.- I. - I. I I.- I III I I III. III II III. I I. I II II I III.- III... J. III.-. .UPIn II. It 3 I I ‘19 IIIIIIVAI I III III- 0‘0th I I. III... III-«I93. It I VII? lllkwI-II III- I .5... Ina-I III-III I-DnII-IIII I. III-IquIII. E. II . CII IIIIII II... .‘llt III. I I, . -I .. {III-III...“ III-III IIIIIIIIQILI. IIluII-IIIIIII- II. I IIuI-I‘IIII! IIIIII III. . cIIIII-II III III I . ...-. I- III-(111.1...- IIIII III..II..II-.I. )I IMIII III-I. -- . 2 .. . .- III-II} .95.“?- .-IIII£III.I.II [In-”III IIh-IIII IIIII III. III- II.I..I..I.I III. II. I I IfiI III..IIkflaI I III. III I I .3. . I... III III I-I-h-IILIBII .III... III Sufi-I, cl I‘N‘IfiI..I¢I-III IIIIII IIVII IIIIIIIICJII; .IIA ‘ I O . III~.\1¢I’I I .‘flIIIPII-IIIIIOI fling-III; I. I! .II . IIUoI‘ .... IzI II .I I- II III IIIU .1 I. III-«Jul II: I I. 1.. I. . I 1. IIIII - III I I II I - II. I I. , .I III I . IIIOII.II I .II hI-D \.4 HI . I I . . IIIIIIW I” IIIIIIIHILIINIJ III III}. «III: IIIIWIIIII III-o: :1.- IJC I- . I I Ificfib". . II“.- IuoIII III )I ‘11.}. I. d II. 5 YR" 1&1 . . I I. UIvI IFI.VIIII VIIIIl-nwlfiflllll I I JICTOFBWnIIuIII a I II . III-IIIIIII I . IIIIIIOOI I .II II a . I III ‘1'! it? I . IdIIIJIIfiIIIIIIaIINImPIhI-VI . II- IJ. c.431- I. II I.. III ”IIIIIIIIII I... III . III .I .I II IIIIU. III.III?§fiII II II IIInIIII I IIIIJIIIIIHI . , I I II..IIWUIIIIIIIIIIIIIIIIH.I. III I III. gang’- I I I I .I I I \o.‘ I . .HI II II II. III - IAIIO I I 7 I III. IIIOIII.IIHII1I. I III“! III I I I U 0 I I III I. I I I . “III“IIII. I I 11“; t I it ‘ I.III I I I - .I IwIIIIV II. IIIIInI $3! I IIIIHIhII-J IIIUHIIHHU‘IHIIIUI..II’II IIIIII- hr I. I .I..3.\ I. I III- IIQIIFIUII IIhI-II- . III-III \IIIEIIIIHNHII -I (Hun-III . IhnhDIIqufi IrIfiflrIJAII) II . I I... .t .I. III IIPI III- -IIIIII IIIIV 1.4.7 I.- -u- goxI II. Iu-IIIIIIIIIIJLI- IIHIHIIIIIII I IIIIIIHvIII I I II I I IIIUQIII I I .II I. I . . III-III It. I .I III! iééT.-IKIIIIIIII.IPEI!I I. .I -. {IIIIIII IPIJHNIPIII. -IIIIIIIII. III I I II IIIIII 3. - . II PIIIIIIJHIIIII IIIII.I III III-III ,IchIII IIIIIIIIIIIIIII I II-I I IflwIIIIIIII I "II II IIEIII III III III II. III IIIIII- IIIIVIIIuIIIII . III I . I .III IIIII. II II III, II . I. II’JIIIC I I! I. III. II III I II . III I. #I III. II I . - IHUHIIUEIIIII IUI III III I III-III - III-III PIII I.I I III. III , III III III .IIIIIIIIIII- I IhII . II III- I . III II III IIIIIIIIIIIJIIII III III IflIIII I I IIIJI III I ha a I II IIVIIIIII “MIIIIII III Ifl-IIWV.‘FIIIIIII IIIII IIIIII .. III IIIII I IIIHIIIIIHII III-In-HIIUIIIIInl II.I.I .IIIIvllIllflIlfivIIIII-Iu I I “NIQMIIIIfiIIIII I IIIIIIIINHIIII I ' I II]!!! III. III IIwIVIIIIIflI-II .III-IIIIIII-IIIIng IIIgIIIIInWMIIIIIII Ila-I III-I III-III IIIUIIIII I II- I I. I .I ., -I I% III III IIIIIIIIIuIII: -. IruII I- IIIoIIIl-I IIIIIIHIIIIIl-I III-III ... .I I . - . It...“ I ...... IIII. IIIIuI I III. II-.IIIIII.I I r . I n I IIIIIIflII-I it II, J” I EICIIIDIIII- IrII’IIIIIIIlI _ IlhanUIFIII‘II-qu II OI”.II.IIII I I . III I I I . .. 1 I.I..I0I.III IIIII I.I«.h.n.(I. InInuIIHIdIgIIUI II III I III-III I g .. I II I III IIII I III- I III. III I. .. III .. IIIIIIIIIIIIJHIIIIIII IIIIIInII- I {III-”Du" IIINHII III III III I I 4:1 I I INII‘ III. III. [fin-WI I III? III IIIIIII III I III IIifIIII Ina/I III-IIIIIII IIIIIIII \III Ibfl. I .III- III-I IIIIIW I. IIIInI-IIII I I.I.flIII.I.IIII...IIIuIIIIII.II2-uIII IFIIIIII. I I III .IIIOIVIII III I IIIII {I III-III IIIIIIII “I III III KIIIIIIII III.IIII¢IINIIJPIIIIIII1IIIIfiIIIIIIIIIII€IIII III III I I I III I n“)! I! IhI I .. I .IIIII-I I .1II..II\III-I. III II“ III I -I I. III-INI. .II ”HHHIIIIHII. I. II I IIIIwIIIIIN“\-II IIIIIIII I .I III ..I III.IIII.nmu III I III III- I I 1 III. II II II -IIIILI- III .1 . . fin“ IIIquII ”IV-IRII IIBfluII-H-INI. III-III «III-III“ III IIIIIIH IIIIIIIdII II .I .II II. I IlfiIII-IIEII IIIII. I .IIl-IIIIIIII "I “III-III I IIIIIIIIIIIII.IIJHHHII III III III. I IIIIIuhurII II III-IIIHIINIMHHIIIIIIIII III-Im III I IIIIMI II III I I III-III!I I‘IIIIIIIIIIJI III.I I I III. I IIIIUIIIIIII III - IFI H .IIu I-I “HUN-I. IIIHVIIIHVII III. nIIIIIIIIII IIIIIIIII .I..II. IIIIIIIIHIIIIHHIIIIII I-.I II IIIIIIIIIIIIIIM III IIIII-IIIHH III-I1 lull I .USKIIIIIIIIIIIIIII.III.III.II II III! I IIIIIIIIIIIII IIIIII III-I II III I III III IIIII .I IIHNIII.II I IIIIIIIII II II II. III-III I I “IIII II. IIIIIIIIIIwIIII- IIIIIIII III IIIouIIIIJII III-II III III I III IHHII- III II I IIIIII‘IIIIIIM I.) .I-IIIIIII IIIII I- III-III IIII IIIIHIIUHflu-HIII. .I-HUII-II.III IIIIIIE III I ,I .III I. IHI I - I I II III IIIIR. I III:IIIIII. In.- . I .IIIPIIIIIIIIInfi. II IInII II-III. IIII-II-IIIIIIIIIIII. I IIIXI-IIIII.I .. I Illl I‘IIIII‘IIC.IIIII I I III III I I I I II IIIII}IIIIIIIIIII" II-IIIIIII I II ...- I. IIIIIIIIII II II .. I I I. I . - - . I-flIIIIIIIPIIIIII-II I II... I udlfl..fl.h«II.I.I.I-II III-INN“? III - - I. .III. I - I- I I - .I .I M” IIII.IILI | IIIH-I-III r-II4.I- . I I II - I . II I. I I IIIIIIIIlfll-HM" III I II .III III I I I I- III I .. II III I IIInilI III I I I I IIIIIIIIII. I I I. \4 II III I. I I \ I I I III III-II I III-III III III! I - . I I J. I II I. UIIIII I I III". I I IIIIIINHIIII - III-III? I .- I” I-III-IIII. IMF-"III. IIIIIIIIIII I I In.- . , - I III- - .IIIII II\I|IIIII.--‘II.II NIIIIIIIIIIIIIIIUIIIIHIIIIIII I .OI-II . II II I I I I III I II III. III I - I III-IIIHIIUOIIIHII I. n . - I I II . -I III “INHIIIIIIIII IIIx-IlIII-n- Jun-IIWIII . I I I I . I III“ II .IIIII“ I IIIJNIIIIIIIIIIII III 1.4..“ III III I I II III I. I7. I, \II-I I’ K. IIIIIIII- .III I II IIIIIIIII III .II I I IIIIIIIQUII.“ - III HUIIIII III -v .IkIIhmIIIIII-I.I.II IIJIIIIIIIMIIIIruII-"WufiIIHII-I III-III III IIILIIIIII III III: II. I . - I 1'. . I II I; wlIIIIIIII flIIIIH ”II IIIHIIIIIIUIIIII. .III II III I. IIIIWIHINIIIIIflrr . IIII IfirIIII-IIIUUPIIMIIIIII III) IIIIII..II I III III} I III II. III III , III IqhhflufiddIz III III I III . I I I .-IIIII.-.I I IIIII IIIIII IIIIIIIII II II .IIN IIuNHHUflIIus-“HIIII. III IIHIHVIUu-IIIIIMNUIVIIIII IIINIIIIIII-I III I [HAHN-III! IIIII III III-In -..I”.II I. IIIlIo-IIII IIIIIII I] III Ii fiIIIIIIIII III I I III-III“ I . IIIII I I 91% II III-IIIINII I I IIIIIlII I141 I I III .IUI‘ I-I IIIIIIII III I I. II III I III I I I l I II I“ II I IIIII. I I . I I - - I I .II III-IL I. I I III .I- h v I .I III . I I 4 I I :I I FII I III I I l I II. III III.“ III. I I III-.IIIIIIPIIII I .- .- - . EIIhhnII I .uIIIIp-MIINMIJII II. I II I I II II II In. Ill“ I grun‘hrI I - "m”.IIII IIKIwI I .MKIIL-I II I LII. II. -- .I r.- ILFIIII II,IIIII I.- III I.IIII'IIII II II II II II - I A,“ | . II. o» l (I III I I I III- I I I . I I I III IIII I .I-I II .I- I I I I I I. VII-“J I I .I .- II I I . I I II I III -, -. -- -I III! II. I. II- I I I I I .I I. “I I l I II I - . IIIIII-IH-I I IIIIIIIquI I 4 “HUI-II I III . I I", III I II I .. I. I I III-II III! ,IIIIHQI‘IIIII III. I. A I. III-III II:- IIAWIIm-I . I - III-II Iv-IIII -I II~IIII III III \ III-II . I I.I-.IIIrI-IIILIII III-III . II.II .- III I , I . :- ..- IIIII-iI-IIIII III-III ..I.I11.1..I.I I IIIVIIIIIIIIIIIUIIEHII. III? :3 I. \ , .1 . . I.IJOI I I . .IIIILII I. I II.WIII-.IlUII.-1.IIIW..I. (I. .VIIIIIIIII)\d III . l L1. (1.-....1 . .I. . I II. III .......I.I.I I III I I . . .I I I IIIII .I I.ILII III .KIII III III II II IIL I. ... «- I r .4 I III. II. ‘1.- no. I 1 -. JIAIL.II.-IIE uh... .IIUUI .I WJJ.-.III)L\1.I..III\ VSIIHHICII. .I I.III-uI..IIII IWIIIIIIJHwI .IIII. IIIIIIIIII I II .I III I .. -: I , _ I . .I-I I... - - . I I I I . .. A -....I-. :- I.-....I.I .I. . It... I I {III II III-NI. £5 . .. .. I I , _ : I T .— U...2.:I.-;I..I.- .. - ..-- .I; .,I-. i. I. 5.... .I I I 3.. I .3 III I.CI-II..IIIII I III . , vi. . - I . . ‘l «III-”I‘fi- Ir.rI.IL 11111111111111111111111111111 WES-Ls LIBRARY , 3 1293 104 mm gen State University This is to certify that the thesis entitled A COMPARISON OF MANAGERS'..ANDSiGONSUMERS" PERCEIVED IMAGES OF DISCOUNT DEPARTMENT STORES: A MULTI-ATTRIBUTE ATTITUDE MODEL presented by Marianne Young Mahoney has been accepted towards fulfillment of the requirements for . MaSEELQLALtS degree in MM Textiles Major prfifessor DateflflfilflhfiLlLlflfiZ. 0.7639 MS U is an Affirmative Action/Equal Opportunity Institution . .10 N0 .. . flCULT FOR Olnt- .. ' NOTE; Hichinar ‘ ‘c . ‘r a mis MSU LlBRARlES r MANY OTHERo EE’IT.- mus _ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is 'returnedeafter the date stamped below. 1&1 1S1 1 1175/16. '1’ N iiijl/llxj. A COMPARISON OF MANAGERS' AND CONSUMERS' PERCEIVED IMAGES OF DISCOUNT DEPARTMENT STORES: A MULTI-ATTRIBUTE ATTITUDE MODEL By Marianne Young Mahoney A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS College of Human Ecology Department of Human Environment and Design 1982 ABSTRACT A COMPARISON OF MANAGERS' AND CONSUMERS' PERCEIVED IMAGES OF DISCOUNT DEPARTMENT STORES: A MULTI-ATTRIBUTE ATTITUDE MODEL By Marianne Young Mahoney The purpose of this study was to investigate the perceived images of three discount department stores and the success with which these retailers offer the attributes desired by the consumer. The research objectives for this study included: (1) identify salient evaluative criteria for discount department stores; (2) measure con- sumers' and managers' professed overall impressions of each discount store in the survey; (3) identify consumers' and managers' perception of the ideal amount of each store attribute; and (4) compare consumers' and managers' perceived images of each store and the success with which these retailers offer the desired amount of each attribute. The Beckwith and Lehmann multiple attribute attitude model was used to evaluate consumers' and managers' professed overall impressions toward three competing discount department stores. Results of multiple tétests indicated that managers and consumers possess significantly different overall impressions of each store. Significant differences were also seen between the ideal amount of each attribute and the amount offered by the retailers. ACKNOWLEDGMENTS Many people assisted in the completion of this thesis. Several people however must be acknowledged due to their unending guidance, support and encouragement. A special thanks is given to Dr. Brenda Sternquist Witter, my major professor, for her support, encouragement, and friendship throughout my graduate program and research. Dr. Dawn. Thorndike Pysarchik, Merchandising Management, through her expertise in Consumer Behavior, gave valuable information concerning the survey instrument and data analysis. Dr. M. Bixby Cooper, Marketing and Transportation, assisted greatly in the methodology and marketing strategies and tactics used by retailers. These three committee members shall always be remembered for their time, encouragement, and friendship throughout my program. Acknowledgment and thanks is given to the retailers who partici- pated in the study. Their cooperation assisted in understanding a typically neglected area of store image research. A special thanks is given to Elaine Johnson-Hahn, Bonnie Davis and Maureen Henry. These three provided friendship and emotional sup- port throughout this study. Finally, to my best friend, Louis Mahoney, to whom this disser- tation is dedicated--words cannot express my thanks for your constant emotional support and assistance. This thesis is just as much a part of you as it is of me. Thank you Lou, for everything. ii ACKNOWLEDGMENTS Many people assisted in the completion of this thesis. Several people however must be acknowledged due to their unending guidance, support and encouragement. A special thanks is given to Dr. Brenda Sternquist Witter, my major professor, for her support, encouragement, and friendship throughout my graduate program and research. Dr. Dawn: Thorndike Rysarchik, Merchandising Management, through her expertise in Consumer Behavior, gave valuable information concerning the survey instrument and data analysis. Dr. M. Bixby Cooper, Marketing and Transportation, assisted greatly in the methodology and marketing strategies and tactics used by retailers. These three committee members shall always be remembered for their time, encouragement, and friendship throughout my program. Acknowledgment and thanks is given to the retailers who partici- pated in the study. Their cooperation assisted in understanding a typically neglected area of store image research. A special thanks is given to Elaine Johnson-Hahn, Bonnie Davis and Maureen Henry. These three provided friendship and emotional sup- port throughout this study. Finally, to my best friend, Louis Mahoney, to whom this disser- tation is dedicated--words cannot express my thanks for your constant emotional support and assistance. This thesis is just as much a part of you as it is of me. Thank you Lou, for everything. ii TABLE OF CONTENTS CHAPTER Page I. INTRODUCTION ........................................... 1 Statement of the Problem ............................ 4 Justification for the Study ......................... 4 Research Objectives ................................. 5 Conceptual Model .................................... 5 Definitions ......................................... 6 Overview ............................................ 7 11. REVIEW OF LITERATURE ................................... 