Measuring the Hedonic and Utilitarian Dimensions of Attitude: a Generally Applicable Scale

ABSTRACT - The first phase of a multi-phase process for developing a generally applicable, reliable, and valid scale for measuring the hedonic and utilitarian (HED-UT) components of attitudes is presented. Six product categories and six brands within those categories were assessed on both HED-UT dimensions. Relationships between the HED-UT scale and involvement, sensation seeking, and need for cognition were examined. Level of involvement was found to have a significant effect on both hedonic and utilitarian dimensions associated with stimuli. The resulting scale should be useful in determining the nature of customer evaluation of products and services and/or related advertising appeals.

Citation:

Eric R. Spangenberg, Kevin E. Voss, and Ayn E. Crowley (1997) ,"Measuring the Hedonic and Utilitarian Dimensions of Attitude: a Generally Applicable Scale", in NA - Advances in Consumer Research Volume 24, eds. Merrie Brucks and Deborah J. MacInnis, Provo, UT : Association for Consumer Research, Pages: 235-241.

Advances in Consumer Research Volume 24, 1997      Pages 235-241

MEASURING THE HEDONIC AND UTILITARIAN DIMENSIONS OF ATTITUDE: A GENERALLY APPLICABLE SCALE

Eric R. Spangenberg, Washington State University

Kevin E. Voss, Washington State University

Ayn E. Crowley, Drake University

ABSTRACT -

The first phase of a multi-phase process for developing a generally applicable, reliable, and valid scale for measuring the hedonic and utilitarian (HED-UT) components of attitudes is presented. Six product categories and six brands within those categories were assessed on both HED-UT dimensions. Relationships between the HED-UT scale and involvement, sensation seeking, and need for cognition were examined. Level of involvement was found to have a significant effect on both hedonic and utilitarian dimensions associated with stimuli. The resulting scale should be useful in determining the nature of customer evaluation of products and services and/or related advertising appeals.

Consumer researchers are increasingly looking for a richer understanding of the traditional information processing view of the various dimensions of attitude. Published efforts to establish a reliable and valid scale which measures the hedonic and utilitarian components of attitudes held by consumers regarding products, services, and activities are becoming more common. However, no widely applicable, satisfactory scale has been developed to date. Batra and Ahtola (1990) developed an eight item semantic differential scale using specific products as targets and suggested it be examined in future research. In further research, Crowley, Spangenberg, and Hughes (1992) found the Batra and Ahtola scale severely lacking when applied to product categories as opposed to specific products. Babin, Darden, and Griffin (1994) reported development of a scale based on the hedonic and utilitarian evaluation of shopping as an activity. The resulting scale, due to the focus only on shopping behavior, lacks general applicability to products, service, and other non-shopping activities. This paper reports the first phase of a multi-phase effort to develop a scale (HED/UT) which more closely captures the hedonic and utilitarian components of attitude experienced by consumers across a variety of situations and attitude targets.

BACKGROUND

Investigation of the hedonic and utilitarian components of consumption has been addressed in such various disciplines as sociology, psychology, and economics. One author in the field of economics stated that, "We use goods in two ways. We use goods as symbols of status and simultaneously as instruments to achieve some end-in-view" (Hamilton, 1987, p. 1541). This view clearly combined the hedonic and utilitarian views of consumption, echoing a parallel theoretical development that has occurred in consumer research (Hirschman and Holbrook, 1982).

Interest in the hedonic/utilitarian construct in the discipline of marketing builds on a series of articles by Hirschman and Holbrook. Based on earlier work in motivation research (e.g., Dichter, 1960), Hirschman and Holbrook (1982) developed an experiential view of hedonic consumption centered primarily around aesthetic products such as novels, movies, and art. Their resulting propositions, however, extend to all product classes. Defining hedonic consumption as "those facets of consumer behavior that relate to the multisensory, fantasy, and emotive aspects of one’s experience with products" (p. 92), several detailed propositions were developed under the general thesis that hedonically consumed products stimulate internal imagery and emotional arousal based on externally sensed, product related stimuli.

