hirschman, Elizabeth C. and Michael R. Solomon (1984) "Utilitarian, aesthetic, and familiarity responses to verbal versus visual advertisement," Advances in Consumer Research, 11, 426-431.
Holbrook, Morris B. (1983) "Product imagery and the illusion of reality: Some insights from consumer esthetics," Advances in Consumer Research, 10, 65-71.
Holbrook, Morris B. and Elizabeth C. Hirschman (1982) "The experiential aspects of consumption: Consumer fantasies, feelings, and fun," Journal of Consumer Research, 2 (September), 132-140.
Holbrook, Morris B. and William L. Moore (1981) "Cue configurality in esthetic responses," in Hirschman and Holbrook (eds.) Symbolic consumer behavior, Ann Arbor: Association for Consumer Behavior. 16-25.
Likert, Rensis (1932) "A technique for the measurement of attitudes," Archives of Psychology, No. 140.
Osgood, Charles E., Suci, George S. and Perry H. Tannenbaum (1957) The measurement of meaning, Urbana: The University of Illinois Press.
Rosenberg, Milton J. (1956) "Cognitive structure and attitudinal affect," Journal of Abnormal and Social Psychology, 53, 367-372.
Unger, Lynette S. and Jerome B. Kernan (1983) "On the meaning of leisure: An investigation of some determinants of the subjective experience," Journal of Consumer Research, 2 (March), 381-392.