9 Store Image ......................................... 9 Department Stores versus Discounters ............. l0 Consumers' Perception of a Store's Image ......... l2 Management's Perceived Image of a Store .......... l4 Image Measurement ................................... 16 Multi-attribute Models ........................... l6 Multi-dimensional Scaling ........................ 18 Multiple Indicator Approach ...................... Zl Multiple-cue Probability Attitude Model .......... 22 Two-factor Theory ................................ 23 Importance of Differential Heights in Attitude Models ........................................ 25 Selection of Image Attributes .................... 27 Summary ............................................. 31 III. METHODS AND PROCEDURES ................................. 32 Selection of Discount Stores ........................ 32 Selection of Salient Store Attributes ............... 33 Multi-attribute Attitude Model ...................... 37 Design of the Questionnaire ......................... 38 Sample Selection .................................... 4O Consumers ........................................ 40 Methods and Procedures ........................... 4] Managers ......................................... 42 Hypotheses and Statistical Analysis ................. 43 iii CHAPTER IV. RESULTS AND DISCUSSION ................................. Introduction ........................................ Demographic Information ............................. Marital Status ................................... Number of Children Currently Being Supported ..... Occupation ....................................... Educational Level of the Head of the Household... Family Income .................................... Age of Respondent ................................ Sex of the Respondent ............................ Multiple Regression Analysis ........................ Consumers ........................................ Store One Managers ............................... Store Two Managers ............................... Correlation Coefficients ............................ Consumers ........................................ Store One Managers ............................... Store Two Managers ............................... Testing the Hypotheses .............................. V. DISCUSSION ............................................. Multiple Regression Analysis ........................ Correlation Coefficients of Salient Store Attributes Hypothesis 1 ........................................ Store One ........................................ Store Two ........................................ Store Three ...................................... Hypothesis 2 ........................................ Hypothesis 3 ........................................ Ideal Store Attributes .............................. Store One ........................................... Store Two ........................................... Store Three ......................................... Summary of Interpretations .......................... VI. SUMMARY AND RECOMMENDATIONS ............................ Limitations ......................................... Recommendations ..................................... LIST OF REFERENCES ............................................ APPENDIX--STUDY QUESTIONNAIRE ................................. iv Page 120 122 124 125 127 134 TABLES 3.1. 4.1. 4.2. 4.3. 4.4. 4.5. 4.6. 4.7. 4.8. 4.9. 4.10. LIST OF TABLES Tabulated Results of Random Sample Interviews to Deter- mine Evaluative Criteria ............................... Summary of Demographic Information ..................... Consumers' Impression of Store One According to the Beckwith and Lehmann Model. Stepwise Regression Analysis ............................................... Consumers' Overall Impression of Store Two According to the Beckwith and Lehmann Model. Stepwise Regression Analysis ............................................... Consumers' Overall Impression of Store Three According to the Beckwith and Lehmann Model. Stepwise Regression Analysis ............................................... Store One Managers' Overall Impression of Store Onei According to the Beckwith and Lehmann Model. Stepwise Regression Analysis .................................... Store Two Managers' Overall Impression of Store One According to the Beckwith and Lehmann Model. Stepwise Regression Analysis .................................... Store Two Managers' Overall Impression of Store Two According to the Beckwity and Lehmann Model. Stepwise Regression Analysis .................................... Summary Table of Correlation Coefficients of Consumer Responses .............................................. Highly Correlated Attributes for Store One as Seen by Store One Managers ..................................... Highly Correlated Attributes for Store Two as.Seen by by Store One Managers .................................. . Highly Correlated Attributes for Store Three as Seen by Store One Managers .................................. . Summary Table of Highly Correlated Attributes as Seen by Store Two Managers.................................. Page 35 48 56 58 59 61 63 64 66 67 68 69 71 TABLE 4.13 4.14 Managers' Overall Impression of Store One. T-test ..... Store One and Store Two Managers' Overall Impression of Store Two. T-test ........................ . ............ . Store One and Store Two Managers' Overall Impression of Store Three. T-test ................................... . Managers Professed Belief of the Ideal Point for Each Attribute Offered by an Hypothetical Ideal Discount Store. T-test ......................................... . Managers' Professed Belief of Each Attribute Offenad by Store One. T-test ..................................... . Managers' Professed Belief of Each Attribute Offered by Store Two. T-test ..................................... . Managers' Professed Relief of Each Attribute Offered by Store Three. T-test ................................... . Consumers' Overall Impression Toward Store One, Store Two, and Store Three. One-way Analysis of Variance.... . Managers' and Consumers' Professed Beliefs of the Amount of Each Attribute Offered by Store Three. One- way Analysis of Variance ............................... . Managers' and Consumers' Ideal Amount of Each Attribute Offered by the Ideal Discount Store. T-test ........... . Managers' and Consumers' Overall Attitude Toward Store One as Defined by the Beckwith and Lehmann Attitude Model. One-way Analysis of Variance ................... . Managers' and Consumers' Overall Attitude Toward Store Two as Defined by the Beckwith and Lehmann Model. One- way Analysis of Variance ............................... . Managers' and Consumers' Overall Attitudes Toward Store Three as Defined by the Beckwith and Lehmann Model. One-way Analysis of Variance ........................... . Managers' and Consumers' Overall Impressions Toward Store One. One-way Analysis of Variance ............... . Managers' and Consumers' Overall Impressions of Store One. T-test ........................................... vi Page 73 74 75 77 78 8O 82 83 85 87 88 89 89 90 91 TABLE 4. 01 01 01010101 euro 28. .29. .30. .31. .32. .33. .34. .35. .36. One-way Analysis of Variance Between Managers' and Consumers' Overall Impressions of Store Two ............ Managers' and Consumers' Overall Impressions of Store Two. T-test ........................................... Managers' and Consumers' Overall Impressions of Store Three. One-way Analysis of Variance ................... Managers' and Consumers' Overall Impressions of Store Three. T-test ......................................... Managers' and Consumers' Professed Beliefs of the Amount of Each Attribute Offered by Store One. One-way Analysis of Variance ................................... Managers' and Consumers' Professed Beliefs of the Amount of Each Attribute Offered by Store One. T-test. Managers' amd Consumers' Professed Beliefs of the Amount of Each Attribute Offered by Store Two. One-way Analysis of Variance ................................... Managers' and Consumers' Professed Beliefs of the Amount of Each Attribute Offered by Store Two. T-test. Managers' and Consumers' Professed Beliefs of the Amount of Each Attribute Offered by Store Three. T-test ................................................. Summary Table of Stepwise Regression Analysis .......... Summary Table of Highly Correlated Attribute .......... Summary Table of Highly Correlated Attributes .......... Correlation Coefficients of Store Two Managers' Responses .............................................. Summary Table of Hypotheses and Statistical Analysis... Summary Table of Respondents' Overall Impression and Attitudes Toward Each Discount Store as Defined by the Beckwith and Lehmann Model ........................ ..... Summary Chart of Managers' and Consumers' Beliefs Con- cerning the Ideal Point and the Attributes Carried by Each Store in Comparison to the Hypothetical Ideal Store .................................................. vii Page 92 92 93 94 95 97 99 101 102 106 108 109 110 112 116 118 CHAPTER I INTRODUCTION Retailers have annually spent billions of dollars in an attempt to portray a positive store image (Singson, 1975). The importance of portraying a positive image can be realized through an analysis of the competitive structure of the retail industry. Rising prices, product shortages, increasing wages, and consum- erism issues are all contributing to a profit squeeze for retailers. The creation of a more favorable store image is one way they can obtain a difficult-to-duplicate differential advantage over competitors. It is logically assumed that the development of a store image consistent with the needs of the target market segment can lead to increased sales and profits. (James, Durand, and Dreves, 1976, p. 3.) The success of portraying a favorable image can be determined by analyzing consumers' and managers' professed images of each store and examining their attitudes concerning the selection and weight of attributes offered by the store. Research on the image protrayed by retailers has been numerous. Pathak et a1. (1974-1975) noted however, that few studies have been conducted which focus on a comparison of how store managers and con- sumers view a store's image. The studies which have been conducted were concerned exclusively with the traditional department store or a composite of the various types of stores such as a department store, specialty store and discount store. Because competition is the most intense among stores within a specific category, it is relevant to examine store image at this level. One segment of retailing, the dis- count department store, has become a major force in American retailing, yet the study of competing discount department stores' image has been relatively unexplored. Many changes in the operation of the discount store have taken place since its beginning in the early 1950's. Originally the discount retailers merchandised hard goods and general merchandise; today they sell items such as apparel, health and beauty aids, garden supplies, household items and foodstuffs. With the expansion of merchandise offerings, services are now being offered which were once unavailable in a discount store. Additional employee service, acceptance of credit, easy exchange policies and in-store restaurant facilities assisted in the wide consumer acceptance and positive image associated with the discount retailer. Due to the various improvements, the number of consumers patron- izing discount department stores is constantly increasing. Consumers from all social classes will at least browse in a discount store (Dreyfus, 1980; Tuhy, 1980). With the rapidly changing consumer atti- tudes toward discounters, it is important to understand what factors affect the selection and weighting of the attributes which form the perceived image of that store. (Bohr, 1980; Dardis and Sandler, 1971; Tuhy, 1980.) Until now, the discounters have been very successful. However, with their move toward trading up on services offered, merchandise selection and quality, they must be careful not to break their traditional characteristic low margins which allow them to offer lower prices. A synthesis of the department store and discounter market segment would result if their prices would rise substantially (Goodman, 1972; Weale, 1969). Because of the many changes in the discount retailers' offerings and subsequently the changes in their marketing tactics, a closer, more accurate assessment of needs and desires of the target market must be made (Hirschman et al., 1978). As a result of the changes in the strategies and tactics of the discount retailer, it is imperative that the management of the traditional department store look closer at the consumer's attitude toward the discount stores. The traditional department stores must evaluate the offerings of the discount store because they must compete with these mass retailers for the limited disposable income of changing consumers (Goodman, 1972). Consumers select one store over another because of specific store offerings which are referred to as store attributes. Some con- sumers shop at a store because of the brand