A STUDY OF DEPARYMENT STORES’ WAGES HELD BY CUSTOMERS AND MANAGEMENT Thesis for the Degree of D. B. A. MICHIGAN STATE UNIVERSITY DEVENDRA S. PATHAK 1972 LIBRARY Michigan Sta te University This is to certify that the thesis entitled A STUDY OF DEPARTMENT STORES' IMAGES HELD BY CUSTOMERS AND MANAGEMENT presented by Devendra S. Pathak has been accepted towards fulfillment of the requirements for D.B.A. Marketing & 419 e in , gre Transportat 10n Major professor / Datea'éj’éfi /g,1 :9‘ 7% 0-7639 ABSTRACT A STUDY OF DEPARTMENT STORES' IMAGES HELD BY CUSTOMERS AND MANAGEMENT BY Devendra S. Pathak The purpose of this study was to investigate and compare the aggregate department store images held by cus— tomers and management and to replicate and extend the Wyckham study.1 The data were collected through group interviews in the Saginaw area from customers and management on five image dimensions of four test stores (Wiechmann, Sears, Federal, and K-Mart) and a hypothetical "Ideal" store. Thirty bipolar semantic differential scales were used to identify five store image dimensions. The first four dimen- sions were: merchandising suitability, sales personnel, store congeniality, and locational convenience. The fifth dimension was an overall dimension used as a summary measure based on all 30 scales in the questionnaire. Data were analyzed to obtain answers to four major questions investigated in this study. The major findings of the analysis were as follows: 1. Different department stores were found to carry differentiable aggregate images among all of their customers Devendra S. Pathak who were members of different social classes, family life cycle stages, races, and sexes. The stores were rated by customers on a continuum from highest to lowest. The regular department stores were rated more favorably than the discount department store. Also, the downtown depart— ment stores were rated less favorably than the shopping center department stores on the locational convenience dimension. 2. Customers classified by social class, stage in family life cycle, attitudes toward shOpping, frequency of shopping, and purchases of different kinds of products did not hold differentiable aggregate images of particular department stores. However, there were a few groups which carried differentiable aggregate images of specific depart— ment stores. For example, Federal was favored by non-whites and Wiechmann, a high status store, was favored by females and by customers whose beliefs about Wiechmann were endorsed by their friends. Recent shoppers and avid readers of advertisements of any department store carried favorable images of that store. Also, differentiable aggregate store images of different locations of particular department stores were held by customers shOpping at those locations. 3. There was communality between the anticipated and the actual aggregate images for high status stores. This was not true for low status stores. Management of all the test stores consistently overrated their stores as compared to their customers. Devendra S. Pathak 4. There was a positive relationship on the mer- chandising suitability dimension and a lack of relationship on all the other image dimensions between customers' satis- faction and marketing orientation of department store management. An additional dimension of this research was updating of the bibliography on images done by Wyckham, Lazer, and Crissy.2 1Robert G. Wyckham, “Aggregate Department Store Images: Social and Experiential Factors" (unpublished Ph.D. thesis, Michigan State University, 1967). 2Robert G. Wyckham, William Lazer, and W. J. E. Crissy, Images and Marketing: A Selected and Annotated Bibliography YChicago: American Marketing Association, 1971). A STUDY OF DEPARTMENT STORES' IMAGES HELD BY CUSTOMERS AND MANAGEMENT BY Devendra S. Pathak A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF BUSINESS ADMINISTRATION Department of Marketing and Transportation Administration 1972 f ACKNOWLEDGMENTS I wish to express my grateful appreciation to Dr. William J. E. Crissy, under whose guidance and direction this study was made, for his close cOOperation, invaluable comments, and more than patient corrections. Upon comple- tion, I more fully realize my indebtedness to his skillful guidance. Thanks are also extended to other members of the com- mittee, Dr. Paul E. Smith and Dr. Donald A. Taylor, for their encouragement and support of this study. A word of thanks and warm appreciation is also extended to Mr. Stuart Gross, Mrs. Sheila Morley, and the members of various groups in the Saginaw area for being helpful in the collection of data. Special appreciation is extended to Miss Sharon Weaver for her help in rough drafts of this study and to Mrs. Ann Brown for her expert advice and skill in typing the final draft of this dissertation. Grateful acknowledgment is extended to Dr. Guy M. Lee, Sr. for his support, friendly counsel, and valuable suggestions. ii I especially wish to acknowledge the perseverance, understanding, and warm support of my wife, Diane, who pro- vided constant inspiration and strength during the course of this research. iii LIST OF TABLES LIST OF FIGURE TABLE OF CONTENTS S. O O O O O O O O O O O O O O O O O O 0 LIST OF APPENDICES O O O O O O O O O O O O O O O O O 0 Chapter I 0 INTRODUCTION 0 O O O O O O O O O O O O O O O 0 Marketing Management and Retailing . . . . . Explanation of the Model . . . . . . . . . The De Depa Statement of the Problem Defi Hypotheses . . . . . . . Need for and Contribution of the Study to Concept of Store Image and partment Stores. . . . . . . . rtment Store Image and the Model nition of the Terms. . . . Limitations of the Study . . Marketing Theory and Practice. . . . . . Procedure for the Investigation. . . . . . . Organization of the Study. . . . . . . . . . II. REVIEW OF RELATED LITERATURE . . . . . . . . . Updating Wyckham's Bibliography. . . . . . . Depa vs rtment Store Image: Management . Customers. . . . . . . . . . . . . . . Models of Department Store Image . . . . Image Dimensions of a Store. . . . . . . The Influence of Social Environment on Department Store Image . . . . . . . . The Influence of Information on Department Store Image . . . . . . . . Images of Branches of Department Stores. Summary of Related Literature. . . . . . . . III. RESEARCH DESIGN AND PROCEDURE. . . . . . . . . Sample Design. . . . . . . . . . . . . . . . The Interview Schedules. . . . . . . . . . . Interview Schedule for Customers . . . . iv Page vi ix 20 21 22 23 26 32 38 39 39 45 45 47 47 The Semantic Differential. . . . Selection of Department Stores . . The Activity of Shepping . . . . ShOpping Behavior Questions. . . Demographic Data . . . . . . . . . Interview Schedule for Management. Techniques Used in Testing Hypotheses. Sample Responses . . . . . . . . . . . Selected Demographic Characteristics of the Sample of Customers. . . . . . . . . . IV. DATA ANALYSES AND RESULTS. . . . . . . . . . ' Hypothesis I . . Hypothesis II. . Hypothesis III . Hypothesis IV. . V. FINDINGS, IMPLICATIONS, AND RECOMMENDATIONS. Findings 0 O O O O O O O O O O O O O O 0 Implications of th Findings . . . . . . . Additional Questions for Future Studies. . BIBLIOGRAPHY O O O O O O O O O O O O O O O O O O O O APPENDICES O O O I O O o I O O O O I O O O O O O O O Page 48 54 55 55 57 59 60 64 65 73 73 87 113 115 124 124 130 133 136 143 Table l. 10. ll. 12. 13. 14. 15. LIST OF TABLES Hypothesized Components and Subcomponents of Retail Store Image . . . . . . . . . . . Variables for Image Analysis. . . . . . . . . Attributes for Store Image Analysis . . . . . Image Dimensions of Retail Stores . . . . . . Determinants of Cognitive Dimensions of Store Image . . . . . . . . . . . . . . . . Kind of Favorite Store of Cleveland Women, by Social Class . . . . . . . . . . . . . . Kind of Department Store Favored by Cleveland women 0 I O O O O O I O I O O O O O O O O 0 Revised House Type Scale. . . . . . . . . . . Classification of Respondents Into Social Classes. . . . . . . . . . . . . . . Classification of Respondents Into Family Life Cycle Stages: This Study and the Wyckham Study . . . . . . . . . . . . . . . Classification of Respondents by Race . . . . Classification of Respondents by Sex. . . . . Classification of Respondents by Age Categories. . . . . . . . . . . . . . . . . Classifications of the Heads of the Households by Education . . . . . . . . . . Probabilities Associated With Wilcoxon T Scores Obtained From the Comparisons of Aggregate Images of Test Department Stores and an "Ideal" Store Among All Customers. . vi Page 27 29 3O 31 33 36 36 58 66 67 68 69 7O 70 74 Table Page 16. Probabilities Associated With Wilcoxon T Scores Obtained From the Comparisons of Aggregate Images of Different Department Stores Among Customers in Different Social Classes. . . . . . . . . . . . . . . . 77 17. Probabilities Associated With Wilcoxon T Scores Obtained From the Comparisons of Aggregate Images of Different Department Stores Among Customers in Different Family Life Cycle Stages . . . . . . . . . . . . . . 80 18. Probabilities Associated With Wilcoxon T Scores Obtained From the Comparisons of Aggregate Images of Different Department Stores Among Customers of Different Races . . 84 19. Probabilities Associated With Wilcoxon T Scores Obtained From the Comparisons of Aggregate Images of Different Department Stores Among Customers of Different Sexes . . 86 20. Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Aggregate Images of Individual Department Stores Among Customers of Different Social Classes. . . . . . . . . . . . . . . . 89 21. Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Aggregate Images of Individual Department Stores Among Customers in Different Life Cycle Stages. . . . . . . . . . . . . . . . . 91 22. Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Aggregate Images of Individual Department Stores Among Customers of Different Races . . 95 23. Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Aggregate Images of Individual Department Stores Among Customers of Different Sexes . . 97 24. Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Aggregate Images of Individual Department Stores Among Customers Grouped by Sh0pping Enjoyment . . . . . . . . . . . . . . . . . . 98 vii Table 25. 26. 27. 28. 29. 30. 31. 32. Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Images of Individual Department Stores Among Customers Grouped by the Social Support for Their Beliefs About the Store . . . . . . . . Probabilities Associated With Mann~Whitney U Scores Obtained From the Comparisons of Images of Individual Department Stores Held by Customers Grouped by Their Advertising Readership for Those Stores . . . . . . . . . Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Images of Individual Department Stores Among Customers Who Differ in the Line of Goods They Buy From These Stores. . . . . . . Probabilities Associated With Mann-Whitney U Scores Obtained From the Comparisons of Images of Individual Department Stores Among Customers Grouped on the Basis of Recency of Their Shopping at Those Stores . . Probabilities Associated With Mann—Whitney U Scores Obtained From the Comparisons of Images of Individual Department Stores Among Customers Grouped on the Basis of Their Frequency of ShOpping at Those Stores . Probabilities Associated With Mann—Whitney U Scores Obtained From the Comparisons of Images of Individual Department Stores Among Customers Grouped on the Basis of Their ShOpping Experience With Different Locations of Those Stores . . . . . . . . . . Probabilities Associated With Mann—Whitney U Scores Obtained From the Comparisons of Anticipated Images Held by Management and Actual Images Held by Customers for Individual Department Stores. . . . . . . . . Aggregate Image Deficiency Scores and Inverse Marketing Orientation Index Scores of Individual Department Stores on All Image Dimensions. . . . . . . . . . . . . . . . . . viii Page 100 102 104 106 108 111 114 117 Figure 1. LIST OF FIGURES Retailing Management System's Strategy Design . . . . . . . . The Concept of Image in Retailing Management System's Strategy Design. . . . A Behavioral Model of Man and "Image" Image Development . . . . . . . . A Scatter Diagram of AID and IMOI Scores for All Test Stores . . . . . . AID and IMOI Scores of All Test Stores on Individual Image Dimensions . . AID and IMOI Scores of Individual Test Stores on All Image Dimensions. ix Page 24 25 118 119 120 LIST OF APPENDICES Appendix Page A. UPDATED BIBLIOGRAPHY O O O O O O O O O O O O O 144 B. LIST OF GROUPS APPROACHED TO OBTAIN DATA . . . 154 C. INTERVIEW SCHEDULE FOR THE CUSTOMERS . . . . . 156 D. INTERVIEW SCHEDULE FOR THE MANAGEMENT. . . . . 171 E. EXPLANATION OF INDEX OF SOCIAL CHARACTERISTICS C C C O O O O O O O O O C O O 176 CHAPTER I INTRODUCTION Marketing Management and Retailing From an ecological vieWpoint, a specie is likely to survive if it has a tenable position in the existing organism-environment system. This position is known as its "ecological niche."l Though it is possible for a system to survive in the short term without every niche being filled, the very existence of an unfilled niche offers an Opportu— nity to be exploited by any species or genus qualified to £111 it.2 Analogously, any retailing firm must have a position or niche within the competitive environment or business milieu if it is to survive and prosper. Application of ecological thinking to marketing provides a useful way of studying competition based on the behavior of firms existing within a common marketing envi- ronment. However, it is based on the assumption of hetero- geneity in the market.3 A customer, for example, may not want just any shirt, but instead may want a sports shirt, a dress shirt, or a turtleneck shirt, or a particular style, color, or material. COping with this heterogeneity is the greatest challenge to marketing practitioners. In this heterogeneous demand structure of the market, the ability to 1 survive and prosper hinges on the firm exploiting and occupy- ing an appropriate niche within its capability and resources. This implies an adaptation of a marketing philosophy on the part of the management.4 The marketing philosophy refers to the top management's acceptance of an orientation to the requirements of the marketplace and the matching of profitable market Opportunities with business efforts. This marketin philosophy is implemente throug the marshalling of a store's resources and capabilities—— its marketing concept.5 It has been suggested that in order for retailers to become more market oriented and to build the marketing phil- OSOphy into their own retail management systems, they should: 1. Engage in more research. 2. Be more alert to changes in customer wants. 3. Consider market integration. 4. Alter the orientation of sales-supporting personnel. 5. Alter the organization structure. 6. Engage in more long—term planning. 7. Charge t0p management with responsibility of creating the proper atmOSphere or state of mind.6 The above suggestions mean an adOption of the market— ing management concept in retailing for those "retail managers concerned with designing a total retail capability to achieve realistic and attainable objectives."7 The marketing manage— ment concept in retailing is characterized by planning, cus— tomer orientation, systems approach, change, and innovation. The total planning and direction of a retail system as per the marketing management concept is shown in Figure 1. This model is based on a few generally accepted models in the literature on marketing and retailing regarding the applica— tion Of the marketing management concept to business firms.9 Explanation Of the Model Retailing firms have some objectives in mind for their Operations. Some Of these objectives are profit, share of the market, sales volume, industry leadership, com— 10 A firm munity status, store image, and channel control. plans its retailing strategy to achieve these objectives. The retailing strategy basically consists of two parts: (1) selecting a target market, and (2) market programming. Selecting a target market, in turn, consists of four stages: 1. Analysis, evaluation, and prediction of the changing environment including changing wants, needs, and technology. 2. Utilization of the above analysis to identify actual and potential markets. 3. Matching these markets with the store's present and future capabilities. 4. Identifying the target market.11 The target market consists Of those customers whose needs and wants are in closest coincidence with the firm's offerings. Once the marketing Opportunities are identified, retail managers go to the next stage of total retail Environmental Factors 'Social oTechnological oChanging Wants & Needs oCompetition oOthers Retailing Objectives °Industry Leadership “Volume 'Community Status °Market Share 'Channel Control ‘Store Image °Profit Planning Retailing Efforts Analysis ++ Evaluation ++ Prediction' \l/ \1' Actual & Present & Potential Future Store Markets Capabilities, Matching V 4 [MarketingOpportunityAssessmentJ Target Market' Market Programming 4/ \ ‘Resources required for competitive job Determining the Retailing Mix fto be done ‘1' ;. I Goods & Service. Mix Communication‘ Mix l 1‘ J 1 Physical Physical Human Distribution Resources Resources Mix .J l J L - l ' l IMarket‘Impactj Figure 1.--Retailing Management System's Strategy Design. strategy--market programming. The major tasks of market programming are to determine the retailing mix together with the mobilization Of resources required to achieve the proper market impact. If the proper planning has taken place, the final output will be in line with the objectives originally planned by the management.12 The Concept of Store Image and Department Stores The above discussion reflects the application Of the marketing concept in prOper planning and direction Of a retail system's efforts to achieve its Objectives. But it fails to include a relatively recent behavioral concept which has profoundly affected retail managers' thinking regarding their retail strategies and how they might per— ceive, analyze, and cultivate a given market opportunity.13 This concept is the store image-~an idea credited in its present form to Pierre Martineau.14 Martineau argued that a consumer's preference for one store over another is not determined by the price, quality, or services Offered by any store, but it is caused by the personality or image projected by the store. He noted: What is it that draws the shOpper to one store or agency rather than another? Clearly there is a force Operative in the determination of a store's customer body besides the functional factors Of location, price ranges, and merchandise Offerings. I shall show that this force is the store personality or image-—the way in which the store is defined in the shopper‘s mind, partly by its functional qualities and partly by an aura of psychological attributes. The concept of a store's personality or a store's image is very important for a retailer to exploit. Only when management is cognizant of the image which the store is projecting can it hOpe to bring its product and service Offerings in alignment with its store image and, in turn, match these with its perceived market Opportunity.16 Weale comments: The closer management's image of its organization is to that held by potential customers, the more effec- tive will be the store's communication through adver- tising and public relations. This sense Of identity will affect the buying, pricing, and service functions of the store. . . . It behooves management to measure its store image periodically to determine the framework within Which it should Operate.17 Thus, the store image is and should be a part of retailing management's strategy design and it should be recognized as such. Figure 2 is an extension of Figure 1. It shows the role of store image in the retail management system comprising the store's total offerings——the store, its merchandise, its service, and its people. Department Store Image and the Model The model is divided into two parts by dotted lines. Part I is the traditional retailing management model with- out the final market impact of the market programming. Part II of the model is the behavioral extension of the traditional model, which includes the concept of store image in the retailing management system's strategy design. As Part I of the model is already explained, only the extension Environmental Factors Retailing Objectives 'Social °Profit °Industry Leadership oTechnological °Volume ' °Community Status oChanging Wants °Market Share 'Channel Control 8 Needs °Store Image 0Competition \l’ oOthers . [ Planning Retailing Efforts j LAnalysis ++ Evaluation +* Prediction] v v Actual & Present & Potential Future Store Markets Capabilities Matching lMarketing,OpportunityiAssessmentJ [Target Market] lMarket Programming W iv Determining the Resources required Retailing Mix for competitive job . k(/////’ to be done 1 I . - l 1 Goods 5 Service H Communication ” Physical Physical Human Mix Mix Distribution Resources Resources Mix L l I j l Aggregate Store Image] ‘_ A g 1 1 Actual, . Anticipated] . Matching ‘ 1 Complete Partial Complete Congruency Congruency Separation v f Y Actual 8 . . Anticipated Anticipated AntiCipated Image Sufficiency Partial'lmage DGfiCiODCY Complete Image Defici- or No Dissatisfaction or Some Dissatisfaction ency or Complete Dis- satisfaction Figure 2.--The Concept of Image in Retailing Management System's Strategy Design. Environmental Factors Retailing Objectives , 'Social °Profit 'Industry Leadership oTechnological . °VO1ume ' ‘Community Status oChanging Wants 'Market Share ‘Channel Control 5 Needs °Store Image oCompetition \L ~0thers [ Planning Retailing Efforts j [Analysis ++ Evaluation ++ Prediction] v v Actual & Present 8 Potential Future Store Markets Capabilities Matching Jr tunity Assessment] ‘ LMarketigg_Qppor LTarget Market] [Market Programming ‘1 4 Determining the Resources required Retailing Mix for competitive job ‘(/////’ to be done 4 1 . . 71 . I ”’7 Goods & Service H Communication 0-. Physical Physical Human Mix Mix Distribution Resources Resources Mix L l A I l ' I J [_Aggregate Store Imagg] {—7 . I 1 l Actual, ? A Anticipated] Matching ‘ I Complete Partial Complete Congruency Congruency W Separatiod I ./ \ Actual G Anticipated . Anticipated Anticipated Image Sufficiency Partial‘Image Deficiency Complete Image Defici- or No Dissatisfaction or Some Dissat13f3°t1on ency or Complete Dis- satisfaction Figure 2.