Investigating the Experiential Dimensions of Product Evaluation

Lawrence J. Marks, Kent State University
Susan Higgins, Kent State University
Michael A. Kamins, University of Southern California
ABSTRACT - The study investigates the application of the Phenomenology of Consciousness Questionnaire (PCQ), an instrument which measures nine experiential dimensions of consciousness, to a product evaluation situation. Subjects with high and low levels of product experience evaluated one of two products (either one which is typically judged on a utilitarian basis or one which is typically judged on a hedonic basis), and reported how they experienced the evaluation situation, their attitude toward the product, and their behavioral intentions. The-results suggest that including a measure of how consumers experience products can improve the relationship between attitudes and behavioral intentions, and in some cases even replace the traditional attitude measure.
[ to cite ]:
Lawrence J. Marks, Susan Higgins, and Michael A. Kamins (1988) ,"Investigating the Experiential Dimensions of Product Evaluation", in NA - Advances in Consumer Research Volume 15, eds. Micheal J. Houston, Provo, UT : Association for Consumer Research, Pages: 114-121.

Advances in Consumer Research Volume 15, 1988      Pages 114-121

INVESTIGATING THE EXPERIENTIAL DIMENSIONS OF PRODUCT EVALUATION

Lawrence J. Marks, Kent State University

Susan Higgins, Kent State University

Michael A. Kamins, University of Southern California

ABSTRACT -

The study investigates the application of the Phenomenology of Consciousness Questionnaire (PCQ), an instrument which measures nine experiential dimensions of consciousness, to a product evaluation situation. Subjects with high and low levels of product experience evaluated one of two products (either one which is typically judged on a utilitarian basis or one which is typically judged on a hedonic basis), and reported how they experienced the evaluation situation, their attitude toward the product, and their behavioral intentions. The-results suggest that including a measure of how consumers experience products can improve the relationship between attitudes and behavioral intentions, and in some cases even replace the traditional attitude measure.

INTRODUCTION

Recently there has been increasing dissatisfaction with the traditional models which have been used to examine and predict consumer behavior. These models have been based on the assumption that consumer decision making is primarily based on rational, cognitive processes (Peterson, Hoyer, and Wilson 1986). Thus, researchers have become interested in other factors influencing consumer decision making, including feeling, emotions, and experiences. This paper focuses on the experiential dimension of product evaluation.

BACKGROUND

Most current models of the consumer product evaluation process are cognitively oriented. That is, they portray the consumer as forming product judgments largely on a conscious and rational basis. For example, Fishbein's (1975) model of attitude toward the object (Ao) has as its foundation consumers' beliefs about the object (bi) and evaluation of those beliefs (ei). The belief component is a cognitive factor. The evaluation of the belief represents the affective component. However, as it has been generally operationalized, this affective component has been largely treated as though it is based on a rational, cognitive evaluation of the beliefs. Petty and Caccioppo (1986) have argued that this type of general evaluation model fails to fully capture the purely affective component of attitude. Miller and Tessar (1986) contend that in many cases, behavior is not based on a general evaluation of the situation, but instead is contingent, to a greater or lesser degree, on the cognitive or affective component of attitude. That is, some types of buyer behavior may be more cognitively driven and other types may be more affectively driven.

In line with Miller and Tessar (1986), Hirschman and Holbrook (1981) have suggested that we need to consider an experiential factor in studying consumers' behavior. They suggest that consumer fantasies and subjective reactions to their general life experiences and to their product consumption experiences have an important influence on their behavior. From this perspective, ignoring the experiential dimension limits our understanding of consumer behavior. For example, with regard to affect, they note that the information processing treatment of affect along a continuum of "like-dislike" only represents a very small subset of the emotions and feelings from the experiential view. This suggests that both marketing academics and strategists may want to consider the potential impact of the experiential dimension of attitude on consumer behavior.

In order to implement a more experiential approach to consumer behavior, it is necessary to find a way to measure how consumers experience their world. Very little effort in marketing has been directed toward this problem. However, Pekala and Levine (1981) have developed a methodology for mapping peoples' subjective experiences through the retrospective completion of a self-report inventory. They call the inventory the Phenomenology of Consciousness Questionnaire (PCQ).