name merchandise carried while others may shop there because of the services offered or the locational convenience. Due to the possible differences in motivating factors associated with shopping behavior, the various store attributes and the importance of store attributes to consumers and retailers should be evaluated. Without accurate information concerning the attributes desired by the target market and the strategies and tactics performed by the competitors or potential competitors, the destruction of the company is a very real possibility (Brown and Fisk, 1965). Statement of the Problem Discrepancies in the perceived store image and that desired to be portrayed by the company can be devastating. With the rapid growth of discount department stores, an accurate analysis of consumers' perceived image of these stores is needed. More specifically, an analysis of managements' versus consumers' image of the discount retailers would assist retailers and marketers in satisfying their target market. Image as perceived by consumers would assist retailers and marketers in satisfying the needs of the consumer. Justification for the Study The majority of attitudinal studies pertaining to a store's image have focused on the consumer's perceived image of traditional department stores. These studies deal with only a small portion of the necessary research. Pathak, Sweitzer, and Crissy (1974-1975) theorize that research concerning management's perceived image of their store is equally important. More specifically, a comparison of con- sumers' and managements' views of discount stores is needed. An analysis of management's professed image of the store as compared with consumers' professed image of the store will assist retailers in determining whether or not the desired store image is ~being effectively portrayed. The results could be used as a guide for the retailer in determining future marketing objectives, goals, strategies and tactics. One method for identifying the consumers' and retailers' pro- fessed store images is through utilizing ideal weights of each attribute offered by the store. The ideal weight will indicate the amount of each attribute desired by the consumer. An understanding of the ideal amount of each attribute will allow retailers to better cater to their customers (Brown and Fisk, 1965). Research Objectives The purpose of this study was to investigate the perceived images of three discount department stores and the success with which these retailers offer the attributes desired by the consumer. The research objectives for this study included: (1) identify salient evaluative criteria for discount department stores, (2) measure consumers' and managers' professed overall impressions of each discount store in the survey, (3) identify consumers' and managers' perceptions of the ideal amount of each store attribute, and (4) compare consumers' and managers' perceived images of each store and the success with which these retailers offer the desired amount of each attribute. Conceptual Model Fishbein's original and extended models have been used exten- sively by researchers to assess attitudes, normative beliefs and con- sumers' motivation to comply. Strengths of Fishbein's model include widespread empirical testing, quantitative measurement and easy adaptation to various situations. A major limitation of this model however is that it does not measure the consumer's ideal point for each attitude attribute. For each attitude attribute, the consumer has an ideal point; a certain amount of an attribute is perfect; any more than the ideal amount of an attribute is saturation whereas any less than the ideal amount represents a deficiency in that attribute. For example, some customers enjoy a certain level of sales service. However, the customer may not necessarily want three sales clerks monitoring the entire shopping experience. At the same time, the customer does not necessarily want to have to search for a clerk for assistance. Because of the relevance of the ideal point of an attribute in evaluating attitudes, Beckwith ahd Lehmann (1973) developed an attitude model which includes the ideal point. The variance between the ideal point and the actual point is determined by subtracting the ideal point for an attribute from the perceived weight of the attribute. The dif- ference will show the degree to which the store successfully satisfies consumers' desired level of each store attribute. Beckwith and Lehmann theorize that this difference will provide additional atti- tudinal information which is not measured in other multi-attribute attitude models. The method for operationalizing this model will be discussed in Chapter III. Definitions Image is an abstract term used to quantify or describe atti- tudes. Store Image consists of distinguishing store attributes which contribute to a person's professed impression of that store. Aggregate Department Store Imagg_is a summation of the sample population's image for a particular store. The aggregate department store image for this study included the summation of consumers' and managers' images for each of the three stores (Pathak, 1971). Store Attribute is a specific characteristic of a store. Store attributes analyzed in this study included merchandise selection. value for price, employee sales service, store atmosphere, locational convenience and quality. Salient Store Attribute is an attribute offered which is viewed to be important but not identical in offering between stores. Attitude is an inferred state of readiness to react in an evalu- ative way toward an object in a situation (Berry, 1969; Crissy, 1971). Attribute is a dimension which contributes to an attitude, impression, or function of an object. An attribute may refer to the entire object or parts/functions of the object. Stimulus is a combination of motivational processes involving emotion, striving and cognitive processes such as thought and memory. Discount Store is a departmentalized retail establishment operat- ing at uniquely low profit margins (Discount Merchandiser, 1967). Overview Chapter II, the Review of Literature, is divided into five sections. Store image is introduced in Section One. The effbct of experiences, attitudes and attributes on the perceived image of a store is addressed. A discussion of store image: department stores versus discounter is presented in Section Two. New trends by these retailers, their competitive fight for the limited consumer dollars and how their strategies affect one another are then outlined. Consumers' per- ception of a store's image is discussed in Section Three and manage- ment's perceived image of a store is addressed in Section Four. The various types of store images, what helps form images and how a store image affects both the behavior of consumers and managers is dis- cussed. The various methods of measuring image are presented in the last section where the characteristics and strengths of image measure- ment models and the selection and weighting systems of attributes to study image are discussed. CHAPTER II REVIEW OF LITERATURE Store Image Image is comprised of tangible and intangible perceptions, iactions and objects. Factors which help form images include actual «:haracteristics, perceived characteristics, past experiences and [Dersonal values. Hirschman, Greenberg and Robertson (1978) define :store image as the "... 'personality' the store presents to the public (or 'a complex of meanings and relationships serving to characterize ‘the store to the populace'" (Hirschman et al., 1978, p. 3). The «consumer's image of a store is formed through experiences associated \with that store. Positive experiences, characteristics and values associated with a store will typically generate a positive store image (as defined by that consumer. Consequently, negative experiences, «characteristics and values, whether perceived or actual, will generate .a negative store image as defined by the consumer (Kunkel and Berry, 1968). More specifically, the brands and products carried, the archi- tectural departmental layouts, sales force, and quality and style of the promotion and advertising influences the consumer's perceived image of the store (Crissy, 1971; Hirschman, 1978; Kunkel and Berry, 1968; Martineau, 1958; Rich, 1969). CHAPTER II REVIEW OF LITERATURE Store Image Image is comprised of tangible and intangible perceptions, auctions and objects. Factors which help form images include actual (:haracteristics, perceived characteristics, past experiences and [Dersonal values. Hirschman, Greenberg and Robertson (1978) define sstore image as the "... 'personality' the store presents to the public «car 'a complex of meanings and relationships serving to characterize the store to the populace'" (Hirschman et al., 1978, p. 3). The (:onsumer's image of a store is formed through experiences associated invith that store. Positive experiences, characteristics and values iassociated with a store will typically generate a positive store image 635 defined by that consumer. Consequently, negative experiences, 1:haracteristics and values, whether perceived or actual, will generate a negative store image as defined by the consumer (Kunkel and Berry, 1968). More specifically, the brands and products carried, the archi- ‘tectural departmental layouts, sales force, and quality and style of the promotion and advertising influences the consumer's perceived image of the store (Crissy, 1971; Hirschman, 1978; Kunkel and Berry, 1968; Martineau, 1958; Rich, 1969). 10 Store image is the result of a composite of attributes; these iattributes are weighed in relation to their importance to the consumer. ‘The store with attributes, real or imagined, which are deemed important ‘to the customer will typically portray a favorable store image as «defined by that consumer. "An aggregate department store's image is ‘the summation of all the customers' images of the store" (Pathak, 1971, p. 32). Many dimensions, such as store atmosphere, merchandise selec- ‘tion, quality and the level of sales service all contribute to a store's image. The product offerings and display techniques are tangible .strategies which can easily be duplicated and manipulated by retailers .and their competitors. The merchandise offering and floor displays of «competing stores are also easily copied. Because of the potential for similarity between stores, each retailing establishment must utilize a strategy which will distinguish itself from its competitor. Accord- ing to James, Durand and Dreves (1976), portraying a favorable store image is perhaps the best technique for differentiating one retailer from the masses. Due to the possible complexity of the various dimen- sions, an analysis of only the aggregate store image is insufficient (Cohen, 1967). Each store attribute and the dimensions must be studied. By doing so, the retailer will have a better comprehension of why the aggregate department store image exists. Department Stores versus Discounters Expansions to additional branch retail locations is the most common form of increasing possible sales and profits. Due to the rising cost in building and maintaining stores, however, the trend toward 11 trapid expansion has slowed dramatically.) Rather than increasing their ssquare footage by building additional branch locations, retailers have ‘increased the space productivity and efficiency of each existing store through remodelled interiors. Traditional department stores have always catered to the customer; plush fitting rooms, abundance of sales service and credit acceptance is the norm. Furthermore, these retailers enhance their merchandise «offerings by creating moods through the use of effective interior design. ‘The effects of store atmosphere, once ignored by discounters, are now being seriously considered by retailers. A trend has begun in which «discount retailers as well as traditional department store retailers pay extremely close attention to space productivity, efficiency of store lay- out, and the effect that lighting and color have on shopping behavior (Chain Store Age Executive, 1978). Due to the move toward renovation as opposed to building additional branch locations, the effect of store iatmosphere, once typically ignored by discounters, is now of concern. ‘The philosophy that more branch locations is better no longer holds true. [fiscounters are trading-up on merchandise, services offered and store iatmosphere and they are slowing down the rapid expansion of additional branch locations (Chain Store Age, 1978). The recent changes made by discounters to trade-up, both in mer- <:handise selection as well as store atmosphere, have strongly affected ‘the traditional department store. The original image that discount :stores cater solely to the lower social classes no longer holds true. 'The increasingly favorable image of discount department stores which cater to many social classes is a result of several factors (Bohr, 1980). 