Consumers and Movies: Some Findings on Experiential Products

ABSTRACT - This paper studies movies as an example of experiential products, defined as those products which consumers choose, buy and use solely to experience and enjoy. The consumption experience, especially its hedonic and aesthetic aspects, is key for understanding experiential products. The results of a survey confirm that involvement with movies is dominantly hedonic. When selecting a movie, subjects consider more alternatives that consume time than ones that consume only money and also consider more subjective than objective features and more global than unidimensional features.

Citation:

Elizabeth Cooper-Martin (1991) ,"Consumers and Movies: Some Findings on Experiential Products", in NA - Advances in Consumer Research Volume 18, eds. Rebecca H. Holman and Michael R. Solomon, Provo, UT : Association for Consumer Research, Pages: 372-378.

Advances in Consumer Research Volume 18, 1991      Pages 372-378

CONSUMERS AND MOVIES: SOME FINDINGS ON EXPERIENTIAL PRODUCTS

Elizabeth Cooper-Martin, Georgetown University

ABSTRACT -

This paper studies movies as an example of experiential products, defined as those products which consumers choose, buy and use solely to experience and enjoy. The consumption experience, especially its hedonic and aesthetic aspects, is key for understanding experiential products. The results of a survey confirm that involvement with movies is dominantly hedonic. When selecting a movie, subjects consider more alternatives that consume time than ones that consume only money and also consider more subjective than objective features and more global than unidimensional features.

All goods and services are consumed, but for some products, like the performing arts, wine, and vacations, the consumption experience is an end in itself. This paper identifies and studies this type of product and labels them experiential products. Consumers choose, acquire and use experiential products solely to experience them and enjoy them. Experiential products include both physical goods, such as wine and recreational drugs, and services, such as sporting events and restaurant meals. The dominant benefit of these products is hedonic consumption, that is the feelings, emotions and sensations experienced during product usage (Hirschman and Holbrook 1982).

Given the special nature of experiential products, consumer behavior for them is likely to differ from that for other types of products. The purpose of this paper is to demonstrate some of the unique aspects of consumer behavior for these special products. Its contribution is to build on the experiential or hedonic perspective (Hirschman and Holbrook 1982; Holbrook and Hirschman 1982) to describe and identify a specific type of product and to begin to understand consumer behavior for it.

Specifically, this paper focuses on movies as an example of the class of experiential products. Movies are a good example because they appeal to a wide range of consumers, compared to the arts (see summary in Capon and Cooper-Martin 1990) or wine (Jobson 1989). Further, movies are purely experiential, as compared to television and radio which may accompany other activities. For example, broadcast music can serve as a background for shopping (Milliman 1982) or a restaurant meal (Milliman 1986). In particular, this study researches two aspects of consumer behavior for movies: alternatives and choice features. Predictions for both aspects are based on the consumption experience, which differentiates experiential products from others.

The next section discusses experiential products in more detail and develops the hypotheses on what types of alternatives and choice features consumers use for movies. The following two sections describe the methodology and results. The discussion section includes suggestions for additional research on experiential products. The final section describes potential implications for managers interested in marketing movies.

EXPERIENTIAL PRODUCTS

Experiential products are defined by their dominant emphasis on the consumption experience. The main benefit from these products is the pleasure or hedonic value in consumption. Experiential products contrast with goods and services that primarily fulfill utilitarian functions. For example, the major benefit of shampoo is to leave hair clean and shiny, blankets keep one warm, a visit to the dentist protects one's teeth. Clearly, there may be hedonic aspects of consuming non-experiential products; for example, the lovely fragrance of a peach-scented shampoo or the fondness for a blanket received from a dear relative. But the primary reason for consuming these products is not to experience them.

Just as utilitarian products may have some hedonic value, experiential products may have some utilitarian functions. For instance, food clearly has the utilitarian benefit of keeping one alive. Thus under certain circumstances, for example grabbing a quick snack to stave off hunger pangs, food is more of a utilitarian product. But under certain circumstances, food can be an experiential product; for example, in a first-class restaurant. The emphasis then is clearly hedonic; to enjoy the sight, aroma, texture and taste of the food. A restaurant meal can be hedonic in other ways: savoring a fine wine, soaking in the elegant surroundings, enjoying the luxury of excellent service.