The PCQ was developed as a phenomenological methodology for assessing and mapping the structure of consciousness with the idea that it might be used to determine the phenomenological parameters associated with any stimulus condition of interest. It was based on the work other researchers had done measuring states of consciousness (e.g., Battista 1978, Krippner 1972, Ludwig 1972, and Tart 1975, 1977) and resulted in a scale measuring nine dimensions using thirty-seven items. The authors conclude that "the thirty-seven items of the...PCQ cannot map all the nuances of subjective experience, [but that] it can do reasonably well in assessing nine major dimensions of consciousness' (Pekala and Levine 1981, p.44). This instrument provides a potential method of measuring how consumers experience products.

Thus, the central purpose of the current study is to perform an exploratory study to investigate whether the experiential dimension of product evaluation might be measured using the PCQ, and how it might help to understand consumers' attitude formation for both cognitively evaluated and affectively evaluated products.

RESEARCH QUESTIONS

Our proposal is to investigate consumer attitude formation for cognitively and affectively evaluated products using both a traditional model of attitude and the experiential dimension. There is no generally agreed on classification of products into cognitively or affectively evaluated ones. However, it would appear that in many circumstances, some product evaluations are based primarily on cognitive judgments about physical brand attributes (consistent with the Fishbein and Ajzen 1975 approach), while in other situations product evaluations are based primarily on phenomena such as the conditioning of affect, mere exposure, nonutilitarian consumption and use effects, and affective reactions to stimulus characteristics. These may be seen as utilitarian/cognitive and hedonic/affective based evaluations (cf. Batra 1986).

A variety of factors may influence whether a product will be evaluated based on a utilitarian or hedonic basis. For example, the visceral anxiety many consumers have about shopping for and using new technologies may be a factor which causes consumers to use a more affectively-driven evaluation of the product; Of course, the degree of anxiety is expected to be less for consumers who already are experienced with the product than for those who have little product-related experience. This suggests experienced consumers might tend to exhibit more cognitively driven decision styles. On the other hand, it seems unlikely that consumers would have a similar fear of evaluating and shopping for low-tech products. Even those who have relatively little experience may feel no anxiety about this process.

Thus, several questions motivated this study:

1. Can an existing instrument developed in another field, for another purpose, be effectively used to measure how consumers experience products. In particular, are the dimensions the same? Does the instrument appear to have internal consistency? Does it work equally well for affectively and cognitively evaluated products and for consumers possessing varying levels of product-related experiences?

2. How does the inclusion of experiential dimensions affect the existing models of attitude? Does it improve the relationship between measures of attitude and measures of behavioral intentions?

3. Can one model of attitude apply equally well to understanding consumers' evaluations of both cognitively and affectively evaluated products?

METHODOLOGY

Subjects

Seventy-three undergraduates from a major Midwest university participated in the study. Thirty-one of the subjects were pre-screened and selected based on their level of experience with computers (i.e., little computer experience versus substantial computer experience). The remaining subjects were exposed to the music treatment. Each group was exposed to a product usage situation for a computer (i.e., a cognitively evaluated product) or a music tape (i.e., an affectively evaluated product), and then asked to complete a set of scales measuring what they experienced during the usage situation and a set of scales evaluating the product. All subjects were recruited and participated strictly on a volunteer basis.

Products

A computer and a music tape were selected as the focus of the study for two reasons. First, these represent two distinctly different types of products. The computer is representative of the type of product which might typically be evaluated on a cognitive basis (i.e., based on the evaluation of product attributes). However, a significant portion of the population is often fearful of such a high-tech product, especially if they have only limited experience with the product. Such consumers may tend to use a more affective evaluation approach. On the other hand, from the consumer's viewpoint, a music tape is a non-threatening, low-tech product. Few consumers have a fear of listening to taped music. In an exploratory study of this nature, it seemed appropriate to use products which might produce maximally different results on the measurement scales. Secondly, both of these products are appropriate for the student subjects used. That is, many students use these two types of products and so it was possible to obtain students who had different levels of experience and interest in the two products.

To reduce the potential biasing influence of prior familiarity with the specific products being evaluated, a Compaq portable computer and a mood music tape were selected.