--The Concept of Image in Retailing Management System's Strategy Design. of the model, i.e., Part II, will be explained in this section. The traditional retail system‘s strategy design suggests the direct relationship between the total market programming and the system's output—-customer satisfaction. However, the final market impact Of the market programming depends upon how the market views the market programming as developed by the store management. The perception of the store and its market programming by the customers is the basis for the development of the actual aggregate image of a store. Similarly, management also carries an aggregate image of the store. This image is the anticipated aggregate image that management is trying to project in order to satisfy those comprising its market. If management of a store is marketing oriented and if prOper planning has taken place, the firm should achieve its objectives.18 The actual and the anticipated aggregate images Of the store should be completely congruent. This complete congruency of the actual and the anticipated images should lead to image sufficiency or no dissatisfaction in the market. However, other possibilities are that matching of the actual and the anticipated aggregate store images may result in partial-congruency or, in the extreme case, in complete separation of the two images. Statement of the Problem Most of the studies on store images have been con- ducted to investigate actual store images held by the customers.19 These studies generally indicate that actual images held by the customers about a particular department store depend on the customers' identification with one or more reference groups. It has been found that groups influ- ence individuals' shOpping and buying decisions.20 Indi- viduals tend to adOpt a value and belief structure similar to that of other members of their socio-economic reference groups.21 These groups for a department store may be classified on the basis Of different factors such as social class, stages Of family life cycle, race, sex, and atti- tudes toward shopping. The studies conducted on store images also emphasize the importance Of the collection of data on store images held by customers and the Opportunity provided by the collection of such data to the management to check their own perceptions of their store images in the eyes Of customers.22 NO study was found which compared the aggregate store images held by customers and management. It is the purpose Of this study tO investigate and compare the aggregate department store images held by customers and management and to replicate and extend the Syckham study.23 The study is directed toward the following ques— tions: 1. Are the aggregate images held by the customers different from store to store?24 10 2. Does the aggregate image of a department store held by the customers vary depending upon their membership in different groups classified on the basis Of their demo- graphic characteristics, various types Of experiences with the store, and attitudes toward shopping?25 3. Does the anticipated aggregate image held by the management of a store match the actual aggregate store image held by the customers of a department store? 4. Is there a relationship between customers‘ satisfaction (or dissatisfaction) and marketing orientation (or lack of it) of management of a department store? Definition of the Terms Actual aggregate department store image--The sum- mation of all the actual images Of the department store held by the customers.26 Anticipated aggregate department store image--The summation Of all the anticipated images Of the department store held by the managers Of the store. Actual image or an individual customer's image of a dgpartment store--The whole of all sensory perceptions and thought interrelationships associated with a department store held by an individual customer.27 Or, it is a con- stellation of attitudes formed regarding different facets Of a department store by an individual customer. Anticipated image or an individual manager's image Of a department store--A manager's perception Of aggregate department store image held by the customers. 11 Aggregate image deficiency--The summation of differ- ences between the actual images of any department store and images of the "Ideal" department store held by the cus- tomers.28 This definition of image deficiency is similar to the one used by Porter and Lawler to measure the degree of satisfaction in their study on Managerial Attitudes and Performance.29 They defined satisfaction as the extent to which a place Of business meets or exceeds the expectations of customers. To the degree that actual facets of a business fail to meet the expectations of a shopper, dissatisfaction was said to occur. Similarly, to the degree that actual aggregate image fails to meet the aggregate image Of customers' "Ideal" store, i.e., customers' expectations to get the complete satisfaction from a department store, the image deficiency or dissatisfaction can be said to occur. Aggregate differentiable department store image-- An image consisting Of attitudes Of direction and intensity distinguishable from other images.3 Communality of aggregate department store images-- The tendency for a collection of images to contain atti- tudes which are similar in direction and intensity.31 Image element--One of the attitudes held by a cus- tomer toward a particular aspect of a department store. Image dimension--A collection of image elements . . 32 centering on a particular facet Of a department store. Social class--"A grOUp of people who share equal or nearly equal social prestige and community status" 12 and who tend to "share the same goals and ways of looking at life."33 Family life cycle stages-—Discrete units of time in a family's existence described in terms Of marital status, age, and the presence of children.34 Manager--Any incumbent of any job that any department store wishes to call a management job.35 Hypotheses I. Different department stores have differentiable aggregate images among their customers classified by 1.1 social class, 1.2 stage in family life cycle, 1.3 race, 1.4 sex. II. Differentiable aggregate store images are held for a particular department store by customers who are classified by 2.1 social class, 2.2 stage in family life cycle, 2.3 race, 2.4 sex, 2.5 their attitudes toward shopping, 2.6 the social support for their beliefs about that store, 2.7 their readership Of advertisements Of that store, 13 2.8 their shOpping practices with that store, 2.9 their preference for shOpping at different locations of that store in a city. III. There is a communality between the anticipated aggregate image Of a department store held by the management and the actual aggregate image Of the same department store held by the customers. IV. There is a significant relationship between customers' satisfaction (or dissatisfaction) and marketing orientation (or lack Of it) of management for a particular department store. Limitations of the Study Generalizations drawn from this study should be interpreted with caution, inasmuch as the department stores and customers investigated in this study are selected from the city Of Saginaw only. Though the study investigates the relationship between the aggregate image deficiency of a department store with the marketing orientation of the management, the normative aspects Of what specifically should be done by the management or the customers faced with the problems of image deficiency regarding a particular department store is considered outside the scope of this study. Need for and Contribution of the Study to Marketing Theory and Practice Different studies have investigated the concept of store image using different instruments and different 14 methodologies. The conclusions reached by each individual researcher are applicable so far as the conditions for his research hold true. Although this is a good exercise in methodologies, it leads to a dangerous situation. On the basis of these conclusions, marketing inferences are drawn and incorporated into the marketing literature without any further verification. This goes on in marketing in spite Of the warning that NO findings in the social and behavioral sciences should be accepted until they have been replicated a number of times. Replications add to the richness of understanding as well as the confidence that can be placed in a conclusion.36 This is the first study in the area of store image that attempts to replicate and extend the work of another study. Data are also provided on the anticipated aggregate store images held by the management. The comparison of the anticipated and the actual aggregate store images should provide the management with insight regarding their under- standing of the customer. The research yields information for the department stores studied regarding their customers classified by demographic characteristics, by their various shopping experiences and practices, and by their attitudes toward shopping. The study investigates the relationship Of the aggregate image deficiency of a department store with the marketing orientation of the management. 15 Procedure for the Investigation The procedure used in this study to test the hypotheses consists Of the following: 1. Development of the questionnaire. 2. Selection of the subjects. 3. Collection and analyses Of the data Obtained from the administration of the questionnaire. Organization of the Study The study consists of five chapters, a selected bibliography, and appendices. Chapter I consists of a model for application of marketing management in retailing, the concept Of store image and department stores, definition of terms, hypotheses, limitations Of the study, need for and contribution Of the investigation, and outline for the organization Of the study. Chapter II updates Wyckham's bibliography and reviews only those articles from the updated bibliography which are directly related to the concept Of image as applied to retailing store management. Chapter III describes the research design and the sample responses for the study. The sample selection pro— cedure, the questionnaire used in the collection of data from customers and management Of test stores, and tech- niques used in testing hypotheses are outlined. The sample 16 responses are presented on the basis of selected demographic characteristics of the respondents. Chapter IV presents the analyses of data. Between and within stores comparisons Of the aggregate store images held by customers in different groups are provided; the anticipated aggregate store image held by management-and the actual aggregate store image held by customers are compared for individual test stores; and finally, the rela- tionships Of customers' satisfaction and marketing orienta- tion Of management for particular department stores are investigated. Chapter V summarizes the findings of the study. Implications of the findings of the study and some additional questions for future studies are presented. FOOTNOTES--CHAPTER I Wroe Alderson, Marketing Behavior and Executive Action (Homewood, Illinois: Richard Irwin, Inc., 1957Y7 p. 29. 21bid., p. 30. 31bid., p. 101. 4 Rom J. Markin, Jr., Retailing Management: A Sys- tems Approach (New York: The Macmillan Company, 1971), p. 33. Surprisingly enough, the marketing philosophy and the mar- keting concept are confused in the literature on marketing and retailing; e.g., Markin mentions: "The marketing concept (or marketing philOSOphy as it is oftentime called). . . ." Also John E. Wakefield, "The Ten Cogs in Marketing for Profits," Sales Management (October, 1959), 3. William Lazer, Marketing Management: A Systems Perspective (New York: John Wiley & Sons, Inc., 1971), pp. 25-260 6Eugene H. Fram, "Application of the Marketing Concept to Retailing," Journal of Retailing (Summer, 1965), 21. 7William Lazer and Eugene J. Kelley, "The Retailing Mix: Planning and Management," Journal Of Retailing, XXXVII, 1 (Spring, 1961), 34. ‘ 81bid.. pp. 34-35. 9Lazer, Op. cit., pp. 48, 88-89; Mark E. Stern, Marketing Planning: A Systems Approach (New York: McGraw- Hill BOOk CO., 1966), pp. 12-20; Thomas A. Staudt and Donald A. Taylor, A Managerial Introduction to Marketing (2nd ed.; Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1970), pp. 40-54; Lazer and Kelley, Op. cit., pp. 34-41. loLazer and Kelley, Op. cit., pp. 34-41. 11Lazer, Op. cit., p. 48. 12Lazer and Kelley, Op. cit., p. 41. 13Markin, Op. cit., pp. 102, 159. Pierre Martineau, "The Personality of the Retail Store," Harvard Business Review, XXXVI, 1 (January- 17 18 lSIbid., p. 47. 16 . . Markin, Op. c1t., p. 106. 17 W. Bruce Weale, "Measuring the Customer's Image of a Department Store," Journal Of Retailing, XXXVII, 2 (Summer, 1961), 41. 18 Supra, p. 6. 19To mention a few: Douglas M. Egan, "Evaluative Bias, Consumer Attitudes and Department Store Image," Southern Journal Of Business, VI, 3 (November, 1971), 73- 81; Weale, Op. cit., pp. 40-48; Stuart U. Rich and Bernard D. Portis, "The Imageries' Of Department Stores," Journal of Marketing, XXVIII (April, 1964), 10-15: John V. Petrof, TiThe Corporate Image of Chain Stores Among Ghetto Residents," Southern Journal of Business, IV, 2 (April, 1969), 33-44; Stuart U. Rich and Bernard A. Portis, "Clues for Action From Shopping Preferences," Harvard Business Review, XLI (March-April, 1963), 132-149; William Lazer and Robert G. Wyckham, "Perceptual Segmentation Of Department Store Markets," Journal of Retailipg, XLV, 2 (Summer, 1969), 3-14. 20M. Venkatesan, "Experimental Study of Consumer Behavior: Conformity and Independence," Journal of Market- ing Research, III (November, 1966), 384-387. 21Francis S. Bourne, Group Influence in Marketing and Public Relations (Ann Arbor: Foundation for Research on Human BehavIOr, 1956). 22Egan, Op. cit., p. 80; Weale, Op. cit., pp. 41, 47; Leonard L. Berry, "The Component Of Department Store Image: A Theoretical and Empirical Analysis," Journal Of Retailing, XLV, 1 (Spring, 1969), 3-20; Rollie Tillman, "Semantic Differential Of Consumer Images of Retail Stores," Southern Journal of Business, II, 2 (April, 1967), 67-74; Michael Perry and Nancy J. Norton, "Dimensions of Store Image," Southern Journal of Business, V, 2 (April, 1970), 1-7 a 23Replication of the Wyckham study. 24Replication of the Wyckham study. 25Robert G. Wyckham, "Aggregate Department Store Images: Social and Experiential Factors" (unpublished Ph.D. thesis, Michigan State University, 1967. 26The term "Aggregate Department Store Image" is adopted from Wyckham, Ibid. 19 27Ben M. Enis, "An Analytical Approach to the Concept Of Image," California Management Review (Summer, 1967). 51. 28The term "Image Deficiency" is adopted from John F. Bolger, Jr., "How to Evaluate Your Company Image,“ Journal Of Marketing (October, 1959), 9. 29Lyman W. Porter and Edward F. Lawler, III, Managerial Attitudes and Performance (Homewood, Illinois: Richard D. Irwin, Inc., 1968), pp. 30-31. 30Wyckham, Op. cit., p. 7. 3lIbid. BZIbid. 33 Richard P. Coleman, "The Significance of Social Stratification in Selling," Proceedings Of the 43rd National Conference of the American Marketing Association, ed. by Martin L. Bell (December, 1960), pp. 157-158. Quoted in Wyckham, Op. cit., p. 8. 34John B. Lansing and Leslie Kish, "Family Life Cycle as an Independent Variable," American Sociological Review, XXII (October, 1957), 512-519. Quoted in Wyckham, op. cit., p. 8. 35Thomas W. Harrell, Manager's Performance and Personality (Chicago: South-Western Publishing CO., 1961), p. 14. 36James F. Engel, David T. Kollat, and Roger D. Blackwell, Consumer Behavior (New York: Holt, Rinehart, and Winston, Inc., 1968), pp. 627-628. CHAPTER II REVIEW OF RELATED LITERATURE Boulding, in his book The Image, suggests that the concept of image is a common thread running through many disciplines.1 Wyckham substantiates Boulding's contention through his exhaustive, but selective, bibliography on the concept Of image.2 He found that the concept of image is used in the military, academic, medical, corporate, and many other settings. The variety of literature on the application of the concept of image since the Wyckham study in 1967 has continued to grow. A few studies representing this variety are: the body image of the "other“ in hypno- analysis,3 the image of electronic publications,4 the mind and the mind's image itself,5 the negro image and the mass media,6 prestige rating of business and other occupations,7 the image of marketing as a field Of study among business students,8 and the image concept and its application in banking.9 Considering the wide variety Of research, the following procedure was used in reviewing the literature on the concept of image in marketing: 1. Use Wyckham's study as a starting point. 2. Update Wyckham's bibliography. 20 21 3. Review only those entries from the updated bib- liography which are directly related to department stores' images. Updating Wyckham‘s Bibliography To update Wyckham's bibliography on the concept of image in marketing, an extensive search of relevant litera- ture since 1967 was made and a selected bibliography was prepared. The criteria used in making the selection were the same as used by Wyckham, Lazer, and Crissy in prepara- tion Of their selected and annotated bibliography on Imagg' and Marketing.10 These criteria were: 1. Does the reference add to a general understanding Of images and marketing? 2. Does the reference contain research findings that are directly applicable to marketing? 3. Does the reference extend the current state of marketing knowledge? 4. Does the reference lead to more effective market— ing practice? 5. Does the reference lead to better measurement of images in marketing? 6. Does the reference suggest areas worthy of future investigation in marketing?11 The bibliography prepared on the basis Of the above criteria is presented in Appendix A. The organization of the bibliography is comparable to the bibliography prepared by Wyckham, Lazer, and Crissy. It consists Of three sec- tions. They are: (1) General concepts and references,' (2) Measurement techniques, and (3) Empirical findings. The remainder of this chapter is based on the ref- erences selected from the above-mentioned bibliography that are directly related to department store images. 22 Department Store Image: Management vs. Customers It is commonly assumed that retailers and retail store managers have the first—hand information about cus- tomers because Of their familiarity and frequency of con- tacts with the customers.12 Yet, all of the studies conducted on retail store image have focused on store images held by customers only and attention is paid to management in a peripheral manner. However, in the area of product image, there is a study conducted to compare retailers' and consumers' perceptions Of selected appliance attributes.l3 McClure and Ryan conducted two simultaneous studies in Indianapolis. They found that retailers con— sistently underestimate the importance of some of the appli— ance attributes to consumers in their purchasing behavior. These attributes were service and warranty, ease Of use, and style. ‘They also found that the differences existed in the retailers' perception of competitive brands as com— pared tO customers' perception. "These images were either over-sensitive or under—sensitive to specific attributes and seem to reflect historic stereotype rather than current consumer brand images."14 If the discrepancy in brand image held by management and customers can be extended to department stores' images, it may partially explain some of the complaints in the market place regarding the present mass merchandising system. 23 Models of Department Store Image The department store image is just a special case Of the corporate image.15 However, the department store image, as compared to the corporate image, held by the customers is based on more first-hand information. Cus- tomers come into more contact with the store and its sales personnel than with the companies manufacturing the products which are sold. It is on the basis of these experiences with the store--actual, imagined, or vicarious-—that image develOpment Of a store takes place. The above contention is the basis Of two existing models in the literature on retail store images. These models, as developed by Kunkel and Berry16 and by Wyckham,l7 are presented on the next two pages as Figures 3 and 4, respectively. Kunkel and Berry's model is based on the learning theory. The basic postulate of their model is that all the behavior patterns evolve from learning. They summarize their thesis and the model as follows: By reinforcing particular behavior patterns when they appear in a specific context, the norms Of a society and various subcultures determine which aspects Of an individual's context will, over time, take on control- ling prOperties. The applicability of this proposition to the problem Of retail store image is evident upon its dissection: by reinforcing Sr (wide selection Of fashionable overcoats), particular behavior patterns R (going to retail store A to buy an overcoat), the norms Of a person's subculture (value placed on wearing fashionable overcoats) determines which aspect of an individual context SD (the image of retail store A) will, over time, take on controlling properties and lead him to shOp at store A.18 24 Societal and Sub- cultural Norms ’ SV r~7Habit SD ’,s ;_,_ 0 SA > R :4: C y 5 “Sa ‘3 . . Extinction Discriminative Behavior Contingent Stimuli Stimuli Figure 3.--A Behavioral Model Of Man and "Image". Relevant terms: R = Any behavior pattern C = Consequences Sr, 53, SO Sr = Rewarding Stimuli (eventually leading to "habit") Sa = Aversive stimuli (eventually leading to "extinction") S0 = Absence Of any consequence. SA = Stimuli in whose presence R has not been reinforced SD = Stimuli in whose presence R has been reinforced (image) SV = State variables (i.e., conditions Of deprivation and dissamation) Source: John H. Kunkel and Leonard L. Berry, "A Behavioral Conception of Retail Image," Journal Of Marketing, Vol. 32, no. 4 (October, 1968), p. 23. 25 .pe .a .Aaoma .suamum>acp cumum savanna: .GOHumuuommflO ..o «no Omamwaasm lacy =mnouomm amflucoflnomxm can Hmfioom “mommaH ououm usefiuummoo mammouomfl: .Emnxoaz .u phenom "condom ucoEQOHo>oa OooEHn|.v ousmflm qaoHooqumsma , muwmua oncommom HMGOmHomueucH fit aTlll wnoum mo 1 messiah Eoflmmonm— All m cofiumafimuoi N. All, muwamflcomcou _ muwoua oncommom V HNGOmHOQHmucH 4f AIIII unowmm m fiuzoosuwuud —mmwmmmouom_aTllll. 3 W I Vfimoflum>wpoz_ku aTlll moaom m I moH>ammm «mwmzH _ avenue oncommom S HMGOMHOQHOUGH Z All. KnUHHflfldflflfim MW II. 1.wmesuanua _moaumoouma annull » m cowuomwumm_ kn AIIIJ omfiocosouoz muwmua ommmmmwm, HMGOmHmmumucH it .Allll mocowco>coo lixoosuauum cowumoouom_ aTIIIl, x bhmofluo>wuox_ma aTIII. GOHHMOOA a pomflno a humane a humane uafimmm wflflufiuufl mcowmcmfifla mmmmmooum OOSUfluud m0 mfisuwuufl mo posse mouse Hmoamoaososma mcoamamsaa consumes nonsense [ mm omH>zm qsoHoqumOS r BzmzzomHgm QUHUOAOHUOM Bzmzz ll|||l||||ll||l||||l|lll|ll1 26 Wyckham's model, as shown in Figure 4, represents the image develOpment of a store taking place through the summation Of attitudes toward various dimensions Of that store. This process of image develOpment takes place within and is affected by the customer's psycholOgical, physio- logical, and sociological environment. However, an indi— vidual's attitudes toward a store are created within these environments through the interaction of individual‘s per— ception, motivation, and interpersonal response traits. Thus, psychological processes, like the Kunkel—Berry model, are an inherent part Of the image develOpment in the Wyckham model. Image Dimensions of a Store As shown in the Wyckham model, the final aggregate image of a store consists Of different image dimensions. The image dimension is "a collection Of image elements centering on a particular facet Of a department store."19 Different studies have found different components or facets of a store important to concentrate on for understanding the aggregate image of a store.20 With the use of unstructured instruments, followed by content analysis and coding responses, Kunkel and Berry found that 99 per cent of the reSpOnses agreed with the prehypothesized components and subcomponents.21 These prehypothesized components and subcomponents were derived from the existing studies conducted by Martineau,2 27 . 2 Fisk, 3 and Collazzo.24 These components are shown in Table 1. Table l.--Hypothesized components and subcomponents Of retail store image. Components Subcomponents 1. Price of Merchandise a. Low prices. b. Fair or competitive prices. c. High or noncompetitive prices. d. Values, except with specific regard to premiums, such as stamps, or quality of mer- chandise. 2. Quality of Merchandise a. Good or poor quality Of merchandise. b. Good or poor department(s), except with respect to assortment, fashion, etc. c. Stock brand names. 3. Assortment of Merchandise a. Breadth of merchandise. b. Depth Of merchandise. c. Carries a brand I like. 4. Fashion Of Merchandise 5. Sales Personnel a. Attitude Of sales personnel. b. Knowledgeability Of sales personnel. c. Number Of sales personnel. d. Good or poor service. 6. Locational Convenience a. Location from home. b. Location from work. c. Access. d. Good or poor location. 