12 (Once known typically for their hard lines at low prices, discount 'retailers cater to the fashion-conscious in soft lines as well. Some «discounters even have restaurant facilities as well as snack areas as an attempt to make the consumer's shopping experience a pleasant one (Goodman, 1972). Additional services such as acceptance of credit cards, layaway, and ample free parking are helping discount retailers portray this positive image (Dreyfus, 1980; Tuhy, 1980). Consumers' Perception of a Store's Image Value systems form the basis for attitudes and beliefs which in 'turn form images. The beliefs are part of a person's attitude (Crissy, 1971). Attitudes and beliefs prepare a person for some kind of action. 4A5 common attitudes toward an object (institution, product, or person) tare formed, an image is created (Nelson, 1962). In essence, the image is an abstract term used to quantify or describe attitudes (Berry, 1969; Crissy, 1971). Pe0p1e perceive an image based upon their beliefs. 'These beliefs help form images through emotional responses and intuition in addition to factual knowledge (Nelson, 1962; Walters, 1978). Because of differences in lifestyles, personalities, needs and desires, people's values, attitudes, beliefs and professed images of an object differ (Crissy, 1971). Rich and Portis (1964) identified three overall store images: high fashion appeal, price appeal and broad appeal. In a survey con- ducted by the researchers, style and quality of merchandise, services offered and store reputation contributed to a strong high fashion store image whereas consumers' image of a price appeal store dealt strictly 12 Once known typically for their hard lines at low prices, discount retailers cater to the fashion-conscious in soft lines as well. Some discounters even have restaurant facilities as well as snack areas as an attempt to make the consumer's shopping experience a pleasant one (Goodman, 1972). Additional services such as acceptance of credit cards, layaway, and ample free parking are helping discount retailers portray this positive image (Dreyfus, 1980; Tuhy, 1980). Consumers' Perception of a Store's Image Value systems form the basis for attitudes and beliefs which in turn form images. The beliefs are part of a person's attitude (Crissy, 1971). Attitudes and beliefs prepare a person for some kind of action. As common attitudes toward an object (institution, product, or person) are formed, an image is created (Nelson, 1962). In essence, the image is an abstract term used to quantify or describe attitudes (Berry, 1969; Crissy, 1971). People perceive an image based upon their beliefs. These beliefs help form images through emotional responses and intuition in addition to factual knowledge (Nelson, 1962; Walters, 1978). Because of differences in lifestyles, personalities, needs and desires, people's values, attitudes, beliefs and professed images of an object differ (Crissy, 1971). Rich and Portis (1964) identified three overall store images: high fashion appeal, price appeal and broad appeal. In a survey con- ducted by the researchers, style and quality of merchandise, services offered and store reputation contributed to a strong high fashion store image whereas consumers' image of a price appeal store dealt strictly 13 with low prices and bargain merchandise. More specifically, consumers envisioned service, quality and style with the high fashion store but related only savings with price appeal stores. Although price appeal stores did not portray a favorable image concerning the level of sales clerk services offered, they were deemed to have a more positive image than department stores in services other than sales clerks such as delivery, telephone orders, charge accounts, reliability and overall service. The researchers hypothesized that the image portrayal for price appeal stores may have been due to the expectations of the cus- tomers for each store. More specifically, consumers who pay more for merchandise demand a higher level of customer service. Broad-appeal stores have characteristics which fall between the fashion—appeal stores and price appeal stores. These stores tend to offer some degree of fashion merchandise as well as bargains. Price, reputation for bargains and store location were key characteristics that consumers stressed in a broad-appeal store. Many researchers (Bellenger, Steinberg, and Stanton, 1976; Grubb and Grathwohl, 1967; Lindquist, 1974-1975; Mason and Mayer, 1970; Myers, 1968; Nelson, 1962; Walters, 1978; Weale, 1961) addressed the relationship between self-concept of consumers, images and refer- ence groups. The concept of self-image is closely related to that of store image and patronage. Pierre Martineau (1958) theorized that the consumer typically will patronize stores which depict an image compli- mentary to that of their desired selfsimage. This desired self-image is typically not congruent with their economic standards. Therefore, in order to be able to afford the higher status and higher priced l4 merchandise, the consumer may purchase higher priced, quality, and/or styled merchandise in smaller quantities or purchase extremely cheap products which are deemed unimportant. As a result, the consumer would have enough money to trade-up on products which are deemed to be sig- nificant indicators of his/her desired self-image (Bohr, 1980; Dreyfus, 1980; Tuhy, 1980; Weale, 1961). Purchasing "no brand" staple food products in order to have enough money to purchase designer jeans is an example of compensating on lower quality and price on one product in order to afford higher status and quality merchandise in a different product category. Management's Perceived Image of a Store Studies dealing with consumers' view of store image have been numerous. The majority of these studies have concentrated on the assessment of consumers' professed image of a store. However, Pathak et a1. (1974-1975) noted that from a list of twenty studies dealing with store image, only a few compared store image from the managerial versus consumer viewpoint. Although the existing studies of consumers' and management's views of a store image have been few, the researchers have drawn similar conclusions. Devendra Pathak (1972) conducted a study which dealt with an analysis of consumers' versus management's image of four department stores. The department stores investigated included one high status store, two medium to low status stores and one discount department store, and an hypothetical "ideal" store. The image attributes: merchandising suitability, sales personnel, store congeniality, and 15 locational convenience were measured by utilizing thirty bipolar semantic differential scales. The hypothesis of no difference between consumers' and management's perceived overall images of the store was tested. The researcher concluded that there was in fact a large difference between managers' and consumers' professed images of each store. The greatest differences in opinion concentrated on intangible attributes such as store congeniality. Pathak concluded that merchandise suitability was the only tangible attribute which showed a positive correlation between consumers' and management's views. In a retail department store image study conducted by Eleanor G. May (1973), the store in question did not have a set of written objectives concerning the desired store image. As a result, the various levels of management were surveyed in order to obtain a description of the store's image. May (1973) included both levels of management (top and middle) for two reasons. T0p management was included because it is this level of management that typically guides middle management's actions and goals. Members of middle management, which included store managers, buyers, divisional sales managers, and divisional merchandise managers, were surveyed because these employees have direct and daily contact with the consumers and merchandise which make-up the dimen- sions and attributes of.a store's image (May, 1973). May (1973) came to conclusions similar to those of Pathak (1972). The researcher concluded that management's view of the store image was different than that of consumers' perceived store image. Furthermore, managers within the store held differing professed beliefs of the 16 image portrayed by their store. The effects of differing viewpoints can be seen by analyzing the store attributes. The combination of store attributes and the amount of each attribute help determine which consumers patronize the store and the amount of their repeat business. When retailers do not offer the appropriate amount and/or combination of attributes for the defined target market, consumers' perceived image of the store will differ from management's. Lost sales, dissatisfied consumers and smaller profits will result from management's inaccurate assessment of what the target market demands (May, 1974-1975). Image Measurement Researchers have agreed on the importance of image studies. Despite this general agreement of the importance of image research, a consensus has not been reached concerning the best measurement tech- nique. In order to obtain a better understanding of the various models, a section of the review of literature has been devoted to image measure- ment. An analysis of the strengths of various attitudinal models and their applicability to store image research is included. Multi-attribute Models Attributes may be economic, structural, functional, psycho— logical or social in nature (Green and Wind, 1973). By studying atti- tudes, a researcher can predict with some degree of certainty the behavior of an individual. Furthermore, an analysis of the attitudes and behavior would enable the researcher to explain why the subjects 17 responded as such, and predict with some degree of certainty how the subject would respond to additional stimuli (Dulany, 1968; Fishbein, 1967; Lutz, 1977). Due to the complex nature of attitudes, Wilkie and Pessemier (1973) theorize that the measurement of something on one overall attribute is insufficient. By measuring an attitude based on one attribute, such as good or bad, a general attitude can be realized, but the dimensions which compose this attitude could not be examined. Multi-attribute models allow an analysis of the various salient attributes used in forming attitudes. As defined by Wilkie and Pessemier (1973), the purpose of a multi-attribute model is to compre- hend the intangible concept of consumer behavior in relation to purchase consumption. The most basic multi-attribute model measures consumers' atti- tudes for an object based upon a specified number of attributes. This model is stated as: Where: > II the consumer's attitude toward store j; III II . the strength of belief i about store j, ‘3 i.e., the probability that store j if associated with attribute i; the evaluation of attribute i (Fishbein, 1967) 93 ll "Attitudes are learned predispositions to respond to an object or class of objects in a favorable or unfavorable way" (Fishbein, 1967, p. 257). In order to assess the importance of normative beliefs and the motivation 18 to comply with the normative beliefs, Fishbein extended the original model. An individual's behavioral intention is a function of: n B = BI [Aact] W0 + [NB (M0)] W1 Where: ' B = overt behavior; BI = behavioral intention; Aact = attitude toward an act; NB normative belief; MC motivation to comply with the normative belief W0 and W1 empirically determined weights. n The [Aact] factor is equal to Aj = Z BiAi in Fishbein's original model. i=i The second factor (NG) and (MC) refers to personal norm, social norm or both. Multi-dimensional Scaling In a unidimensional scale, possible opinion responses (positive, negative, or neutral) are placed along a single bipolar continuum and overall attitudes can be inferred. Many researchers theorize, however, that an analysis of an overall attitude is insufficient information when evaluating the professed image of a store (Bass and Wilkie, 1973; Beckwith and Lehmann, 1973; Berry, 1969; Cohen, Fishbein, and Ahtola, 1972; Dulany, 1968; Enis, 1967; Green and Wind, 1973; Kusher, 1972). Store image is comprised of many attributes which vary in importance as viewed by the consumer. By utilizing a multidimensional model, as opposed to a unidimensional scaling technique, the effect of each attribute on the perceived image can be analyzed. The strength of the multidimensional model lies in the fact that it takes into 19 consideration that the consumer's impression of a product or store is based upon a combination of attributes. Each dimension of an attribute is weighted differently by consumers according to the perceived import- ance of that attribute. The weighted differences of each attribute which form attitudes, impressions or images are computed by utilizing multidimensional models (Kusher, 1972). In addition to having the ability to measure multiple attributes of varying weights, the multidimensional scaling model allows for a visual representation of an individual's perceptions. Products or services may be visualized as having both objective and subjective attributes. Objective attributes include tangible services or proper- ties of a product. Subjective attributes include the congeniality, ease of shopping and atmosphere offered by the store. The perceived space constitutes the consumer's perceived product/service attributes which are used in decision making. These objective attribute "spaces" may not be consistent with the consumer's perceived "space". The dif- ferences