To savor the feelings, emotions and sensations of using an experiential product requires the expenditure of time. Likewise, the experiential perspective argues that the key resource which consumers expend in a transaction is time, rather than money (Holbrook and Hirschman 1982). Thus one intrinsic quality of experiential products is that they demand the consumption of time. Another useful way to describe products is to distinguish between those directed to people and those directed to their possessions (Lovelock 1983). Products directed at possessions include laundry detergent which cleans clothes and banking services which protect money. Experiential products, such as movies, restaurant meals, and vacations, are directed towards people, not towards their possessions. This description again suggests that consuming an experiential product involves allocating time.

Because experiential products require expenditures of time, the alternatives for them are likely to be other products that utilize the same resource. To consume such products, consumers must actively devote some level of cognitive or physical effort, as when reading a book or playing basketball. Some products do not require the consumer's time or attention during consumption but are consumed passively, for example wearing clothes or sitting on a couch. For a movie, alternatives that require expenditures of time would include other movies, as well as other product classes, as in the following hypothesis:

H1: As alternatives to a movie, consumers will consider more alternatives that require expenditures of time than alternatives that only require expenditures of money.

If consumers do consider other product classes as alternatives to movies, what types of features do they use for comparisons? Research using other types of products reveals that as consumers choose from alternatives in increasingly dissimilar product classes, the abstractness of product comparisons used increases (Johnson 1984). More abstract features describe a greater number of alternatives than concrete features (Johnson and Kiselius 1985). For instance, exclusivity is an abstract feature which could be applied to alternatives in several product classes: wine, performing arts, sporting events, restaurant meals. Thus it is likely that consumers use features that are abstract to select experiential products.

Because the benefit from experiential products is pleasure in consumption, consumers should choose movies and similar products based on what they like and enjoy, on what pleases them, in other words, based on intrinsic preference (O'Shaughnessy 1987). Therefore, consumers use subjective features, such as "funny", "suspenseful, "scary", "romantic", to select movies. ("Feature" is used here to refer to a product characteristic as defined by consumers. It is not used in the sense that such characteristics are dichotomous.) Subjective features reflect the consumer's viewpoint, i.e., the personal nature of the experience, and describe the consumption experience, for example, what it is like to see and hear the movie. The subjective features of products are important from an experiential perspective (Holbrook and Hirschman 1982) and by extension are important for experiential products. In fact, such features can be described as aspects of consumption (Holbrook and Hirschman 1982).

Consumers also consider objective features of experiential products; such features can be externally verified. Examples for a movie are the director, theater location, admission price or schedule convenience. Objective features are often tangible and utilitarian (Holbrook and Hirschman 1982). They should be less important for experiential products than subjective features because such products, by definition, do not fulfill utilitarian functions.

Subjective features are more generalizable and more abstract than objective ones. Similarly, Johnson and Kiselius (1985) argue that experiential (i.e., subjective) versus tangible (i.e., objective) features reflect differences in the dimension of abstractness-concreteness. Subjective features can be used to describe different types of products. For example, subjective features such as fun, relaxing, and exciting are appropriate for comparing one movie to another, as well as to several, other product classes that require an expenditure of time: watching TV or a video, or going to a restaurant or club. By contrast, leading actor and the director are objective features of a movie. Though useful for comparisons with other movies, they are less so with non-movie alternatives. Thus to choose a movie, subjective features will be more useful due to their abstractness and their reflection of the hedonic experience, as per the following hypothesis.

H2a: Consumers will consider more subjective features than objective ones when choosing movies.

The experiential view of consumption emphasizes both the hedonic and aesthetic nature of products (Holbrook and Hirschman 1982). For the aesthetic qualities of products, there appears to be a relationship between stimulus complexity and hedonic value (Berlyne 1971). More specifically, for aesthetic products, consumer evaluations seem to depend on the configuralities or interactions among product features (Holbrook 1983). While exploring aesthetic alternatives, consumers are more likely to consider global features, which refer to the entire product, not just one dimension. Global features reflect the entire product and thus capture the product's complexity or interactions. Global features should be more useful than unidimensional features which apply only to a single, specific characteristic of a product. For instance, plot/storyline and genre (e.g., action, comedy, science fiction) seem to be the two most important considerations in deciding what movie to see (Austin 1981; Variety 1981). These are complex descriptors which capture the interactions among features such as setting, character development, pacing, special effects.