7. Other Convenience Factors a. Parking b. Hours store is Open. c. Convenience with regard to other stores. d. Store layout with respect to convenience. e. Convenience (in general). Table 1.--Continued. 28 Components Subcomponents 8. Services a. Credit. b. Delivery. 9. Sales Promotions 10. Advertising 11. Store Atmosphere 12. Reputation on Adjustments Restaurant facilities. Other services (gift consul- tants, layaway plans, baby strollers, escalators, etc.). Special sales, including quality or assortment of sales merchandise. Stamps and other promotions. Fashion shows and other special events. Style and quality of advertising. Media and vehicles used. Reliability of advertising. Layout of store without respect to convenience. External and internal decor of store. Merchandise display. Customer type. Congestion. Good for gifts, except with respect to quality, assort— ment or fashion of mer- chandise. "Prestige" store. Returns. Exchange. Reputation for fairness. Egan administered a questionnaire (see Table 2) con- taining 12 preselected characteristics relevant to different facets of a department store to 172 randomly selected 25 housewives in a large eastern city. From a factor analysis of the data, six components were discovered as important for 29 studying retail store image. These components were: 1. Merchandising factor, with emphasis on service and style; Product factor, primarily composed of quality and selection; Economic factor——I, with emphasis on expenses; Dependability factor; Economic factor—-II, with emphasis on credit policy and shOppers' wealth; Accessibility factor, with emphasis on conven— ience Of location. These six factors accounted for 67 per cent Of the variance. Table 2.--Variables for image analysis (shown in bipolar adjective form). Variable Number Variable 1 Dependable --Undependable 2 Modern --Old Fashioned 3 Expensive --Inexpensive 4 Friendly --Unfriendly 5 Good Quality --Poor Quality 6 Courteous --Discourteous 7 Good Service --Poor Service 8 Old ShOppers --Young Shoppers 9 Good Selection--Poor Selection 10 Rich Shoppers --Poor ShOppers 11 Easy to get to--Hard to get to 12 Good Credit --Poor Credit Policy Policy 30 Perry and Norton used a random sample of 100 female students attending Oklahoma State University to discover dimensions upon which women evaluate fashion stores.26 They utilized stores' attirubtes as well as stores themselves as their variables in the factor analysis. From the rotated fattor matrix of responses to seven attributes (see Table 3) Of the six stores, they found three factors. In the order of importance, these factors were: 1. Sales-personnel factor. 2. Price—quality factor. 3. Congeniality factor. When they used the stores as variables and performed the factor analysis on the dimension Of overall impression, a two factor solution was found. These two factors were labeled as Sales-personnel factor and the Economic (or budget) factor. Table 3.--Attributes for store image analysis. Courteous l 2 3 4 5 6 7 Salespeople High Low Service Good Bad Knowledgeable Salespeople High Low Price High Low Quality ~ High Low Atmosphere Good Bad Overall Impression Good Bad Kelly and Stephenson used a semantic differential instrument with 30 items and found eight image dimensions. (See Table 4.) 27 Table 4.-—Image dimensions of retail stores. Image Dimension I Advertising by the Store Informativeness Of adsa Helpfulness Of ads Appeal Of ads Believability of ads Frequency of ads Image Dimension II Physical Characteristics of -the Store Cleanliness of store Attractiveness Of decor Ease of finding items Ease of moving through store Speed of checkout Image Dimension III Convenience Of Reaching the Store Nearness Of location Time required to reach store Ease of drive Convenience Of other stores Image Dimension IV Your Friends and the Store Known to friends Liked by friends Recommended by friends Number of friends patronizing Image Dimension V Merchandise Selection Degree Of selection Level Of stocks Number of brands Image Dimension VI Store Personnel Courtesy of personnel Friendliness of personnel Helpfulness of personnel Number Of personnel Image Dimension VII Prices Charged by the Store Relative level of price Level of value Number of special prices Image Dimension VIII Dependability of the Store Dependability of products Quality Of products Well-known brands Level of value aKey words are used to indicate the subject of the particular scale, that is, "informativeness Of ads" refers to the scale uninformative--informative. Source: Robert F. Kelly and Ronald P. Stephenson, "Semantic Differential: An Information Source for Designing Retail Patronage Appeals," Journal of Marketing, XXXI (October, 1967), 32 Wyckham based his research on Fisk's findings Of six cognitive dimensiOns Of the department store image (see Table 5). Wyckham compressed these six dimensions into four by combining value Of price and post—transaction satisfaction into merchandising suitability.28 The Influence of Social Environment on Department Store Image Wyckham, in his review of literature, found that "intimate primary groups, secondary groups and even statis— tical groups have an impact on the image a person has of a department store."29 He also suggests that if these groups are rated on the basis of their strength of controlling their members' attitudes, the rating will be: primary groups (family, friends, colleagues)—-the most powerful; secondary groups (social class, church, school)--next in strength; and statistical (demographic) groups (age, sex, and race)--the least powerful.30 Primary Groups: Family and Friends.--The family is considered a very important primary reference group affect— ing an individual's attitude toward an Object. One of the important concepts that has emerged out Of the research on the family as a primary group is the family life cycle.31 Attitude patterns have been claimed tO vary as the family passes through various stages of the family life cycle.32 Wyckham reports from the Collazzo study that family size and stages in the family life cycle affect attitudes toward 33 Table 5.--Determinants Of cognitive dimensions Of store image. Cognitive Dimensions Determinants l. Locational Convenience 2. Merchandise Suitability 3. Value for Price 4. Sales Effort and Store Services 5. Congeniality 6. Post-transaction Satisfaction (a) access routes (b) traffic barriers (c) traveling time (d) parking avail- ability (a) number of brands stocked (b) quality Of lines (c) breadth of assortment (d) depth Of assortment (e) number Of outstanding departments in the store (a) price of a particular item in a particular store (b) price Of same item in another store (c) price of same item in substitute store (d) trading stamps and discounts (a) courtesy Of sales clerks (b) help- fulness Of salesclerks (c) reliability and usefulness of advertising (d) billing procedures (e) adequacy Of credit arrangements (f) delivery promptness and care (g) eating facilities (a) store layout (b) store decor (c) merchandise displays (d) class of customers (e) store traffic and congestion (a) satisfaction with goods in use (b) satisfaction with returns and adjustments (c) satisfaction with price paid (d) satisfaction with accessibility to store Source: George Fisk, "A Conceptual Model for Studying Customer Image," Journal Of Retailing, XXXVII (Winter, 1961-62), 5. 34 shopping and retail institutions.33 However, Wyckham, in his own study, rejected the hypothesis that differentiable aggregate images of particular department stores are held by the members of groups which are at different stages in the family life cycle.34 Approximately only 20 per cent Of the comparisons between the family life cycle stages were found significant. Even this significance was between nonadjacent life cycle stages. Rich and Jain found that regular department stores ranked high among women at any stage of life cycle except the younger women with children.35 Friends, also, play an important role in shOpping decisions. Rich and Jain report friends as the most impor- tant source Of influence in all the social classes.36 However, when Wyckham compared the store image of consumers whose friends agreed with their beliefs about a store with those whose friends disagreed with them, the results were not clear cut. It was found that the differences were significant on store congeniality and locational convenience, but the differences were not significant on merchandise 37 suitability and sales personnel. Secondary Groups: Social Class.--Socia1 class, though classified as a secondary group, due to its inter- relationship with primary groups is a very important factor in the image analysis.38 Different studies have reported the influence of social class on consumers' attitudes toward different retail stores.39 Wyckham found only one 35 study that mentioned: "The image of a [department store]. . . does not vary from group to group [i.e., social classes]. . . and users to non-users."40 Several other studies conducted since Wyckham's study agree with the above contention and question the usefulness of "social class concept in under- standing consumer behavior in View of recent changes in income, education, leisure time, movement to suburbia and other factors."41 Though the doubts expressed by Rich and Jain in the above quotation may be true for the weakening effect of social class in relation to consumer behavior in general, it does not specifically deal with the effects Of social class on different individuals' perceptions of var— ious department stores. This can be seen from Tables 6 and 7, adOpted from the same Rich and Jain study from which the above quotation on weakening effect of social class on consumer behavior was taken.42 Table 6 shows that the majority Of women from dif- ferent social classes mentioned the regular department store as their favorite shOpping source. Among the favorite department stores mentioned by name, only three department stores were found to be mentioned most frequently.‘ These stores were called a high fashion store, a price appeal, and a brand appeal store. Table 7 shows that the high fashion store was favored progressively more by each higher class. The price appeal store was inversely related to social class, and the broad appeal store was mentioned most frequently 36 Table 6.--Kind of favorite store of Cleveland women, by social class. Social Class Kind of Store L-L U-L L-M U-M L-U U-U Regular department 51% 60% 77% 83% 88% 91% Discount department 14 ll 6 2 .. 9 Variety and junior department 2 6 6 5 .. .. Mail order 9 14 5 2 3 .. Medium to low specialty 2 2 1 .. 6 .. Neighborhood ll 2 1 1 3 .. Others 11 5 4 7 .. .. Total 100% 100% 100% 100% 100% 100% Number Of cases 132 346 265 206 36 ll Source: Stuart U. Rich and Subhash Jain, "Social Class and Life Cycle as Predictors Of ShOpping Behavior," Journal of Marketing Research, V (February, 1968), 47. Table 7.--Kind Of department store favored by Cleveland women. Social Class Kind of Department Store L-L U-L L-M U-M L-U U-U High fashion store 4% 7% 22% 34% 70% 67% Price appeal store 74 63 36 24 19 18 Broad appeal store 22 30 42 42 11 15 Total 100% 100% 100% 100% 100% 100% Number of cases 67 208 204 71 32 10 Source: Stuart U. Rich and Subhash Jain, "Social Class and Life Cycle as Predictors of Shopping Behavior," Journal of Marketing Research, V (February, 1968), 47. by the middle classes. Thus, when viewed in the context of department store images, the Rich and Jain study shows a strong relationship between social class and kind of 37 department store. Wyckham accepted the hypothesis that differentiable aggregate images of particular department stores are held by members of different social classes. He concluded: "The social class effect appears to be most strong on the merchandise suitability, store congeniality and locational convenience dimensions and most weak on sales personnel."43 Statistical Groups: Race.--The least powerful group affecting an individual's attitude is the statistical group. These groups are defined by basic demographic data such as age, race, and sex. It has been held that blacks hold dif— ferent attitudes toward a retail store than whites do.44 The bigness or just mere size of the department store elicits feeling of insecurity among negroes for it places them on the defensive. Bullock mentions: One reason for this is what some call their "high visibility," that is, in telling us Of their reac- tions to downtown shOpping, many negroes confessed that they felt conspicuous—-as if all eyes were on them.45 However, for whites, "The atmosphere Of bigness and for— mality characterizing most department stores presents no awe which they cannot overcome."46 Whatever the reasons, there are differences between the perception Of department stores held by negroes and whites. Wyckham concluded that differentiable aggregate images of particular department stores are held by members of different races.47 When the difference in the store image held by different races for 37 department store. Wyckham accepted the hypothesis that differentiable aggregate images of particular department stores are held by members of different social classes. He concluded: "The social class effect appears to be most strong on the merchandise suitability, store congeniality and locational convenience dimensions and most weak on 43 sales personnel." Statistical Groups: Race.-—The least powerful group affecting an individual's attitude is the statistical group. These groups are defined by basic demographic data such as age, race, and sex. It has been held that blacks hold dif— ferent attitudes toward a retail store than whites do.44 The bigness or just mere size Of the department store elicits feeling of insecurity among negroes for it places them on the defensive. Bullock mentions: One reason for this is what some call their "high visibility," that is, in telling us Of their reac— tions to downtown shOpping, many negroes confessed that they felt conspicuous-—as if all eyes were on them.45 However, for whites, "The atmOSphere of bigness and for— mality characterizing most department stores presents no awe which they cannot overcome."46 Whatever the reasons, there are differences between the perception Of department stores held by negroes and whites. Wyckham concluded that differentiable aggregate images of particular department‘ stores are held by members Of different races.47 When the difference in the store image held by different races for 38 each individual store were investigated, non-whites were found to prefer Sears and Federal and no preference by race was found for Hudsons. Thus, differences in the store image held by different races depend on the individual department store in question. The Influence of Information on Department Store Image The process of perceiving any department store by a customer can be simply described as an information transfor— mation process.48 The customer takes in the information regarding the store, transforms it within his own environ- ment, and puts out his judgments. Most of the information a customer Obtains regarding a department store is from his own experience with the store, from the advertisements of the store, and by word of mouth. The more experience the customer has with the store, the better basis for his judg- ments. 'Frequency and recency Of shopping at a store are two important factors affecting customers' experiences with department stores. Advertisements also have been found to shift the image of the store "in the direction Of more favorable image for viewers."49 Wyckham found that more favorable images Of the department stores were held by those who were frequent customers or who always read the adver— tisements as compared to non-frequent or sometimes and never-reader customers.50 The effect Of recency of shopping on department store images was not found significant.51 39 Images of Branches of Department Stores The research on multi-unit department stores' branch images does not provide clear-cut findings.52 This lack Of clarity of the department store's branch images is repre- sented by the following statement quoted by Wyckham from the Rich and Portis study: Martineau's belief that the branch stores take on the personality and characteristics of their downtown units was borne out to some degree. . . . However, they also found that the images Of branch stores are weaker and that there is considerable similarity among suburban branches Of downtown stores. Summary Of Related Literature An aggregate department store's image is a summation of all the customers' images Of the store. The develOpment Of this image takes place through customers' attitudes toward different store dimensions. As no consensus exists as to which store dimensions are important in image formu- lation, this study utilizes the same dimensions used by Wyckham. These dimensions are: Locational Convenience, Merchandising Suitability, Store Congeniality, and Sales Personnel. The process of image develOpment takes place within a psychological and sociological environment and, in turn, is affected by this environment. As a part of this study is a replication Of the Wyckham study, emphasis is placed on the effects Of social and experiential factors on the department store image. In support of the above emphasis, 40 Kunkel and Berry state: "Since psychological (learning) principles do not vary over time, . . . differences in beha— vior must be due to the person's present social context and his experience."54 However, as an individual's group affil- iations, needs, and his experiences with the store are slow to change, his image of that store tends to remain stable over a period of time. This stability in the store's image is Of strategic value to the management of the store. It can provide the store with the marketing niche so necessary for the store's survival and growth in our competitive environment. FOOTNOTES-~CHAPTER II 1K. E. Boulding, The Image (Ann Arbor, Mich: The University Of Michigan Press, 1956), pp. 5—6. 2Wyckham, Op. cit., pp. 193-212. 3Edith Klemperer, "The Body Image Of the 'Other' in Hypnoanalysis," American Journal of Clinical Hypnosis, XI, 2 (1968), 63-68. 4Warren K. Schoonmaker, "The Image of Electronic Publications," Mainly Marketing (December, 1970), 8-9. 5Richard Wolleheim, "The Mind and the Mind's Image of Itself," International Journal Of Psychoanalysis, L, 2 (April, 1969), 209-220. 6Royal D. Colle, "The Negro Image and the Mass Media" (unpublished Ph.D. thesis, Cornell University, 1967). 7A. L. Delbecq and J. Vigen, "Prestige Rating Of Business and Other Occupations," Personnel Journal (February, 1970), 111-116. 8W. D. Rountree, "The Image Of Marketing as a Field Of Study Among Business Students" (unpublished Ph.D. thesis, The Louisiana State University and Agricultural and Mechan— ical College, 1970). 9Leonard Berry, "Banking, Marketing, and the Image Concept," Arizona Business Bulletin (November, 1968), 239-244. 10Robert G. Wyckham, William Lazer, and W. J. E. Crissy, eds., Images and Marketing: A Selected and Annotated Biblio- graphy (Chicago: American Marketing Association, 1971)} 11 Ibid., pp. v-vi. 12Peter J. McClure and John K. Ryans, "Differences Between Retailers' and Consumers' Perceptions," Journal Of Marketing Research, V (February, 1968), 35; RonaId R. Gist, Retailing: Concepts and Decisions (New York: John Wiley and Sons, Inc., 1968), pp. 30-31; Delbert J. Duncan, Charles F. Philips and Stanley C. Hollander, Modern Retailing Manage- ment: Basic Concepts and Practices (8th ed.; Homewood, 111.: Richard D. Irwin, Inc., 1972), pp. 40-41. l3McClure and Ryans, Op. cit. l4Ibid., p. 40. 41 42 15Wyckham, Op. cit., p. 61. 16John H. Kunkel and Leonard L. Berry, "A Behavioral Conception Of Retail Image," Journal Of Marketing, XXXII, 4 l7Wyckham, Op. cit., p. 46. l8Kunkel and Berry, Op. cit., p. 24. 19Supra, Chapter I, p. 14. 20Kunkel and Berry, Op. cit., pp. 25-27; Petrof, Op. cit., pp. 33-44; Berry, Op. cit., pp. 19-20; Ronald P. Stephenson, "Identifying Determinants of Retail Patronage," Journal of Marketing (July, 1969), 58-59; Perry and Norton, Op. cit., pp. 4-6; George Fisk, "A Conceptual Model for Studying Customer Image," Journal Of Retailing, XXXVII (Winter, 1961-62), 1—8; Lazer and Wyckham, Op. cit., p. 5. 21Kunkel and Berry, Op. cit., pp. 48-49. 2Pierre Martineau, Motivation in Advertising: Motives That Make People Buy_(New York: McGraw—Hill Book Company, Inc., 1957). 23Fisk, Op. cit. 24Charles J. Collazzo, Consumer Attitudes and Frustrations in ShOpping (New York: National Retail Mer- chants Association, 1963). 25 . Egan, op. Cit. 26Perry and Norton, Op. cit. 27Robert F. Kelly and Ronald P. Stephenson, "Semantic Differential: An Information Source for Designing Retail Patronage Appeals," Journal Of Marketing, XXXI 28Wyckham, Op. cit., p. 84. 29Ibid., p. 67. 3OIbid., p. 28. 31John B. Lansing and Leslie Kish, "Family Life Cycle as an Independent Variable," American Sociological Review (October, 1957), 512-519; William D. Wells and George Guber, "Life Cycle Concept in Marketing Research," Journal Of Marketing Research (November, 1966), 355-363. 43 32 . . . . Wyckham, Op. c1t., p. 30; Martin ZOber, Principles of Marketing (Boston: Allyn and Bacon, Inc., 1971)7'pp. 111- 112. 33Collazzo, Op. cit., p. 63, quoted in Wyckham, Op. cit., p. 67. 34Wyckham, Op. cit., pp. 178-179. 35Stuart U. Rich and Subhash Jain, "Social Class and Life Cycle as Predictors Of Shopping Behavior," Journal Of Marketing Research, V (February, 1968), 41—49. 36Ibid., pp. 44-45. 37Wyckham, Op. cit., p. 183. 381bid., p. 183. 39ZOber, Op. cit., pp. 143-144; Robert F. Hartley, Marketing: Management and Social Change (San Francisco: Intext Educational deIisHérs, 1971), pp. 244-245; H. Robert Dodge and Harry N. Summer, "Choosing Between Retail Stores," Journal of Retailing (Fall, 1969), 11—21; David J. Rachman and Marvin Levin, "Blue Collar Workers Shape Suburban Markets," Journal of Retailing (Winter, 1966-67), 5-13; James H. Myers and William H. Reynolds, Consumer Behavior and Marketing Management (Boston: Houghton Mifflin, 1967), p. 181. 0Burleigh B. Gardner, "Behavioral Sciences as Related to Image Building," New Direction in Marketing, Proceedings Of the American Marketing Association, ed. by F. S. Webster (Chicago: June, 1965), p. 147, quoted in Wyckham, Op. cit., p. 68. 41Rich and Jain, Op. cit., p. 41; Frederick C. Klein, "Rising Pay Lifts More Blue Collar Men Into a New Affluent Class," The Wall Street Journal, CLXV (April 5, 1965), 1; Philip George Kuehl, "An Examination Of the Influ- ence of Family Life Cycle and Social Class on Information Seeking in Self-Medication Behavior" (unpublished Ph.D. thesis, The Ohio State University, 1970); Robert F. Hartley, OE. Cite, ppa 346-2470 42Rich and Jain, Op. cit. 43Wyckham, Op. cit., p. 177. 44Henry Allen Bullock, "Consumer Motivation in Black and White--Part I," Harvard Business Review, XXXIX, 3 (May-June, 1961), 89-104; Hartley, Op. cit., p. 249. 44 4SIbid., p. 103. 46Ibid. 47 . Wyckham, Op. c1t., pp. 180-181. 48 H. J. Schaffer and F. E. Brown, "Perception and Misperception Of Marketing Phenomena," Marketing Involvement in Society and the Economy, Proceedings Of the American Marketing Association, ed. by Philip R. McDonald (Chicago: August, 1969), pp. 3-4. 49Leon Arons, "Does TV Viewing Influence Retail Store Image and ShOpping Frequency?" Journal of Retailing, XXXVIII (Fall, 1961), 11. 50Wyckham, Op. cit., pp. 182-183. 51Ibid. 52Ibid., pp. 69-70. 53Rich and Portis, Op. cit., p. 15. 