between the objective attributes and perceived space may be due in part to advertisements, previous experience, reputation of the product and/or company. This perceived space will in turn influence the consumer's perceived image of that product, brand, service or company. Through an analysis of the perceived space, researchers can determine which attributes are those attributes which differ between stores, products or brands according to the consumer. One such model was developed by Ben Enis (1967). Enis (1967) developed a three-dimensional model which would "blend" the various 20 components which helped form images. This three-dimensional model consists of a configuration in which to measure three opposite pairs of salient characteristics for an entity. These characteristics are then "blended" together to form an image. The characteristics include the tangible versus intangibility of the entity, the significance versus the insignificance of the entity to the person, and the degree to which the entity is conceived as being consonant or dissonant with his self-image (Enis, 1967). The model was developed because Enis believed that the three pairs of traits are "universal characteristics" of attitudes. That is, each person experiences a mixture of three entities. For representa- tive purposes, the three pairs of characteristics are presented as: Intangibility Diss nance Sign'ficance Insignificance / Consonance Tangib1lity Stefflre (1978) hypothesized that multidimensional scaling as a model for studying perceptions, cognitions and attitudes is limited. The researcher claimed that this model cannot effectively measure the complex dimensions of a person's attitude. Stefflre felt that by 21 utilizing a two- or three-dimensional model, the marketer/user will have a clear, concise, and tangible view of the results. Although the clarity of the perceptual map is advantageous, the importance of relevancy of the dimensions may be underestimated depend- ing upon the complexity of the concept being measured. For example, the geometric representation may lead the researcher to conclude that this representation encompasses all aspects and implications of con- sumers' perceived store image, which is not always true. Another dis- advantage of the multidimensional model is the complex procedures used to convert the theory of the model into actual research (Fishbein, 1967). Multiple Indicator Approach The multiple indicator approach to measure attitudes include utilizing self-reports, observing reactions, and measuring physiological reactions to stimuli (Cook and Selltiz, 1964). Self-reports are reports filled out by the respondent concerning the person's feelings, attitudes and beliefs. Attitude scales which could be used in self- reports include: a Thurstone scale, which measures the amount of agreement; a Likert scale, which typically measures the level of agree- ment, based upon a five-point scale; and the semantic differential bipolar scale. Observing behavior is another commonly used method for assessing attitudes. Subjects' reactions to stimuli are observed and recorded and then analyzed. Based upon the subjects' behavioral reactions to stimuli, inferences concerning the subjects' attitudes are made. Lazarfeld (1959) pointed out that a given behavior is not necessarily 22 produced by a given stimulus. The researcher must define the various stimuli and possible corresponding behaviors. The hypotheses can then be tested by analyzing the reactions. Physiological reactions to stimuli include measuring the rate of heartbeat, the amount of perspiration, and the adrenalin flow. Consumers' perception of a store can be analyzed by measuring consumers' reactions to stimuli such as the visual presentation of merchandise, the psychological impact of store layout and the design. By testing consumers' responses to bright colors and lighting, color schemes, layout and type of decorations, retailers are able to manipulate their store interiors according to the desired mood (Chain Store Age, 1978; Hardware Retailing, 1981; International Management, 1977). The multiple indicator methods mentioned all have common limita- tions. First, methods for measuring attitudes using a multiple indi- cator approach are time consuming. The amount of time needed greatly reduces the number of willing respondents and increases the cost of the study. Another limitation of this model is the potential for vari- ability in its application. Without a specified quantitative measure, the results can be interpreted in a number of ways, depending upon the technique for analysis. Multiple-cue Probability Attitude Model The multiple-cue probability model, developed by K. R. Hammond (1967) is another procedure for measuring consumer attitudes. The model is based on the theory that events have a certain probability of occurring. Each event will have a varying probability depending upon 23 the interaction between people, the environment and events. Hammond refers to the events as cues. The multiple—cue probability model was employed by Will and Hasty (1971) to examine consumer attitudes of convenience food products. The stimulus cues used were price, brand familiarity, and quality. Would purchse/would not purchase was the dependent variable used in relation to the stimulus cues. The data were analyzed using multiple correlation coefficients. The results of the data showed that every correlation coefficient was significantly different at the .05 level. The researchers concluded that the multiple-cue probability model accurately measured consumers' attitudes (Will and Hasty, 1971). Two-factor Theory Rosenberg (1956) developed a two-factor theory model which measures an individual's attitude toward an object. The formula for the model is: A. = f 1k (PI.. . VI..) 1 13k 13 "M: 3 Where: _ A. = attitude toward an object k expressed in terms 'k of an individual i's degree of like-dislike (affect) of_that object; PIi'k = individual i's perceived instrumentality of the J kth object toward attaining or blocking the jth goal or value; VI.k = value importance to an individual i of the jth 3 goal or value (Rosenberg, 1958). In this model, the attitude toward an object (k) is a "function of weighted sum of beliefs about the object (perceived instrumentality) as to whether it blocks or helps the attainment of certain goals." 24 (Sheth and Talarzyk, 1972, p. 6). The weights act as a symbol pertain- ing to the importance of each goal. In a study by Sheth and Talarzyk (1972), an attempt was made to determine whether factor, perceived instrumentality (PI) or value importance (V1) is more significant in determining consumer attitudes. The researchers selected 2000 female heads of households from the Consumer Mail Panel of Market Facts, Inc. Six product categories, five product attributes and five product brands were used in the survey. Three simple regressions were employed to test the factors. The first regression predicting variance in attitude from the weighted sum of scores of respondents were derived by multiplying each perceived instrumentality (PIijk) with the value importance (Vlij) of a character- istic j and then summing across all characteristics" (Sheth and Talarzyk, 1971, p. 7). The second regression formula is stated as: n . = 2 PI.. Alk f j=l 13k The summing of the beliefs (representing perceived instrumentality) allow the prediction of variance in affect. The last regression was utilized to predict variance in affect derived from the weighted sum of importance characteristics (VIij) of a product class. The formula used to determine value importance was: ik j=1 ij The researchers concluded that perceived instrumentality was more important than value importance for determining a consumer's 25 affect toward a brand. The continuums are measured on a zero to one hundred point basis. The point of intersecting continuums is a neutral point. An image of an object is determined by measuring the degree of strength of each pair of characteristics. Once the strength of each characteristic per- taining to an object is determined, the consumer's image can be mapped and studied. Through an analysis of a consumer's three characteris- tics, behavioral and attitudinal information can be attained (Enis, 1967). Importance of Differential Weights in Attitude Models Through a review of research dealing with multi-attribute attitude models, Bass and Wilkie (1973) concluded that attitudes and their measured weights are significantly correlated. This relationship will in turn allow researchers to measure and assess attitudes by utilizing uniform weights (Werbel, 1978). One such model which employs uniform weights is the Fishbein extended model. The uniform weights are compen- sated by the "value" component of the model. The value component incorporates both attribute importance and the degree of deviation from an ideal point (Cohen, Fishbein, and Ahtola, 1972). Just as attitudes toward an object vary among individuals, several researchers (Beckwith and Lehmann, 1973; Pekelman and Sen, 1974; Werbel, 1978) theorize that the importance (weights) of attributes also vary. More specifically, each individual has a specific preferred weight for each product, brand or store. The desired weight of each attribute is called the "ideal" point. A weight higher or lower than the "ideal“ point contributed to 26 the dissonance in the consumer's mind. By ranking the attributes in relation to other similar stimuli, an overall preference is determined (Beckwith and Lehmann, 1973). In order to provide empirical evidence concerning the importance of weighted attributes, Beckwith and Lehmann (1973), conducted a survey of 20 television shows using six attributes. The model used is stated as: ID) II "M 3 Z W I H r 3 i where: A. = estimated overall attitude toward show j; W. = weight of the ith attribute; B.. = perception of the jth show on the ith attribute; I. = ideal point on the ith attribute; B..-I. = the absolute value of the difference between the perception of the jth show on the ith attribute and the ideal point on the ith attribute; n = number of attributes; a parameter (Beckwith and Lehmann, 1973, p. 142). x W The perception (Bji) of each show (j) in relation to each attribute (i) was measured. The ideal point (I) for each attribute (i) was specified by employing a six-point scale. The subjects then indicated the atti- tude (Aj) toward each show and the weight (W1) of each dimension. Several conclusions were reached concerning the inclusion of weights in a multi-attribute attitude model. First, the model accur- ately defined the subjects' television show preferences. Secondly, although the difference between the ideal point and the differential weights and the ideal point for each attitude, a small improvement for 27 explaining attitudes may be realized by utilizing this model over other multi-attribute models. The Beckwith and Lehmann model will be used to study retail store image of discount department stores. The methodology for this study, the population sample and the attributes to be used will be discussed in Chapter III. Selection of Image Attributes Employing attributes which succinctly depict the various dimensions of store image is essential to the validity of image research. Marks (1976) addressed the issue of the interrelationship and saliency of attributes. Marks theorized that attributes may be viewed differently among consumers. Attributes such as style, type of display and fashion may mean the same thing to one individual, while the terms describe different attributes to another consumer. Researchers have emphasized the importance of using only salient attributes (Fishbein, 1967; Marks, 1967; Rosenberg, 1957). The diffi- culty in identifying salient attributes is due to the fact that the saliency of attributes vary between markets and objects studied (Hirschman et al., 1978). Hirschman et a1. (1978), conducted a store image study to determine whether salient attributes vary between target markets and the type of store. The researchers conducted a frequency count of attributes used in recent research. Through an analysis of the data, the researchers concluded that the saliency of attributes did vary between target markets and the type of store being investigated. Attributes were selected for the study based upon the degree of fre- quency with which consumers mentioned the attributes. The attributes 28 used included salesclerk service, location of the store, merchandise pricing, credit or billing policies, layout and atmosphere, quality of merchandise, variety/assortment of merchandise, merchandise display, guarantee, exchange and adjustment policies, and real savings repre- sented in sales. Two components can be included in the study of store image. The first component used is the attributes associated with store image. The second component includes dimensions of each attribute. Dimensions are elements of an attribute which further define the attribute, thus explaining the interrelationship of each attribute. For example, service is a tangible store attribute. Dimensions of this attribute may include salesclerk service, self-service, ease of returns, credits, delivery or telephone orders. Perhaps the most difficult and controversial step