ThE THEoRY OF RE:ooL MARKEr STRUcruRE: APPUCATIONS 10 SPECIFIC MERCHANDISE LINEs William C. Rodgers Kenneth C. Schneider St. Cloud State University 8t. Cloud, Minnesota Introduction In what has become a classic article in the retailing literature, Pierre Martineau de- fined the personality or image of a retail store as "the way in which the store is defined in the shopper's mind, partly by its functional qualities and partly by an aura of psycho- logical attributes" [14, p. 47]. Spurred by Martineau's conceptualizations, research into store image has continued virtually unabated ever since. One stream of research has in- vestigated the use of various methods and procedures for measuring the image of par- ticular stores. At least two such measurement techniques have achieved widespread ac- ceptance; semantic differential and multidimensional scaling. A second stream of research has investigated the extent to which competing retail outlets in any market are defined by shoppers as similar to or different from one another. The primary goal of that research has been to better understand the overall structure of a retail market. Managers can then use that understanding to propose positioning strategies for one store (or store type) vis-a-vis competing stores (or store types). Several interesting and useful theories of the structure of retail markets have resulted from this research. For example, Malcolm McNair ([16], [17]) devised a theory explaining how new competitors in a market can gain entry by eliminating most or all ancillary services, thereby under- cutting prevailing price levels. Known as the "Wheel of Retailing," this theory has proved useful in studying the dynamics of retail positioning strategies over time. Also, Eliza- beth Hirschman [8] devised a theory explaining how retail competitors tend to cluster at one of relatively few points along a price-quality continuum. This theory again provides useful insight into appropriate positioning strategies for any given store. The purpose of this study was to apply the Hirschman theory of retail market struc- ture to the purchase process for a single product line - greeting cards. The study was undertaken to examine whether that theory, largely developed from the perspective of general merchandise retailing, could be applied on a micro level to one specific product line. Greeting cards were targeted in this research for several reasons. First, as part of a general trend toward scrambled merchandising, greeting cards are now available in a variety of retail outlets. Second, greeting cards are somewhat unique in posing a pur- chase situation involving minimal economic risk but substantial social risk (i.e., the risk that the consumer's choice will affect the way others think of himlher). Third, most consumers likely have at least some recent experience purchasing greeting cards because of their applicability in so many interpersonal relationships and social settings. Fall 1991 Rodgers & Schneider: Retail Market Structure Background, Hypotheses and Research Design 121 Hirschman [8], in a seminal discussion of the structure of competing retail institutions, postulated that consumers differentiate between types of general merchandise retailers on the basis of perceived quality and perceived price. According to Hirschman, discount department stores are seen by consumers as low price, low quality stores. Hence, they primarily appeal to price sensitive shoppers. At the other end of the price-quality con- tinuum, traditional department stores are seen by consumers as high price, high quality stores. Hence, they primarily appeal to quality conscious shoppers. National chain de- partment stores are seen as roughly mid-way between discount department stores and traditional department stores on both dimensions. These relationships are depicted graphi- cally in Exhibit 1. Hirschman further postulated that specialty stores in any given market can be similarly differentiated on the basis of price and quality; in essence, clustering around one or an- other of the three types of department stores. Thus, she concludes, "the three types of department stores form the essential core of a market's retail structure with the specialty stores serving as adjuncts to their functioning" [8, p. 33]. Despite the compelling logical appeal of Hirschman's theory (or maybe because of that appeal), there have been few attempts to empirically verify it. Specifically, it would be interesting to examine the extent to which the ideas depicted in Exhibit 1 transfer to specific lines of merchandise. Intuitively, it seems that consumers would rank sources of specific products such as greeting cards in much the same way that they do general merchandise. (Indeed, it is even plausible that general merchandise stores might be ranked at different points along the price-quality continuum for different products when the theory is applied to individual product lines. For example, Sears might be classified as a mid- price, mid-quality source of clothing but a high-price, high-quality source of tools.) Unfortunately, such extensions of the Hirschman theory have not yet been