An Integrated View of the Store Choice/Patronage Process

ABSTRACT - Three papers, each touching on one or more aspects of store choice/patronage behavior, are reviewed here with the goal of relating each to a simple phase model. Contributions, methodological issues and ideas for future research growing out of this review are presented.

Citation:

Donald Granbois (1981) ,"An Integrated View of the Store Choice/Patronage Process", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 693-695.

Advances in Consumer Research Volume 8, 1981      Pages 693-695

AN INTEGRATED VIEW OF THE STORE CHOICE/PATRONAGE PROCESS

Donald Granbois, Indiana University

ABSTRACT -

Three papers, each touching on one or more aspects of store choice/patronage behavior, are reviewed here with the goal of relating each to a simple phase model. Contributions, methodological issues and ideas for future research growing out of this review are presented.

INTRODUCTION

Beyond their common concern with consumers' recall patronage behavior, the three papers presented vary considerably in method, the underlying bodies of theory and research drawn on, and the degree of innovativeness the authors have shown in conceptualizing the behaviors studied. Each paper reports an investigation of one or more phases of a sequence of mental processes and overt behaviors consumers exhibit in their store choice/patronage behavior. Beyond the usual compliments, criticism and suggestions triggered by the individual papers under review which are expected in a discussion paper, each paper's relation to a simple phase model of store choice/patronage will be shown. Suggestions for the authors' future research growing out of this attempt at integration will be made.

THE STORE CHOICE/PATRONAGE PROCESS

Elements in the store choice/patronage process can be categorized as four linked phases, including:

Image Formation -> Patronage -> Post-patronage -> Evaluation Further Behavior

Image formation includes those mental processes whereby information and experience are processed and evaluated resulting in predispositions which generally guide patronage. Patronage refers both to customers' actual shopping trip behavior -- number and sequence of score visits -- and their in-store behavior, as goods are acquired, interactions with sales personnel and other customers are entered into, etc. Post-patronage evaluation has been thought of both as a comparison of actual and expected behavior and as an attribution process through which consumers attempt to assess the causes of the outcomes of their patronage behavior. Further behavior includes remedy-seeking among unhappy consumers, word-of-mouth (both positive and negative) reflecting the outcome of the post-patronage evaluation process, and even punitive behavior directed at retailers.

Retail Shopping Area Image

Consumers probably have at least rudimentary knowledge about available stores and shopping clusters prior to most store visits, although the extent and character of this knowledge is seldom if ever the subject of research. As in the literature on brand and product choice, there has been a tendency for researchers to assume that an evaluation process based on two or more attributes underlies an overall affective orientation towards individual stores.

Houston and Nevin, like earlier researchers, refer to the output of this evaluation process as the score's "image," the "complex of a consumer's perceptions of a store on attributes." It seems possible, however, that an effective orientation towards a store sufficient to trigger a store visit can develop with little or no cognitive evaluation, a position consistent with a considerable body of evidence in both psychology (Zajonc 1980) and consumer behavior (Olshavsky and Granbois 1979). Respondents' seeming ability to assign ratings to listed alternatives in terms of criteria specified on a questionnaire does not support the conclusion that we carry such neatly-structured images around in our brains! There is ample precedent for Houston and Nevin's assumption, however, although studies asking if and how such images are formed would be quite interesting in the context of retail shopping clusters.

Houston and Nevin's research on Madison shopping clusters assumed a multi-attribute structure comprised of 16 dimensions underlies shoppers' images of the five clusters studied. Respondents provided up to 80 judgments (depending upon their familiarity with the clusters). The 16 cluster attributes were represented by short, very abstract phrases of 2 or 3 words. Again, there is much precedent for this approach, but one wonders what meanings respondents associate with such terse statements.

To the extent that consumers do not have complete knowledge about stores or shopping clusters, "images" may best be thought of as expectations or predictions. As such, these may be biased by information processing errors and deficiencies such as the incorrect application of "surrogate indicators." There is, in fact, already evidence in the literature that such biases do indeed occur in judgments about individual stores' price levels and locations (Brown 1969, Thompson 1963, Olshavsky, MacKay and Sentell 1975).

Finally, the 16 shopping cluster dimensions were taken from the store image literature, with some modifications apparently based on the judgment of the researchers after consulting with shopping center managers. Surprisingly, two attributes of great potential importance -- location (distance and/or driving time from home or other trip origin) and depth of assortment -- were excluded (assortment depth may be captured somewhat by the dimension "Product Selection," although other meanings may well have been interpreted by respondents). Gravitational studies cited by the authors have found these two variables to effectively predict aggregate movement of fashion goods shoppers even when measured with crude proxy variables such as map distance and shopping goods stores' square footage.