Study Procedures

Approximately 130 undergraduate students were given a screening questionnaire to evaluate their level of computer experience. Sixteen subjects who indicated a high level of computer experience and sixteen subjects indicating a low level were contacted to schedule a time when they could participate in the study. Only one student failed to keep the appointment, yielding a sample of 31 computer subjects. The same screening instrument measured level of experience with music tapes. Because the results indicated a small degree of variation in experience for this product compared to computers, a convenience sample of forty-one students from a marketing class were exposed to the music tape. The music tape treatment was performed in small groups of five to six people. These procedures for the two products were used as it was not possible to have more than one person use and evaluate the computer at a time, and it did not seem necessary to have individual exposure to the music tape. Prior to the experiment, all subjects read and signed an informed consent form.

In both cases the subjects were told that they would have a product usage experience, and be asked to evaluate that experience and the product itself. The administrators stressed that the subjects were to evaluate the computer itself, not the software; or the music tape itself, not the tape player.

A step-by-step instruction guide was created to lead the subjects through the computer evaluation experience. The guide had them insert two floppy disks, turn the computer on, load a word processing program, and type a brief letter. All subjects followed the same procedure, which took about five minutes. For the music tape, all subjects heard the same two minute section of the tape.

After the product usage experience, both groups were asked to fill out a questionnaire which contained a modified experiential measurement scale (described below), a set of product evaluation questions, and some demographic items.

Independent Variables

Product Type: As previously noted, the computer was selected as representative of products which usually require a cognitive evaluation. The music tape was selected to represent a non-threatening product, which would be evaluated on a more affective basis.

Level of Experience: Level of experience was judged by examining the subjects' responses to a series of questions in the screening questionnaire. To disguise the product category of interest, the screening questionnaire asked about cars, compact disk players, computers, and music tapes. Subjects were asked to indicate on a five point scale how interested they were in each product (no interest--extremely interested) and their degree of expertise with the product (I am a novice--I am an expert). They were also asked to indicate which of the products they owned and which they used frequently. Finally, they were asked to indicate on a five point scale their level of agreement (strongly agree--strongly disagree) with four statements. The two key questions were "The idea of using a computer frightens me somewhat" and "I like just about all kinds of music." The results indicated that there were strong differences in level of experience with computers and much weaker differences with respect to experience with music tapes. Subjects were selected for the high and low experience computer groups on the basis of the results from the first five items. Cronbach's alpha coefficient for these items was .88 which compares favorably with the base criterion of .65 suggested for exploratory research by Nunally (1978). As he screening instrument did not indicate very large differences in music tape experience, that part of the study was administered to a class of forty-two students (in small groups) with the intention of selecting the highest and lowest levels of experience available within the group.

As the experimental procedures for the computer subjects yielded fifteen subjects with a high level of computer experience and sixteen subjects with a lower level of experience, the same number of subjects were selected from the forty-two music subjects to represent high and low level of music tape experience. This was done after the data was collected by calculating a "level of experience" factor on the basis of three questions -asked in the main experiment. These questions asked the subjects to rate themselves (on seven point scales) as to their level of expertise (total expert--complete novice), interest (extremely interested in--no interest at all in), and opinion leadership (people frequently ask for my opinion about what music tape to buy--people never ask for my opinion about what music tape to buy) with regard to music tapes. The level of experience factor was created as a numerical average of the answers to the three questions (alpha=.88). The subjects were then ranked on the experience factor and the sixteen highest and fifteen lowest ranked subjects were selected. These procedures provided relatively equal cell sizes of consumers with a naturally occurring range of product usage experience.

Manipulation Checks

The main study asked four questions of the computer subjects to act as "manipulation checks" for the level of experience variable. In addition to the three questions noted above (expertise, interest, and opinion leadership) the computer subjects were also asked to indicate on a seven point scale how frequently they worked on a computer (very frequently--never). An average of these four measures was used to create an experience factor for the computer subjects (alpha=.92). As expected there was a significant difference for this variable between the high level of experience computer subjects (x=4.8, s=1.11, n=15) and the low level of experience computer subjects (x=2.0, s=0.62, n=16, t=8.72, pc.00, one-tailed test). Likewise, there was a significant difference in the experience factor for the high (x=5.9, s=.40, n=15) and low (x=33.5, s=.58, n=16, t-13.22, p<.00, one-tailed test) level of experience music subjects.