in an image study is in deciding which attributes and dimensions, if any, should be used. A frequently used method of determining salient attributes is through focus group interviews (Alpert, 1971; Berkowitz, Deutscher and Hansen, 1978; Cohen, 1967; McDougall and Fry, 1974-1975; Pessemier, 1980; Vaugh, Pitlik and Hansotis, 1973). A subsample from the popula- tion is randomly selected for the interview. Questions concerning attributes, substitutability and ranking of importance in relation to a specific situation are asked. The results are analyzed to determine which attributes are deemed salient according to the subsample. In order to determine which image attributes to use in a study, James et a1. (1976) conducted focus group interviews. A subsample of the population was selected to answer an open-ended question concerning 29 important attributes of a men's clothing store. The attributes chosen for the actual study was based on a frequency count of attributes. The researchers calculated an importance score for each attribute. The order of importance of each attribute (according to the sample) was quality (with a value importance score of 6.37), price (6.13), assortment (6.11), service (5.63), personnel (5.15) and atmosphere (4.84). Robert Wyckham (1967) conducted a study of aggregate department store images. The attributes used in his study were merchandise suitability, sales personnel, store congeniality and locational con- venience. As part of a research study, Lindquist (1974-1975) grouped store image attributes according to recent literature of 10 researchers' empirical evidence. The attributes used by the researchers are listed in Figure 1. Based upon a review of empirical evidence, Lindquist concluded that merchandise, service and locational factors are important attributes when studying the consumer's perception of store image. Although some researchers (James at al., 1976; Pathak, 1967; Wyckham, 1967) utilize relatively few attributes (under five), as many as forty-two attributes have been used in image research (Marks, 1976). Researchers such as Berry (1969), Fisk (1961-1962), Kunkel and Berry (1968), and May (1971) have employed the following attributes: sales atmosphere and location. In addition, the researchers used as many as seven dimensions for each attribute. 3O Attributes Researchers Dimensions Alderson, Sessions (1964) Martineau (1957) (1962) Rich, Portis (1964) Myers (1960) Rachman, Kemp (1963) x x xxx Kunkel, Berry (1968) xx Tillman (1967) Rich >< ><><>< May (1967) >< Wyckham Merchandise Quality Selection/Assortment Styling/Fashion Guarantee Pricing Service General Salesclerk Service Self-Service Ease of Return Credit Delivery Phone Orders Clientele Social Class Appeal Self-Image Congruency Store Personnel Physical Facilities Physical Facilities Store Layout Shopping Ease Architecture Convenience Convenience X X Locational Convenience X X X X Parking X X X Promotion Sales Promotion Advertising/Display X X X Advertising Trading Stamps Symbols and Colors X Store Atmosphere Atmosphere/Congeniality X X X InstitutiOnal Conservative/Modern Re utation X X X X Re iability X X X XX xxx X X xxx X X XX XX X XX ><><><><><><>< >< ><><><>< >< ><><><>< X XX X XX xxx X X X X X ><><><>< Source: Adapted from Lingquist, Jay 0. Meaning of Image, Journal of Retailing, 1974-1975, 59, 29-38, 116. (1967) 31 Summary Store image is defined as "... 'the personality' the store pre- sents to the public or 'a complex of meanings and relationships serving to characterize the store to the populace'" (Hirschman et al., 1978, p. 3). Many research studies have dealt with a store's image and an examination of how consumers perceive attributes offered. The majority of studies conducted have focused on the consumer's perceived image of the traditional department store. Increasing interest and awareness has developed concerning the importance of research related to discount store image as perceived by management as well as consumers. May (1971), Pathak (1971) and Wyckham (1969) have conducted studies on managements' and consumers' perceived images of several stores. Despite the number of store image studies the measurement tech- niques have varied. Various multi-attribute attitude models have been used extensively. One such model was developed by Beckwith and Lehmann (1973). This model incorporates an ideal point which is the desired weight of an attribute. The Beckwith and Lehman model facilitates a comparison of consumers versus retailers perceptions desired attributes. Chapter III is a descussion of the methodology and statistical techniques used to analyze consumers' and management's perceived image of three discount stores. The theoretical basis for the Beckwith and Lehmann model will serve as the tool for analyzing store image. CHAPTER III METHODS AND PROCEDURES Chapter III is divided into six sections. The First Section addresses the selection of discount stores which were analyzed and the rationale for including an analysis of an hypothetical ideal dis- count store. The selection of salient store attributes is discussed in the Second Section. The reasons for the selection of certain attrib- utes for the study are discussed. The mathematical model used in this study to determine respondents' image of each store is presented in Section Three. Section Four addresses how the sample population was chosen. Section Five addresses the design of the questionnaire; the hypotheses and statistical analyses are presented in Section Six. Selection of Discount Stores Respondents' hypothetical ideal discount store and three compet- ing discount department stores were analyzed. Respondents' professed ideal store was included in this research study for several reasons. Few image studies have included an analysis of respondents' ideal store. Researchers agree that a comparison of the hypothetical ideal store and actual store leads to a deeper, richer understanding of the success to which retailers offer desired attributes. Specifically, the comparison 32 33 of attributes offered by the ideal store and those of the stores patronized by the target market will assist retailers in understanding (1) what attributes consumers desire in a store, (2) the level of each attribute desired, and (3) how successful their store is in satisfying consumers' desires. The respondents' ideal discount store was examined. The perfect discount store does not exist. However, consumers have general ideas concerning their ideal or perfect store. An hypothetical ideal store is defined as a store which offers the exact amount of each attribute desired by the consumer. In addition to the analysis of the respondents' ideal discount stores, an evaluation was made concerning the degree to which three competing discount department stores offered attributes desired by the consumer. The three discount stores under investigation were chosen for several reasons. They (1) are major competitors, (2) are well-known discount chain operations, (3) included value for price as a major factor in promotional tactics, (4) offered similar general services such as sales assistance, ample free parking and restaurant facilities, and (5) are located in Okemos, Michigan. Selection of Salient Store Attributes In order to determine which attributes would be included in the survey, telephone interviews were conducted. Using a random numbers table and prefixes relevant to the area, twenty-seven consumers from the Lansing/East Lansing, Michigan, areas were randomly chosen. Respondents were asked a series of questions concerning discount store 34 characteristics. To determine the saliency of each attribute, the respondent was asked if the store attribute mentioned was considered to be identical for the three discount stores under investigation. Once completed, respondents ranked each attribute according to their professed belief which reflected the importance of each attribute in a discount store. Responses were manually recorded and a frequency count was tabulated. A summary of the results of these interviews is presented in Table 3.1. From the 27 respondents, a total of nine attributes associated with discount stores were mentioned. The attributes used in the study were chosen due to the frequency response and saliency. When asked whether the amount of an attribute varied across the three discount stores studied, value for price, quality and store atmosphere were seen to have no variability. However, due to the substantially greater number of respondents who declared that these attributes did vary, these attributes were included in this study. The six most frequently mentioned attributes were included in the research. The store attributes examined in this study included: (1) value for price, (2) quality, (3) employee sales service, (4) mer- chandise selection, (5) locational convenience and (6) store atmosphere. Due to the relatively low response rate of the attributes: brand name merchandise (7.4 percent), Services offered (7.4 percent), and store reputation (3.7 percent), these attributes were not included in this study. From the 27 respondents, 21 mentioned price as an important store attribute. These respondents noted that low and/or competitive price 35 F meF F eoFFa F eeFoeoeeem eeeom F noeooz N mo> F ocooom N oocoewo mooF>gom F cacao; omFooozocoz N oz F pmch N oEoz oomcm F gonzo; F oz F othe o moF m umeFu m economosu< ocoum F zueood N ochF F ocooom oocoFoo>coo m mm» F umcFm m FoooFuoooo o ochF F ocooom :onooFom o moF F umch o omFooozogoz F ochF o ocooom ooF>com m moF o umeFd m moFom omzoFosm zucoom F ongF m oz m ocooom m mo> m pmcFo mF zpFFooo o ochF o oz m ocooom ooFco NF mm» m omcFu FN com ooFo> zuFFFooFco> motoum mouooFcop< to moF mmoooFcoo< omcoowoz oFcochu mo mocoooocm mmoeo< -zcoz mucoucoosF to to mchcom zocmoooeu o>Fpo=Fo>m anFFooFeo> omooomoz zocoooocd oucoucoosF oFcostu m>FpooFo>m ooFELouoo op mzoF>cmucF oFoEom Eoocoz to muFomom omquoooh .F.m oFooF 36 was the most significant factor for determining shopping behavior. More specifically, these respondents typically patronize a discount store because of the lower prices rather than other store attributes such as services offered and store atmosphere. The second most frequently mentioned store characteristic was quality of merchandise. In order for merchandise to be considered for purchase, consumers stated that it must meet a certain level of quality. This acceptance level will typically vary between products, consumers and the quoted price of the merchandise. Furthermore, respondents mentioned that the quality of merchandise greatly influenced the perceived image of the store in general. Store service was the third most frequently mentioned store attribute. Information concerning products or departments, fast and friendly check-out service and help with packages were mentioned as important aspects of sales service. Selection of merchandise was mentioned by six of the 27 respond- ents. Consumers noted that breadth of merchandise is extremely important for everyday, non-sale merchandise. Depth of merchandise, as opposed to breadth, is important for sale merchandise or advertised specials. Locational convenience was discussed by five respondents. The consumers stated that locational convenience was important for several reasons. These reasons included: (1) price savings offered by dis- count retailers would be eliminated due to gasoline expense if the store were a considerable distance from home; (2) time saved by shopping close to home was more important than dollar savings of a particular 37 discount store. Store atmosphere is the last attribute to be used in the study. Wide aisles, bright lighting, and the ability to see every department at a glance were stated as positive store attributes. From the five respondents who mentioned store attributes, three stated that the ability to browse through the various departments without being ques- tioned for help greatly contributed to a positive image of the store atmosphere. Multi-attribute Attitude Model The Beckwith and Lehmann attitude model was used to measure con- sumers' and managers' professed images of an hypothetical ideal store and three actual competing discount stores. The model is stated as: n A. = W .. - 3 1:1 '8“ I" . 