adequately addressed Accordingly, as applied to greeting card outlets, the first hypothesis is: H1: Consumers classify retail sources of greeting cards along a price-quality con- tinuum, with concentrations of sources at a low price, low quality extreme, a high price, high quality extreme, and a mid-range point. Dawson [4] enhanced Hirschman's theory of the structure of retail markets by propos- ing and empirically verifying that the continuum from the upper-right to lower-left ex- tremes in Exlnbit 1 can be viewed as a store prestige hierarchy. A sample of consumers was asked to compare the "social standing or level of prestige" of fifteen clothing stores representing a cross-section of Hirschman's store types (ie., local and national depart- ment stores, chain or mass merchandise stores, discount stores and specialty stores) (Ex- hibit 1). The resultant rankings were roughly as anticipated by Hirschman's price-qual- ity continuum. Department stores were perceived as having higher prestige than mass merchandise stores which, in turn, were perceived as having higher prestige than discount stores. Interestingly, national department stores were seen as having higher prestige than local department stores, with specialty clothing stores interspersed among these traditional 122 Journal ofBusiness Strategies Vol. 8, No.2 department stores. On the basis of these findings, Dawson concluded that "a prestige hierarchy does exist and that it closely approximates the price-quality dimension underly- ing Hirschman's [8] store typology" [4, pp. 142-143]. Thus, there is substantial conceptual rationale and growing empirical support for the contention that consumers classify retail stores on the basis of their prestige or social standing which, in turn, maps directly onto a price-quality continuum. Indeed, retailers are often advised to use such a continuum in selecting a positioning strategy. In one insightful discussion, Lucas and Gresham [13], after renaming the continuum as price- service rather than price-quality, warn retailers that positioning strategies off the price- quality/service diagonal in Exhibit 1 are doomed to failure in the long-run. Low price, high service positions trap many new specialty stores because they seem to offer a means of building a customer base. The problem, of course, is that such positions generally lead to bottom-line financial disasters. The converse strategy employing high price, low service positions would nicely solve the bottom-line problem. But, those positions typi- cally result in so small a customer base that top-line revenues are insufficient to cover operating expenses. Q>nsequently, Lucas and Gresham argue that the best chance for long-run survival and growth lies along the price-service continuum. Indeed, they are not even much enam- ored with the mid-range strategy typified by mass merchandisers, preferring instead ei- ther the low price, low service position of the price appeal discounter, or the equally clear- cut high price, high quality prestige position of the traditional department store. The problem with a mid price, mid quality position is that it entails straddling two well-de- fined positions. Unfortunately, adopting such a ''be all things to all shoppers" strategy runs the risk of leaving the store with an unclear, muddled image that appeals to nobody and results in too small a customer base. The national chains' management teams evi- dently concur with this assessment; at least it appears so given J. C. Penney's reposition- ing as a national department store (i.e., repositioning up the diagonal and more directly against traditional department stores) and Sears' recent stressing of its low everyday prices (Le., repositioning down the diagonal and more directly against discounters). Applying these arguments to greeting card purchases, one would expect to see more consumers patronizing outlets representing the extremes of the price-quality continuum in Exhibit 1 than outlets representing the mid-range of the continuum. Accordingly, the associated hypothesis is: HZ: More greeting card customers will be attracted to low price, low quality sources and to high price, high quality sources than to sources located at the mid-range along the price-quality continuum. The notion that consumers evaluate the prestige of various retail outlets based on price/ quality comparisons can be linked to several important issues in consumer behavior. For example, such prestige ratings are likely to be related to consumer strategies for reduc- ing perceived risk in shopping situations. Most behaviorists believe that consumers pur- chase "well-known" brands (i.e., brands of known quality, reputation or prestige) when facing what are thought to be risky purchase situations (see, for example, [19], [7]). H Fall 1991 Rodgers & Schneider: Retail Market Structure 123 purchasing high price, high quality brands represents a means of reducing perceived risk, so also would patronizing high price, high quality stores. Indeed, Hirschman [9] relied heavily on risk concepts in explaining why her three types of department stores tended to dominate certain lines of