Dependent Variables

Experiential Dimensions: The way in which the consumers experienced the product evaluation situation was measured using a modified version of the Phenomenology of Consciousness Questionnaire (PCQ) (see by Pekala and Levine, 1981). The original instrument measures nine dimensions of consciousness: internal dialogue, awareness (both self awareness and state of awareness), imagery (amount and vividness), positive affect, volition, altered experience (meaning, perception, time, body image), attention (direction and absorption), negative affect, and memory. The instrument contains thirty-seven items, including five pairs of duplicate item questions used to measure internal dialogue, state of awareness, imagery amount, positive affect, and direction of attention. Correlations for the duplicate pair items ranged from .47 (for state of awareness) to .83 (for internal dialogue) with an average correlation of .67 across the five duplicate item pairs. This result suggests that people asked to answer the same question twice on a questionnaire tend to answer it in a fairly stable way. Under the assumption that our subjects would be likely to respond to identical questions on the questionnaire with rather similar answers, we decided to modify the questionnaire by using only one item of the duplicate item pairs for all but one question. Given that the lowest correlation in the original study (.47) was found for the state of awareness item pair, this one pair was selected for inclusion in the modified instrument. In the current study, the correlation for this duplicate item pair across all the subjects was .59, suggesting that these subjects indeed tended to answer the same question with similar responses.

In further modification of the PCQ, one of the questions involving positive affect was reworded. The original question asked about awareness of sexual feelings (I was not aware of any strong sexual feelings--l experienced very strong sexual feelings). As it seemed fairly unlikely that subjects evaluating a computer would experience sexual feeling, this question was modified to ask about awareness of positive feelings. A second question was dropped, as the experimenters could devise DO appropriate reworded question The original question dealt with meaning, and was phrased "I had an experience which I would label as very religious, spiritual, or transcendental." Thus, the modified questionnaire had thirty-two questions measuring nine experiential dimensions.

Importance of product characteristics: Based on a preliminary investigation, five product characteristics were selected for each of the products as ones which would be expected to influence the subjects' attitudes about the products. For the computer these were: speed, keyboard layout, clarity of screen, keyboard touch, and portability. For the music tape, the characteristics were: rhythm or beat, mood of the music, melody, variety of sound, and effect it has on me. The subjects were asked if they were interested in buying the product and how important each characteristic would be in their purchase decision. The characteristics were rated on seven point scales ranging from extremely important to not very important at all.

Beliefs about the product: The subjects were then asked to rate the product which they had evaluated on each of the five characteristics. The instrument used a seven point scale ranging from very good to very poor.

Attitude toward the product: The subjects' product attitudes were calculated in terms of a multiattribute attitude model. The attribute importance ratings were multiplied by the beliefs about the product and summated to produce an attitude score.

Behavioral intentions: Given that it was not possible to actually have the subjects purchase the computer, a measure of behavioral intention for each product was taken. The subjects were asked if they were interested in buying this type of product and how likely it is that they would actually buy the particular product they evaluated (very likely-not very likely).

RESULTS

Question 1--Applicability of the PCQ to product evaluation situations

The first research question deals with how effectively the PCQ could be applied to the study of consumer behavior. Three specific issues were raised: are the dimensions the same? does it have internal consistency? and does it work equally well for cognitively and affectively evaluated products?

To investigate the first part of the question, two confirmatory factor analysis procedures were performed. The first used a LISREL model (Joreskog and Sorbom 1984) for all of the subjects and each of the four sub-sets (i.e., computer, music, high experience, low experience), specifying a nine factor solution. However, in each case the results indicated that the model was mis-specified and so interpretable results were not obtained. Because the LISREL procedure did not provide useful results, a factor analysis program was run which specified nine factors be extracted using a principal components analysis and a Varimax rotation.