9 Where: /"We ' Aj = estimated overall attitude toward store j; Wi = weight of the ith store attribute; .th th B'i = the perception of the J store on the i J attribute; I = the ideal point on the ith attribute. This model was chosen due to the inclusion of the ideal point for each attribute. Researchers have noted the importance of studying the ideal level for each attribute (Pathak, 1971; Wyckham, 1969). An analysis of the ideal store in comparison to the actual store would allow the researcher to assess the success to which retailers offer the amount of each attribute desired by the target market. In addition, 38 this model allowed the researcher to analyze what consumers' and managers' beliefs of what the "ideal" or perfect discount store should be. Design of the Questionnaire The questionnaire was designed to test the Beckwith and Lehmann multiple attribute attitude model. An initial questionnaire was developed and pretested by employing a random telephone survey of 31 consumers from the East Lansing and Okemos, Michigan areas. Consumers were asked to mention specific areas of the questionnaire which they felt were difficult to understand or difficult to complete due to the telephone format. The questionnaire was revised based upon their comments. The questionnaire was then approved by the Human Subjects Committee, Michigan State University. The questionnaire used to survey the managers is included in Appendix A. The consumers' questionnaire is similar in length and format. The only difference between the two questionnaires pertained to additional information concerning the managers' position within the store and their length of employment with the company. As stated earlier, the questionnaire was developed to measure the Beckwith and Lehmann model. Questions 19 through 25 were developed for another study using Fishbein's Extended Model (1967). The data collected from questions 19 through 25 will not be analyzed in this study. The components of the questionnaire and their corresponding question numbers are listed below: 39 Beckwith and Lehmann Component Questionnaire Ii = ideal point for each store Questions 1-6 attribute; B'i = perceived weight of each Questions 7-12 3 attribute for each store; Wi = weight of each attribute; Questions 7-12* Aj = attitude toward each store. Questions 7-12 *An explanation of this component of the model is dis- cussed below. The ideal point for each store attribute was measured by allocat- ing a total of 100 points among the six store attributes. The greater number of points received, the greater the value of that attribute in a discount store according to the respondent. The lowest value any attribute could receive was 1; the highest ideal attribute value was 95. The weight of importance of each store attribute was measured in relation to the other attributes offered by the store. More specific- ally, the weight or importance of each attribute is found by the sum of the attributes for a given store, and dividing the sum by that attribute. The formula is stated as such: 0 N1. =yi/ 2 ya“ j=l Where: W1 = the weight of the store attribute; yi = the attribute; j = the index. 'This weighting factor allowed the researcher to investigate what attrib- Lrtes were most important to consumers and managers. 40 Lemon (1973) addressed the issue of utilizing a ratio scale when measuring attitudes and impressions. The researcher noted that indi- viduals may have the same opinion (i.e., good or bad), but the intens— ity of their opinion may differ significantly. By employing a ratio scale, the intensity of each individual's response toward each attrib- ute and store can be analyzed. Because a ratio scale allows the intens- ity of a response to be shown, the perceived weight of each attribute offered by the stores was measured by using a ratio scale of l to 100. Unlike the ideal store, the summation of attributes for each actual discount store under investigation did not have to equal 100 points. Rather than totaling 100 points, the perceived weight of each attribute was given a number from 1 to 100 based upon its comparison to the ideal point for that attribute. If the perceived amount of an individual store attribute were higher than the ideal amount of that attribute, the store supplied too much of that store attribute. Likewise, if the perceived amount of an attribute had a value lower than the ideal amount, the retailer did not supply enough of that attribute according to the respondent. Sample Selection Consumers Consumers from the Okemos, East Lansing, and Williamston, Michigan areas were chosen as the sample population for this study. Prefix tele- phone numbers for these areas were used. The last four digits of the telephone numbers were determined by using a random numbers table (Beyer, 1978). 41 The telephone prefix numbers for these areas were chosen due to the fact that store attributes may differ between the various branch stores; these differences may, in turn, have resulted in biased responses by consumers. By researching specific branch locations and utilizing consumers who lived closest to the location studied, percep- tions of different branch locations were not felt to be an intervening variable. Consumers 18 years old and over were eligible to participate in the survey. Two reasons exist for a prerequisite of being at least 18 years old. First, parental consent, which is required by law, would be difficult to obtain and prove due to the nature of the survey; and secondly, consumers under 18 years old are not typically the major family purchaser or decision maker. Professed overall impression of each store based on past experiences of those other than the major or frequent purchaser(s) and/or decision maker could have biased the results. Methods and Procedures The nature of the design limited the sample to consumers with a telephone. However, by using prefix numbers and a random numbers table, unlisted telephone numbers as well as listed telephone numbers were included in the survey. The telephone surveys took place between July 5, 1982 and July 27, 1982. Random calls were made from 10:00 a.m. to 4:30 p.m. and from 7:00 p.m. to 8:30 p.m. By calling in the evening, possible bias of one income families in which the housewife answered the survey, thus 42 increasing the female to male ratio, was reduced. Furthermore, the random phone calls made in the evening increased the probability of dual income families. Non-response telephone numbers would be called twice during the day and once that same night. After the third attempt to reach a respondent, that number was eliminated. The random selec- tion of telephone numbers continued after each successful call. The final sample population of consumers for the study consisted of 186 respondents. These figures indicated that the response rate was 16 percent. From the 186 respondents, 9 percent felt unsure about rat- ing one or more of the stores and therefore did not fully complete the questionnaire. To be included in the sample consumers were required to complete the rating of the hypothetical ideal store and the comparison of the ideal store with at least one of the three discount stores. If respondents did not rate at least one store, their responses were immediately eliminated from the sample. The total number of responses associated with each store is: Ideal Store ......... n = 186 Store One ........... n = 180 Store Two ........... n = 169 Store Three ......... n = 163 Total consumer Sample ........... n = 186 Managers Personnel from the Market Research Department and Personnel Department at Store One and Store Two gave permission to survey the management team at their Okemos, Michigan branch location. Top manage- ment from Store Three did not give permission for their employees to be surveyed. Despite their lack of participation, consumers and managers 43 from Store One and Store Two answered questions concerning their over- all impressions of Store Three. Due to the relatively small sub-sample population of managers at each store, it was impossible to obtain a random sample. The sub-sample population of managers was a convenience sample. All managers from Store One participated in the survey. The final sub-sample consisted of nine managers from Store One with a participation rate of 100%. Twenty-nine managers from Store Two were asked to participate in the survey. Five managers from Store Two were not available for answering the survey. One manager decided not to participate, thus making the participation rate of Store Two managers 82.4%. Questions one through twelve were stated by the researcher, and then each manager filled in his/her responses. By verbally asking all respondents the first 12 questions, these questions were presented in the same form as those posed to the consumer respondent. Therefore, bias due to one sub-sample reading the questionnaire as opposed to hearing the questions would be reduced. After completing the first twelve questions, the managers completed the survey at their own pace. Hypotheses and Statistical Analysis Three hypotheses were tested in this survey. Each hypothesis was broken down into sub-hypotheses and individual tests were used to analyze each sub-hypothesis. H1: No significant differences exist between Store One managers' and Store Two managers' professed image of_ Store One, Store Two, Store Three and the hypothetical ideal store. 44 A t-test was used to test the differences between managers' perceived image of the respective stores. Multiple t-tests were con- ducted as the method for analysis due to several reasons. First of all, the mean difference between two populations can be studied. The populations in this study included Store One managers versus Store Two managers and consumers versus all store managers. Secondly, results of t-tests can assist in evaluating whether significant differences exist between two population means a key variable. The key variables studied included the ideal point for each attribute, the perceived point for each attribute and respondents' overall impressions of Store One, Store Two and Store Three. Lastly, when comparing two population means, the t-test is the strongest test. No significant differences exist between: Hl-l: Store One managers' and Store Two managers' overall impression of Store One. Hl-2: Store One and Store Two managers' overall impression of Store Two. Hl-3: Store One and Store Two managers' overall impression of Store Three. Hl-4: Store One and Store Two managers' professed belief of each attribute offered by an hypothetical ideal discount store. H1-5: Store One and Store Two managers' professed belief of the level of each attribute offered by Store One. Hl-6: Store One and Store Two managers' professed belief of the level of each attribute offered by Store Two. H1-7: Store One and Store Two managers' professed belief of the level of each attribute offered by Store Three. 45 H2: No significant differences exist between consumers' pro- fessed image of Store One, Store Two, Store Three and the hypothetical Ideal Store as as defined by the Beckwith and Lehmann multiple attribute model. The statistical test for the second hypothesis was an analysis of variance (ANOVA). An analysis of variance was used to test the differences between consumers' perceived image of each store. Sub-hypothesis H2-l: No Significant differences exist between consumers' impressions of Store One, Store Two, and Store Three. Sub-hypothesis H2-2: No significant differences exist between consumers' ideal point of each point of each attribute and their professed beliefs of the amount offered by each discount store. H3: No significant differences exist between consumers' and managers' professed image of Store One, Store Two, Store Three and the hypothetical Ideal Store as defined by the Beckwith and Lehmann multiple attribute model. ANOVA and multiple t-tests were conducted to test the third null hypothesis. Analysis of variance and multiple t-tests were used to test the differences between all managers for each store and consumers perceived image of each store and ideal store. No significant differences exist between: H3-l: Consumers' and managers' ideal amount of each attribute offered by an hypothetical ideal discount store. H3-2: Managers' and consumers' overall attitude toward Store One as defined by the Beckwith and Lehmann model. H3-3: Managers' and consumers' overall attitudes toward Store Two as defined by the Beckwith and Lehmann model. H3-4: Managers' and consumers' overall attitudes toward Store Three as defined by the Beckwith and Lehmann model. H3-5: Managers' and consumers' professed overall impression of Store One. 46 H3-6: Managers' and consumers' professed overall impression of Store Two. H3-7: Mfinagers' and consumers' overall impression of Store ree. H3-8: Managers' and consumers' professed belief concerning the amount of each attribute offered by Store One. H3-9: Managers' and consumers' professed belief concerning the amount of each attribute offered by Store Two. H3-lO: Managers' and consumers' professed belief concerning the amount of each attribute offered by Store Three. H3—6: H3-7: H3-8: H3-9: H3-lO: Managers' and of Store Two. Managers' and Three. Managers' and the amount of Managers' and the amount of Managers' and the amount of 46 consumers' professed overall impression consumers' overall impression of Store consumers' professed belief concerning each attribute offered by Store One. consumers' professed belief concerning each attribute offered by Store Two. consumers' professed belief concerning each attribute offered by Store Three. CHAPTER IV RESULTS AND DISCUSSION Introduction This chapter is a summary of findings from the collected data. Chapter IV is divided into several sections. Section One addresses the demographic characteristics of the sample. In Section Two, the results of the stepwise regression equation for defining the best one term model for identifying respondents' overall impressions of Store One, Store Two and Store Three are presented. Results of the corre- lation coefficients are discussed in the Third Section and the con- clusions for the hypotheses are included in Section Four. Demographic Information A summary of demographic information is shown in Table 4.1. The summary table is broken down by demographic variables and sub- samples. For analysis purposes, each demographic variable is addressed and similarities between sub-samples are discussed. Marital Status The majority of the respondents from the total sample were married (see Table 4.1). Fifty-seven percent of the consumers indicated 47 48 ooooFuooo O O 0 VNri-MKOF- GOMNr—d'r—m MF- cozcoz ooF>com coeoooo Egomuooz copogooo ochooz comcoomuwocu FooFeoFu moFom o>FpoLuchFEo<\FoFLomocoz FooFczooF\FocomeoFoLo ooFooooooo thxmmoomm F \OF- .. omcoomoz oz F o>Fu m Loom mF omczF NN ozF NF 0.5 mNF ocoz ooucoooom moFom promccou :oLoFon to goosoz N omooomom oz m omoco>Fo\oopoeoomm N oozoon moF ooFecoz 8 £95 . mouoam Fochoz N .oz N am u 2 .oz ozh ogoom moo oeoom. mcomocoz N .oz moFooFto> mwF n z meoEomcou coFuoELomcF onoocmosmn to zuoEE=m .F.o mFooH 49 ooooncoo mmmomomoom GNOVNNDMQ 1011'st N mmcoomoz oz coon: oco mom.oa moo.