54Kunkel and Berry, Op. cit., p. 23. CHAPTER III RESEARCH DESIGN AND PROCEDURE This chapter describes the research design and the sample responses for the study. The discussion is divided into four sections: sample design, a description of the interview schedules, techniques to be used in testing hypotheses, and the sample responses. Sample Design The sample was designed: (a) to test the effect of social and experiential factors on individuals' percep- tions of department stores and (b) to investigate the department store managements' understanding Of their cus- tomers. The sample Of customers Of different department stores was selected by the following procedure: 1. A list Of different groups existing within the Saginaw Standard MetrOpOlitan Statistical Area (S.M.S.A.) was develOped with the help Of Mr. Mickey Warren, Executive Vice-President of the Greater Saginaw Chamber of Commerce; Mrs. Esther Way, Society Editor Of Saginaw News; and Mr. Stuart Gross, Director Of Community Affairs Of Saginaw Valley College. The list comprised 300 groups. 45 4o 2. After the list of groups in the Saginaw S.M.S.A. was completed, a sample of groups was selected to provide the study with a cross-sectional representation Of Saginaw S.M.S.A. Mrs. Sheila Morley, the District Extension Con- sumer Marketing Information Agent Of Michigan State Univer- sity Cooperative Extension Services Center in Saginaw, provided assistance in establishing a list of 31 groups from the original listing of 300 groups in the Saginaw S.M.S.A. In the development of the list, special attention was given to Obtain groups representing different age cate- gories, races, and social classes to provide a cross—section Of Saginaw S.M.S.A. population. 3. The President and the Secretary of each group listed in Appendix B were approached and their permission was asked to administer the questionnaire to their reSpective group at their regular meeting. Only the groups marked with an asterisk against their name in Appendix B filled out the questionnaires. Reasons for other groups not filling out the questionnaire were: not meeting during Spring or Summer, unable to provide 20 minutes of the meeting time, the fear of being identified as a group patronizing one or the other department store, and not interested in participating in any surveys. The data from members of the management of the selected department stores were collected with the assis- tance of the store managers of individual stores. The questionnaire was administered to the management group 47 during a special meeting called by the store manager or dur- ing a routine management meeting of each store. This pro- cedure was not followed in the case of Federal Department Store's branch on the west side of Saginaw. At Federal the store manager distributed four questionnaires to his managers personally and a week later the responses were collected. The Interview Schedules To collect data for this study, separate interview schedules were developed for customers and management. Interview Schedule for Customers The interview schedule for customers is divided into two parts (see Appendix C): 1. Introduction of the interviewer and the instruc- tions to be announced by the interviewer. 2. The questionnaire to collect data on customers' evaluation of selected department stores in the Saginaw area. This questionnaire is comprised of four parts. The first part consists Of instructions and examples to help the subjects in filling out the questionnaire. The second part (questions 1 through 6) contains a number of semantic differ- ential scales designed to measure the subjects' attitudes. toward four actual department stores, one hypothetical "Ideal" department store, and the activity Of shopping. The third part (questions 7 through 13) consists of a set of questions to determine the subjects' attitudes toward actual 48 department stores on the basis of their experiences with the store. The fourth part (questions 14 through 29) is designed to Obtain demographic information about the subjects. The Semantic Differential Among many different ways of collecting image data in marketing, "probably the most pOpular method . . . is the time-honored rating scale."1 However, among different available scaling techniques, one widely used in marketing is the semantic differential technique developed by Osgood, Suci, and Tannenbaum.2 The technique of semantic differential was devised with the idea of measuring connotative meaning Of various concepts in a multidimensional semantic space. The semantic differential technique, as used in image and attitudinal research, essentially consists of a concept or a set of con— cepts to be rated on a series of bipolar adjective scales with cues separating the polarized adjectives.3 These cues separating the bipolar adjectives may be numerical, Verbal, graphic, or some combination. The bipolar nature of the adjectives used provides the measurement of both the direc- tion and intensity of attitude toward a concept being mea- sured. In the develOpment Of the instrument using semantic differential technique, the following were given special attention: (a) the selection and presentation of the con- cepts and bipolar adjectives related to these concepts; 49 (b) the type Of cues separating the bipolar adjectives; (c) the number of cues separating the bipolar adjectives; (d) the problem of contextual contamination; and (e) the stability and meaning of an ideal rating. The Selection and Presentation of the Concepts ang Scales.--One purpose of this study is to replicate the Wyckham study cited previously. Hence, the concepts used to measure department stores' images are the same as used by Wyckham. These concepts (or store dimensions) are: merchan— dising suitability, sales personnel, congeniality, and loca- tional convenience. The scales are also the same as used by Wyckham with the exception Of the form used in presenting the scales (see Appendix C). Osgood, et al. have suggested two basic forms for presenting semantic differential scales. They are: Form I and Form II.4 These forms differ chiefly in the way in which concepts are paired with scales. Form I presents the differ- ent scales as follows: dishonest Voice rough : ~ unfair,etc. Teacher fair : Each scale carries a capitalized concept against which it is judged by the subject. All the scales are ordered in such a way that the maximum number of scales occur between the repetition Of each concept. Form II presents the concepts as follows: 50 Teacher honest ° dishonest fair unfair, etc. In this form, all the scales for each concept are presented under the heading of that concept. Wyckham, following Mindak's modification,5 used descriptive Opposite phrases (instead of adjectives as used by Osgood, et a1.) selected from the studies reported in the literature on department stores. Due to the self-explanatory nature of phrases, he did not use capitalized concepts together with each scale as shown in Form I above. However, to make phrases self-explanatory he used complete long sen- tences. The lengthy phrases used for each semantic scale forced him to use very small print so that he could include all 30 scales on one page for each department store. The small size of the print used for the semantic differential scales was found by Wyckham to be one of the most signifi- cant factors in number of refusals and termination of inter- views before completion.6 To eliminate this problem, this study utilized Form II with modification of Wyckham's scales. This modification permitted the printing of all 30 scales in elite type on one page for each department store. No complaints were received regarding the size of the print used for the scales. The Type of Cues.--Osgood, et al. in their original research on the development of the technique of semantic 51 differential used the following graphic form:7 fair : ° unfair Subjects were asked to mark an x on the space that repre— sented their attitude. Numerical and verbal cues are also frequently combined to clarify the intent of the scale for the respondent. For example: Very Quite Slightly Neutral Slightly Quite Very Quality high ® ® ® ® (5) @ C7) low Though numerical values apparently suggest inter- val scaling, Coombs points out that this may not be true.8 Crissy suggests: "Generally speaking, ordinal scaling yields more reliable results than interval scaling."9 Following Crissy's suggestion, this study uses only verbal cues that separate the bipolar phrases and obtains ordinal scales. The Number of Cues.--A puzzling question attached to rating scale devices is what number Of cues should be used? There are two schools of thought on this issue.10 One school advocates that a fine set of response categories or a large number of cues, e.g. 11 or more, should be used to obtain reliable results. The other school of thought holds that two or three cues are adequate. However, Osgood, et a1. advocate neither too fine nor too coarse scales, i.e., 11 neither many nor a few cues. Their review of the 52 literature shows: Over a large number Of different subjects in many dif- ferent experiments it has been found that with seven alternatives all of them tend to be used and wigh roughly, if not exactly, in equal frequencies. Similarly, Green and Rao, from their experiment on relation- ship between number Of cues and information recovery, conclude: What usually happens as the response categories are coarsened is that the configuration Obtained from the scaling procedures tends to look more and more like a circle. This type of "degeneracy" reflects the fact that . . . 3-point or 2-point scales result in poor recovery of the original (synthetic) configuration. Moreover, recovery is NOT enhanced appreciably by merely including more rating scales; diminishing : returns set in rather rapidly beyond the level of eight vectors (simulated scales). . . . This latter finding is heartening for the advocates of 7-point rating scales.13 On the basis of the above evidence, this study uses seven cues for each scale. In quantifying the results, numerical scores were assigned to the responses. The scores are assigned to the cues for each scale, from the score of one for the most desirable descriptive phrase (from the point of view of department store management) to the score of seven for the least desirable descriptive phrase. The Problem of Contextual Contamination.4-Whenever a subject responds to more than one question, the problem of contextual contamination, i.e., the problem of latter responses being biased by the previous responses arises. Thus, long questionnaires, like the ones used in this research, are very susceptible to the order bias. 53 The order bias may occur in this research in the following ways: 1. The responses made to the scales of the first concept or dimension of any store may affect the responses made to the scales Of the other dimensions of that store. This is known as anchoring effect.14 In addition, there may be strong motivation for the respondent to make his responses internally consis- tent even if this results in some inaccurate ratings; this strain toward gonsistency has been labeled "Socratic effect."l Osgood, et al. report a study to test the independence of concepts. They report: "The results of this study showed no significant differences in the scalar locations of the test concepts as a function of the context in which they are imbedded."l6 Similar results regarding robustness of semantic differential technique are reported by Clevenger, Lazier, and Clark in their study on "Measurement of Cor- 17 Also, the porate Images by the Semantic Differential." store dimensions selected to study the images of each store in this research are adapted from Fisk's findings. These findings were based on his factor analysis of different responses. As each store dimension represents a factor consisting of a group of scales which are statistically independent of all the others, the order effect in the eval- uation of store dimensions for each store should be at a minimum. 2. The anchoring effect may also take place in this study through the ordering of stores in each 54 questionnaire. To reduce this kind Of order bias, the pages of the questionnaires were arranged in the manner that each store, including the "Ideal" store, appeared first, second, third, fourth, and fifth in an equal number Of question— naires distributed. The Stability and Meaning Of an Ideal Rating.-—A hypothetical "Ideal? department store is used in the study as a control concept18 and as a comparison point with the test stores to Obtain a measure of image deficiency. The stability and meaning of such ideal rating has been questioned and tested by Landon.19 The results Of his study show that the "Ideal" does remain stable over time and over subjects. Selection Of Department Stores The Wyckham study selected the J. L. Hudson Company, Sears Roebuck and Company, and Federal Department Stores as the test stores. These stores were selected and judged to represent high (Hudsons), medium (Sears), and low (Federal) scales on a socio-economic continuum.20 To keep the compar- ability, this study involved Sears and Federal to represent medium and low scale on a socio-economic continuum. As there is no Hudsons store in Saginaw, Wiechmann's store was selected to represent the high end of the socio-economic continuum. The decision to consider Wiechmann's as a rep- resentative of the high end Of socio-economic continuum was based on the researcher's consultation with his colleagues at Saginaw Valley College. Only those colleagues were 55 consulted who have lived in Saginaw for at least five years. K-Mart was also included in this study to examine the images of a discount department store as compared to the images of regular department stores. The Activity Of ShOpping Question 6 consists of five bipolar adjectives selected to measure customers' attitudes toward the activity of shOpping. These scales adopted from the Wyckham study are included in the questionnaire to provide the information on the images held by customers who enjoy shopping and those who do not enjoy shopping. Shopping Behavior Questions Questions 7 through 13 are designed to Obtain infor— mation regarding customers' shopping behavior in the selected department stores included in the questionnaire. Question 7 seeks information regarding the respon— dents who have shopped and those who have never shopped at the test stores, and is used to compare the images Of each department store held by those who have actually shOpped at the stores as compared to those who have never shOpped at that store. In Question 8, the subject is requested to list three Of his favorite department stores. This information is Obtained in order to compare the images of each test store held by those customers who have declared that store 56 as their favorite as compared to the images held by those customers who have declared other stores as their favorite. Question 9 is devised to secure information regard- ing recency Of shopping experience of the customers at the test stores, and is used to compare the images Of each department store held by customers who have shopped recently and those who have not shOpped recently. Question 10 is concerned with frequency of shOpping by customers at the test stores. This information is obtained in order to compare the images Of each department store held by frequent, occasional, and rare customers.21 Question 11 provides information to allow comparison of the images Of each test store held by customers buying hard-line vs. soft-line goods from the test stores. Question 12 explores the social support for cus— tomers' vieWpOints regarding the test stores and provides information to compare the images of each department store held by customers who have social support and those who do not. Question 13 Obtains information regarding the respondents' frequency of reading advertisements by the test stores, and is used to compare the images Of each department store held by frequent readers, occasional readers, and non- readers of advertisements of that store. 57 Demographic Data The major purpose of the demographic data obtained through questions 14 through 29 is to classify the respon- dents into groups as to sex, race, social class, and family life cycle stages. §e§,--Question 14 allows comparison Of the images of the test stores held by men and women. Rage.--Question 15 Obtains data for comparison of the images of the test stores held by whites and non-whites. Family Life Cycle.--Questions 16, 17, 18, and 19 are devised to provide information on respondent's marital status, number Of children, age, and sex of the head of the household. These questions provide information to classify respondents into various stages of family life cycle. The family life cycle stages are adapted from Lansing and Morgan.22 These stages are: a . b 1. Young, Single 2. Young, married, no children 3. Young, married, with childrenc 4. o1der,d married, with children 5. Older, married, no children 6. Older, single (a) By young is meant, head of spending unit (male or female) under 40; (b) by single is meant, head Of spending unit is not married, widowed, divorced, or separated; (c) by children is meant, children living at home or away at school and supported by the head of spending unit, and (d) by older is meant, head of 23 Spending unit (male or female) is 40 years or older. Social Class.--Questions 19 through 29 provide data to classify subjects into social classes. Warner's Index of Social Characteristics24 is used to classify subjects into only four social classes instead of six social classes 58 suggested by Warner. This is done for the sake of conven- ience in analyzing the data and comparability with the Wyckham study. The four.socia1 classes used in the study are upper (including Warner's upper upper and lower upper), upper middle, lower middle, and lower (including Warner's upper lower and lower lower). Questions 20 and 22 provide information to classify respondents according to the occupation and the source of income scale used by Warner.25 Questions 24, 25, and 26 are designed to classify respondents on a revised house type scale develOped by Wyckham.26 The revised scale overcomes the problem of lack of objectivity of interviewer classify— ing the respondent's home by his subjective judgment and instead, uses the market value Of the house (as seen by the respondent) if owned, or the rent paid for the house or apartment if rented. The revised house type scale as devised by Wyckham is shown in Table 8. Table 8.--Revised house type scale. Rating Market Value Rent Paid Monthly 1 more than $50,000 more than $500 2 $35,000 - 49,999 $250 - 499 3 $20,000 - 34,999 $150 - 249 4 $15,000 - 19,999 $100 - 149 5 $12,500 - 14,999 $ 75 - 99 6 $10,000 - 12,499 $ 50 - 74 7 less than $10,000 less than $50 Source: Robert G. Wyckham, "Aggregate Department Store Images: Social and Experiential Factors" (unpub- lished Ph.D. thesis, Michigan State University, 1967). s~" 59 Questions 28 and 29 Obtain data to classify each respon- dent's dwelling area according to Warner's scales shown in Appendix E. Instead Of using interviewer's subjective judg- ment, the respondent's Opinion is Obtained regarding his dwelling area. Question 29 asks the respondents to provide the addresses on an Optional baSis. If the addresses are provided, these addresses are used to verify information provided in Question 28 through personal Observation and with the help of Mrs. Morley, District Extension Consumer Marketing Information Agent Of Michigan State Cooperative Extension Center Of Saginaw. Interview Schedule for Management The interview schedule for the management is divided into two parts (see Appendix D): 1. Introduction of the interviewer and the instruc- tions to be announced by the interviewer. 2. The questionnaire to collect the data on manage- ment's perception Of customers' attitudes toward their department store. This questionnaire consists of instruc- tions and examples to help the subjects in filling out the questionnaire, number of semantic differential scales designed to measure the subjects' vieWpoints about the store's image as held by the store's customers, and a set of questions to Obtain demographic information about the subjects and their experience with the store. 60 Techniques Used in Testing Hypotheses As noted earlier, to quantify the results the seven cues of each scale are numbered from one to seven; one representing the most desirable judgment on the scale (from the management's VieWpOint) and seven representing the most undesirable judgment. The 30 bipolar semantic scales used to evaluate each department store are separated into four store image dimensions. These four.dimensions are: merchandising suitability, sales personnel, store congeniality, and loca— tional convenience. An overall dimension is also used in this analysis as a summary measure based on all the 30 scales used in the questionnaire. Each respondent's score for the scales comprising each store dimension is totaled and the mean is obtained. This information provides data to test the first three of the four hypotheses being investigated. The first hypothesis compares aggregate store images of the test stores held by the members of various groups. For example, the aggregate image of Sears is compared with the aggregate image of K-Mart or the "Ideal" store for the members of upper class. As this kind of comparison provides matched pairs (i.e., images of two department stores held by the same group of peOple), the Wilcoxon Matched-Pairs Signed Ranks Test is used for the comparisons.27 The Wilcoxon test is a non—parametric test used to determine the probability that two matched samples have come from identical pOpulations. The Wilcoxon test results in 61 T scores for samples less than 25. If the sample size is greater than 25, T scores are converted into Z scores. The tied pairs are eliminated from the computation of T scores. For these T scores, corresponding probabilities are Obtained from the tables to test the null hypothesis that these two matched samples come from identical pOpulations. The null hypothesis in this study is rejected at .05 level of sig- nificance in each comparison. The Computer Institute of Social Science Research at Michigan State University has a program on tape for Wilcoxon test.28 This program not only provides signifi- cance probabilities associated with Wilcoxin T scores obtained from the comparison of data from matched samples, but also indicates the higher and the lower groups to pro- vide directional information. Due to the quantification procedure used in this study, the higher scores represent unfavorable rating for the store and the lower scores represent favorable rating for the store. Thus, the lower group is the one which is favorably rated in every comparison. All significant comparisons are marked With directional notation to indicate the lower group. If the comparison is not significant at .05 level, only NS (not significant) is shown instead of exact probabilities. If the compari- sons are significant, either .05 or .01 probabilities are assigned to them. The .01 level of significance is shown to emphasize higher significance of the differences between the samples. 