Contrasting the experiential perspective to the more traditional information-processing view, Holbrook and Hirschman (1982) concluded that a synthesis is necessary if the science of marketing is to develop an "enlarged view" of consumer behavior. One aspect of their article was problematic, however, since they apparently viewed attitude as a component of affect. In their discussion of affect they stated that the traditional information processing approach has focused on: "only one aspect of hedonic response C namely, like or dislike of a particular brand (attitude) or its rank relative to other brands (preference). This attitudinal component represents only a tiny subset of the emotions and feelings of interest to the experiential view (Holbrook and Hirschman, 1982, p. 136)."

The aforementioned conception of affect is contrary to the more recent suggestion of Cohen and Areni (1991) that affect is a valenced feeling state. Emotion and mood are examples of these states. On the other hand, attitude, being evaluative in nature, contains a cognitive element and should be distinct from affect. Much work has accumulated regarding the affective and cognitive dimensions of attitude (see e.g., Abelson et al., 1982; Ajzen and Driver, 1992; Bagozzi and Burnkrant, 1979; Breckler and Wiggins, 1989; Fishbein and Ajzen, 1974; and Madden et al., 1988). We offer a competing hypothesis. We suggest that hedonic value is experienced on both affective and cognitive levels while the utilitarian component, which also may include both affective and cognitive dimensions, is dominated by the cognitive element. The affective and cognitive components of hedonic consumption imply that emotional desires compete with utilitarian motives in purchase and consumption decisions. This line of reasoning is linked to the growing body of research concerning compulsive consumption (see e.g., Hirschman, 1992). Compulsive consumers abuse purchasing the same way alcoholics abuse alcohol. They will purchase to achieve emotional arousal without regard to cognitively arrived at conclusions regarding financing, use, impact on others, etc. Short term arousal is sought at the cost of "significant negative consequences" and a disruption of "normal life functioning" (Faber and O’Guinn,1989, p. 738). It may be possible that compulsive consumers represent the extreme end of a continuum. That is, all consumers may engage in an internal negotiation, motivated by need, to satisfy both hedonic and utilitarian consumption desires.

Hirschman (1983) proposed that some offerings, such as art and ideologies, are almost entirely hedonic. Value for the consumer stems from the subjective response evoked, there being little utility derived from, say, a painting. Establishing a more developed theory, Hirschman (1984) considered all consumption an "experience seeking" phenomenon. Accordingly, Hirschman decomposed experience seeking into three alternatives: 1) Cognition seeking, 2) Sensation seeking and 3) Novelty seeking. Cognitive experience seekers desire to stimulate, or activate, thinking. Sensation seekers desire to experience consumption through one or more of the five senses. Novelty seekers are looking for unique, fresh sources of stimulation through consumption. That is, the perceived uniqueness of a given product, service, or activity may create hedonic value for the consumer. A novel experience realized through consumption, such as finding a new restaurant or "trekking," may serve more than one hedonic benefit by acting as a status symbol in the consumer’s reference group as well as providing temporal enjoyment. A consumer may realize hedonic value from novel experiences on a cognitive and/or emotional level (Hirschman, 1984). Hirschman seems to have adopted the position that consumers occupy either a high, low, or average posture as a cognition seeker, sensation seeker, or novelty seeker. These postures are hypothesized to be static over time. This position would seem to be contrary to the body of thought on hedonic consumption which implies shifting value orientations dependent on the specific stimulus and the consumer’s involvement with that stimulus (see e.g., Holbrook, 1986). Havlena and Holbrook (1986) argued that hedonic and utilitarian components of attitude are situational when they asserted that although "consumption experiences vary in their mix of utilitarian/hedonic, tangible/intangible, or objective/subjective components, the latter, more emotional aspects of consumption experiences occur to a greater or lesser degree in almost all consuming situations (p. 394)."