merchandise: Traditional department stores will dominate fashion goods and luxury items where the prestige of the store and the product's social conspicuousness are of integral importance (Le., there is high social risk).... National chain department stores, in contrast, will dominate staple hard and soft goods ... (that) are non-conspicuous, have low social risk, and possess moderate to high economic risk. ... Discount department stores will dominate branded hard goods (such as automotive supplies and small electrics) where economic risk is reduced by low prices, and social risk is overcome by brand name. [9, pp. 402-3] Several empirical studies confirm just such a link between store patronage and perceived risk. For example, Granzin and Schjeldrup [5] found that consumers make different decisions about automobile repair services when in a situation of high risk (e.g., stranded 150 miles from home) than when in a situation of low risk (e.g., at home). Similarly, Hawes and Lumpkin [6] found that consumers tend to rate discount stores as riskier than department or specialty stores when shopping for apparel, a purchase decision involving considerable social risk. They also found that price/quality perceptions are used heavily by consumers as risk reducing strategies in store patronage decisions for apparel. These findings suggest that consumers are relatively more likely to patronize higher prestige than lower prestige stores for purchases involving significant risk. Greeting cards pose relatively little economic risk. They have gotten somewhat more expensive in recent years, with most now priced at $1.00 and up. This, of course, can add up to substantial expense over the course of a year, especially for the frequent giver. Still, on a per unit basis, greeting cards are among the least expensive purchases made by consumers. However, greeting cards pose substantial social risk, as do other gift purchases. That greeting cards are gifts has been well-established in the literature. Belk [1] referred to greeting cards as gifts several times while discussing the four functions served by gift~ giving; communication, social exchange, economic exchange and socialization. The com~ munication function would appear especially well-served by greeting cards. Levi-Strauss [12] suggested that recipients understand well the importance of greeting cards as gifts when, for example, a person publicly displays Christmas cards on a fireplace mantel. Davis [3] argued that some objects, including greeting cards, are perceived solely as gifts, being of no inherent utilitarian value to the recipient Warshaw [22], in an empirical investigation into the determinants of purchase intentions regarding gifts, used Valentine's Day cards to operationalize an inexpensive gift condition. Thus, there is substantial agree- ment that greeting cards often serve one or another gift function. That gift-giving entails considerable social risk is also well-established. Vmcent and Zikmond [21] found that purchasing a kitchen knife entailed much more social risk when 124 Journal ofBusiness Strategies Vol. 8, No.2 bought as a gift than when bought for use at home. This suggests that behaviors such as buying a more expensive item, buying a well-known brand or shopping at a higher prestige store might be employed as risk reducing strategies intended to alleviate the perceived social risk implicit in selecting gifts. Indeed, Ryans [20J found that consum- ers are more likely to purchase small electric appliances in department stores (i.e., higher prestige outlets) when the purchase is for a gift than when it is for personal use. Mattson [15] found the same was true for gift versus personal use purchases of apparel. Finally, Warshaw [22] found that subjective social norms (i.e., an individual's perception of what others feel is appropriate behavior) are closely linked to purchase intentions regarding inexpensive gifts such as Valentine's Day cards and boxed chocolates. These various findings imply that the selection of gifts, including greeting cards, is fraught with social risk. H so, one would expect to see higher patronage of higher pres- tige than lower prestige sources of greeting cards. Accordingly, the appropriate hypoth- esIS IS: H3: More consumers will be attracted to high price, high quality sources than to low price, low quality sources of greeting cards. Finally, a prestige ordering of retail store types also can be conceptually linked to the social status of shoppers. That consumers tend to patronize stores whose images they feel most closely match their own self-images has long been known (see, for example, [23]). Burstiner [2], in discussing the results of a comparative image study of New York department stores, concludes: "Human nature being what it is, it seems reasonable to assume that store shopping preferences tend to be functions, more or less, of the degree of congruency or "match" between the shopper's perceptions of each store's ''personality'' and his own self-perceptions." [2, pp. 33-34] When applied to individuals in society, prestige refers to the degree of social rank, respect or esteem bestowed on