Global features refer to the entire product, and so may not be product-specific; for example, good, awful, moving. If a consumer applies such terms to a movie, it is not clear which facet of the film portrayed each of these features. Further, these descriptors could be used to describe other products, such as a play or novel. Thus global features tend to be abstract and for this reason, along with their ability to capture complexity and feature interactions, should be useful when selecting experiential products. This leads to the following hypothesis.

H2b: Consumers will consider more global features than unidimensional features when choosing movies.

The next section describes the research conducted to test these hypotheses.

METHODOLOGY

The study had two phases. Phase one was conducted to confirm the hedonic nature of experiential products, such as movies. Phase two tested the hypotheses on alternatives and choice features for movies.

Phase 1

The importance of hedonic value for movies and other experiential products was assumed by definition and used to develop the hypotheses on alternatives and choice features. Thus, it seemed important to test this assumption before testing the hypotheses.

To confirm the importance of the hedonic consumption for experiential products, it seemed appropriate to use existing measures of hedonic or pleasure value. These measures are part of a well-researched and tested scale that measures product class involvement, i.e.the consumer's concern for a product class (Kapferer and Laurent 1985; Laurent and Kapferer 1985). The scale includes pleasure/hedonic as one of five antecedents to or facets of involvement. The remaining four facets of involvement are: perceived importance of the product, perceived importance of the consequences of a mispurchase, subjective probability of a mispurchase, and symbolic value. For movies, and other experiential products, involvement should primarily be hedonic and should be the dominant one of these five facets.

This paper asserts that experiential products are unique because of their emphasis on consumption and therefore differ from other products. To test this assumption for movies, it would be most appropriate to compare them to similar, but utilitarian, products. Recall that experiential products, including movies, are directed to people, not their possessions. Another important distinction between types of products is physical goods versus services; a movie is a service. Examples of utilitarian, people directed, services are health care, transportation, haircuts, exercise clinics and education. The experience of consuming these products is not necessarily pleasant and so the consumption experience is not the dominant benefit. Compared to an experiential product, the result of consuming these utilitarian ones is more permanent and long term, for example, a cured illness, a new hair style, an educated mind. For utilitarian, people-directed services, the involvement facet reflecting the consequences of a mispurchase should be very high. Also hedonic involvement should be greater for experiential, people-directed services than for these similar, but utilitarian, ones.

Subjects. The subjects were 160 undergraduate business students from a major university in the Mid-Atlantic region. The sample was 49% male and 51% female; mean age was 20.5. This age group is very appropriate for a study about movies, because 15-24 years old accounted for 48% of movie admissions in 1983 (Newspaper Advertising Bureau 1984).

Procedure Each subject answered a written questionnaire and so provided self-reports on hedonic/pleasure value and the other four facets of product class involvement: product class importance, symbolic value, risk consequences of purchase, and subjective risk probability of a bad purchase. The questionnaire was based on existing scales (Laurent and Kapferer 1985) and had been used reliably previously (Cooper-Martin 1989); it included four items on risk probability and three items on each other facet of involvement. Each item had a 5-point Likert-type response format (fully disagree to fully agree). Each facet's scale was the mean of its items and ranged from 1-5. The hedonic scale consisted of the following three questions (using movies as an example): For me, a movie is a real pleasure; When you go to the movies, it's a way to reward yourself; and I don't particularly like movies. The score on the latter item was reversed to form the hedonic scale. The involvement questionnaire concerned four product classes: movies; another experiential product, wine; and two utilitarian, people-oriented services: haircuts and dentists. Subjects who never used a product class (e.g., wine) did not answer items on it.

For the hedonic involvement scale, the mean Cronbach's alpha over all products was .71. For the other scales related to movies, Cronbach's alpha was .78 for importance, .68 for symbolic, .77 for risk consequences, and .41 for risk probability. Except for the latter, reliability coefficients of all scales analyzed exceeded .6, and so their reliability was judged acceptable (Peter 1979).

Results. Mean values for all facets of involvement for each product class are in the Table. The importance of pleasure and hedonic responses for experiential products is supported. For movies, subjects report a higher level of involvement on the hedonic facet than on any of the other four facets of involvement: importance (t (160) = 6.60, p < .0001), symbolic (t (160) = 6.20, p < .0001), risk consequences (t (160) = 16.96, p < .0001), and risk probability (t (159) = 15.59, p < .0001). (Although these different mean values are interpreted as significant differences across facet of involvement, they may possibly be due to item differences across the five scales.)