In the hedonic consumption view, the high interest and involvement generated by aesthetic products is strongly emphasized. Thus, a strong link between involvement and hedonic consumption is implied (Hirschman and Holbrook, 1982; Holbrook and Hirschman, 1982). Stimuli that are highly involving should result in more intense and arousing affective reactions as compared to more peripheral stimuli (Cohen and Areni, 1991). Involvement develops as the consumer experiences intrinsic and extrinsic satisfaction from interaction with the product (Bloch and Bruce, 1984). Stimuli are not inherently involving, consumers possess the capacity to be involved and this involvement will necessarily fluctuate from consumer to consumer and product to product (Traylor and Joseph, 1984). Therefore, high levels of involvement should logically be associated with high levels of hedonic response.

Two additional components, mood and mental imagery, are relevant in any discussion of the hedonic and utilitarian components of attitude. Mood may be a key variable in determining specific hedonic consumption behavior. Mood is generally described as a feeling state of longer duration and lower intensity than emotions. Where emotions are stimuli specific, consumers may experience a mood without being aware of its time of origin, cause, influence, or perhaps even its presence (Cohen and Areni, 1991). It has been suggested that consumers might use consumption as a means to change undesirable moods or achieve highly desired moods (e.g., Babin, Darden, and Griffin, 1994). Further, it has been reported that consumers who experienced negative prior-to-purchase moods, and implemented an information processing approach to purchase behavior, experienced positive post-purchase moods (Gardner and Hill, 1988). It is possible that products judged to be primarily hedonic in consumption valu may be an instrument through which consumers manage mood states.

Mental imagery may be stimulated by the hedonic component of the consumption experience. Hirschman and Holbrook (1982) identified two possible forms: 1) historical imagery based on actual events in the subject’s life and 2) fantasy (projective) imagery of events that, while perhaps referenced to historical events, have not actually taken place. This aspect would appear especially important in advertising contexts. Advertisers have historically used the power of communication to relate their products to images thought to appeal to consumers. It is, therefore, possible that successful measurement of hedonic consumption may also help to gauge the extent to which such images are adopted by consumers. Among other applications, a reliable and valid HED/UT scale would be useful in evaluating the effectiveness of advertisers in convincing consumers to assign a higher, or perhaps a lower, hedonic value to their offering.

EXISTING MEASURES OF THE HEDONIC/UTILITARIAN CONSTRUCTS

Batra and Ahtola (1991) reported their development of an eight-item semantic differential scale purportedly measuring the hedonic and utilitarian components of attitude toward products. The scale was developed primarily with specific brands; despite performing acceptably regarding tests of reliability and validity , it suffered inconsistent loadings on stimuli, perhaps due to a concept x scale interaction. The scale also failed to incorporate many of the theoretical concepts (e.g., involvement) outlined by Hirschman and Holbrook (1982) as relevant components of hedonic consumption. Further, as noted earlier, the scale also failed to perform adequately in a subsequent generalization to product categories (Crowley, Spangenberg, and Hughes, 1992).

Crowley, Spangenberg, and Hughes (1992) showed some interesting results mapping product ratings in a two-dimensional space using the eight Batra and Ahtola items. Specifically, almost all of the product categories appeared in the same quadrant indicating relatively high levels of both hedonic and utilitarian value. This may be an artifact of the same concept x scale interaction observed in the original Batra and Ahtola article. Moreover, the eight adjectives chosen may not have adequately reflected the construct since some apparently utilitarian products appeared in very close proximity to items that would seem more hedonic. As an example, kitchen utensils were positioned very close to vacation resorts on the two-dimensional map reported by Crowley, Spangenberg, and Hughes (1992).

More recently, Babin, Darden, and Griffin (1994) reported development of a scale for assessing the hedonic and utilitarian components of shopping behavior (Personal Shopping Value Scale). Their scale is highly specific to retail shopping activity and is not generalizable to products, services, or other activities. Similar to the Batra and Ahtola (1991) work, this study failed to include some of the theoretical constructs (e.g., involvement) proposed by Hirschman and Holbrook (1982). Perhaps more troublesome was the procedure employed to validate the scale within the general population. Students accompanied respondents on a shopping trip to a local mall to observe the respondents shopping behavior and administer the survey. Respondents were informed that students were receiving course credit for the investigation. Upon examination, it appears a strong possibility that this procedure could have introduced demand artifacts into the scale development process. The Personal Shopping Value Scale should be appropriately validated before confidence can be placed in its ability to assess the hedonic and utilitarian value experienced by shoppers.