Houston and Nevin's factor analysis findings provide what seems to me to be rather indirect evidence of an unsurprising conclusion; that is, downtown Madison as a shopping cluster is perceived differently from the four planned shopping centers in the area. The nature of these differences is only hinted at in Table 4, where downtown is shown somewhat lower in respondents' evaluations than all four shopping centers.

Since even summary measures of the Madison shoppers patronage behavior are not reported, it is not possible to assess the predictive validity of the study's results, so we are not able to evaluate the importance, or the nature of the less favorable evaluation given downtown. For all we know, Madison patrons could be shopping downtown despite lower evaluations on the study's 16 attributes. The paper's conclusion, then, about the need for repositioning downtown through promotion seems premature, since it is by no means clear that the image dimensions studied are salient ones, driving actual patronage behavior. Beyond this, mere knowledge that downtown is rated lower on "General Price Level," "Friendly Atmosphere," "Helpful Store Personnel" and "Conservative" seems insufficient guidance for creating an effective promotional campaign to stimulate patronage. Indeed, price and personnel policy changes might well prove to be more appropriate responses by downtown merchants.

Consumers' Marketplace Problems

Shuptrine and Wenglorz, in another city, report a study fitting our score choice/patronage model somewhat less directly. Like many earlier studies, their findings reveal some consumers having some problems seeking remedy from sellers, which overwhelmingly were the retailers selling the product (81.3% of all complaints). This behavior, an important type of "Further Behavior" identified in our phase model, was found to be considerably more prevalent than in earlier studies, perhaps because questions were purposely included to probe and encourage reports of problems. The study's primary contribution is its conclusion that consumer problems may be more widespread than has been shown in earlier research, thus focusing even greater concern on the retailer as an important factor in remedying consumer dissatisfaction. Beyond this, the study is totally descriptive in character, its many cross-classifications revealing product and consumer-type variations in both problem incidence and tendency to complain quite similar to many earlier studies.

While the research serves to heighten our concern for the level of consumer problems reported (even though over 2/3 were reportedly adjusted satisfactorily) it provides no insights into the evaluation process whereby the consumer concludes a problem exists. Post-patronage evaluation processes are hardly touched on in consumer research, yet this seems to be a necessary next step if we are to begin to understand and explain the phenomenon of consumer dissatisfaction and subsequent behaviors. Such understanding and explanation need to precede serious recommendations for policy action. For example, possible consumer education efforts are difficult to conceive of without far greater understanding of consumer expectations formation; how product experience is interpreted and evaluated; and the nature of the attribution process as consumers attempt to decide whether they themselves, the manufacturer, the retailer, or some of the party should be blamed for less-than-satisfactory experience.

Deviance and Dissatisfaction

Mills' research fits our phase model too in that the five deviant behaviors studied occur either during the shopping process itself (part of the Patronage phase) or as part of Further Behavior. Since respondents predicted both their own and others' likely behavior in response to "scenarios" comprised of store descriptions and photographs, the perceptions evoked as the settings in which the deviant behaviors of interest were to be imagined by respondents seem to correspond to the concept of store image as it is normally defined. An interesting theoretical twist distinguishes the study in that the retail unit's power relative to that of customers is added as a significant dimension of recall image. Self-reported and projected deviant behavior vas hypothesized to vary with the three power "types" represented by the various scenarios, as was consumer satisfaction.

Relating the perceptions evoked by Mills' scenarios to the concept of store image focuses attention on the interesting issue of how consumers form impressions of retail power. Apparently the image elements included in the written scenarios and photographs were successful in evoking the "proper" perception of power among respondents; from the written sample scenario presented, these elements seemed to include building size and type, market segment served, price policy, credit policy, and parking. Since we are shown just one scenario and don't know the nature of the process used in developing the written and photographic stimuli, we can only speculate about why the scenarios contained the elements that they did. Given the results of the study, which tended to confirm the importance of perceived power as an influence on self-reported and projected deviant behavior, the issue raised here has much practical significance. Other things equal, manipulating those elements perceived as indicators of retail power or changing consumers' beliefs about the relationships which exist between store attributes or policies and store power could conceivably both increase consumer satisfaction and reduce deviant behavior directed at stores.

SOME METHODOLOGICAL CONCERNS

The surveys by Shuptrine and Wenglorz and by Houston and Nevin both utilized good-sized samples selected randomly from city populations, a method still not common in academic research. The cost and effort in such relatively ambitious surveys is substantial, placing a burden on the researchers to be sure the research question not only requires but is worthy of such a costly setting. Judging by the conclusions and recommendations offered by these two papers, neither study seem highly cost efficient, but it is possible that additional findings not reported here were obtained in each case.