TABLE 1

NINE FACTOR CONFIRMATORY FACTOR ANALYSIS-THE FACTOR EACH ITEM LOADED ON BY GROUP

Due to space limitations tables containing the factor loadings for each group are not included in this paper, however, they are available from the first author upon request. The results of this analysis are presented in a summary fashion in Table 1 which shows the factor that each item loaded on (using factor loadings of at least .50). Inspection of this table clearly indicates that the items do not all load on the dimensions which they are expected to. However, the results are not entirely discouraging, as further inspection indicates that many of the items which should load together do so. For example, in three cases (for the entire group, the music subjects and the high experience subjects) the positive affect items load together in a single dimension. Likewise, several of the items in each of the other dimensions (and their sub-scales) load together for one or more of the subject groups. These results suggest that, while the items of the PCQ scale do not load on the original nine dimensions, they do retain some of the proposed relationships. Based on these results, it seems appropriate to use exploratory factor analyses to discover more about the number of dimensions and the relationships among the items (the results for all groups are in Table 2).

TABLE 2

EXPLORATORY FACTOR ANALYSIS-THE FACTOR EACH ITEM LOADED ON BY GROUP

First, a factor analysis was performed for the full set of subjects. The procedure included a principal components analysis followed by a Varimax rotation. The criterion for retaining factors was an eigenvalue greater than or equal to 1.0. The analysis yielded ten orthogonal factors. The first factor loaded heavily on imagery, positive affect and meaning. The second consisted of the time elements (part of the altered experience dimension). The third factor represented state of awareness, the fourth is most interpretable as absorption, and the fifth as negative affect. The other factors did not match up very well with the original dimensions. These results provide moderate evidence that the PCQ captures some of the same dimensions in the domain of consumer product evaluations as it did in its original application.

Next, factor analyses were performed for the computer subjects and the music subjects respectively. Although there are a large number of items relative to the number of subjects, it has been argued by Adelman (1983) that the factor analytic results are stable even in this situation (see Arrindell and van der Ende 1985 and Bumb 1982 for further discussion). A subjective interpretation of the results indicates that the PCQ performed better for the music subjects than for the computer subjects. For the music subjects, the first factor loads heavily on positive affect, however for the computer subjects this first factor is difficult to interpret. For the music subjects, the second factor shows a strong grouping on the altered experience dimension, the third factor represents imagery, the fourth awareness, the fifth is uninterpretable, and the sixth loads on negative affect. The seventh through the tenth factors do not match up well with any dimension. For the computer subjects, the second factor represents state of awareness, the third represents aspects of the altered experience dimension, with the clearest loading on the time variables, the fourth factor is not interpretable, the fifth loads most heavily on absorption, and the remaining factors do not match up well with any of the dimensions. Similar results occur for the other two groups.

While the results of the factor analyses do not match perfectly with the dimensions proposed by Pekala and Levine, they do suggest that the PCQ captures some important experiential factors that occur during a product evaluation situation. It appears, however, that the PCQ may perform somewhat better for a product which requires more affective evaluation than for a one requiring a more cognitive approach.

TABLE 3

DIMENSIONS OF CONSCIOUSNESS AND CORRESPONDING CRONBACH'S ALPHA COEFFICIENTS FOR THE FULL PCQ

Another way of examining question 1 is to investigate the internal consistency of the items measuring each of the nine dimensions and the instrument's overall consistency. A series of Cronbach's alpha correlations was run on the items within each dimension. These items should be strongly correlated with one another. As can be seen in column 1 of Table 3, the correlations for several of the dimensions are reasonably strong. For example, the correlation coefficient for the three measures of state of awareness is .68, for imagery it is .73, and for positive affect it is .79. On the other hand, the coefficient for the items measuring volition is .32, for altered experience it is .39, and for attention it is .34. These results indicate that some of the dimensions measured by the PCQ scale are more appropriate in a product evaluation situation than are others.

Columns 2 and 3 in Table 3 show the alpha coefficients for the computer subjects and the music subjects. The overall average alpha coefficients show no difference between these two groups of subjects. However, investigation of the individual dimensions does suggest differences in how well some dimensions applied to the groups. For example, the awareness items show a stronger correlation for the computer group (.72) than for the music group (.60). In contrast, the music group has a stronger correlation in both the positive (.81) and negative (.53) affect dimensions than does the computer group (.70 and .35 respectively). The volition and altered experience dimensions do not perform very well for either group.