various people. While prestige derives from various sources, in advanced industrial and information based societies it is most often a func- tion of one's occupation, income, wealth, education, or other measure of social status [4]. Thus, it is reasonable to anticipate that consumers will tend to patronize retail outlets that most closely match their own level of prestige. For example, Prasad [18] found a rela- tionship between social status and attitudes toward shopping at discount department stores, but only for higher social risk products. To wit: "For products that may be generally categorized as of high social risk, on the other hand, consumers in the upper socioeconomic strata appear to hold significantly different and relatively less favorable patronage attitudes toward discount stores as compared to those consumers in the lower socioeconomic strata" [18, pp. 46-47]. As applied to greeting card purchases, the associated hypothesis is: H4: The social status of shoppers will be directly related to patronage of greeting card stores by prestige ordering. That is, higher social status shoppers will patronize higher prestige stores, lower social status shoppers will patronize Fall 1991 Rodgers & Schneider: Retail Market Structure 125 lower prestige stores. Methodology The research was conducted as part of a larger study of gift giving behavior. Specifi- cally, telephone interviews were conducted with a sample of 440 householders randomly selected from a mid-sized, upper Midwest metropolitan area. All relevant information concerning 1) perceptions of differences in price and quality levels across various retail sources of greeting cards, 2) patronage rates across these outlets, and 3) shoppers' social status, was obtained through these interviews. Greeting cards are generally available in five types of outlets in the retail trade area covered by this study, including card/gift stores, department stores, discount stores, drug stores and supermarkets. Respondents were asked to rate these types of greeting card outlets as to perceived price and quality level using selected five-point Likert scale items, as shown in Exhibit 2(A). The first of these items, "greeting cards sold at [STORE TYPE] are expensive," was designed to measure perceived price level. Perceived quality was measured with two items, "the selection of greeting cards available at [STORE TYPE] is excellent" and "[STORE TYPE] greeting cards look nicer than those in other stores." lWo items were used to measure quality because there was some concern that perceived qual- ity might entail not only the physical product but also the breadth of selection at a store type. The results of a preliminary factor analysis (see Exlnbit 2(B» confirm that these three Likert scale items comprise two separate factors. A quality/selection factor with high loadings from the second and third statements was obtained, and a separate price factor with high loading from just the first statement was obtained. Consequently, responses from these three statements were used to classify the five store types on perceived price and perceived quality/selection, with the latter measure consisting of a simple average of the responses to the second and third Iikert scale statements. Elsewhere in the telephone interview, respondents were asked approximately how many birthday, anniversary, wedding and general cards expressing friendship or love had been purchased within the past thirty days. They were then asked whether most of these greet- ing cards had been purchased at a supermarket, department store, and so on. Responses to this question were used as a measure of store patronage. Finally, as part of a standard battery of demographic questions, respondents were asked their education level, household income and occupation. Responses to these questions were used to construct three separate measures of social status; education (college de- gree, some college, no college), income ($35,000 and above, $20,000 to $34,999, under $20,(00) and occupation status (higher level, mid-range, lower level). The last of these measures was derived by condensing into three categories the occupation scale developed by August Hollingshead [10] as part of his "Index of Social Position," which is now widely accepted as a measure of social class. 126 Journal ofBusiness Strategies M=gor Findings and Discussion Vol. 8, No.2 Using the results concerning perceived quality/selection and perceived price, the five store types can be plotted in two-dimensional space, as in ExbIlJit 3(A). The relative placement of the store types in this exhibit conforms extremely well to that expected from the Hirschman model of retail structure. Thus, H1 is strongly supported by these find- ings. Consumers do classify retail sources of greeting cards along a price-quality con- tinuum, though perhaps not a strictly linear one. More precisely, cardlgift stores are perceived alone as a relatively high quality, high price retail source of greeting cards. Relative to cardlgift stores, supermarkets and discount stores are perceived as low price, low quality types of outlets. Finally, drug stores and