Shuptrine and Wenglorz's research largely replicated the research question already investigated in a number of earlier studies. The justification appeared to be their desire to provide a more valid estimate of the true incidence of consumer problems growing out of their realization that earlier research may have understated the importance of consumer dissatisfaction. Assuming this objective to be the purpose of the research, one questions their estimate of the incidence of various consumer problem, since each respondent apparently reported on a single problem. An alternative method would have identified product categories first, then established the incidence of recent purchases in each, followed by probing questions getting at types and frequencies of problem experiences.

Houston and Nevin appeared to miss an opportunity to get at true differences in consumers' perceptions of downtown versus shopping center attributes by imposing the same set of image dimensions on responses for both kinds of shopping cluster. Unless questions not reported here covered both familiarity and actual patronage of the shopping clusters investigated, some very important validating data were excluded from the study.

While factor analysis simplified interpretation by reducing large numbers of variables into a much smaller number of dimensions, the nature of the question pursued here seems to call for more rather than less detailed analysis. Each of the 16 initial image dimensions represents a somewhat complex set of variables (how many aspects of "atmosphere' characterize downtown Madison?) each worthy of investigation if workable proposals for stimulating increased downtown patronage are to be derived. At the least, examination of the profiles of each shopping cluster on all 16 attributes should be made.

Mills' study recruited a group of shoppers in malls and elsewhere with no apparent attempt to sample a defined population. Details about the scenarios used and their method of development are not known. In Mills' design, hypothesis testing was based on totally hypothetical situations presented on the basis of quite artificial stimuli. There are real questions of external validity in this approach. As a preliminary (and perhaps not very expensive) test, the study is of value. Judgments of Mills' results, though, should probably await verification in research relying less on artificial scenarios. (Admittedly, given the sensitive nature of the issues investigated, more realistic designs don't come easily to mind). Since the behaviors in question are deviant acts, the validity problem is still more serious. It would be useful to know if actual levels of deviant acts by consumers experienced by retail outlets in each of the three power types vary as would be predicted by Mills' hypothesis. Measures of satisfaction with actual stores, classified similarly into the three types, would provide a stronger test of the hypothesis relating satisfaction to perceived power.

FUTURE RESEARCH

The linkage between consumers' perceptions of shopping clusters' attributes and actual patronage behavior has seldom been investigated, and Houston and Nevin's study represents an important start in what should be a fruitful stream of research. More detailed study of perceptions and misperceptions of shopping cluster characteristics (including location/convenience factors!) seems called for, followed by validating studies in which relationships between shoppers' evaluations of shopping clusters are compared with their actual patronage behavior.

The next step in Shuptrine and Wenglorz's dissatisfaction and complaining behavior research might well follow up their finding that probing questions appear to stimulate recall of dissatisfying experiences. Perhaps different levels of probing questions could be devised and the level of incidence of dissatisfaction indicated by each level could be experimentally determined. Further evidence that dissatisfaction is in part an artifact of the persistence of the interviewer would certainly force some reinterpretation of earlier research, and provide further incentive for researchers in this increasingly-popular area to devise more comprehensive (and perhaps less overt) measures of dissatisfaction. A second direction worthy of much additional effort is the exploration of evaluation processes giving rise to varying levels of satisfaction and dissatisfaction.

Deviant consumer behavior is clearly important to study but the sensitivity of the issues involved serve to magnify our usual problems in devising valid self-report measures. Nonetheless, Mills' hypotheses are of sufficient interest to warrant his continued efforts on the topic. More thorough exploration of the retail firm's attributes underlying consumers' perceptions of the firm's power type might give insights into aspects of store design, advertising media and copy, salespersons' behaviors and other controllable dimensions of the retail "mix" that could be manipulated to influence consumers' perceptions.

REFERENCES

Brown, F. E. (1969), "Price Image versus Price Reality," Journal of Marketing Research, 6, 185-91.

Olshavsky, R. W. and Granbois, D. H. (1979), "Consumer Decision Making -- Fact or Fiction?", Journal of Consumer Research, 6, 93-100.

Olshavsky, R. W., MacKay, D., and Sentell, G. (1975), "Perceptual Naps of Supermarket Locations," Journal of Applied Psychology, 60, 80-86.

Thompson, D. L. (1963), "New Concept: 'Subjective Distance' or Store Impressions Affect Estimates of Travel Time," Journal of Retailing, 39, 1-6.

Zajonc, R. B. (1980), "Feeling and Thinking: Preferences Need No Inferences," American Psychologist, 35, 151-175.

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