A generally applicable scale useful for measuring the hedonic and utilitarian components of attitude toward products and services has yet to be developed. This paper reports the first phase in the development of a valid and reliable scale tha can fill this void in consumer research.

SCALE DEVELOPMENT

A pool of adjective pairs was generated in two steps. First a thorough review of the literature suggested many possibilities. Then, pretest subjects were asked to describe their attitudes towards products and services. The results of both processes were reviewed and duplicate pairs were dropped. In total, 27 distinct adjective pairs were generated for use in a seven-point semantic differential scale format. Since, as discussed above, there are many factors which may interact in determining the level of hedonic and/or utilitarian attitudes held by consumers, a reliable and valid scale must isolate the hedonic and utilitarian values from possible confounding variables. Accordingly, in addition to the 27 adjective pairs, three additional multiple item scales, to be used for validation checks within a larger nomological network, were jointly administered to the respondents. These scales included a measure of involvement, sensation seeking, and need for cognition.

The notion that product involvement is closely linked to the level of hedonic meaning the consumer associates with the stimuli has strong face validity (Hirschman and Holbrook, 1982). As discussed above, high levels of involvement should lead to stronger hedonic values. It seems reasonable that a scale measuring consumer involvement should be positively associated with a scale seeking to measure hedonic components of attitude. Accordingly, the 10-item reduced form of the Personal Involvement Inventory (PII) (Zaichkowsky, 1990) was administered concurrently with the pool of hedonic and utilitarian semantic differential adjective pairs.

In each of the two existing hedonic and utilitarian scales presented above, positive emotion is characterized strictly as a hedonic value while negative emotions have been attributed to utilitarian value. This is not the meaning ascribed to hedonic consumption by Hirschman and Holbrook (1982), who pointed out that emotions can be positive or negative and still be hedonic; horror movies or stomach turning roller coaster rides, for example, provide little in the way of utilitarian benefits. Both of these activities provide high emotional arousal through what might be considered a negative emotion C fear. Perhaps a more accurate word is one commonly used by bungee jumpers C a "rush". Accordingly, a valid measure of hedonic consumption seeks to identify strong emotional reactions to stimuli, but is sympathetic to the idea that consumers may consume for emotional reactions (sensation seekers) that might not generally be considered pleasant. Therefore, an adaptation of Zuckerman’s (1979) 38-item Sensation Seeking Scale (SS) was included to assess subjects’ predilection to affective responses.

The evaluative judgment necessary to form attitudes regarding stimuli theoretically contains a strong cognitive component. Emotional reactions to stimuli have been linked to judgments regarding the desirability of the outcome as well as the appeal of the activity (Ortony, Clore, and Collins, 1988). Further, stimuli that match expectations may interact with the consumer’s cognitive structure to determine the affective response (Cohen and Areni, 1991). Thus, cognition and affect are intertwined in the evaluative process necessary to develop attitudes about products, services, and activities. Cognitive processing then is potentially present in both hedonic and utilitarian product evaluations. Accordingly, the 18-item short form of the Need for Cognition Scale (NFC) (Cacioppo, Petty, and Kao, 1984) was also administered.

PROCEDURE AND SUBJECTS

Questionnaire booklets including our initial 27 items and the three additional multi-item scales,were prepared for rating the following branded products: IBM Personal Computers, Lay’s Potato Chip’s, Hilton Vacation Resorts, Wesson Cooking Oil, Dawn Dish Detergent, and Nike Athletic Shoes. Product categories and specific brands were selected in a separate, 2-stage pretest: 1) List products or services purchased predominantly for pleasure/utility, and 2) Within category "x", what brand first comes to mind? The most commonly mentioned brands within each product category were selected to ensure that respondents would be familiar with, and hence able to render judgement on, the target brands. Booklets were also prepared for the product category associated with each brand. Thus, there were a total of twelve versions of the initial combined measures. Subjects in both the pretest and main study were students at Washington State University. In the main study, subjects completed only one of the twelve versions of our instrument in return for course credit in a single lab session. After removing booklets exhibiting non-response problems, six hundred and eight useable responses were obtained.