department stores are seen as oc- cupying what could be called a mid-range point: exactly mid-range in terms of perceived price, with perceived quality/selection higher than but nearer to discount stores and su- permarkets than to card/gift stores. If the five store types are reclassified as quality appeal stores (Le., card/gift stores), price appeal stores (Le., discount stores and supermarkets) and mixed appeal stores (i.e., drug stores and department stores), (Exhibit 3) the patronage results appear as follows: 'JYpe of Store Quality Appeal Mixed Appeal Price Appeal Total Percent of Customers 51.8% 18.3% 29.9% 100.0% These results, shown graphically in Exhibit 3(B), can be used to examine both H2 and H3. H2 implies that both quality appeal and price appeal stores should enjoy greater patronage than mixed appeal stores because of the enhanced clarity of their respective positions. The graphical results in Exhibit 3(B) tend to confirm such an implication. Indeed, using 95 percent lower confidence limits for the ratio of multinomial probabili- ties it can be shown that the patronage ratio to mixed appeal stores is not less than 2.25:1. Similarly, the partronage ratio of price appeal to mixed appeal stores is not less than 127:1. Thus, it is inferred that for every 100 consumers who patronize mixed appeal stores there are at least 127 who patronize price appeal stores and 225 who patronize quality appeal stores. Hence, H2 is supported by these findings; a mid-range position on the priceI quality or prestige continuum is not a particularly effective one in terms of generating patronage. Using a similarly derived patronage ratio, it can also be inferred that for every 100 consumers who patronize price appeal stores there are at least 143 who patronize a qual- ity appeal store. Thus, H3, which suggests that quality appeal stores will attract more consumers than price appeal stores because of the perceived social risk inherent in giv- ing greeting cards, is also supported by these results. Together, of course, H2 and H3 lead to a striking hierarchy among these three types of outlets for greeting cards. Owing both to its clarity of position and the security of- Fa111991 Rodgers & Schneider: Retail Market Structure 127 fered in the purchase of a socially risky product, quality appeal stores are able to garner a relatively large share of the greeting card market. Offering similar clarity but much less security against perceived risk, price appeal stores attract the next largest share of the market. Finally, lacking even a well defined position, mixed appeal stores struggle with the lowest market share. Finally, Exhibit 4 presents findings appropriate to H4, linking the status or prestige of shoppers to the prestige ordering of types of outlets for greeting cards. As can be seen in Exhibit 4, H4 is only partly supported by these findings. A significant relationship was found between education level and patronage across these types of stores. Specifi- cally, patronage of quality appeal stores increases with increasing education, while the converse occurs in mixed appeal stores. A similar but weaker relationship occurs with respect to occupation status. While not statistically significant, patronage of quality ap- peal stores shows a tendency to increase with increasing occupation status while, again, the converse occurs in both price and mixed appeal stores. Findings relative to income are both statistically insignificant and directionally incon- sistent with H4. This inconsistency, especially the V-shaped patronage of quality appeal stores across income, is at least partly explained by an associated result. In particular, it was found that relatively young, unmarried consumers show a disproportionate tendency to patronize quality appeal stores (details not shown). While speculative, it is probable that this consumer segment patronizes quality appeal stores despite their generally lower incomes because of the especially intense social risk involved in greeting card exchange in a dating or courtship relationship. (Exlnbit 4) This, coupled with the status expectations from H4, could explain why quality appeal stores tend to draw disproportionately from both the "Under $20,000" and the "$35,000 and Above" income groups. Finally, it is also plausible that because greeting cards pose minimal economic risk, any link between the prestige of stores patronized and income would be much less strong than for more expensive lines of merchandise. On balance, then, support for H4 is weak. The most appropriate conclusion is that there may be a mild patronage relationship between greeting card store prestige and consumer social sta- tus. But, even if there is such a link, it depends on other characteristics of the con- sumer and/or purchase situation, including possible intervening links to consumer life stage characteristics and related variables. Concluding Observations Perhaps the most important conclusion to be gleaned from this study is that the gen- eral theory of retail structure, originally developed by Hirschman ([8], [9]) in the context of general merchandise retailing, does indeed apply