Columns 4 and 5 show the alpha coefficients for the low versus high experience groups. The overall average alpha suggests that the scales worked somewhat better for the low experience subjects than for the high experience ones. Inspection of the individual dimensions shows that the alpha coefficients for the low experience subjects are much higher on the dimensions of volition (.51 vs. -.04), altered experience (.49 vs. .26), attention (.59 vs. -.06), and negative affect (.59 vs. .39). The memory dimension does much better for the high experience subjects (.70) than for the low experience ones (.32), and the positive affect dimension does somewhat better (.83 vs .72).

Finally, the last column shows the results of the -Pekala and Levine (1981) study. Although their study used a different number of scale items in several dimensions, the results indicate a generally poorer performance in the current study than in their research. The primary exceptions to this are the positive affect dimension and the memory dimension for the high experience subjects.

Overall, it appears that the awareness, imagery, and positive affect dimensions worked quite well for both types of products and for both levels of experience. The memory dimension worked moderately well for all but the low experience group, and the negative affect dimension worked rather well for the music subjects and the low experience subjects. The attention dimension did not work very well for the computer subjects or the high experience group, but did work for the music subjects and the low experience group. The volition scales only worked well for the low experience subjects, and the altered experience scales did not do especially well. In general, the PCQ appears to have worked moderately well when applied to a product evaluation experience.

Questions 2 and 3

The second research question involves whether including experiential dimensions improves the relationship between traditional measures of attitude and behavioral intentions. As this is an exploratory study and because there is little in the way of theory to suggest precisely which of the PCQ dimensions should be related to which treatment condition, a regression approach was used. With behavioral intentions as the criterion variable, a model was specified so that the attitude measure was first included in the model and then, using a forward selection procedure, the nine dimensions of the PCQ were considered. Thus, in each case, the model included attitude and any of the dimensions which significantly contributed to the relationship with behavioral intentions.

The results for the full group of subjects had an adjusted R2 of .35 (p'.01) using attitude and attention as predictor variables (both significant at the .01 level). The weakest result was for computer subjects with an adjusted R2 of only .12 (p<.10) using attitude and positive affect (due to multicolinearity, neither of these made a significant contribution individually to the relationship). For low experience subjects, the adjusted R2 was .31 (p<.01) using attitude and positive affect. However, only positive affect made a significant contribution to the relationship (p<.05). The adjusted R2 for the high experience subjects was .51 (p<.01) using attitude, awareness, and memory (all contributing significantly to the relationship, p<.01). Finally, for the music subjects, the adjusted R !was .52 (p<.01) with both attitude and attention making a significant contribution (p<.01 and p'.05, respectively).

These results are quite encouraging, indicating that dimensions of the PCQ can produce improved results in the relationship between attitude and behavioral intention, and may even provide a useful replacement for the attitude measure itself. Likewise, in answer to the third research question, they suggest that a single attitude model may not apply very well to describing both affectively and cognitively evaluated products. The model did much better for the affectively evaluated product (i.e., music), and a different dimension of the PCQ scale was selected for the two cases. Also, the results indicate that different models were useful for the low and high experience groups, in fact suggesting that the traditional measure of attitude worked poorly for the low experience subjects.

As an interesting aside, the analysis described above was performed for level of experience across product type. The poorest results were for the high experienced computer subjects with a non-significant (p<.35) adjusted R2 of .53, using attitude and all nine dimensions of the PCQ. The low experience computer subjects had an adjusted R2 of .34 (p<.02) using attitude (p>.95) and altered experiences ( 15.01). For the music tape, the low experience subjects had an adjusted R2 of 53 (p<.01) using attitude (p<.10) and positive affect (p<.03), while the adjusted R2 for the high experience music tape subjects was .70 (p<.01) with attitude (p<.02) and volition (p<.01). Note that for the low experience computer subjects the cognitively driven measure of attitude did not work very well. Presumably, these subjects were unable to evaluate the computer on a logical, rational basis and may have used a more experientially driven approach. It is rather disconcerting that neither the attitude measure nor the experiential measure worked well for the high experience computer subjects, certainly indicating a problem with the measurement scales for this group. However, the fact that the experiential dimension added significantly to the music subjects certainly seems to indicate the music tape was being evaluated on an experiential as well as a cognitive basis.