The importance of hedonic value was also supported when tested with wine, a second experiential product included on the questionnaire. For wines, subjects report a higher level of involvement on the hedonic facet than on any of the other four facets of involvement: importance (t (145) = 9.14, p < .0001), symbolic (t (145) = 6.65, p < .0001), risk consequences (t (145) = 8.94, p < .0001), and risk probability (t (144) = 3.29, p < .005).

The findings also support and confirm the greater importance of hedonic responses for movies, than for similar (i.e., people-directed services) but non-experiential products. Subjects report higher hedonic involvement for movies than for either of the two, people-oriented services: haircuts (t (160) = 11.15, p < .0001) and visits to the dentist (t (159) = 24.47, p < .0001). The analysis with wines confirms this result; hedonic involvement is greater for wine than for haircuts (t (145) = 3.66, p < .0005) or visits to the dentist (t (144) = 16.37, p < .0001 ).

TABLE

MEAN VALUES OF INVOLVEMENT FACETS FOR FOUR PRODUCT CLASSES

Because of these confirmatory results on the hedonic nature of movies, further study on consumer behavior for these products seemed justified.

Phase 2

This study tested the hypotheses on alternatives and choice features for movies.

Subjects. The subjects were 181 undergraduate business students. This sample was 56% women and 44% men; mean age was 20.4 years. The sample was taken from the same university as that used for the questionnaire on involvement and so was from the same population.

Procedure. Subjects responded to a written survey; they were tested in groups but each filled out the survey individually. Each subject answered the following question: What alternatives to this movie did you consider? Each subject also gave a written report on whatever he/she could remember about choosing the last movie attended. The directions asked the subjects to include all thoughts, observations and feelings. Since these protocols are retrospective, they are not as reliable as concurrent ones. However, it seems unlikely that the subjects altered their responses in favor of the hypotheses (i.e., by increasing the number of subjective and global features reported).

A judge who was unaware of the hypotheses coded each alternative as either one that requires expenditure of time or one that only requires expenditure of money. Features in the written protocol were identified; features were broadly defined as any characteristic of the movie itself (e.g., actress) or of going to the movies (e.g., theater location) or any criteria used in the choice process (e.g., my companion wanted to see the movie). The same judge then coded each feature in every written protocol as subjective or objective. An objective feature was defined as one that is externally verifiable. In other words, people would agree on whether or not a particular product has this feature, for example, whether a particular actor is in a movie. The most frequently mentioned objective features of movies were location, not seen before, price, schedule convenience, setting, and a particular actor. By contrast, subjective features were defined as those for which each consumer is likely to have his/her own opinion about whether a product has this feature. The most common examples were good, liked the actors, funny, others wanted to see it, and I wanted to see it.

The same judge also coded each feature as global or unidimensional. Global descriptors refer to the entire product, not to just one part that can be delineated. For example, one subject described the latest remake of 'The Fly" as "1 1/2 hours of a gross and-stupid movie"; "gross" and "stupid" are global descriptors. The most frequent examples were descriptions like "I heard it would be good", comedy, "I wanted to see this movie", and "I hadn't seen it before". Unidimensional features refer to one specific part of a product. Subjects most frequently mentioned specific actors or producers, theater location, price, show time, and the movie's setting.

Analysis Because every hypothesis involved a comparison between two measures for the same subject, each hypothesis was analyzed by paired t-tests.

RESULTS

H1 is supported; when choosing a movie, subjects consider more alternatives that consume time than alternatives that only consume money. There is a significant difference between the number of alternatives that consume time (mean = 1.2) and those that only consume money (mean =0.1) t (181) = 16.94, p < .0001.

Of the 181 subjects, 80% (145) listed an alternative to the last movie attended. Of this group, 41% (59) considered only other movies, 36% (52) considered non-movie alternatives, and 23% (34) considered both other movies and non-movie alternative. They mentioned 140 alternatives that consume time. The majority (67%) of these alternatives involved other experiential products, for example, watching TV or a video, drinking, going to a restaurant or to a club. The rest (33%) were other activities, like staying home, studying or partying.

As predicted by H la, subjects use more subjective features (mean =1.77) than objective ones (mean = 1.08) in selecting movies, t (181) = 4.96, p < .0001. Also, as hypothesized by H2b, the subjects use more global (mean = 1.77) than unidimensional features (mean = 1.04) in choosing a movie, t (181) = 5.17, p < .0001.