to at least some individual merchan- dise lines as well. This is a significant finding in these days of unbridled scrambling of merchandise lines across types of retail outlets. Consumers still manage to simplify their thinking about outlets for products like greeting cards by categorizing stores into one or another of Hirschman's three groups based on perceived quality and price; quality appeal stores versus price appeal stores versus those in the middle. It would, of course, be most instructive to examine the extent to which these findings 128 Journal ofBusiness Strategies Vol. 8, No.2 hold in still other product lines, especially those which have been subjected to intense scrambling during the past decade or so. Among others, grocery products have special appeal as a topic for such research. Packaged foods and related grocery items are now available in a dizzying array of outlets, from traditional supermarkets to wave after wave of food discounters, and from established convenience stores to a host of other outlets capitalizing on the place utility inherent in food retailing (e.g., drug stores, service sta- tions, car washes and video rental stores). It would be most interesting to see if these and other emerging grocery outlets are similarly classified into the same three price/qual- ity categories. Second, this study adds further empirical support to the contention that success in re- tailing tends to go to those stores and/or store types with a distinctive personality. While this has been said many times before, its importance cannot be overstressed. Retail outlets that attempt to operate from a "middle of the road" position designed to appeal to wide segments of the consuming public generally end up with indistinct and undistinguished images that appeal to no one. How much better it is to set the store apart as one that attempts to accomplish something for someone; whether it be differential advantage on the quality side or on the price side. Finally, the results of the study suggest that a quality appeal might be somewhat more effective than a price appeal if the purchase decision is one that entails sizable social risk. lbis, too, carries important strategic implications. Consumer behaviorists have don~ considerable work in identifying and studying the phenomenon of perceived risk. Retail managers, however, have been slow to make use of that knowledge. Here, though, is an area of direct application. It seems that, when faced with a purchase decision carrying significant social risk, many consumers gravitate toward the security of the quality ap- peal store more so than the thrift of the price appeal store. This study, of course, was concerned only with social risk. It would be equally in- structive to see whether or not choosing a quality appeal store becomes increasingly more likely as other types of purchase risk increase. For example, Jacoby and Kaplan [11] postulated the existence of at least four other risks besides social risk that consumers might face in any given purchase situation. These include financial risk (i.e., the purchase price), performance risk (ie., whether the product works as it's supposed to), physical risk (i.e., risk of harm or injury in using the product), and psychological risk (ie., how the pur- chase affects the consumer's self-image). Each of these mayor may not be related to store choice in the same sense that social risk appears to be so related. Finally, interesting issues pertaining to consumer behavior have been raised, especially in the finding that, while many consumers turn to quality appeal sources of greeting cards to reduce perceived risk, not all consumers do so. In addition, the selection of greeting card sources cannot be adequately explained just by the social status of consumers. Thus, there are likely other demographic or psychological underpinnings to this choice. Do some consumers perceive relatively less social risk in card and gift giving situations than others? Are some consumers relatively more driven than others by different store choice criteria, possibly including shopping convenience or store proxirnity?Is perceived risk in purchasing greeting cards highly situational for some consumers, increasing and de- creasing by oocasion and recipient? Each of these would be an interesting avenue for Fa111991 Rodgers & Schneider: Retail Market Structure 129 further research into perceived risk and retail store choice. References 1. Belk, R. "Gift-Giving Behavior." Research Marketing. Sheth, J. (ed.) Greenwich. Cf: JAI Press (1979) pp. 94-126. 2. Burstiner, I. "A Three-Way Mirror: Comparative Images of the Clienteles of Macy's, Bloomingdale's, and Korvettes." Journal of Retailing. Vol. 50 (Spring 1974), pp. 24-36. 3. Davis, J. "Gifts and the V.K. Economy." ~ Vol. 7, No.3 (1972), pp. 408429. 4. Dawson, S. "An Exploration of the Store Prestige Hierarchy: Reification, Power, and Perceptions." Journal of Retailing. Vol. 64 (Summer 1988), pp. 133-152. 5. Granzin, K.L. and Schjelderup, lUI. "Situationally-Determined Risk as an Influence on Buyer Behavior." Evolving Marketing thought for 1980. Summey, JR. and Taylor, R.D. (eds.) Carbondale, IL: Southern Marketing Association (1980), pp. 127-131. 