RELIABILITY AND ITEM REDUCTION

Principal components exploratory factor analysis using the intial 27 items was conducted separately for both specific branded products and product categories. Rotation and the scree plots indicated that a two-factor solution best explained the data. All of the items loaded, as expected, on the first two factors with the exception of the hedonistic/not hedonistic and utilitarian/not utilitarian adjective pairs which were subsequently dropped (note that these terms may have theoretical meanings different than generally understood meanings) due to low item-to-total correlations and factor loadings. Internal consistency reliability was assessed on both the brand and product category responses. In the case of the brands, the 27 adjective pairs produced a coefficient alpha of .89. For the product categories, alpha was .92. Because we are proposing two dimensions, it is appropriate to compute alpha for each (Churchill, 1979). When used to measure brands both the 13 item hedonic and 14-item utilitarian subscales produced an alpha of .94. For product categories, alpha was .95 and .93 for the hedonic and utilitarian components respectively. Item-to-total correlations were examined separately for each dimension with both the brand and product category data. Deleted adjective pairs with values less than .50 were the hedonistic/not hedonistic and utilitarian/not utilitarian pairs. Recalculation of item-to-total correlations revealed another unsatisfactory value for the convenient/not convenient adjective pair which was also dropped.

Thus, the scale was reduced to 24 adjective pairs with 12 pairs in each of the HED/UT subscales. Dropping the three adjective pairs with low item-to-total correlations did not substantially change coefficient alpha. Thus, initial reliability estimates for the HED/UT scale were encouraging.

CONSTRUCT VALIDITY

Confirmatory factor analysis (CFA) was conducted using EQS (Bentler 1993). For the unidimensional, single-factor, brand name model: c²=885.0 (df=250, p<.001); Bentler-Bonett fit indexes were normed=.86, nonnormed=.88, and CFI=.87. For the single factor model regarding product categories: c²=940.8 (df=250, p<.001); Bentler-Bonett indexes were normed=.84, nonnormed=.86, and CFI=.87. Although the overall the overall c² statistic was significant, it was within the rule of 2.5 to 3 times the number of degrees of freedom, suggested as acceptable by Bollen (1989), a significant improvement over the null model for both brand names and product categories [c²=6110.2 (276 df) for brands; 5777.0 (284 df) for product categories]. For both thebrand names and product categories, the items loaded, as predicted by the exploratory factor analysis, on two factors providing strong evidence of construct validity (Gerbing and Anderson, 1988). CFA factor loadings and item-to-total correlations are presented in Table 1. [EFA results are available upon request from the first author.] Also shown in Table 1 are reliabilities and average variance explained (AVE) calculated according to Fornell and Larcker (1981). In all cases reliability was>.91 and AVE ranged from .47 to .50.

In order to test the relationship between involvement and the HED/UT scale the adjective pairs were separated into their two respective 12-item subscales. We fit the data to ANCOVA models using the summated hedonic subscale and the summated utilitarian subscales as dependent variables. In addition to the classification variable level of involvement, summated results from the modified Sensation Seeking scale (SS) and the short form of the Need for Cognition scale (NFC) were included as covariates. The hedonic subscale showed a non-significant regression relationship (F=. 14, p=.87). Level of involvement, however, was a significant explanatory variable (F=63.4, p=.0001). Against the utilitarian subscale, the ANCOVA showed a similar significant result when testing level of involvement (F=62.5, p=.0001). The regression relationship was, again, not significant (F=3.3, p=.12). In order to more completely explore the relation between the HED/UT scale and the two covariate scales (SS and NFC), reduced models were tested. That is, a series of ANCOVA models were fit dropping either the SS or NFC covariate. The only significant covariance relationship was in the model which included the SS scale as a covariate with level of involvement against the utilitarian subscale (F=4.3, p=.04). Based on the weight of statistical evidence, it was concluded that the SS and NFC scales did not significantly covary with our HED/UT scale. Accordingly, we fit regression lines for each summated subscale with the summated PII scale as the independent variable (Hedonic R²=.29; Utilitarian R²=.31). Figure 1 clearly shows that high scores on the PII led to extreme responses on both dimensions of the HED/UT scale. Low levels of involvement are then associated with lower perceived levels of hedonic and utilitarian attitude components. Interestingly, personal computers proved to be the most involving to our sample while potato chips and cooking oil were the least involving (dishwashing liquid=4.25; athletic shoes=5.56; cooking Oil=4.02; personal computers=5.73; potato chips=4.03; vacation resorts=5.43).