62 The second hypothesis compares aggregate images of individual department stores held by the members of various groups. For example, the aggregate images of Sears held by the members Of upper and lower class are compared. As respondents classified into upper and lower class create two independent groups, the store images held by these or any other two separate groups of people are compared using the Mann-Whitney U Test.29 The Mann-Whitney test is a non-parametric test used to determine the probability that two independent samples have come from identical pOpulations. The Mann-Whitney test results in U scores from samples less than 20. If the sam— ple size is greater than 20, U scores are converted into Z scores. For these U scores, corresponding probabilities are Obtained from the tables to test the null hypothesis that these two independent samples come from identical pOpulations. The null hypothesis is again rejected at .05 level of significance. The Computer Institute of Social Science Research at Michigan State University has a program on tape for the Mann- Whitney test.30 The program provides probabilities asso- ciated with U scores Obtained from the comparisons of data from paired independent samples and the higher and the lower group for each comparison. Due to the quantification pro- cedure used in this study, the higher score represents unfavorable rating for the store. Hence, the directional information is provided for each significant comparison in 63 all the tables in this study. The probabilities for U scores are presented in the same manner as probabilities for Wilcoxon T scores. The third hypothesis tests the communality Of manage- ment's perception of customers' images and the actual cus- tomers' images to prOvide an insight into management's understanding Of customers. As management and customers consist Of two independent groups of people, the Mann-Whitney U test is again utilized to test this hypothesis. The fourth hypothesis is focused on the relationship between customers' satisfaction (or dissatisfaction) with a department store and marketing orientation (or lack of it) of management Of that store. For the purpose of this study, relationship between customers' dissatisfaction and lack of marketing orientation of management for a department store is examined. Customers' dissatisfaction with a store is measured by the Aggregate Image Deficiency (AID) score. The AID scores are Obtained by first subtracting each cus- tomer's test store rating of each scale from the "ideal" rating on the same scale. All Of these subtractions for each scale on each dimension are summed for all customers and means are Obtained to provide a measure Of aggregate image deficiency on each image dimension for all the cus- tomers for each department store. The AID scores, inter- preted disregarding the signs, show that the higher the score, the higher the dissatisfaction among customers On that image dimension of a department store. The lack Of 64 marketing orientation on the part of management of a store is measured by Inverse Marketing Orientation Index (IMOI). IMOI is measured by the difference between the actual and the anticipated aggregate store images. The larger the difference, the lower the marketing orientation of the management of a department store. Thus, low (or high) scores on AID and IMOI show a positive relationship between the two indexes or between customers' dissatisfaction and the lack Of marketing orientation on the part Of management for a department store. The relationship between the scores of these two indexes for each store dimension is tested with the help of only descriptive statistics and graphical analy- sis. The use Of derived statistics is considered not war- ranted due to the nature of the data. Sample Responses The subjects were selected through group interviews and not with the help of random selection procedure. The lack of randOmness in the Selection Of the sample restricted the use of high power statistical tools in the analysis of the data. The group interviews resulted in response to 208 questionnaires. Of these 208 questionnaires, those question- naires considered usable satisfied the following two cri- teria: 1. All the scales should be filled in for the "Ideal" department store and at least one test department store. 65 2. All the demographic information necessary for classifying respondents into various social classes and family life cycles should be provided. Also, all the other necessary demographic characteristics tO classify customers into different groups should be provided. On the basis of the above criteria, 178 question- naires were considered usable for customers. The management inverviews Of four test department stores resulted in 72 questionnaires. Of these question- ‘ naires, only seven were considered unusable. Of these seven unusable questionnaires, five were from K-Mart (East) and two were from Federal (East). For management, a question- naire was considered usable if all the scales for their store were properly marked. As the study does not necessitate the use Of demographic information obtained from the management, no questionnaire filled out by the management was rejected due to their errors in, or lack of, providing information on demographic questions. The following represents the distri- bution of questionnaires from the management of different department stores: 14 from Wiechmann; 5 from Sears; 22 from KeMart (East) and 10 from K-Mart (West); 10 from Federal (East) and 4 from Federal (West). Selected Demographic Characteristics of the Sample of Customers This section discusses the following demographic characteristics of the sample Of customers: social class, family life cycle, race, sex, age, and education. 66 Warner's Index of Social Characteristics (see Appendix E) was used to classify the respondents into social classes. The classification is shown in Table 9. Table 9.--Classification Of respondents into social classes. Social Class Number Percentage Upper 23 13 Upper Middle 44 25 Lower Middle 75 42 Lower _32_ 20 Total 178 It was hoped that there would be at least 20 respon- dents in each social class category so that Wilcoxon and Mann-Whitney tests would not be adversely affected by small cell entries.31 As shown in Table 9, the goal of at least 20 respondents for each class was achieved. The two extreme classes, i.e., upper and lower classes, are represented in the sample by approximately 33 per cent, whereas the middle classes are represented by approximately 67 per cent of the total respondents. The sample selection procedure and dif— ficulties encountered in finding groups representing high income and social status willing to cooperate during the summer may also have contributed to the high prOportion of middle class in the sample. Table 10 shows the classification of respondents into various family life cycle stages. Here, too, there were 67 at least 20 respondents found in each cell; hence the derived statistics for both the Wilcoxon and Mann-Whitney tests were unimpaired. Table 10.--Classification of respondents into family life cycle stages: this study and the Wyckham study. Number Percentages Family Life Cycle This Wyckham This Wyckham Stages Study Study Study Study 1. Young, Single & Young, Married, NO Children 46 36 26 5.4 2. Young, Married, Children 54 214 30 32.8 3. Older, Married, Children 22 210 12 32.2 4. Older, Married, No Children 28 127 16 19.4 5. Older, Single 28 66 16 9.9 Total 178 653 aRobert G. Wyckham, "Aggregate Department Store Images: Social and Experiential Factors" (unpublished Ph.D. thesis, Michigan State University, 1967). As suggested earlier, special attention was given to Obtain groups for interviews to provide the study with a prOper representation of races. Table 11 shows that this objective is partially achieved. As compared tO Saginaw S.M.S.A., this research contains 4 per cent more non-whites. However, a comparison with race proportions in Saginaw 68 .ma .a .laama .OOHmmO mcflucflam unsecuo>ow .m .D «.0 .Q .coumcfinmmzv cmmfl£OHz umOflumflHODosumno sofiumammom Homecou .mdmcmo mo smousm .m .Dm Nmm.nva mwh.mam mha Houoa ma ma ha hm¢.nm mmo.mm om suacslcoz Hm mm mm mmo.oma Hmw.HmH mva cues: «and .<.m.s.m spasm memes m.«.m.2.m spasm poem OONHGOQHD 3mcflmmm mane UoNHcmoHD sesflmmw mane 3mcflmmm 3mcwmmm momsucooumm Hmnfidz , .Oomn mo mucmpcommou mo cOHDMOHmammMHUII.HH OHQMB 69 Urbanized Area in Table 11 shows that races are represented prOportionately in this study. As shown in Table 12, 81 per cent Of the respondents were female and only 19 per cent were male. NO specific efforts were made to Obtain respondents prOportionate to sex in the Saginaw S.M.S.A. area. On the contrary, due to the sample selection procedure, it was expected that females would be represented by a high proportion in the sample, just as they are among shOppers generally. Table 12.--Classification Of respondents by sex. Sex Number Percentages Male 35 20 Female 143 Total 178 Table 13 shows that proportions Of respondents in different age categories in this research and the Wyckham study are quite similar except in the second (20 but less than 30 years) and fourth (40 but less than 50 years) age categories. Table 14 shows the classification Of the heads of the households of the sample responses on the basis of their academic background. The data indicate that the sample has a higher education level than might be expected in the general pOpulation. This may be because Of sample selection through group interviews. 70 Table l3.--C1assification of respondents by age categories. Number Percentages This Wyckham This Wyckham Age Study Studya Study Study Less than 20 years, 7 9 4 1.4 20 but less than 30 years 50 88 28 13.4 30 but less than 40 years 43 156 24 23.8 40 but less than 50 years 30 208 17 32.1 50 but less than 65 years 36 131 20 20.3 65 years and more 12 57 7 8.7 Total 178 649 aRobert G. Wyckham, "Aggregate Department Store Images: Social and Experiential Factors" (unpublished Ph.D. thesis, Michigan State University, 1967). Table 14.--C1assification of the heads of the households by education. Education Number Percentages Grade School 5 Some High School 15 Graduated from High School 64 36 Some College 52 29 Graduated from College 22 12 Advanced Degree 20 11 Total 178 FOOTNOTES--CHAPTER III 1Paul E. Green and Vithala R. Rao, "Rating Scales and Information Recovery--How Many Scales and Response Categories to Use?" Journal of Marketing, XXXIV (July, 1970), 33. 2C. E. Osgood, P. H. Tannenbaum, and G. J. Suci, The Measurement of Meaning (Urbana: University Of Illinois Press, 1957). 3The term cues is used by G. David Hughes, Attitude Measurement for Marketing Strategies (Glenview, Illinois: Scott, Foresman and Company, 1971), pp. 91-94. 4Osgood, et al., Op. cit., pp. 81-82. 5William A. Mindak, "Fitting the Semantic Differen- tial to the Marketing Problems," Journal of Marketing 6Wyckham, Op. cit., p. 97. 7Osgood, et al., Op. cit., p. 83. 8Quoted in Edger Gunther, "Evaluating Corporate Image Measurements: A Review Of Techniques," Journal of Advertising Research ( ), p. 65. 9William J. E. Crissy, "Image: What Is It?" MSU Business TOpics (Winter, 1971), 80. - 10 Green and Rao, Op. cit., p. 33. 11Osgood, et al., op. cit., p. 85; Hughes, Op. cit., p. 94. 1ZOngOd, et al., Op. cit., p. 85. 13 Green and Rao, Op. cit., p. 38. l4Osgood, et al., Op. cit., p. 84. 15E. Laird Landon, Jr., "Order Bias, the Ideal Rating, and the Semantic Differential," Journal of Marketing Research, VIII (August, 1971), 375. 16Osgood, et al., Op. cit., p. 84. 71 72 l7Theodore Clevenger, Jr., Gilbert A. Lazier, and Margaret Leitner Clark," Measurement of Corporate Images by the Semantic Differential," Journal of Marketing Research 18Mindak, Op. cit., p. 30. 19Clevenger, Lazier, and Clark, Op. cit., p. 81. 2OWyckham, op. cit., pp. 84-85. 21Arons, Op. cit., p. 8. 22Lansing and Morgan, Op. cit. 23Wyckham, Op. cit., p. 88. 24 W. Lloyd Warner, Marcia Meeker, and Kenneth Eells, Social Class in America: A Manual of Procedure for the Measurement of Social Status (New York: ‘Harper & Row, 1 1960), Chapters 8-11. See Appendix E for a brief explana- ‘ tion of Warner's Index of Social Characteristics. 25 Ibid. 26Wyckham, Op. cit., p. 94. 27Sidney Siegel, Non Parametric Statistics (New York: McGraw-Hill Book Company, 1956), pp. 75-83. 28John Morris, "Mann-Whitney and Wilcoxon Tests, Randomization, and Sign Tests, Parametric T Test and F Test," Unpublished Technical Report 45 (East Lansing, Michigan: Computer Institute of Social Science Research, Michigan State University, September 15, 1967). 29Siegel, Op. cit., pp. 116—127. 30 . . Morris, Op. Cit. 31Siegel, Op. cit., p. 126. CHAPTER IV DATA ANALYSIS AND RESULTS This chapter is comprised of four sections, each containing the data associated with testing one Of the four main hypotheses as set forth previously. Hypothesis I Hypothesis I compares aggregate store images Of the test stores and an "Ideal" store held by the members of various groups. It consists of between stores comparisons of aggregate store images held by customers in different groups. This hypothesis is divided into five sub-hypotheses. Each one Of them is tested separately in this section. 1.1 Different department stores carry differen- tiable aggregate images among all of their customers. Table 15 shows probabilities associated with the null hypothesis that there are no differences between aggre- gate images Of different department stores held by all customers. All the test stores and the "ideal" store are compared on five store dimensions: merchandising suitabil- ity, sales personnel, store congeniality, locational con— venience, and overall. Of the 50 between stores comparisons, 48 differences are significant at the .05 level. The 73 74 .mmAZO .uonmas Oops“ mm3 ououm pcooomo .mflmosuomwn Hana osuu mo muflaflomnoumb .muosoumso ecu Sn momma: pouch mp3 GOmwquEOo some ca ououm umaflw .Oouoc OHO£3 umooxmm Odo.v oao.v omz OHo.v omz Houston I DumEIM Ho.v OHo.v Ho.v Ho.v Ho.v Houopom I musom Ho.v Odo.v ac.v Ho.v Ho.v unmzlx I mnmom Ho.v oHo.v Ho.v Ho.v Ho.v Houopom I cemenomfis Ho.v OHo.v Ho.v Ho.v Ho.v mumom I essences: Ho.v OHo.v Ho.v Ho.v Ho.v Dumznm I casesomwz Ho.v Ho.v Ho.v Ho.v Ho.v Houmpom I :HtOOH: Ho.v Ho.v Ho.v Ho.v Ho.v Dumzlx I rammOH: Ho.v Ho.v Ho.v Ho.v Ho.v mumom I =HMOOH= Ho.v Ho.v Ho.v Ho.v oao.v comesooflz I =HMOOH= + smououm Hamum>o OOGOHcO>GOU SDMHMHcomcoo Amadomuom muaaflpmuwsm t ucofiuusmma HchflDmooq muoum moamm mcfimfipcmnoumz mcoflmcmfiflo mmmEH OmHofioumso Had .mHOEoumso Ham moose onoum =Hmopa= so was monoum usefiuummop and» mo mommfiw ousmmumms mo mnemflummsoo ms» Eoum pocflsuno moHOOm B GOMOOHHS suds pousaoomms mmwuwaanmooumII.mH OHQMB 75 results indicate that different department stores carry differ- ential aggregate images among all the customers. A comparison Of all the stores on the overall dimension in Table 15 shows that the stores are rated in the following order: "Ideal," Wiechmann, Sears, Federal, and K-Mart. These data support the assumption under which the stores were selected for this study that the three regu- lar department stores can be ranked on a continuum to repre- sent Wiechmann as high, Sears as medium, and Federal as low.1 K-Mart was included in the study "to examine the images Of a discount department store."2 The results in Table 15 show that KeMart, a discount department store, is ranked lower than all the three regular department stores selected for this study. The rating in the order of "Ideal," Wiechmann, Sears, K-Mart, and Federal is maintained. NO differences were found between K—Mart and Federal on merchandising suitability and store congeniality. These results show that differences between the emerging discount department stores and some of the lower ranked regular department stores may be getting blurred on their merchandising mix and store congeniality, but the regular department stores are still able to maintain an image advantage on the basis Of the quality of services and of their sales personnel. The results of the comparisons on locational con- venience indicate that though the differences between the stores are significant, the rating Of the stores changes 76 from "Ideal," Wiechmann, Sears, Federal, and K—Mart to the order Of "Ideal," Federal, K-Mart, Sears, and Wiechmann. The change in the order Of rating may be due to the differ— ences between the shopping center department stores (Federal- and K-Mart) and the downtown department stores (Sears and Wiechmann). The shOpping center stores are rated higher than the downtown department stores on this dimension. This may be due to better parking facilities and easy access to other stores for customers without being subjected to seasonal hazards in the shOpping centers. 1.2 Different department stores have differentiable aggregate images among their customers who are members Of different social classes. Table 16 presents the probabilities associated with the null hypothesis that there are no differences-between the aggregate images Of different department stores held by customers in different social classes. Of 200 comparisons, 83 per cent of the comparisons are significant at the .05 level. These results indicate that customers in different social classes do vary in their perceptions of department stores. All the comparisons between the "Ideal" and the four stores are found significant on all five image dimen- sions used in this study. A look at comparisons Of different department stores Ion the overall dimension among customers in different social classes reveals that different social classes do not rate Table 16.--Probabilities associated with Wilcoxon T score obtained from the '77 comparisons of aggregate images of different department stores among customers in different social classes. Image Dimensions Merchandising Sales Store Locational Departmegt + Suitability Personnel Congeniality Convenience Overall Stores + . d UPPER CLASS "Ideal" - Wiechmann <.01b <.01 <.01 <.01 <.01 "Ideal" — Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K—Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 NS <.05 NSc <.01 Wiechmann - K-Mart <.01 <.01 <.01 <.01c <.01 Wiechmann - Federal <.01c <.01 <.01 <.01c <.01 Sears - K-Mart NS <.01 NS <.01C NS Sears - Federal NSC <.05 <.05c <.01C NSc K-Mart - Federal NS <.05 NS <.05 NS UPPER MIDDLEd "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 “Ideal" - Federal <.01 <.01 <.01 <.01C <.01 Wiechmann - Sears <.01 NS <.05 NSC <.05 Wiechmann - K-Mart <.01 <.01 <.01 <.01C <.05 Wiechmann - Federal <.01 <.01 <.01 <.01C <.05 Sears - K-Mart <.01 <.01 NS <.01C <.05c Sears - Federal NSc <.05C <.05C <.01C NSc K-Mart - Federal <.05 <.01 NS <.01 <.05 LOWER MIDDLEd "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 <.05 NS NSc <.05 Wiechmann - K-Mart <.01 <.01 <.01 <.01c <.01 Wiechmann - Federal <.01 <.01 <.01 <.01c <.01 Sears - K—Mart <.01 <.01 <.01 <.05C <.01 Sears - Federal <.01C <.01C <.01c <.01c <.01c K-Mart - Federal NS <.01 NS <.05 NS LOWERd "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears_ <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.05 NS NS <.01c NS Wiechmann - K-Mart <.01 <.01 NS <.01c NSC Wiechmann - Federal <.01 <.05 NS <.01C NS Sears - K-Mart <.01 <.01 NS <.01c NSc Sears - Federal <.05c NSc NS <.01C NSc K-Mart - Federal NS <.01 NS NS NS aExcept where noted, the customers. first store in each comparison was rated higher by bProbability of true null hypothesis. cSecond store was rated higher. dN>25. 78 all the stores in the same order. Upper class members rate Wiechmann the highest and find no significant differences among Sears, K-Mart, and Federal: upper middle class mem- bers rate Wiechmann as the highest, Sears and Federal as the combined middle, and K—Mart as the lowest; lower middle class members rate Wiechmann and Sears as the top two and find nO significant differences between K-Mart and Federal; and finally, lower class members find no significant differ- ences among any of the test stores. The lack of differen- tiation in the perception held by the upper class members among the test stores, with the exception of Wiechmann, may be due to a lack Of interest in low mark-up and high turnover kinds of stores by this group; the lack of differ- entiation among the different stores among the lower class members may be due to their lack of experience and their common fear Of bigness and formality characterizing all the department stores.3 The highest differentiation among dif- ferent department stores is seen by customers in the middle class. This may be due to the very presence of the factors which contributed to the lack Of differentiation among test stores by customers in upper and lower class. All the comparisons on all the other store dimensions lead to similar results as discussed in Hypothesis 1.1 for all social classes with the exception Of two special cases: (1) In contrast to the lack Of perceived difference on the overall image score among test stores by the customers in the lower class; they do perceive differences among test 79 stores on merchandising suitability, sales personnel, and locational convenience. (2) In all but one comparison on locational convenience, members of all social classes con- sistently rated the test stores in the order Of Federal, K-Mart, Sears, and Wiechmann. This result is similar to the one discussed before in terms of all customers, and was eXplained earlier.4 1.3 Different department stores carry differen- tiable aggregate images among their customers who are in different family life cycle stages. Table 17 indicates the probabilities associated with the null hypothesis that there are no differences between aggregate images of different department stores among customers in different life cycle stages. Of all the comparisons, 83 per cent are significant at the .05 level. This result indicates that, overall, customers at different family life cycle stages do perceive differences among dif- ferent department stores. All comparisons between "Ideal" and test department stores are found to be significant at the .05 level in each family life cycle stage on all the image dimensions. A lOOk at the comparisons of stores on overall dimension among customers in various life cycle stages reveals that different stages of family life cycle do not rate all the stores in the same manner. The customers in stage one Of the family life cycle rate Wiechmann as high, Sears and Federal in the middle as equals, and K-Mart as low. However, the customers in stage two and three rate Table l7.-—Probabilities associated with Wilcoxon T scores Obtained from the 8C) comparisons of aggregate images of different department stores among customers in different family life cycle stages. Image Dimensions Merchandising Sales Store Locational Department; * Suitability Personnel Congeniality Convenience Overall Storesa STAGE 1d (Young, Single and Young, Married--NO Children) "Ideal" — Wiechmann <.01b <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 <.05 <.01 NSc <.01 Wiechmann - K-Mart <.01 <.01 <.01 <.01c <.01 Wiechmann - Federal <.01 <.01 <.01 <.01C <.01 Sears - K-Mart <.01 <.01 <.05 <.01 <.01 Sears - Federal <.01C <.01c NSc <.01: NSC K-Mart - Federal NS <.01 <.05 <.01 <.05 STAGE 2 (Young, Married--Children) "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01C <.01 Wiechmann - Sears <.01 NS NS <.05c NS Wiechmann - K-Mart <.01 <.01 <.01 <.01C <.05 Wiechmann - Federal <.01 <.01 NS <.01C <.05 Sears - K-Mart <.01 <.01 <.01 <.01c <.01 Sears - Federal <.01 <.01C <.05c <.01C NSC K-Mart - Federal NS <.01 <.01 <.01 <.05 STAGE 3d (Old, Married--Children) "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 NS NS <.05c NS Wiechmann - K-Mart <.01 <.01 <.01 <.05c <.01 Wiechmann - Federal <.01 <.05 <.01 <.01c NS Sears - K-Mart <.01 <.01 <.01 NS <.01 Sears - Federal NS NS <.01 <.05 NSc K-Mart - Federal <.01C <.01c NSC <.01C <.01 Table l7.-—Continued. 