DISCUSSION

As expected, tests for both movies and wines confirmed the importance of hedonic value for experiential products and its greater importance for these products compared to more utilitarian ones. The importance of hedonic and pleasure value was used for predictions on alternatives and choice features for movies, as examples of experiential products. The results of a study on these aspects of choice behavior support the predictions; when choosing a movie, subjects predominantly considered alternatives that require time allocations and examined subjective and global product features. These two findings are consistent, as follows. Both these types of features are abstract. As such, they facilitate comparisons with other product classes (as shown by Johnson 1984). Because consumers consider as alternatives products that require expenditures of time, they are likely to compare movies to other product classes (e.g., TV shows, restaurants).

Of the four product classes researched, the two experiential products, wine and movies, had the highest levels of hedonic involvement. This suggests that the consumer's relationship with these products is largely hedonic. This confirms a study on movies which tested the relationship of 30 lifestyle characteristics to film-going; only four were related but one of those was hedonism/optimism (Knapp and Sherman 1986). Another study found very high levels of hedonic involvement for champagne, which is similar to wine (Laurent and Kapferer 1985). Laurent and Kapferer did not focus on experiential products but did include three products which could be so classified: champagne, chocolate, and TV sets (assuming that consumers rated TV shows not simply the sets). These three products (plus dresses) clearly had the highest scores for hedonic/pleasure value of the products studied (no tests for significant differences between products were reported).

Laurent and Kapferer (1985) also found that hedonic involvement was not related to the reported extensiveness of the choice process. Extensiveness reflected time, effort and number of features used. Thus, given the importance of hedonic value for experiential products, the choice process for such products is likely to involve less time and less effort than the process for people-directed, nonexperiential products. Future research using decision time and/or concurrent protocols of the choice process for various products (experiential and utilitarian) could be used to test this hypothesis.

Despite the evidence for the importance of hedonic value for experiential products, including movies, there is one way in which movies can be utilitarian. In American society, movie-going is a social activity; 96% of movie-goers attend with at least one other person (Johnson 1981). Likewise, of the 181 subjects who answered questionnaires on features and alternatives, 97% went with someone else to the last movie attended. It is possible that a consumer would attend a movie simply to join friends or take someone else on a date and so has little interest in the movie itself; the movie might thus be utilitarian in the sense of furthering or enhancing a social relationship. (Eight of the 181 subjects reported in their protocols that they went along due to an invitation by friends.) Given the social nature of movies, joint decision-making, in which the final selection must be acceptable to all those involved, is very likely. (Of the 181 subjects, 68 or 38% had evidence of joint decision-making in their protocols.) Also, the interaction between consumers who go together, as well as among all consumers in a movie theater, may well be an important dimension of the consumption process for this product. Thus another topic for future research is to go beyond the individual consumer to understand both choice and consumption behavior for movies and other experiential products that are also leisure activities.

The current results suggest that consumers rely more on subjective than objective features to select a movie. Consumers may search different sources for these two types of features. Faber and O'Guinn's (1984) survey of movie consumers included information sources; their subjects rated previews and friends' comments as more useful and more credible for evaluating movies than advertisements or critics' reviews. Thus, to learn about movies, it appears that consumers may turn to experiential sources of information; these sources are distinguished by their ability to convey a sense of the consumption experience, that is, of what it is like to see the movie. Previews are a chance to try the movie; friends who have seen the movie can describe it. Non-experiential sources, such as ads are less useful because they don't concern the consumption experience. Although reviews should be an-experiential source, the subjects found them less useful.

These different types of information sources, experiential versus non-experiential, relate to subjective and objective features as follows. The usefulness and credibility of the information source may depend on what information it provides. Specifically, a non-experiential source (e.g., a magazine ad) should be just as credible and useful as a friend to determine an objective features (e.g., the movie's director). However, for subjective features (e.g., how funny the movie is), experiential sources (e.g., a friend) should be more credible and useful. Building on the findings from the current study, I am developing another to address whether experiential information sources are more credible and useful than non-experiential sources for both objective and subjective features.

CONCLUSIONS

This paper focused on the special nature of experiential products and began the process of understanding consumer behavior for them. The results confirm the hedonic nature of these products and show that as alternatives to movies, subjects considered more products that require expenditure of time than ones that only require expenditures of money. Further, they examined more subjective than objective features and more global than unidimensional ones.