TABLE 1

HED/UT SCALE ITEMS" ITEM IDENTIFICATION AND DIMENSIONALITY INDICES

FIGURE 1

REGRESSION OF SUMMATED HEDONIC AND UTILITARIAN SUBSCALES ON SUMMATED PII SCALE

Our results suggest that level of involvement is positively and significantly related to the level of hedonic and/or utilitarian value. Combining all 24 questions would hide this effect since low utilitarian ratings can offset high hedonic ratings. Specifically, there is a strong positive linear relationship between the summated PII and the hedonic and utilitarian subscales. Although we expected high levels of involvement to lead to high levels of hedonic value, it was interesting to find that high levels of involvement also lead to high levels of utilitarian value. It is worth noting that this effect could be mistaken for standard deviation bias. Standard deviation bias is the tendency for respondents to use either wide or narrow ranges of response intervals (Greenleaf, 1992). In Figure 1 it is clear that respondents with extreme responses on the utilitarian subscale also gave extreme responses on the hedonic subscale (i.e. these respondents used a wide range of response intervals). In the present study however, the PII has provided a surprising explanation for this potential artifact.

The HED/UT scale also discriminates between levels of the construct for branded products and product categories. The two dimensional plot in Figure 2 shows the mean responses for each stimulus for both the brand and product category response data. Some observations become immediately clear. First, both Hilton Vacation Resorts and the product category Vacation Resorts were consitently rated highest on the hedonic adjectives. Also, IBM Personal Computers and the product category Personal Computers were consistently rated the most utilitarian. Finally, Lay’s Potato Chips and the product category Potato Chips were seen as the least utilitarian products. Surprisingly, these products were rated approximately equal to the IBM Personal Computer/Personal Computers in Hedonic value. In general, there were no large differences between ratings of branded products and their respective general product categories.

DISCUSSION

The two-dimensional HED/UT scale developed here appears to be a reliable tool for measuring the attitudes held by consumers with respect to specific products and their related categories. Initial tests of construct validity are consistent with our two-dimensional proposition. As our theoretical background suggests, and our results support, involvement appears to have a strong correlation with level of hedonic response. Somewhat unexpectedly, involvement also shows a strong relationship to the utilitarian component of attitude. Thus, it is suggested that product involvement be measured concurrently with the HED/UT scale in future administrations. This is consistent with Crowley, Spangenberg, and Hughes’ (1992) criticism of Batra and Ahtolas’ (1991) lack of accounting for involvement in their work.

Affect, as measured by our adaptation of the SS scale, shows mixed results. Our review suggests that affect would strongly influence the hedonic construct. Our results, however, indicate a link to the utilitarian construct. It is possible, as suggested by the negative slope of the regression coefficient in the ANCOVA model, that respondents with a high emotional predisposition evaluate more functional and mundane products in strictly utilitarian terms. That is, they may be willing to judge products which provide arousal along a broad continuum while non-arousing products are judged in a constricted utilitarian space. A less interesting explanation questions the validity of the SS scale.

This study failed to find any relation between cognition, as measured by the NFC scale, and the hedonic or utilitarian components of attitude as measured by our HED/UT scale. It is possible that since, as we suggested, cognition is involved in both hedonic and utilitarian assessments, the effects were evenly distributed. This, however, is an explanation that our data do not address.