8]. Image Dimensions Merchandising Sales Store Locational Department + Suitability Personnel Congeniality Convenience Overall Storesa + . STAGE 4d (Old, Married--No Children) "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal“ - K—Mart <.01 <.01 <.01 <.01 <.01 “Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 <.05 <.01 NSc <.05 Wiechmann - K-Mart <.01 <.01 NS <.01C <.05 Wiechmann - Federal <.01 <.01 <.01 <.01C NS Sears - K-Mart NS <.01 NS <.01C NS Sears - Federal NS <.01 <.05 <.01C NS K-Mart - Federal NS NS <.01 NsC NS STAGE 5d (Old—~Married, Single) "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" — K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 <.01 NS NSc <.05 Wiechmann - K-Mart <.01 <.01 <.05 <.05C <.05 Wiechmann — Federal <.01 <.05 <.01 <.05c <.05 Sears - K-Mart <.01 NS NS Nsc NS Sears - Federal NS NS NS NSc NS NS NSC NS NSC Nsc K—Mart - Federal aExcept where noted, the customers. bProbability of true null hypothesis. CSecond store was rated higher. dN>2s. first store in each comparison was rated higher by 82 the three regular department stores (Wiechmann, Sears, and Federal) as equals and higher than the discount department store (K-Mart). The customers in stage four and five of family life cycle rated Wiechmann as high, and all the other department stores—~Sears, K—Mart, and Federal--as equals and lower than Wiechmann. The fact that older customers in stage four and five rated Wiechmann significantly higher than other stores may be due to a "halo effect" resulting from their long association with that store. Overall, the three regular stores are rated ahead Of the discount depart- ment store. The ratings Of the test stores on the basis of com- parisons on locational conveniences is consistent with the findings in Hypotheses 1.1 and 1.2.5 The stores are again ranked in the order Of Federal, K-Mart, Sears, and Wiechmann. The rationale for this sequence was described earlier.6 A comparison Of merchandising suitability factor shows that Wiechmann and Sears rated higher than K—Mart and Federal (the last two being rated as equals). Regular department stores (Wiechmann, Sears, and Federal) are rated higher than the discount department store (K-Mart) on sales personnel dimension in all stages but the last one. These results are consistent with previous results in Hypotheses 1.1 and 1.2 and suggest the superiority of regular depart- ment stores over discount department stores in rendering personal services. 83 1.4 Different department stores have differential aggregate images among their customers who are members of different races. Table 18 provides the probabilities associated with the null hypothesis that there are no differences between aggregate images Of different department stores held by customers who are members of different races. The results indicate that the null hypothesis should be rejected for the comparisons between "Ideal" and test stores for either race. All the comparisons between "Ideal" and test department stores are found significant at the .05 level on all the image dimensions used in this study. The comparisons of test stores on all five image dimensions for non-whites show that more than 50 per cent of the comparisons of test stores are not significant. However, non-whites do carry differentiable aggregate images of test stores on merchandising suitability and locational conveni- ence dimensions. Non-whites rate Wiechmann and Sears higher than K-Mart and Federal, the last two being rated as equals on merchandising suitability. The rating by non-whites on locational convenience is completely reversed as compared to the rating Of stores on merchandising suitability. Non- whites rated K-Mart and Federal as equals and higher than Sears and Wiechmann on locational convenience. This finding is consistent with other results Obtained earlier on the same dimension. The comparisons Of test stores on all the dimensions in this study for whites shows that all but two comparisons E34 Table 18.--Probabilities associated with Wilcoxon T scores obtained from the comparisons of aggregate images of different department stores among customers of different races. Image imensions Merchandising Sales Store Locational Depa::2:2t* + Suitability Personnel Congeniality Convenience Overall WHITEd "Ideal" - Wiechmann <.01b <.01 <.01 <.01 <.01 "Ideal" — Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 <.01 <.01 <.05c <.01 Wiechmann - K-Mart <.01 <.01 <.01 <.01c <.01 Wiechmann - Federal <.01 <.01 <.01 <.01c <.01 Sears — K-Mart <.01 <.01 <.01 <.01c <.01 Sears - Federal <.01 <.01 <.01 <.01C <.05 K—Mart - Federal NSC <.01C NSC <.01c <.01c NON-WHITEd "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" — Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K—Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears NS NS NS NSc NS Wiechmann - K-Mart <.05 <.05 NS <.01C NS Wiechmann - Federal <.01 NS NS <.01C NS Sears - K-Mart <.01 <.01 NS <.05C NS Sears - Federal <.05 NS NS <.01c NSc K-Mart - Federal NSC <.01C <.05C NSC <.01C aExcept where noted, first store in each comparison was rated higher by the customers. bProbability of true null hypothesis. CSecond store was rated higher. dN>25. 85 are significant and hence, the null hypothesis should be rejected. They consistently rate test stores in the order of Wiechmann, Sears, K-Mart, and Federal respectively on all the dimensions except on locational convenience dimen- sion. On this dimension, the ordering is altered to Federal, K-Mart, Sears, and Wiechmann, respectively. This is con- sistent with the earlier results. Overall, the data indicate that different department stores do carry differentiable aggregate images among their customers Of different races. 1.5 Different stores carry differentiable aggregate images among their customers who are members of different sexes. The probabilities associated with the null hypothesis that there are no differences between aggregate images Of different department stores among customers who are members of different sexes are presented in Table 19. The data indicate that the null hypothesis should be rejected for the comparisons between "Ideal" and test stores for either sex. The "Ideal" store is consistently rated higher than all the test stores. The comparison Of test stores on all the five image dimensions for males revealed the following: 1. Fifty per cent Of the comparisons of test stores are found not significant. 2. Males do not carry differentiable images of test Stores on the overall dimension or on the store congeniality dimension. Table 19.--Probabilities associated with Wilcoxon T scores obtained from the 865 comparisons of aggregate images of different department stores among customers Of different sexes. Image Dimensions Merchandising Sales Store Locational Department + Suitability Personnel Congeniality Convenience Overall Storesa + ‘ MALEd "Ideal" - Wiechmann <.01b <.01 <.01 <.01 <.01 "Ideal" - Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears NS NS NS NSc NS Wiechmann - K-Mart <.01 <.01 NS <.01c <.05 Wiechmann - Federal <.01 <.01 NS <.01C NS Sears - K-Mart <.05 <.01 NS <.01c NS Sears - Federal <.05 <.05 NS <.01c NS K-Mart - Federal NSC <.01c NS <.05C <.05C FEMALE "Ideal" - Wiechmann <.01 <.01 <.01 <.01 <.01 "Ideal" — Sears <.01 <.01 <.01 <.01 <.01 "Ideal" - K-Mart <.01 <.01 <.01 <.01 <.01 "Ideal" - Federal <.01 <.01 <.01 <.01 <.01 Wiechmann - Sears <.01 <.01 <.01 <.01c <.01 Wiechmann - K-Mdrt <.01 <.01 <.01 <.01C <.01 Wiechmann — Federal <.01 <.01 <.01 <.01C <.01 Sears - K-Mart <.01 <.01 <.01 <.01C <.01 Sears - Federal <.01 <.01 <.01 <.01C <.01 K-Mart - Federal <.01c <.01C NSC <.01c <.01C aExcept where noted, the customers. first score in each comparison was rated higher by bProbability of true null hypothesis. CSecond store was rated higher. dN>25. 87 3. On the locational convenience dimension, males rate Wiechmann and Sears as equals but lower than Federal and K-Mart. This is consistent with earlier results. 4. Males do not differentiate between Wiechmann and Sears on all the image dimensions. In general, the inability Of males to differentiate between the test stores may be due to their lack Of exper- ience in shOpping. The comparisons Of test stores on all the image dimensions used in this study for females show all but one comparison significant at the .05 level and hence, the null hypothesis should be rejected for females. They rate test stores in the order of Wiechmann, Sears, K—Mart, and Federal on all the dimensions except on locational convenience. With this dimension, the ranking is changed to Federal, K—Mart, Sears, and Wiechmann. This fact is again consistent with the earlier results. Hypothesis II Hypothesis II compared aggregate store images Of individual test stores held by members Of various groups. It consists Of within stores comparisons Of the aggregate store images held by customers in different groups. The discussion on this hypothesis is divided into nine sub- hypotheses. Each sub-hypothesis is analyzed separately in this section. 88 2.1 Differentiable aggregate store images are held for a particular department store by customers who are members of different social classes. , Table 20 provides the probabilities associated with the null hypothesis that no significant differences exist between the images held by customers in different social classes for individual test stores and the "Ideal" store. The majority of the comparisons among customers in different social classes do not yield statistically different results for the department stores in this study. But there are noticeable differences in the images of individual stores held by non-adjacent social classes. Of those comparisons among social classes which resulted in significant differ- ences, 57 per cent were in non-adjacent social classes. It is interesting to observe that among comparisons with significant statistical difference, the lower classes rated the store in question higher in all but one comparison. Relatively, the upper class is shown to be more critical of the department stores. This may be an indication of the decline in status of mass merchandising stores among the upper classes. However, this apparent criticalness of the upper classes should not be interpreted as limited social appeal of department stores. There is no indication that a particular department store attracts members of only one social class. All the department stores seem to be appeal- ingvto a mass market. The comparisons for "Ideal" store on all the five image dimensions among all social classes are not significant 89 Table 20.--Probabilities associated with Mann-Whitney U scores obtained from the comparisons of aggregate images of individual department stores among customers of different social classes. Image Dimensions Merchandising Sales Store Locational Social a d + Suitability Personnel Congeniality Convenience Overall Classes ’ v "Ideal" Upper - Upper Middle NSb NS NS NS NS Upper - Lower Middle NS NS NS NS NS Upper - Lower NS NS NS NS NS Upper Middle - Lower Middle NS NS NS NS NS Upper Middle - Lower NS NS NS NS NS Lower Middle - Lower NS NS NS NS NS Wiechmann Upper - Upper Middle NS NS <.05 NS NS Upper - Lower Middle NS NS NS NS NS Upper - Lower NSC NSC NSc NSC NSc Upper Middle - Lower Middle <.05C <.05C <.05C <.05C <.05C Upper Middle - Lower <.05 <.05 <.01 <.05 <.05 Lower Middle - Lower NS NS NS NS NS Sears Upper - Upper Middle NSc NS NS NS NS Upper - Lower Middle <.05C NS NS NS NSC Upper - Lower <.01C NS NSc NSC <.05C Upper Middle - Lower Middle <.05c <.05 <.05C <.01C <.01C Upper Middle - Lower <.01C <.05 <.05 <.Ol <.01 Lower Middle - Lower <.05 NS NS NS NS K-Mart Upper - Upper Middle NS NS NS NS NS Upper - Lower Middle NS NSC NS NS NS Upper - Lower NS <.01 NS NS NS Upper Middle - Lower Middle NSc NSC NSC NS NSC Upper Middle - Lower <.01 <.01C <.05 <.01 <.01C Lower Middle - Lower NS <.01 NS <.05 <.05 Federal Upper - Upper Middle NS NS NS NS NS Upper - Lower Middle NSc NSC NS NS NSC Upper - Lower \.01 <.05 NS NS <.05 Upper Middle - Lower Middle NSC NS NSC NSC NS Upper Middle - Lower <.05c <.01C <.01c <.05C <.01c Lower Middle - Lower <.05 <.01 <.05 <.05 <.01 a . . . . . . . First soc1al class in each comparison rated store in question higher except where noted. bProbability of true null hypothesis. cSecond social class rated store in question higher. dN2>20. 90 at the .05 level. This result together with results of Hypothesis 1.2 can be interpreted to mean that customers have equal and high standards for their "Ideal" store regardless of their social class. Another interesting result is among the lower middle- lower classes comparisons. The comparison of images held by these two classes on any store dimensions for Wiechmann and Sears are found significant, but their images on the same five dimensions are significantly different for Federal and K-Mart. On the different dimensions of Federal and K-Mart, the lower class rated the stores higher than did the lower middle class. This result may be due to higher association of customers in lower middle and lower class with these two lower status stores. In comparison with upper middle, both lower middle and lower class rated Wiechmann and Sears higher on all store dimensions, but only the lower class rated K-Mart and Federal higher than did the upper middle. This discussion reinforces an earlier conclusion of the critical outlook of members of the upper class regarding relatively higher status stores. 2.2 Differentiable aggregate images for a particular department store are held by customers who are members of different stages of the family life cycle. Table 21 provides probabilities associated with the null hypothesis that no significant differences exist between the images of individual department stores held by customers in different stages of the family life cycle. The results 9]. Table 21.--Probabilities associated with Mann-Whitney U scores obtained from the comparisons of aggregate images of individual department stores among customers in different family life cycle stages. Image Dimensions Family Life Cycle Stages Merchandising Sales Store Locational a,d,e Suitability Personnel Congeniality Convenience Overall "Ideal" 1 - 2 NSb NS NS NS ms 3 NS NS NS NS NS 4 NS NS NS NS NS 5 NS NS NS NS us 2 — 3 NS NS NS NS us 4 NS NS ms NS us 5 NS NS NS NS NS 3 - 4 NS NS NS NS NS 5 NS NS NS NS NS 4 - 5 NS NS NS ms Ns Wiechmann 1 - 2 NS NS NSC NSC NS 3 NS NS <.05 <.05 NS 4 vs NS NS NS NS 5 NS <.05c < 01: <.05: <.01: 2 — 3 NS NS <.01 <.01 <.01 4 NSC NSC NSC NSC NSC 5 <.01 <.05 <.01 <.01 <.01 3 - 4 NS NS NS <.05 NS 5 NS NS NS NS NS 4 - 5 <.05C NS <.05C <.05C <.01C Sears 1 - 2 NS NS NS NS NS 3 NS <.05c <.01C <.01c <.01c 4 NS NS NS NS NS C C C 5 <.05 NS <.05C <.05C NSC 2 - 3 NS NS <.05 <.01 <.01 4 NSc NS NS NSC NS 5 <.05 NS NS <.05 NS 3 - 4 NS <.05 <.05 <.01 <.01 5 NsC NS NS NSC 145C 4 - 5 <.05 NS NS <.05 <.05 532 Table 21.--Continued. Image Dimensions Family Life a d e Merchandising Sales Store Locational cycle Stages ' ' Suitability Personnel Congeniality Convenience Overall K-Mart l - 2 NS NS NS NS NS 3 NS NS NS NS NS 4 NS .NSC NSc NS NSc 5 NS <.01 <.05 NS <.01 2 - 3 NS NS NS NS NS 4 NS NSC NS NS NS 5 NS <.05 NS NS < 01c 3 - 4 NS NSC NS NS NS 5 NS <.01 NS NS NS 4 - 5 NS NS NS NS NS Federal 1 - 2 NSC NSC NS NS NSC 3 <.01 <.0l NS NS <.Ol 4 NSc NSc NS NS NSc 5 <.01 <.01C NS NS <.05 2 - 3 NS <.05 NS NS NS 4 NSC NSC NS NS NS 5 <.05 < 01 NS NS NS 3 — 4 NS <.01 <.05 NS NS 5 NSC NSc NS NS NS 4 - 5 <.05 < 01 NS NS NS aFirst family life stage in question rated store in question higher,- except where noted. bProbability of true null hypothesis. cSecond family life cycle stage in comparison rated store in question higher. dN2 20. 8Family life cycle stages: 1. Young - single and Young - married, no children; 2. Young - married with children: 3. Older,-married with children; 4. Older married, no children; 5. Older - single. 93 indicate that the null hypothesis should not be rejected. It shows that customers classified by family life cycle stages do not carry differentiable images of any of the test stores or the "Ideal" store. This finding is in line with Wyckham's finding cited earlier. Of all those comparisons which are significant, 90 per cent of the comparisons show that customers in the second life cycle stage in each pair rate the store in ques- tion higher and 62 per cent of the comparisons show signifi- cant differences in non—adjacent stages. Also, 75 per cent of all those comparisons which are significant involve fifth stage of life cycle and in each case the fifth stage (Old, single) rated the store in question higher. This indicates that customers who are old and single are less critical of individual test stores than are younger customers. The comparisons on all the five image dimensions for "Ideal" store among all life cycle stages are not sig- nificant at the .05 level. This result together with the results of Hypothesis 1.3 can be interpreted to mean that regardless of the membership of life cycle stage, customers have equal and high standards for their "Ideal" store. This result is consistent with the findings on social classes. This finding may indicate that aggregate images held by customers are not influenced by family life cycle stage, or that all the test stores have a mass appeal among cus— tomers regardless of their membership in any family life cycle stage. 94 Data on comparisons on individual dimensions for each store show some interesting differences. The maximum number of significant comparisons for Wiechmann and Sears are on store congeniality and locational convenience; for K-Mart on sales personnel; and for Federal on merchandise suitability and sales personnel. These findings together with the result that older peOple are less critical of the stores in question suggest that Wiechmann and Sears should improve their image on locational convenience and store con- geniality to attract younger families and to enhance their appeal to customers at all stages of family life cycle; and K-Mart and Sears should improve on sales personnel and mer— chandising mix to widen their appeal in all family life cycle stages. It is interesting to observe that all the improvements to be made on different dimensions for each individual store to enhance their mass appeal are to attract customers who are relatively younger. 2.3 Different aggregate images are held for a particular department store by customers who are members of different races. Table 22 provides the probabilities associated with the null hypothesis that there are no perceived differences between races on image dimensions of the test and "Ideal" stores. Of 25 comparisons among races, only four compari- sons are significant at the .05 level. However, all four comparisons showing significant differences occur on Sears and Federal. In each case non—whites rate the stores higher. Whites are critical of Sears on the merchandising suitability 95 .omAmze .uo:mfln coaummsw cfl muoum poumu omcommmu muflnslcozo .mflmmsuommn Has: osuu mo mmfluflafinmnoumn .pmuoc mum£3 ummoxm .nogmHQ soflpmmsq sfl ououm poumu GOmHHmmEoo comm CH mucmpcommmn mufinzm Omo.v mz m2 m2 m2 Hamuw>o Omo.v m2 m2 m2 m2 mosmflsm>cou Hmcoflumooq m2 m2 m2 m2 m2 mafiamncmmcoo muoum 0mo.v m2 m2 m2 m2 Hmsc0mumm mmamm mz mz omo.v mz nmz suflaflnmufism maflmflecmzonmz amnmpmm pumzux mumom Gamecomflz :HmmpH: mGOHmcmeflo mmmEH monoum ucmEuumme .moomu ucmummmwp mo mumEoumso w.mwuflrsucoz .m> muses mcoEm mmuoum DcoEunmmmp HMSUfl>HUdH mo mommEfl mummmummm mo mcomflummfioo onu Eoum pocflmuno mmuoom D moduHQZIccmz guaz pmNMNOOmmm mmauwafinmnoumll.mm magma 96 dimension and of Federal on sales personnel, locational convenience, and overall dimensions. Thus, it shows that the lower status regular department store (Federal) does not have wide spread appeal and needs improvements in all facets of the store to attract whites along with non-whites. 2.4 Differentiable aggregate images of a particular department store are held by customers who are members of different sexes. Table 23 showsthe probabilities associated with the null hypothesis that there are no perceived differences between sexes on the image dimensions of the test stores and the "Ideal" store. The results indicate that the null hypothesis cannot be rejected for "Ideal," K—Mart, and Federal department stores. However, the null hypothesis should be rejected for Wiechmann and Sears. In each case of significant difference between sexes, females rated the store in question higher. The sex effect is very high on (Histomers' image of Wiechmann and Sears. There is an abun— dance of evidence that the housewife plays a very important r0112 in the purchase decision making for the family. If thits is true, it may provide a competitive advantage of Wiechmann and Sears over K—Mart and Federal. 2.5 Differentiable aggregate images of a particular department store are held by customers who are members of groups which differ in their attitude toward shopping. Table 24 presents the probabilities associated with the ruill hypothesis that there are no differences in the images of a particular department store held by customers 97 . N ONA ZU .uogmfls coaumosv ca ououm mpmu mucopsommmu mHmEmmo .mwmocuomws Adan can» mo mmfluflaflnmnoumn .popoc muons umooxm .uoanQ coaummsv cw muoum woumu mwsmocommmu oamZm mz mz omo.v oao.v mz Hamuo>o m2 m2 omo.v m2 m2 mosoflso>coo HmcoHumooq m2 m2 0mo.v oao.v mz wuflamflsomsoo ououm m2 m2 0mo.v 0mo.v mz HoGCOmnom moamm m2 m2 m2 oao.v sz wpwaflnmuflsm mcfimflpcmzoumz Hmumpom pumzlm mummm camenooflz zamopH: mcofimcmeflo mmmEH mmHOpm pcmfiuummma .mmxmm ucmHOMMHp mo mHoEOpmso Urmmamamm was mam: msOEm monoum ucwEuummop Hmspfl>apaw no mommfiw mummmummm mo mGOmHHmmEoo on» Eon“ pocflmuno mouoom D mocuflnztcdmz saws poumNUOmmm mowuflaflnmnoumul.mm magma 98 N . ONA ZU .mo.A mcflcmoe mZIumflmmnu0d>£ Has: mum» mo zuflaflnmnoumn .umnmflz cofiummsv ca muoum pmumu msflmmonm momsm 0:3 mpcmpcommomm m2 m2 m2 m2 m2 Hamum>o m2 m2 m2 m2 m2 mommaco>soo Hmcoflumooq m2 m2 m2 m2 m2 muwamwsomcoo muoum m2 m2 m2 m2 m2 Hmccomumm moamm m2 m2 m2 m2 nmz muHHHQMDNSm mcflmflpcmnoumz Hmnopom unmzux mumom ccmsnomwz zammpH= mcoflmcmaflo mmmEH monoum ucofigummmo 040%0msm u.coollwonsm .ucofimoflam mcwmmonm ma pompoum mHmEoumso mcoEm monoum usmeuummmp Hmsww>fipcfl mo mommfifl mummmummm mo chmwummEoo map Eoum pmGMMDQo mOHOOm D mosuficztssmz spas meMHoommm mmfluflaflnmnoumun.vm magma 99 who differ in their attitudes toward shopping. Only one out of the 25 comparisons is found significant at the .05 level. Thus, the null hypothesis is rejected. It shows that cus- tomers grouped on the basis of shOpping enjoyment have simi- lar images of individual department stores. This may indicate that none of the test stores provide shopping excitement to customers who enjoy shopping. 2.6 Differentiable aggregate images are held for a particular department store by customers who are members of groups which differ in social support for their beliefs about that store. Table 25 provides the probabilities associated with the null hypothesis that there are no differences in the images of a particular department store among customers grouped on the basis of social support they receive for their beliefs about that store. The results show that the null hypothesis should be rejected for Wiechmann on all the image dimensions; however, the null hypothesis is not rejected for Sears, K-Mart, and Federal. These findings may indicate that Wiechmann is the only test store which effectively utilizes interpersonal channels of communication. It shows that customers with social support do have a defi- nite favorable image of a high status department store and this may be the reason why images of Sears, Federal, and K-Mart are getting blurred on some variables whereas Wiechmann is consistently separated from other stores and rated high throughout the study. Probably, such an image may be necessary for a high status store like Wiechmann, lOO .omAmzo .mwmmnuom>£ Has: osun mo muaawnmnoumb .Hwbmfln coaumosv as muoum pmumn .oHODm may usonm mwmflHOQ Mamba buss pomnmm mpsmfiuw moons mucopsommmmm m2 m2 mz mo.v Hamum>o mz m2 m2 mo.v mosmflcm>coo Hmsofiumooq m2 m2 m2 mo.v muHHMHcmmcou muoum mz mo.v mz mo.v HchOmumm moamm m2 m2 m2 bmo.v wuflafibmufism mcflmflpsmsouoz Hmuopom bumzlx mumom acmfinooflz .mcoflmsoeflo mmmEH monoum usofiuummmo o.momummmHD .m> mmumé umpcmflum .muoum may usonm mmowamb Mamba How uuommsm Hmwoom any Sb pomsoum mHmEoumso mGOEm monoum psmEDMMQOU Hmspfi>flpcfl mo mommaa mo chmHHMQEoo or» Eoum pmcflmubo mouoom D wocuflbzxscmz spas pmumfl00mmm mmwuwawbmnoumll.mm magma 101 but it may not be a good competitive strategy for national mass merchandising stores like Sears and Federal as such a definite image may reduce their mass appeal. 2.7 Differentiable aggregate store images are held for a particular department store by customers who are members of groups which differ in their advertising readership of that store. Table 26 shows the probabilities associated with the null hypothesis that there are no differences in the images of a particular department store among customers grouped on the basis of their advertising readership of that store. The findings indicate that "always" readers of advertise— ments of a department store do carry a favorable image of that store as compared to occasional or non-readers of those advertisements. This may be due to the fact that avid readers may be trying to avoid cognitive dissonance by seeking information to support their views and behavior. 