DISCUSSION

Overall the results of the study are promising. There appears to be an indication that the dimensions proposed by the PCQ can be usefully applied to a product evaluation situation. However, the instrument seems to work better for the affectively evaluated product than for the cognitively evaluated one. In addition, because the model test showed that for the low experience subjects the positive affect dimension was significant and the attitude measure was not, the results suggested that low experience subjects probably used affect to evaluate the products to a larger extent than did the high experience subjects. Thus, there is limited support for the notion that we should consider the ways in which consumers experience products and we should measure these using instruments that go beyond the traditional cognitive oriented measures of attitude.

On the other hand, the PCQ instrument should not be applied in its current form to consumer product evaluation situations. The scales were developed without regard to the experiences a consumer might have in an evaluation or a consumption situation. Rather the results of this study should encourage further research into the dimensions of consciousness which are related to product situations. For example, in the computer situation anxiety might be an experiential dimension which would have been useful. In addition, many dimensions of emotion have been discussed which may be important in consumption experiences (see Holbrook 1986) that are not tapped by the PCQ. However, the results of this study provide a strong indication that a questionnaire which measures important dimensions of consumer experiences can be developed.

REFERENCES

Adelman, Irma (1983), "Factor Analysis and Development: A Reply," Journal of Development Economics, 11, 109-112.

Arrindell, W.A. and Jan van der Ende(1985), "An Empirical Test of the Utility of The Observations-To-Variables Ratio in Factor and Components Analysis," Applied Psychological Measurement,9(June), 165-178.

Batra, Rajeev (1986), "Affective Advertising Role, Process, and Measurement," in Peterson, Hoyer, and Wilson (eds.), The Role of Affect in Consumer Behavior, Lexington, Mass: Lexington Books, 53-85.

Battista, J. R. (1978), "The Science of Consciousness," K.S. Pope and l.L. Singer(eds.), The Stream of Consciousness, Plenum Press, New York.

Bumb, Balu (1982), "Factor Analysis and Development," Journal of Economic Development, 11,109-112.

Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research, Reading, Mass: Addison- Wesley.

Hirschman, Elizabeth C. and Morris Holbrook eds. (1981), Symbolic Consumer Behavior, Ann Arbor MI: Association for Consumer Research.

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

Joreskog, Karl G. and Dag Sorbom (1984), LISREL VI: Analysis of linear structural relationships by the method of maximum likelihood, (3rd ed.), Mooresville, IN: Scientific Software, Inc.

Krippner, S. (1972), "Altered States of Consciousness," J. White (ed), The Highest State of Consciousness, New York: Doubleday.

Ludwig, A. H. (1972), "Altered States of Consciousness," C.T. Tart (ed), Altered States of Consciousness, New York: Wiley.

Miller, Murray G. and Abraham Tesser (1986), "Effects of Affective and Cognitive Focus on the Attitude-Behavior Relation," Journal of Personality and Social Psychology, 5(2), 270-276.

Nunally, Jum C. (1978), Psychometric Theory, New York: McGraw-Hill Book Co.

Pekala, Ronald J. and Ralph L. Levine (1981), "Mapping Consciousness: Development of an Empirical Phenomenological Approach," Imagination, Cognition, and Personality, 1(1), 29-47.

Peterson, Robert A., Wayne D. Hoyer, and William R. Wilson (1986), The Role of Affect in Consumer Behavior, Lexington, Mass: Lexington Books.

Petty, Richard E. and John T. Cacioppo (1986), "The Elaboration Likelihood Model of Persuasion," L. Berkowitz (ed), Advances in Experimental Social Psychology, Vol. 19, 123-205.

Tart, C.T. (1977), "Discrete States of Consciousness," P.R. Lee, R.E. Ornstein, D. Galin, A. Derkman, and C.T. Tart (eds), Symposium on Consciousness, New York: Penguin.

Tart, C.T. (1975) States of Consciousness, New York: Dulton.

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