FIGURE 2

CONCLUSION

We developed a scale to measure both the hedonic and utilitarian dimensions of attitude and showed strong indicators regarding reliability and validity. The scale was internally consistent with a stable factor structure across several product categories and specific brands within those categories; explained variance and average variance extracted was higher than reported for most published scales. The scale had the hypothesized correlation with involvement but did not perform as expected in relation to sensation seeking and need for cognition.

The study was limited in that development was based on student subjects; extension to a non-student population in further tests of validity is desirable. We are conducting further research which incorporates scales measuring imagery, negative emotions, mood, novelty, and compulsive consumption in conjunction with the HED/UT scale. We would expect HED/UT to be highly correlated with the aforementioned constructs indicating further convergent validity.

In addition to further validation efforts, we should like to see future research examine the determinants of the hedonic and utilitarian components of attitude. And, while we see several possibilities for practical application of our scale, the most obvious benefits woud accrue to advertisers. If a target market seeks primarily hedonic or utilitarian benefits, appeals could be directed respectively. In fact, marketing researchers may want to investigate whether hedonic- or utilitarian-oriented consumers are more or less responsive to differing methods of sales promotion. Finally, it is possible that our Hed/ut scale may measure attitudes that could explain variety seeking behavior (see e.g., Van Trijp, Hoyer, and Inman, 1996).

REFERENCES

Abelson, Robert P., Donald R. Kinder, Mark D. Peters, and Susan T. Fiske (1992), "Affective and Semantic Components in Political Person Perception," Journal of Personality and Social Psychology, 42 (4), 619-630.

Ajzen, Icek and B. L. Driver (1992), "Contingent Value Measurement: On the Nature and Meaning of Willingness to Pay," Journal of Consumer Psychology, 1 (4), 297-316.

Bagozzi, Richard P. and Robert E. Burnkrant (1979), "Attitude Organization and the Attitude-Behavior Relationship, Journal of Personal and Social Psychology, 37 (6) 913-929.

Batra, Rajeev and Olli T. Ahtola (1990), "Measuring the Hedonic and Utilitarian Sources of Consumer Attitudes," Marketing Letters, 2(2), 159-170.

Babin, Barry J., William R. Darden, and Mitch Griffin (1994), "Work and/or Fun: Measuring Hedonic and Utilitarian Shopping Value," Journal of Consumer Research, 20 (March), 644-656.

Bentler, Peter (1993), EQS Structural Equations Program, Ver 3.0. Los Angeles, CA: BMPD Statistical Software, Inc.

Bloch, Peter H. and Grady D. Bruce (1984), "Product Involvement as Leisure Behavior," in Thomas C. Kinnear (ed.), Advances in Consumer Research, Vol. 11 (pp. 197-202). Ann Arbor, MI: Association for Consumer Research.

Breckler, Steven J. and Elizabeth C. Wiggins (1989), "Affect versus Evaluation in the Structure of Attitudes," Journal of Experimental Social Psychology, 25, 253-271.

Bollen, Kenneth A. (1989), Structural Equations with Latent Variables. New York: John Wiley, Inc.

Cacioppo, John T., Richard E. Petty, and Feng Chuan Kao (1984), "The Efficient Assessment of Need for Cognition," Journal of Personality Assessment, 48 (3), 306-307.

Churchill, Gilbert A. Jr. (1979), "A Paradigm for Developing Better Measures of Marketing Constructs," Journal of Marketing Research, 16 (February), 67-73.

Cohen, Joel B. and Charles S. Areni (1991), "Affect and Consumer Behavior," in Thomas S. Robertson and Harold H. Kassarjian (Eds.), Handbook of Consumer Behavior. Englewood Cliffs, NJ: Prentice Hall.

Crowley, Ayn E., Eric R. Spangenberg and Kevin R. Hughes (1992), "Measuring the Hedonic and Utilitarian Dimensions of Attitudes toward Product Categories," Marketing Letters, 3: (3), 239-249.

Dichter, Ernest (1960), The Strategy of Desire. New York, NY: Doubleday.