2.8 Differentiable aggregate images are held for a particular department store by customers who are members of groups which differ in their shOpping practices with that store. In this study, special attention is given to four shopping practices: the kind of goods customers buy from the store, the frequency of visits to the store, the recency Of visit to the store, and the particular branch of the Stuores preferred for shOpping. The impact of each one of t1'lese shopping practices on the images carried by customers 111 different groups is discussed under four separate sub- hYpotheses . 102 Table 26.-~Probabilities associated with Mann—Whitney U scores obtained from the comparisons of images of individual department stores held by customers grouped by their advertising readership for those stores. Image Dimensions Advertising Merchandising Sales Store Locational Readershipaic Suitability Personnel Congeniality Convenience Overall Wiechmann Always - Sometimes <.01b <.01 <.01 <.01 <.01 Always - Never <.01 <.01 <.01 NS <.01 Sometimes - Never NS NS NS NS NS Sears Always - Sometimes <.01 <.01 <.01 <.01 <.01 Always - Never < 01 <.Ol <.01 <.01 <.01 Sometimes - Never NS NS <.05 NS .05 K—Mart Always - Sometimes <.01 < 01 <.01 <.01 <.01 Always - Never < 01 <.01 <.01 <.05 <.01 Sometimes - Never NS NS NS NS NS Federal Always — Sometimes <.01 <.01 <.01 <.01 <.01 Always - Never <.01 <.01 <.01 <.01 <.01 Sometimes — Never NS NS NS NS NS aRespondents in first group in each comparison rated store in question higher, except where noted. bProbability of true null hypothesis. CN2>20. 103 2.8:1 Differentiable aggregate images are held for a particular department store by customers who are members of groups which differ in the kind of goods they buy from the store. To test this hypothesis, all the customers are divided in this study into three groups on the basis of lines of goods they buy from the store: (1) soft-line buyers, (2) hard- line buyers, and (3) buyers of both lines. Table 27 provides the probabilities associated with the null hypothesis that there are no differences in the images of a particular depart- ment store among customers grouped on the basis of the line of goods they buy from the store. The results indicate that the null hypothesis can not be rejected as all the comparisons on all image dimensions for any individual store show no significant difference among customers buying different kinds of products from that store. This may be as a result of mass appeal of those department stores carrying similar lines of goods. This is contrary to the writer‘s expectation. It was supposed that there would be differences between the images held by hard-line buyers and soft-line buyers because hard-line buyers are more involved in such purchases. 2.8:2 Differentiable aggregate images are held for a particular department store by customers who are members of groups which differ in their recency of shOpping. To test this hypothesis, all the customers are divided into three groups on the basis of their recency of Shopping at an individual store: (1) those who have shOpped ‘vithin last week (very recent), (2) those who have shopped mOre than a week ago but less than a month ago (intermediate 104 Table 27.--Probabilities associated with Mann-Whitney U scores obtained from the comparisons of images of individual department stores among customers who differ in the line of goods they buy from these stores. Image Dimensions Merchandising Sales Store Locational Line of Goodsa,c Suitability Personnel Congeniality Convenience Overall Wiechmann Soft Hard NSb NS NS NS NS Soft Both NS NS NS NS NS Hard Both NS NS NS NS NS Sears Soft Hard NS NS NS NS NS Soft Both NS NS NS NS NS Hard Both NS NS NS NS NS K-Mart Soft Hard NS NS NS NS NS Soft Both NS NS NS NS NS Hard Both NS NS NS NS NS Federal Soft Hard NS NS NS NS NS Soft Both NS NS NS NS NS Hard Both NS NS NS NS NS aReSpondents in first group in each comparison rated store in question higher. bProbability of true null hypothesis. cN2>20. 105 recency), and (3) those who shOpped more than a month ago (least recent). Table 28 presents the probabilities asso— ciated with the null hypothesis that there are no differ- ences in the images of a particular department store among customers grouped on the basis of recency of their shopping at that store. Of 60 comparisons, 29 are found significant at the .05 level. Although the results do not show overwhelming proof of the effect of recency of shopping on the images of all the test stores viewed collectively, the effect does show significant differences for individual stores. For Wiechmann, all the comparisons with significant differences occur between group 1 (very recent) and group 3 (least recent). This may indicate that recency is an important factor affecting the images of a department store held by its recent customers. It could mean that reinforcement of experience is necessary for customers to maintain a favorable impression of Wiechmann. For Federal, the comparisons involving very recent shoppers are significant and in favor of Federal on all the image dimensions. For K-Mart, all but one of the comparisons involving relatively recent shoppers are significant. For both of these stores, recency of shopping experience with the stores is an important factor in maintaining a favorable image of the stores. No significant differences are found for Sears on any of the image dimensions among customers grouped by their irecency of shopping experience. This may indicate that 106 Table 28.--Probabilities associated with Mann-Whitney U scores obtained from the comparisons of images of individual department stores among customers grouped on the basis of recency of their shopping at those stores. Image Dimensions Recency of Merchandising Sales Store Locational Shoppinga'c'd Suitability Personnel Congeniality Convenience Overall Wiechmann l - 2 NS NS NS NS NS - 3 <.01 <.01 <.01 <.01 <.01 2 - 3 NS NS NS NS NS Sears 1 - 2 NS NS NS NS NS - 3 NS NS NS NS NS 2 - 3 NS NS NS NS NS K—Mart l - 2 <.0l <.01 <.05 <.05 <.01 - 3 <.01 <.01 <.01 < 01 < Cl 2 - 3 < 05 NS <.01 < 05 <.05 Federal l - 2 <.05 <.05 <.05 NS <.05 - 3 <.01 <.Ol <.01 <.01 <.01 2 - 3 NS NS NS NS NS aRespondents in first group in eacn comparison rated store in question higher, except where noted. bProbability of true null hypothesis. CN2>20. dRecency of shopping: 1) less than one week ago; 2) more than a week ago but less than a month ago; 3) more than a month ago. 107 Sears, as the oldest department store among the test stores, has such an established image on all store dimensions that the time-gap involved in customers' shopping experiences with Sears has no effect on the image customers carry of Sears. Overall, the results indicate that very recent shoppers do carry favorable images of individual stores. 2.8:3 Differentiable aggregate images are held for a particular department store by customers who are members of groups which differ in their frequency of shopping at that store. For the purpose of investigating this hypothesis, all the customers are divided into four groups on the basis of their frequency of shOpping with an individual store. These groups are: (l) frequent shoppers (more than three times a month), (2) occasional shoppers (two or three times a month), (3) infrequent ShOppers (once a month); rare Shoppers (less than once a month). Table 29 displays the Lnnobabilities associated with the null hypothesis that there are Iu320. 109 ShOppers in comparison to all the other shoppers rated the test stores unfavorably. This may be the very reason why they are rare shoppers. Rare shoppers' lack of interest or attraction to the store may have led them to stOp at the store less than once a month and consequently is reflected in their less favorable images of the test stores. The above results are true for all the comparisons on all the image dimensions for all the test stores with an interesting exception for K-Mart on the sales personnel dimension. With this dimension, no significant differences are found for K-Mart among customers classified by their frequency of shOpping at K-Mart. This may be because of lower rating of K-Mart on sales personnel dimension among all the customer groups classified by their frequency of shOpping. This is consistent with previously discussed .results of lower rating of discount department stores on the sales personnel dimension. 2.8:4 Differentiable aggregate images are held for particular department stores by customers who are members of groups which differ in their shopping at different locations of those stores within a city. This hypothesis specifically compares differences in tflie images of different locations of individual depart- meni: stores within a city. In this study, only K-Mart and Federal have more than one location in Saginaw. Both the storxes have one location on the East side and the other on the VWest side of Saginaw. All the customers are classified intC> three groups for each store. These groups are: 110 (1) those who shop at the East side location of K-Mart or Federal in Saginaw, i.e., K-Mart (East) or Federal (East) ShOppers; (2) those who shOp at the West side location of K-Mart or Federal in Saginaw, i.e., K—Mart (West) or Federal (West) shoppers; and (3) those who shop at both the locations, i.e., K-Mart (Both) or Federal (Both) shoppers. The comparisons among these groups are shown in Table 30 for both the stores. Table 30 shows the probabilities asso- ciated with the null hypothesis that there are no differences in the images of a particular department store among cus- tomers who shOp at different locations of that store in a city. The results show that East side ShOppers consistently rate the East side location of both the stores favorably. This may have been caused by demographic characteristics of the residents of areas where these stores are located. The JEast.side of Saginaw is typically characterized by lower :hncome, non-whites, and factory workers. Considering these (fliaracteristics and the lower status of Federal and K-Mart, iJ:.is not surprising to find that the East side locations Of tflnese stores are found attractive and are rated higher by Ekist side ShOppers. The West side of Saginaw is char- actexrized by higher income, whites, and businessmen. The lowen: status stores located in this area may find them- selxnes relatively unattractive to the shoppers and hence, may hue rated low. Thus, higher rating by East side ShOppers combiJued with lower rating by West side ShOppers of the 111 .omANzo .mflmmbpomms Hans map» mo huflafibmboumb .pmboc muons ummoxm Norman coaummsq cfl ououm pouch comflummaoo boom ca dsoum umuflw CH mucmpcommmmm m2 m2 m2 m2 m2 nuom I ummz mo.v mo.v mo.v mo.v mo.v nuom mo.v mz mo.v mo.v mo.v umoz I ummm Hmnopmm m2 m2 m2 m2 mo.v guom I ummz m2 m2 m2 m2 m2 gpom Ho.v Ho.v mz Ho.v bao.v ummz I ummm “HMSIM Hamum>o mocoflco>coo SDHHMHcomsoo amazemuom wuaaflbwufism o.mmuoum may Hmcoflumooq ououm mmHMm msflmflosmnoumz mo morocmum mDONmGOEflQ mmmEH .mwuoum mmonu mo mcofiumooH ucmummwflp rufi3 mocmwuomxm moammonm Mamba mo mammn on» so pompoum mumfioumso mcofim mmuoum pamfiuummmp Hmsoa>aocfl mo momma“ mo mGOmflHmmEoo map Eon“ pocflmubo monoom D mosanZIscwz nufl3 omuma00mmm mmflpfiafibmooumII.om mabme 112 respective locations of Federal and K—Mart may have led to consistently favorable rating of their East side locations. When the images of customers shOpping at either location of K-Mart are compared with the images of customers shOpping at both the locations of K—Mart, the results are not significant. This indicates that the overall image of K-Mart is not significantly different from the images held of its separate locations. However, this is not true for Federal. No differences are found between the images of West side ShOppers and customers shopping at both the loca— tions of Federal; but images held by East side customers are significantly in favor of Federal (East) as compared to customers shopping at both the locations of Federal. The East side is rated significantly higher than either the West side location or the overall image of Federal. This is contrary to expectation. As all the chain stores are cen— trally merchandised and are directed from the main office, it was expected that the overall image may be similar to the image of individual locations of the store. This is also contrary to Rich's conclusion.lo All the data indicate that customers do hold dif- ferential images of different locations of department stores; however, there is a lack of clarity on the images of separate locations of a department store compared to the overall image 0f their store. 113 Hypothesis III 3.1 There is communality between the anticipated aggregate image of a department store held by the management and the actual aggregate image of the same department store held by the customers. Table 31 presents probabilities associated with the null hypothesis that there are no differences between the anticipated and the actual images of a particular department store. The results show that the null hypothesis should be rejected for Wiechmann, K-Mart, and Federal; however, the null hypothesis cannot be rejected for Sears. Of all the comparisons, approximately 66 per cent are found significant. The direction of the ratings for the comparisons with sig— nificant differences indicates that the management con- sistently rated their store higher than did their customers. The management of Wiechmann anticipated customers' images correctly on merchandising suitability and sales per- sonnel but overrated their store on store congeniality, locational convenience, and overall dimension. Sears man- agement anticipated customers' images of their store correctly on all the image dimensions with the exception of locational convenience. Surprisingly, both high status stores over- rated the locational convenience of their store with their downtown locations. The management of K—Mart and Federal anticipated customers' images of their stores incorrectly and consistently overrated their store on all image dimen- sions. 114 . N omA zc .Hmbmfln soflummsw ca muoum omumu usmEommcmzo .mflmonuom>£ Has: wasp mo awaHQMQoumn .ompo: muons umooxm Norman coflumosv ca mHODm poumu somaummeoo boom ca mumEODmsom oao.v oao.v omz omo.v Hamum>o oao.v oao.v 0mo.v oao.v mosmwcm>soo Hmcowumooq oao.v oao.v omz omo.v muHHMHcmmsoo muoum oao.v mao.v omz omz Hmcaomumm mmamm omz oao.v omz o.bmz muHaHDMDHsm msfimwocmnonoz Hmuooom pnmzrm mummm somenomfiz msoflmcmsflo mmmEH monoum usmfiaummoo AucmammmcmfiImumEoumsov U.mpmummwoflpcHpsH MOM mumeoumso an tam: mommEH Hmsuom tam pcmfiomMGME Sb pawn mommafi omummflofiucm mo chmHHmmEoo may Eouw omsflmubo mouoom D mmcuHSSIssmz rues meMMUOmmm meuHHHQMQOHmII.Hm magma 115 It is interesting to observe that all but one of the not significant differences between the anticipated and the actual images of department stores occur on the image dimensions of the two high status stores, Wiechmann and Sears. All but one of the significant differences on com- parisons between the anticipated and the actual images occur on the image dimensions of the two low status stores, Federal and K-Mart. These results may indicate that cus— tomers rate those stores higher where they find that manage- ment of the store understand them better. Another interesting finding is that management of all the regular department stores correctly anticipated cus- tomers' images on merchandising suitability dimensions, whereas the management of the discount department store overrated merchandising suitability of their store as com- pared to their customers. This may be an additional reason why K-Mart, a discount department store, is rated lowest among the test stores. Hypothesis IV 4.1 Customers' satisfaction and marketing orienta- tion of management for a particular department store are positively related. This hypothesis is in line with the arguments pre- ssented in the present marketing literature. The literature ssuggests that if management of a store is marketing oriented 61nd if proper planning has taken place in the implementation c>f the marketing concept, customers' dissatisfaction for 116 that store will be at a minimum. The hypothesis is tested with the help of Aggregate Image Deficiency (AID) and Inverse Marketing Orientation Index (IMOI). AID measures customers' dissatisfaction and IMOI measures lack of marketing orien— tation of the management of a store. If the hypothesis is true, AID and IMOI scores should be positively related. The scores on AID and IMOI on each image dimension for all the test stores are presented in Table 32. The results in Table 32 show that AID scores increase with the increase in IMOI scores on merchandise suitability dimension for all the test stores. The results on other dimensions do not indicate positive relationship between the two variables. When viewed Erma the point of view of each individual store, the relationship between AID scores and IMOI scores seems weak or nonexistent for all the stores. A visual picture of this relationship between AID scores and IMOI scores can be seen from the plottings in Figures 5, 6, and 7. A look at Figure 5 shows that the points are scattered. This indicates that the relationship between AID and IMOI scores is very weak or nonexistent. JFigure 6 shows the relationship between AID and IMOI scores can individual image dimensions for all the stores; i.e., it shows between the stores relationship of AID and IMOI on each .individual image dimension. Figure 7 relates AID and IMOI scores for individual stores on all image dimensions; i.e., it shows within the store relationship of AID and IMOI Ecores on all the image dimensions. If there is expected 117 .muoom xoocH coaumusmflno msfluoxumz mmum>sH "mHoan .muoom SocoHOHmmo mmmEH mummoummm "moHém mm.a+ m>.HI «m. + om.HI Ho.a+ mm.HI on.H+ em.HI mm.a+ «m.HI unszx hm. + om.HI ma.a+ om. I wa.a+ mH.HI mm.H+ mm.HI em. + mm.HI Hmuoomm we. + ov.aI mn.a+ mo.mI mm. + mm. I mo. + mm.HI mm. + b¢.HI mummm om.a+ mH.HI Ho.a+ mm.mI mm. + em. I mm. + om.HI mo. + mm. I camasomfiz mHozH moH< mHozH moH¢ mHozH mDHd mHozH mon bmHozH MmDHd monoum Hflmum>o mocmficm>coo mufiamficomcoo Hmccomuom Speaflbmuasm ucmEpummoo Hmcofiumooq ououm moamm mmfiocmzouoz mcoflwsofiflo mmmEH mmHOOm xmmuCfl .mcofimcmaflo mmmEH Has no mououm uswEuummmp stpfl>flpcfi mo sODDmpcofino msfluoxuma tam mmuoom wocmHOHmmo mamas mummoummmII.Nm magma AID Scores 118 20401 W1 o 81 2.004 0 K 0 o Fm m Fp °K o 1.60 4 S ' .Fo m ‘S O .K 'Sp Kl «KS p 1.20 . .W o . p Fs W0 '8 s . W 'F m .W l .80. s .40. 0 . , . . . 40 80 1.20 1.60 2.00 IMOI Scores IFigure 5.--A Scatter Diagram of AID and IMOI Scores for All Test Stores. Wiechmann Symbols: W = m = Merchandising suitability S = Sears p = Sales personnel K = K-mart s = Store congeniality F = Federal l = Locational conveniency o = Overall 2.40 2.00 1.60 1.20 0‘) 0 H O U U) D H “ .80 .40 o Merchandising Suitability—______—_ Locational Convenience ———-A-—- Sales Personnel Store Congeniality Overall ___ _____ U .40 v .80 r 1.20 1.60 2.00 IMOI Scores Figure 6.--AID and IMOI Scores of All Test Stores on Individual Image Dimensions. Syn&mls: W S K F Wiechmann Sears K-mart Federal o H m'u B Merchandising suitability Sales personnel store congeniality Locational conveniency Overall AID Scores 120 2 O 40* 2.00. 1.60- 1.20« .80' . 604 Wiechmann ....... Sears K-mart —A—" Federal ___... 0*h V I 1 I v .40 .80 1.20 1.60 2.00 Figure 7.--AID and IMOI Scores Dimensions. W Wiechmann S = Sears K = K-mart F Federal Symbols: IMOI Scores of Individual Test Stores on All Image Merchandising suitability Sales personnel Store congeniality Locational conveniency Overall 0 h'm'U S n 121 positive relationship between AID and IMOI scores, all the lines in Figures 6 and 7 indicating relationship between these two variables should move from bottom left corner to top right corner of the graph at a positive angle from the origin. But if the lines move from the top left corner to bottom right corner of the graph, it indicates a negative relationship. Hence, if a line for a store in Figure 7 or for an image dimension in Figure 6 moves up and down, this sheds doubts on a positive relationship between the two variables and may indicate a weak or no relationship between the two variables. ‘ An analysis of charting in Figures 6 and 7 indicates that out of nine charted lines, only one line (on merchan- dising suitability) moves in the direction of a consistent relationship. On the basis of the fluctuation on all the other charted lines for between and within store compari- sons, the hypothesis is rejected and it is concluded that there is weak or no relationship between AID and IMOI scores. The lack of relationship between customer satisfac- tion and marketing orientation of management should not be interpreted to mean that marketing orientation is not a proper tool for management to achieve customer satisfaction. Customers' dissatisfaction with a store is not a function of only the marketing orientation of management of that store. The implementation of marketing orientation through proper planning is one of the other important factors that may affect customers' dissatisfaction. Lack of proper 122 planning for the implementation of marketing orientation may have caused a lack of relationship between AID and IMOI scores. (Furthermore, the proper implementation of marketing orientation requires that the target market be identified before the implementation of marketing strategy. The AID scores are found for all, not just target customers in this study and then related to IMOI scores. This procedure of using AID scores for all customers may also have weakened its relationship with IMOI scores. The results of positive relationship between AID and IMOI on merchandising suitability indicate that mass mer- chandising stores are merchandise oriented. The lack of relationship on other dimensions-~sales personnel, store congeniality, and locational convenience—-may indicate lack of marketing orientation, specifically on the intangible aspects of business on the part of management, or as dis— cussed above, it may indicate imprOper planning in the implementation of the marketing orientation by the store management. The lack of relationship between AID and IMOI on the overall dimension reinforces the above discussion. FOOTNOTES--CHAPTER IV Su ra, pp. 55-56. 2Ibid. 1: 3Supra, pp. 38—39. 4Supra, p. 78. Ibid. Su ra, p. 76. I: 7Su ra, p. 78. I: 8William Kenkel, "Husband—Wife Interaction in Deci- sion Making and Decision Choices," Journal of Social Psychol- ggy_(August, 1961), 2555-2562; Elizabeth Wolgst, "Do Husbands or Wives Make the Purchasing Decisions?" Journal of Market- ing (October, 1958), 151-158; J. N. Fry and F. H. Siller, "Comparison of Housewife Decision Making in Two Social Classes," Journal of Marketing Research, VII (August, 1970), 333-337. 9Supra, p. 76. 10Rich and Portis, 0p. cit., p. 15. 