6. Hawes, J.M. and Lumpkin, J.R. "Perceived Risk and the Selection of a Retail Patronage Mode." Journal of the Academy of Marketing Science, Vol. 14 (Winter 1986), 3742. 7. Hawkins, D.I., Best, RJ., and Coney, K.A. Consumer Behavior: Implications for Mar- keting Strategy. Fourth Edition, Plano, TX: Business Publications, Inc. (1989). 8. Hirschman, E.C. "A Descriptive Theory of Retail Market Structure." Journal of Retail- ing. Vol. 54 (Winter 1978), pp. 2948. 9. . "Retail Competitive Structure: Present and Potential." 1979 Educators Conference Proceedings. N. Beckwith et al. (eds.) Chicago, IL: American Marketing Association (1979), pp. 401405. 10. Hollingshead, A.B. and Redlich, F.C. Social Oass and Mental lliness. New York, NY: John Wiley & Sons, Inc. (1958). 11. Jacoby, J. and Kaplan, L. ''The Components of Perceived Risk." Proceedings of the Third Annual Conference of the Association for Consumer Research. Venkatesan, M. (ed.) Chicago, IL: Association for Consumer Research (1972), pp. 382-393. 12. Levi-Strauss, C. "The Principle of Reciprocity." Sociological Theory. Coser, L.A. and Rosenberg, B. (eds) New York, NY: Macmillan Company (1965). 13. Lucas, G.H. Jr. and Gresham, L.G. "How to Position for Retail Success." Business. Vol. XX (April-June 1988), pp. 3-13. 14. Martineau, P. "The Personality of the Retail Store." Harvard Business Review. Vol. 36 (January-February 1958), pp. 47-55. 130 Journal ofBusiness Strategies Vol. 8, No.2 15. Mattson, B.E. "Situational Influences on Store Choice." Journal of Retailing. Vol. 58 (Fall 1982), pp. 46-58. 16. McNair, M.P. "Significant Trends in the Postwar Period" Competitive Distribution in a Free High Leyel Economy and Its Implications for the University. Smith, A.B. (ed.) Pittsburgh, PA: University of Pittsburgh Press (1958), pp. 17-18. 17. and May, E.G. "Next Revolution of the Wheel of Retailing." Harvard Business Review, Vol. 56 (September-October 1978), pp. 81-90. 18. Prasad, V.K. "Socioeconomic Product Risk and Patronage Preferences of Retail Shoppers." Journal of Marketing. Vol. 39 (July 1975), pp. 42-47. 19. Roselius, T. "Consumer Rankings of Risk Reduction Methods:' Journal of Marketing. Vol. 35 (Janwuy 1971), pp. 56-61. 20. Ryans, A "Consumer Gift Buying Behavior: An Exploratory Analysis:' Contemporary Marketing Thought. BeUenger, D. and Greenberg, B. (eds.), Chicago, 1L: American Mar- keting Association (1977), pp. 99-104. 21. Vincent, M. and Zikmond, W. "An Experimental Investigation of Situational Effects on Risk Perception." Advans;es in Consumer Behavior. Scblinger, MJ. (00.), Chicago, 1L: Association for Consumer Research (1975). 22. Warshaw, P.R. "Buying a Gift: Product Price Moderation of Social Normative Influences on Gift Purchase Intentions." Personality and Social Psychology Bulletin. Vol. 6 (March 1980), pp. 143-147. 23. Weale, W.B. "Measuring the Customer's Image of a Department Store:' Journal of Retailing. Vol. 37 (Summer 1961), pp. 40-48. Fall 1991 Rodgers & Schneider: Retail Market Structure 131 Exhibitl Retail Market StnJctDre Highest Quality Traditional ~ Department Stores ,~ ,,,,,,,,,,,,,,,, Lowest Quality ,,,,,,,,,,,,,, @Discount Department Stores Lowest Price Highest Price -------ooo{.~----:::--- National Chain Department Store Source: Redrawn from Hirschman, Elizabeth C. (1978), itA Descriptive Theory of Retail Market Structure," Journal of Retailing, 54 (Winter), p. 32. Exhibit 2 Measurement of Store Prestige A. Perceptions of Store Types - Index Items Item Verbal Description 1 Greeting Cards Sold at [Store Type] Are Expensive 2 The Selection of Greeting Cards Available at [Store Type] is Excellent 3 [Store Type] Greeting Cards Look Nicer Than Those In Other Stores Item B. Factor Analysis Results Quality/Selection (Factor 1) Price (Factor 2) 1 2 3 -0.006 0.780 0.785 0.992 -0.106 0.096 Eigenvalue Pet of variance 1.225 40.8% 1.004 335% 132 Journal ofBusiness Strategies Exhibit 3 A Two~Dimensional Plot of Greeting Card Outlets Vol. 8, No.2 Highest Selection/Quality Lowest Card/Gift Store 0 r L,....J D D DrugStore Discount Store Department Store D Supermarket Lowest Price Highest B. Market Share by Type of Greeting Card Outlet Highest Quality Appeal Stores Selection/Quality Lowest o~ Appeal Stores Lowest Price Mixed Appeal Stores Highest Fall 1991 Rodgers & Schneider: Retail Market Structure 133 Exhibit 4 Relationship Between Greeting Card Store Patronage and Socioeconomic Characteristics Store Type Choice by Socioeconomic Price Mixed Quality Chi-Square Characteristics Appeal Appeal Appeal Total P-Value Totals, All Respondents 29.9% 18.3% 51.8% 100.0% By Education Level College Degree 30.8% 9.6% 59.6% 100.0% <.05 Some College 30.9% 16.3% 52.8% 100.0% No College 27.6% 25.5% 46.9% 100.0% By Occupation Status Higher Level 28.1% 14.9% 57.0% 100.0% n.S. Mid Range 26.5% 21.7% 51.8% 100.0% Lower Level 34.3% 19.0% 46.7% 100.0% By Income $35,000 and up 28.1% 16.5% 55.4% 100.0% n.s. $20,000-$34,999 30.9% 23.6% 45.5% 100.0% Under $20,000 29.1% 15.1% 55.8% 100.0% The Theory of Retail Market Structure: Applications to Specific Merchandise Lines