The current findings are limited by a nonrandom, relatively small sample that focuses on the most important age group for movie-goers but neglects other age groups, including teenagers. If a study with a bigger, projectable sample confirms the -current findings, then certain managerial implications would result. For example, communication efforts for movies and others types of experiential products should include subjective features, visually or verbally, and not simply focus on objective ones, such as the star. The results concerning alternatives to movies indicate that competition for movies includes other activities or products, as well as movies. This information may help to better position movies relative to the competition or to direct management's efforts on information gathering about competition. For example, if drinking and eating out are popular alternatives (as in this study), then there may be a ready market for movie theaters that serve alcoholic beverages or meals. Likewise, it seems reasonable that movie theaters locate in malls where consumers can have a meal and/or a drink before or after seeing a film.

The results of this study are encouraging and give us the courage to go on to do the future research described above. They show that experiential products have specific patterns of consumer behavior. It appears that hedonic consumption can define a category of products and further, is key for understanding them. This emphasis thus contributed towards the need in the field of consumer behavior to address the experiential aspects of consumption.

REFERENCES

Austin, Bruce A. (1981), "Film Attendance: Why College Students Chose to See Their Most Recent Film," The Journal of Popular Film and Television, 9 (Spring), 43-49.

Capon, Noel and Elizabeth Cooper-Martin (1990), "Public and Nonprofit Marketing: A Review and Directions for Research," in Annual Review of Marketing 1990, ed. Valarie Zeithaml, Chicago: American Marketing Association.

Cooper-Martin, Elizabeth (1989), "The Effect of Three Contingency Factors on Consider Choice Strategies: A Test of Awareness of Costs and Benefit," in Advances in Consumer Research, Vol. 16, ed. Thomas Srull, Provo, UT: Association for Consumer Research.

Faber, Ronald J. and Thomas C. O'Guinn (1984), "Effect of Media Advertising and Other Sources on Movie Selection," Journalism Quarterly, 61 (Summer), 371-377.

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.

Jobson (1989), Jobson's Wine Marketing Handbook, New York: Jobson Publishing.

Johnson, Keith F. (1981), "Cinema Advertising," Journal of Advertising, 10 (4), 11-19.

Johnson, Michael D. (1984), "Consumer Choice Strategies for Comparing Noncomparable Alternatives," Journal of Consumer Research, 11 (December), 741-753.

Johnson, Michael D. and Jolita Kiselius (1985), "Concreteness-Abstractness and the Feature Dimension Distinction," in Advances in Consumer Research, Vol. 12, eds. Morris Holbrook and Elizabeth C. Hirschman, Provo, UT: Association for Consumer Research, 325328.

Kapferer, Jean-Noel and Gilles Laurent (1985), "Consumers' Involvement Profiles: New Empirical Results," in Advances in Consumer Research, Vol. 12, eds. Morris Holbrook and Elizabeth C. Hirschman, Provo, UT: Association for Consumer Research, 290-295.

Knapp, Steven, and Barry L. Sherman (1986), "Motion Picture Attendance: A Market Segmentation Approach," in Current Research in Film: Audiences, Economics, and Law,, Volume 2, ed. Bruce A. Austin, Norwood, NJ: Ablex Publishing Corporation, 3546.

Laurent, Gilles and Jean-Noel Kapferer (1985), "Measuring Consumer Involvement Profiles,'' Journal of Marketing Research, 22 (February), 41 -53.

Lovelock, Christopher H. (1983), "Classifying Services to Gain Strategic Marketing Insights," Journal of Marketing, 47 (Summer), 9-20.

Milliman, Ronald E. (1982), "Using Background Music to Affect the Behavior of Supermarket Shoppers," Journal of Marketing, 46 (Summer), 86-91.

Milliman, Ronald E. (1986), "The Influence of Background Music on the Behavior of Restaurant Patrons," Journal of Consumer Research, 13 (September), 286-289.

Newspaper Advertising Bureau (1984), Demographic Characteristics of Frequent Movie-goers, New York: Newspaper Advertising Bureau.

O'Shaughnessy, John (1987), Why People Buy, New York: Oxford University Press.

Peter, J. Paul (1979), "Reliability: A Review of Psychometric Basics and Recent Marketing Practice," Journal of Marketing Research, 16 (February), 6-17.

Variety (1981), "Film Subject Matter Looms Large in Stay-Away; Ticket prices are Related to Age; Income Strata," Variety, 305 (November 18), 5 and 32.

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