123 CHAPTER V FINDINGS, IMPLICATIONS, AND RECOMMENDATIONS The purpose of this study was to investigate and compare the aggregate department store images held by cus~ tomers and management and to replicate and extend the Wyckham study. In the pursuit of this purpose, the data were col- lected from customers and management on five image dimensions of four test stores and a hypothetical "Ideal" store. Findings .The findings obtained from the analysis of the data collected are discussed in this section in answer to four major questions presented in Chapter I of this study. In the order of their presentation in Chapter I, the questions and the resulting summary of data are presented below. The findings on the first two questions are also compared with the Wyckham study. I. Are the aggregate images held by customers different from store to store? To answer and investigate this question, it was hypothesized that different department stores carry differ— -entiable aggregate images among all of their customers who are members of different social classes, family life cycle Stages, races, and sexes. The analysis of the data on the 124 125 above-mentioned variables indicates that the hypothesis is not to be rejected; i.e., customers do carry differential aggregate images of different stores. These findings support the Wyckham study.l Other findings obtained in the investigation of this question are as follows: 1. The stores are rated on a continuum with Wiechmann highest and K-Mart lowest on an overall image dimension and on all the other image dimensions with the exception of loca— tional convenience. 2. The three regular department stores (Wiechmann, Sears, and Federal) are rated higher than the only discount department store (K—Mart) investigated in this study. 3. The downtown department stores are rated lower than the shOpping center department stores on the locational convenience dimension. 4. The "Ideal" store is rated higher than any test store on all image dimensions. 5. The regular department stores are rated higher on the sales personnel dimension. II. Do the aggregate images of a particular depart— ment store vary among the various groups of customers? To answer and investigate this question, it was hypothesized that differentiable aggregate store images are held for a particular department store by customers who are members of different groups. Different groups were obtained 126' by using eight different variables. These variables were: social class, family life cycle, race, sex, attitude toward shopping, social support, advertising readership, and shop— ping practices. The findings on each one are as follows: 1. There is no conclusive evidence that differen— tiable aggregate images of a particular department store are held among customers in different social classes. This result is contrary to Wyckham‘s findings2 and agrees with the conclusions of Kuehl,3 Klein,4 and Rich and Jain.5 The study also agrees with the Wyckham study that the majority of the significant differences occur in non-adjacent social classes.6 Further investigation of the data reveals that relatively upper classes are more critical of the depart— ment stores in general. 2. Customers classified by family life cycle stages do not have differentiable images of any of the department stores. These findings support Wyckham‘s conclusion of the lack of family life cycle effect on the aggregate image of individual stores.7 Data also indicate that old and single customers are less critical of individual test stores than relatively younger customers. 3. Different races carry differentiable aggregate images of only Federal department store. However, there is a communality in the aggregate images of other individual test stores held by whites and non—whites. Thus, it is con- cluded that differences in the store images held by different :races depend on the individual department store in question. 127 This result is contrary to the findings of Bullock8 and the conclusion of Wyckham that ". . . the hypothesis, the differ- entiable aggregate images of particular department stores are held by members of different races, should be accepted."9 4. Differences in store images held by members of different sexes depend on the individual department store in question. This result is contrary to Wyckham's cate— gorical conclusion that members of different sexes do not carry differentiable aggregate images of particular depart- ment stores.10 The females are found to rate higher status stores favorably. 5. Customers' attitudes toward shopping do not affect their perception of the department stores. This find— ing is contrary to the conclusion of Wyckham that "Customers who enjoy shopping rate a store higher, and perceive test stores differently, than do those who do not."11 6. Differentiable aggregate images are held for high status stores by customers with social support. This is not true for other department stores. Again, this result is quite contrary to Wyckham's findings.12 He found that customers who are members of groups which differ in the social support for their beliefs for a high status store (Hudsons in the Wyckham study) do not have differentiable aggregate images of that store. 7. This study is in agreement with Wyckham's con- <:lusion that avid readers of advertisements of a department :store do carry favorable images of that store as compared 128 to occasional or non-readers of advertisements of that store.13 8. Comparisons of individual store images held by customers who differ in their shOpping practices show that: a. There are no differences in the aggregate images of a department store held by customers buying different kinds of products from individual department stores. b. Unlike the findings of the Wyckham study,l4 very recent customers carry amore favorable image of a department store than least recent shOppers. c. The effect of frequency of shOpping on customers‘ aggregate images of a particular department store is very weak. However, rare shoppers are more critical of the indi- vidual test stores than frequent, occasional, or infrequent shoppers. . d. Customers do hold differentiable images of various locations of particular department stores on all image dimensions. The shoppers of East side locations of K—Mart or Federal rated those stores more favorably than the shOppers of West side locations of K-Mart or Federal. This is contrary to the finding of Rich and Portis that there is considerable similarity among suburban branches.15 III. Does the anticipated aggregate image held by the management of a store match the actual aggregate store image held by the customers of that department store? 129 It was hypothesized that there is a communality between the anticipated aggregate image of a department store held by the management and the actual aggregate image of the same department store held by the customers. This hypothesis is based on the assumption that retail managers, due to their first-hand knowledge of their customers, can anticipate the aggregate image of the store held by customers and hence, there should be communality between the anticipated and the actual aggregate images. The hypothesis is rejected for low status stores, but it is not rejected for high status stores. On all the significant comparisons, management of the test stores overrated their store. These results support the conclusion of Ryan and McClure. They conclude: "These [retailers'] images . . . seem to reflect historic stereo— type rather than current consumer brand images."l6 IV. Is there a relationship between the marketing orientation of the management and customers‘ satisfaction for any department store? To investigate this question, it was hypothesized that there is a significant relationship between customers‘ satisfaction and marketing orientation of management for any department store. This hypothesis was tested with the help of descriptive statistics and graphical analysis of between and within the stores relationship of AID and IMOI scores on all the image dimensions. Except on the merchandising suit- ability dimension, the hypothesis is rejected and it is con- cluded that there is weak or no relationship between AID and 130 IMOI scores for all the stores, i.e., no significant relation- ship was discovered between the marketing orientation of the management and customers' satisfaction for any department store. Implications of the Findings The finding that customers do hold differentiable images of department stores and they do rate stores on a continuum from highest to lowest on the image dimensions is of importance to marketing strategists. The marketing strategists have devised various ways to segment a market. However, no method of market segmentation is found completely satisfactory for all the firms. This is why it is suggested that ". . . creative market segmentation involves the search for new ways . . . in the hOpe of discovering fresh marketing opportunities."17 The segmentation of market on the basis of perception of stores may be a new creative method worth exploring. For example, though all the chain stores operate on central merchandising policy and are directed from the main office to cater to a mass market, it is found that Sears, K-Mart, and Federal are perceived differently with Sears as a high status store and Federal and K-Mart as lower status stores. The analysis of data also provides an insight into competitive strengths and weaknesses of individual stores, e.g., the strength of Wiechmann is in its sales personnel whereas its weakness is in its locational conveniences. 131 This is found true in general :flxr downtown department stores as compared to shOpping center department stores. K-Mart and 'Federal are seen strong in their locational convenience and merchandising suitability for lower class customers. The management of these stores should aim their strategies at removing their weaknesses. While the data for this study were being analyzed, Wiechmann and Sears opened up their stores in the Fashion Square ShOpping Mall outside of Saginaw. This may indicate realization on the part of downtown stores of their limitations on the locational convenience dimension. The move by Wiechmann and Sears also implies that if the downtown stores are to remain a shopping attraction for their customers, they will have to improve their accessibility and parking facilities. The study also provides a method for identifying areas of deficiency of the store in the eyes of customers. The comparison of "Ideal" with the test stores indicates that customers are dissatisfied with the present mass merchan— dising department stores. The comparisons of "Ideal" in different groups does not show significant differences. This lends more support to the stability of the "Ideal" rating. This provides an argument for using "Ideal" ratings as a criterion against which the actual image of the store may be ,judged by the store managers. Though the results on the effect of family life cycle Humouo Homm H cofluommmflumm on ma mm>Hm mcflmmonm A. ,6 moan H cons mammsoo 1. m>mn ou mesa H mcHuHoxo mH mcHQQOSm we won xuos mH mchmonm we mom CH mchmocm muoooum pummmuwwp mmmmHm spm> «page sHunmaHm enemas sHuamaHm muflso sum> swm> muwso sHuanHm asaemz sHuamaHm mango mum> sum> «page .sauamaam assess sauraaHm muaso spm> sum> mango sHuanHm agave: anamaHm mnaso sum>. sum> ouaso sauamaam endows sHuamHHm muaso sum> .mchmonm uconm Hmmm com 30: mnHHommp.ou mmHmom ecu xumz mchmocm cm H cons m>Hummuo Hmmm u.coc H .m.m coHuommmHumm we mm>Hm mchmonm .v.m mcon moan _ ou mxHH H .m.m HHSp mH mchdosm we now .m.o cam mH mcwmmonm we now .H.m .mmHmom omen» mcHxHME .cucumu 50> HHucs mmdemsn mo oomHm no mEon uso» m>moH so» mEHu mzu Ecumllmchmonm Ho >UH>Huom may no xchu mmmmHm 10. 165 Have you ever shOpped at: YES NO Sears Roebuck & Company Wiechmann's Federal (East Side) (West Side) K—Mart (East Side) (West Side) List in order your favorite department stores in Saginaw. (If one of your favorite stores has two locations, indicate which one.) 1. 2. 3. When did you last shOp at: (skip those not shopped at all) SEARS WIECHMANN'S FEDERAL ' K-MART East West East West Less than a week ago. More than a week ago, but less than a month ago. More than a month ago. Could you estimate how many times, on an average, during this past year, you shopped at: (skip those not shopped at all) SEARS WIECHMANN'S FEDERAL K-MART East West East West Once a month Two or three times a month More than three times a month None of the above 11. 12. 13. 166 Which of the following stores have you shopped for these products? (skip those not shOpped at all) SEARS WIECHMANN'S FEDERAL K-MART East West East West Cosmetics _____ _____ _____ Clothing ____ ____ ____ Health-aid _____ _____ _____ Hardware Sporting goods Electronic Equipment (TV, stereo,other) Furniture Appliances (range, re- frigerator, dishwasher, etc.) How do your friends feel about the following department stores? SEARS WIECHMANN'S FEDERAL K-MART East West East West Same as I do Differently than I do All of the following department stores advertise in the daily and Sunday newspapers. How often do you read these advertisements? SEARS WIECHMANN'S FEDERAL K-MART East West East West Always Sometimes Never 167 DEMOGRAPHIC DATA: 14. 15. 16. 17. 18. 19. Sex: Race: Male Female White Negro Mexican Other (spec1fy) Marital Status: 1. Single 2. Married 3. Widowed 4. Divorced — Separated 5. Other Do you have any children living at home (include any children supported by parents who are away at school)? 1. Yes 2. No How Many? Please indicate which letter corresponds to your age category: A. Less than 20 years. B. More than 20 but less than 30 years. C. More than 30 but less than 40 years. D. More than 40 but less than 50 years. E. More than 50 but less than 65 years. F. 65 years and older. Who is the head of the household? Male Female 20. 21. 22. 23. 168 Please indicate which letter correSponds to the job of the male (female) head of your household. A. B. '0 D. E. F. G. Please Professional and prOprietor of large business. Semiprofessional; supervisory position in a large business. Clerical and kindred workers. Skilled production work. Proprietors of a small business. Semi-skilled production work. Other ' indicate which letter correSponds to your total income before taxes last year. Less than $4,000 $4,000 _ $7,999 $8,000 - $9,999 $10,000 - $12,499 $12,500 - $14,999 $15,000 - $24,999 $25,000 and more indicate which letter corresponds to the source greatest part of your family income. Inherited wealth Profits and dividends Earned salary Earned hourly wage Other Please indicate which letter correSponds to the last year of school completed by male (female) head of your household. A. B. C. D. E. I'd Grade school or less Some high school Graduated from high school Some college Graduated from college Advanced degree 24. 25. 26. 27. Do you own your place of residence? Yes No If you 169 do not own a house, please go to the next question. If you own a house, how much do you think you could get for your house, if it were put up for sale today? A. More than $50,000 $35,000 - $49,999 $20,000 - $34,999 $15,000 - $19,999 $12,500 - $14,999 $10,000 - $12,499 Less than $10,000 rent a place, what is the monthly payment? More than $500 $250 - $499 $150 - $249 $100 - $149 $75 - $99 $50 - $74 Less than $50 indicate which letter corresponds to your dwelling type: A. Detached single dwelling Two family dwelling Multiple family dwelling Other 170 28. Which of the following best describes your dwelling area; i.e., the area in which you are presently residing? A. Comparable to area behind Saginaw Country Club Golf Course. B. Good suburban and apartment house area comparable to Saginaw Township areas. C. All residential area, larger than average spaces around houses, apartment area in good condition. D. Average residential neighborhood, no deteriora— tion in the area. E. Area not quite holding its own, beginning to deteriorate, business entering. F. Considerable deterioration in the area, run-down houses, semi-slum. G. Slum area. 29. Address: (optional) Thank you very much for your cooperation. All the information provided by you will be kept confidential. Your assistance in this research is greatly appreciated. APPENDIX D INTERVIEW SCHEDULE FOR THE MANAGEMENT 171 INTERVIEW SCHEDULE FOR THE MANAGEMENT Hello. My name is . I am conduct- ing this interview to gather data for a research project conducted with the Marketing Department of Michigan State University. The basic purpose of this interview is to obtain your vieWpoint regarding the meaning of certain items to the cus— tomers of your store. You will be asked to judge the meaning of these items on a series of descriptive scales. In giving your answers, please make your judgments on the basis of what you think these items mean to the customers of your store. Specific instructions are given at the beginning of the questionnaire. The important thing is that you try to answer each question as honestly and candidly as possible. Your opinions and only your Opinions are important. Under no circumstances will your individual responses be made available to anyone. If you have any questions while filling out this questionnaire, please feel free to ask. 172 173 MANAGEMENT‘S PERCEPTION OF CUSTOMERS' EVALUATION OF THEIR STORE IN THE SAGINAW AREA The following pages contain several items for you to judge in relation to what they mean to the customers of your department store. Each item is to be judged for a given set of scales. Here are some examples of how to use these scales in judging the items given to you. If you are rating the meaning of the size of your store to your customer and the scale is BIG-SMALL, it would look as follows: SIZE Very Quite Slightly Medium Slightly Quite Very BIG : : : : : : : SMALL If you believe that the customers view the store as very small, you would mark the scale as follows: Very Quite Slightly Medium Slightly Quite Very BIG : : : : : : : X SMALL If you believe that the customers view the store as slightly big, you would mark the scale as follows: Very Quite Slightly Medium Slightly Quite Very BIG : : : x : : . : : 'SMALL If you believe that the customers view the size of the store as medium, i.e. neither big nor small, you would mark the scale as follows: Very Quite Slightly Medium Slightly Quite Very BIG : : : : X : : : SMALL IMPORTANT: 1. Please place your marks in the middle of the scale spaces, not on the boundaries. : X, X THIS NOT THIS 2. Never put more than one mark for any one scale. 3. Be sure to mark every scale for each item--please do not omit any. 4. Please move directly from one scale to the next-edo not check back and forth. 5. Please try to fill in the scales as quickly as possible. 174 When you fill in the following scales for the items given below, please think of what they mean to the customers of your store in the Saginaw are Lower quality. Fully stocked. High prices com- pared to other stores. Merchandise on sale of regular quality. Stocks merchan- dise I like to buy. Leader in newest styles. Wide selection. Dependable products. Always satisfied with my purchases. Know their mer- chandise. Make one feel important. Courteous. Helpful, even if one is browsing. Available. Liberal in cashing checks. Liberal on exchange or returns. Warm, friendly store. Likely to meet friends in the store. Friendly to minorities. Likely to attract upper-class people. Likely to attract lower-class. Likely to attract middle-class. Believable advertisements. Easy to get a charge account. a. Very Never too crowded.: Easy to get theref Excellent parking. Save time by shopping there. Easy to find items one wants. Convenient to other stores for shopping. Quite Slightly MERCHANDISING SUITABILITY Medium Slightly I . Quite O. O. 0. CI SALES PERSONNEL .0 O. STORE CONGENIALITY LOCATIONAL CONVENIENCE I. High quality. Under stocked. Low prices com- pared to other stores. Merchandise on sale of below regular quality. Do not stock merchandise I like to buy. Absence of newest styles. Limited selection. Undependable products. Never satisfied with my purchases. Don't know their merchandise. Make one feel unimportant. Discourteous. Unhelpful, if one is browsing. Unavailable. Strict in cash- ing checks. Strict on exchange or returns. Cold, unfriendly store. Unlikely to meet friends in the store. Unfriendly to minorities. Unlikely to attract upper- class people. Unlikely to at- tract lower clas 5. Unlikely to at- tractmiddleclass. Misleading advertisements. Hard to get a charge account. Often too crowded Hard to get there. Poor parking. Can't save time by shopping. Hard to find items one wants. Inconvenient to other stores for shOpping. W-l-l- _u 175 Your position in the store: How long have you been employed with your store? Less than one year. More than one year, but less than three years. More than three years. Sex: Male Female Race: White Negro Mexican Other (8990375375— Please indicate which letter corresponds to your age category: A. Less than 20 years. B. More than 20 but less than 30 years. C. More than 30 but less than 40 years. D. More than 40 but less than 50 years. E. More than 50 but less than 65 years. ‘WJI . 65 years and older. Marital Status: Single Married Widowed Divorced-Separated Other Please indicate which letter corresponds to the last year of school completed by you. A. Grade school or less B. Some high school C. Graduated from high school D. Some college E. Graduated from college F. Advanced degree APPENDIX V EXPLANATION OF INDEX OF SOCIAL CHARACTERISTICS 176 _JL‘ . 44 EXPLANATION OF INDEX OF SOCIAL CHARACTERISTICS The Index of Social Characteristics uses four char— acteristics to classify individuals into various social classes. These characteristics are: occupation, source of income, house type, and dwelling area. For each character— istic rating is obtained on a scale from one to seven, The different ratings for each characteristic are shown in Table A on the following page. After obtaining rating for each variable for a respondent, the respondent's rating is multiplied by a weight and each product is totaled to obtain a weighted total rating. The weights used for each variable to develOp the index scores were obtained by Warner with the help of regression analysis. These weights are as follows: Occupation 4 Source of Income 3 House Type 3 Dwelling Area 2 The weighted total rating obtained through the above pro- cedure is converted into its social class equivalents with the help of Table B of this appendix. 177 .v,_.\ . A: 178 Table A.--Social class equivalents for I.S.C. ratings. Weighted Total of Ratings Social Class Equivalents 12 17' Upper class 18 22 Upper class probably, with some possibility of upper—middle class 23 24 Intermediate: either upper or upper—middle class 25 33 Upper-middle class 34 37 Intermediate: either upper—middle or lower—middle class 38 50 Lower-middle class 51 53 Intermediate: either lower—middle or upper-lower class 54 62 Upper-lower class 63 66 Intermediate: either upper-lower class or lower-lower class 67 69 Lower-lower class probably, with some possibility of uppernlower class 70 84 Lower-lower class Source: W. Lloyd Warner, Marcia Meeker, and Kenneth Eells, Social Class in America: A Manual of Procedure for the Measurement of Social Status (New York: Harper & Row, Publishers, 1960), p. 127. 179 Table B.--Scales for making primary ratings of the status characteristics of the I.S.C. Occupation Professionals and prOprietors of large businesses Semi-professionals and smaller officials of large businesses Clerks and kindred workers Skilled workers Proprietors of small businesses Semi-skilled workers Unskilled workers NH \lONU'lbL») o o o o O Source of Income Inherited wealth Earned wealth Profits and fees Salary Wages Private relief Public relief and non-respectable income gonna-comb: o o o o o c 0 House Type Excellent houses Very good houses Good houses Average houses Fair houses Poor houses Very poor houses \lO‘U‘luh-WNH o c o o o o o Dwelling Area 1. Very high; comparable to area behind Saginaw Country Club Golf Course 2. High; the better suburbs and apartment house areas comparable to Saginaw Township areas 3. Above average; areas all residential, larger than average space around houses; apartment areas in good condition 4. Average; residential neighborhoods, no deterioration in area 5. Below average; area not quite holding its own, beginning to deteriorate, business entering . Low; considerably deteriorated, runedown and semi-slum 7. Very low; slum Source: W. Lloyd Warner, Marcia Meeker, and Kenneth Eells, Social Class in America: A Manual of Procedure for the Measurement of Social Status (New York: Harper & Row, Puinshers, 1960), p. 123} q III...- III-II "‘I1111111111111“