Faber, Ronald J. and Thomas C. O’Guinn (1989), "Classifying Compulsive Consumers: Advances in the Development of a Diagnostic Tool,’ in M. Wallendorf and P. Anderson (Eds.), Advances in Consumer Research, Vol. 16 (pp. 132-135). Provo, UT: Association for Consumer Research.

Fishbein, Martin and Icek Ajzen (1974), "Attitudes towards Objects as Predictors of Single and Multiple Behavioral Criteria," Psychological Review, 81 (1), 59-74.

Fornell, Claes and David F. Larcker (1981), "Evaluating Structural Equation Models with Unobservable Variables and Measurement Error," ournal of Marketing Research, 28 (February), 39-50.

Gardner, Meryl P. and Ronald Paul Hill (1988), "Consumer’s Mood States: Antecedents and Consequences of Experiential versus Informational Strategies for Brand Choice," Psychology and Marketing, 5 (Summer), 169-182.

Gerbing, David W. and James C. Anderson (1988), "An Updated Paradigm for Scale Development Incorporating Unidimensionality and its Assessment," Journal of Marketing Research, 25 (May), 186-192.

Greenleaf, Eric A. (1992), "Improving Rating Scale Measures by Detecting and Correcting Bias Components in Some Response Styles," Journal of Marketing Research, 29 (May), 176-188.

Hamilton, David B. (1987), "Institutional Economics and Consumption," Journal of Economic Issues, 21 (December), 1531-1553.

Havlena, William J. and Morris B. Holbrook (1986), "The Varieties of Consumption Experience: Comparing Two Typologies of Emotion in Consumer Behavior," Journal of Consumer Research, 13 (December), 394-404.

Hirschman, Elizabeth C. (1983), "Aesthetics, Ideologies and the Limits of the Marketing Concept," Journal of Marketing, 47 (Summer), 45-55.

Hirschman, Elizabeth C. (1984), "Experience Seeking: A Subjectivist Perspective of Consumption," Journal of Business Research, 12, 115-136.

Hirschman, Elizabeth C. (1992), "The Consciousness of Addiction: Toward a General Theory of Compulsive Consumption," Journal of Consumer Research, 19, 155-179.

Hirschman, Elizabeth C. and Morris B. Holbrook (1982), "Hedonic Consumption: Emerging Concepts, Methods and Propositions," Journal of Marketing, 46 (Summer), 92-101.

Holbrook, Morris B. (1986), "Emotion in the Consumption Experience: Toward a New Model of Consumer Behavior," in R. A. Peterson, W.D. Hoyer, and W. R. Wilson (Eds.), The Role of Affect in Consumer Behavior: Emerging Theories and Applications. Lexington, MA: Heath.

Holbrook, Morris B. and Elizabeth C. Hirschman (1982), "The Experiential Aspects of Consumption: Consumer Fantasies, Feelings, and Fun," Journal Of Consumer Research, 9 (September), 132-140.

Madden, Thomas J., Chris T. Allen, and Jacquelyn L. Twible (1988), "Attitude toward the Ad: An Assessment of Diverse Measurement Indices Under Different Processing 'Sets’," Journal of Marketing Research, 25 (August), 242-52.

Ortony, A., G. L. Clore, and A. Collins (1988), The Cognitive Structure of Emotions. New York, NY: Cambridge University Press.

Traylor, Mark B. and W. Beanie Joseph (1984), "Measuring Consumer Involvement in Products: Developing a General Scale," Psychology and Marketing, 1 ( Summer), 65-77.

Van Trijp, Hans C. M., Wayne D. Hoyer, and J. Jeffrey Inman (1996), "Why Switch? Product Category-Level Explanations for True Variety Seeking Behavior," Journal of Marketing Research, 33 (August), 281-292.

Zaichkowsy, Judith Lynne (1990), "The Personal Involvement Inventory: Reduction and Application to Advertising," Unpublished manuscript, Burnaby, BC: Simon Fraser University.

Zuckerman, Marvin (1979), Sensation Seeking: Beyond the Optimal Level of Arousal. Hillsdale, NJ: Lawrence Erlbaum Associates.

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