~» op ooo.mw mom.m» on oom.Nm mmm.oF» op ooo.oFa mmm.mFa op ooo.mFm mom.¢Nm o» ooo.ONm mmm.omm op ooo.me co>o one coo.mme osoooF mmcoomoz oz Foozom zgoucoEoFu meow Foozom agouooEoFm omquoEou Foeeem zon meow Feeeem emF: eeoeFeeeo omoFFou meow oocmmo omoFFoo Foozom opooootu meow Foozom opoooocu :onoooom omooommz oz Lozpo oozoFosooz oochoz goztoz Econ Fo.ucoov conoooouo wN u z ozh mcoum moo ocoum & mmF u z meoEomcou moFooFco> ooocFucoo 1 F.o oFooF 50 N.oF m 11 11 11 11 omooomoz oz N.No NF N.om o N.Nm oNF oFosou ¢.mo mF m.mm m m.Nm om oFoz xom N.oF m 11 11 F.N o oncoomoz oz 11 11 11 11 m.oF oN L_o>o ooo meoo> mm 11 11 11 11 F.o NF om o» mm N.oF m F.FF F N.m NF om o» me N.NF m . F.FF F N.m NF we op mm ¢.mo mF m.mm m m.oN on em op mN N.oF o o.oe o m.mN mm oN op NF 11 11 11 11 11 11 . NF Loocz om< N .oz N .oz N .oz moFooFeo> wN u z m u z mmF u 2 oz» mcoum moo ogopm mcomoooz mcosomoou oaseFoeoo 1 F.e eFeeF 51 that they were married, while 66.7 percent of managers from Store One were married. Approximately 75 percent of Store Two managers also stated that they were currently married. Number of Children Currently Being Supported The number of children in the family were determined by the number of children currently being supported including those living away from home at a boarding school or university. When consumers were asked to state how many children they were currently supporting, 67.2 percent indicated that they did not support any children at the present time. The majority of managers from Store Two (66.7 percent) stated that they were not currently supporting any children, while one-half (50.0 percent) of Store Two managers were not supporting children. A commonality existed among respondents who were currently sup- porting children. From the total sample, 15.2 percent were supporting two children. Approximately 14.0 percent of the consumers supported two children. Twenty-two percent of Store One managers had two children while 17.9 percent of Store Two managers were supporting two children. Occupation Respondent's occupation was divided into thirteen categories as defined by the United States Census Bureau. A summary of these cate- gories is presented in Table 4.1 of demographic information. Thirty-four percent of the consumer respondents indicated that they held professional/technical positions (Table 4.1). With 15.6 per- cent of the consumer sample retired, this category held the second largest number of respondents. Managerial/administrative positions 52 were held by 10.2 percent of the consumer sample, making this the third largest occupation category. When Store One managers were asked to state the occupation held by the head of the household, 33.3 percent indicated that they were professional/technical positions. An additional 33.3 percent of managers from Store One indicated that the head of their household held a managerial/administrative job. As was the case in the Store One managers' responses, the majority of Store Two managers stated that the heads of household held managerialladministrative positions or professional/technical positions. Sixty-seven percent of the Store Two managers indicated that the occupa- tion of the head of their household was managerial/administrative oriented. Approximately 17.8 percent of the Store Two managers stated that the head of their household had a professional/technical occupa- tion. Educational Level of the Head of the Household Respondents were asked to state the amount of education completed by the head of the family. Twenty-nine percent of the consumers sur- veyed had some college education, with 22.6 percent having attained a four year college degree. An additional 15.1 percent of the consumer sample stated that the head of their household had successfully com- pleted a graduate program. When comparing the educational level of the head of the household for managers, Store One managers had attained a higher educational level than Store Two managers. Forty-four percent of the Store One managers 53 had a four year college degree whereas only 26.9 percent of Store Two managers held a four year college degree. Unlike that of consumers, none of the managers from either store indicated that the head of his/her household completed a graduate program. Family Income Family income was calculated based on the income level before taxes. The educational level was based only upon the head of the family. However, the family income is based upon all income earned by all family members. When consumers were asked to state their income for 1980 before taxes, 19.3 percent earned $35,000 and over, thus making this income level the most common. The second highest percentage of consumers earned between $10,000 to $14,999, consisting of 17.7 percent of the sub-sample. Only 3.8 percent of the consumers indicated that their income before taxes was under $5,000 a year. The average family income level before taxes for Store One mana- gers was lower than that of Store Two managers. Thirty-three percent of the Store One managers indicated a family income of $35,000 and over. Forty-six percent of the managers from Store Two, however, estimated this same income. The second highest reported income level for Store . One managers was $15,000 to $19,000 with 11.1 percent of the respondents in this income bracket. The second most frequent income level for Store Two managers was $20,000 to $24,999, consisting of 14.3 percent of the sub-sample. 54 Age of Respondent When respondents were asked to specify their ages at their last birthday, twenty-eight and one-half percent of the consumers and 44.4 percent of the managers from Store One indicated that they were in the age bracket from 18 to 24 years old; this age bracket constituted the mode for each of these sub-samples. The most common age bracket for Store Two managers was between the ages of 24 to 34 years old. Sex of the Respondent Sixty-seven percent of the consumers and 66.7 percent of the managers surveyed at Store One were female. Forty-two and nine tenths percent of the Store Two managers were female and 46.0 percent male. The remaining 10.7 percent of Store Two managers did not indicate their sex. Forty-eight and seven tenths percent of the population in Ingham County in 1970 were male. This percentage of males in the population indicates that male consumers and managers were underrepresented in the sample. Female consumers and Store One managers were overrepre- sented in the survey by 16.4% and 15.4% respectively. Female Store Two managers were underrepresented by 8.5%. Multiple Regression Analysis The Beckwith and Lehmann multiple attribute model was used throughout the study. Using respondents' overall impression of Store One, Store Two and Store Three as the dependent variables, a stepwise regression equation was calculated. The purpose for employing the 55 stepwise regression equation was to determine the best one term model for each sub-sample when utilizing the Beckwith and Lehmann multiple attribute model. The method for operationalizing this model is: n A. = z W |B.. - I l J ._ 1 31 1 1-1 Where: Aj = estimated overall attitude toward show j; Wi = weight of the ith attribute; Bji = perception of the jth show on the ith attribute; Ii = ideal point on the ith attribute; n = number of attributes (Beckwith and Lehmann, 1973, p. 142). Consumers Results of the stepwise regression analysis of consumers' responses for Store One are presented in Table 4.2. The significance level for the best one term model through the best six term model indicate that all term models are excellent predictors of consumers' overall impression of Store One. The best one term model for Store One is quality, with a signifi- cance level of p<:.OOl. Merchandise selection, employee sales service and store atmosphere were the best two term, three term and four term models respectively, for consumers' overall impression of Store One (p<:.001). Locational convenience was the best six term model for consumers' overall impression of Store One (P<:.001). When utilizing the Beckwith and Lehmann model, the best one term through the best five term model proved to be significant as predictors of consumers' overall impression of Store Two. The best one term model 56 So . v 93% 8. v9.5 mo. VAT. mFmNF. eetooFmo.o m moomFom>oou Fmoos FoooFuoooo somu m>FF pmmm FmNNF. «eeNmmmo.o o momoomosu< mooum Fmooe somu Loom ummm ooNFF. eeemmFNN.N m moF>omm FmooE mmFom mmzoFosm Eomu mmoou pmmm NFmoF. eeeNooom.oF N ooFuomme Fmooe mmFoooooomz Eomp o3» ommm ooFNo. eetmNoom.mF F Fmoos zuFFooo somp moo pmmm moooom a o FFoom>o no muooFoup< Fmooz Eomh pmmm owF u z mFmaFoo< oomemommz mszompm .Fmooz oooEzmo ooo oonzomm moo op ooFooooo< moo moomm Fo oomemooeF .momsomoou .N.o mFooH 57 for consumers' overall impression of Store Two was identical to that of Store One; quality was the best one term model (p<:.05) (see Table 4.2). Employee sales service and merchandise selection were the best two term and the best three term models respectively, in predicting consumers' overall impression of Store Two. The best four term and the best five term models for predicting consumers' overall impression of Store Two were store atmosphere and value for price (p<:.05) respectively. The addition of each term for the prediction of consumers' over- all impression of Store Two did not significantly increase the R square. The R square for the best one term model was .02719 whereas the R square for the best five term model was .06491, an increase of .03772 (see Table 4.3). Results of the stepwise regression of consumers' overall impres- sion of Store Three according to the Beckwith and Lehmann model are presented in Table 4.4. The best one term model through the best six term model for predicting consumers' overall impression of Store Three were highly significant (p<:.001). The store attribute, quality, was the best one term model for explaining consumer's overall impression of Store Three as well as Store One and Store Two. Merchandise selection, value for price and locational convenience were the best two, three and four term models respectively, for expressing consumers' overall impression of Store Three. The best five and six term models were employee sales service and store atmosphere respectively. The R square for the best one term model of consumers' overall impression of Store Three was .0877; the R square for the best six term 58 So . v at; 5. v9.5 mo. v91 Fmoos Fmomo. ewwNmN.N m moFoo com moFo> Eomp m>FF ummm Fmooo oomoo. eoFmom.N e meeeomoooo meoom see» teem ommm mFNmo. emNNom.m m ooFuomme Fmooe mmFoooooomz somp mmoou pmmm Nmmmo. tommNm.m N moF>omm Fmooo mmFom mmonosm somu o3» pmmm Fmooe mFNNo. emmNmm.o F zpFFooo somu moo ummm moooom z u FFoom>o on muooFopu< Fmooz some pmmm moF u z mmeFoo< oomemommz mszompm .quoz oooEomo ooo oszxomm mop om moFooooo< oz» moomm mo oomemooEF FFoom>o .momsomoou .m.o mFooh 59 So . v no.3. Fo. v93, mo. vat Fmoos NoMNN. eeemoomN.o o meeeomoooo oeeom some me ommm moF>omm Fmoos FmomN. eeemomoo.oF m meFem mmNoFeom some m>Fm omem NeFoN. eeeoNFNm.NF o moeeFee>eoo Fmooo Fooonoooo Eomp Loom pmmm Fmooe NFNFN. eeeoomNN.oF m moFee toe meFe> some meeeo omem ommNF. eeeoomFo.NF N eeFoomme Fmooo mmFoooooLmz somp o3» ummm Fmoos NNNmo. etemoooo.m_ _ NoFFeeo some moo ommm meeeom o o FFeem>o on eoeeFeooo Fmooz somF omen mmF u z mszFoo< oomemommm mszomum .Feeez ooosomo ooo oszxomm mom op moFooooo< mmooh mooum Fo oomemooeF FFoom>o .momsomooo .o.o mFooF 60 model was .27342. The increase of .18565 in the R square for consumers' overall impression of Store Three was larger than that of their overall impression of Store One and Store Two. The R square of .27342, however, indicates a low level of total explanation of variance. Store One Managers The results of the stepwise regression model for Store One managers' responses for their store are presented in Table 4.5. With a significance level of p<=.01 and an R square value of .62167, value for price was the best one term model. The best two term model was value for price and store atmosphere (p<:.001) with an R square of .84769. The R square increased by .22602 between the best one term model and the best two term model. The best three term model and four term model were significant at the p<:.Ol and .05 levels respectively. The increase in the R square, however, was very small, thus indicating that the addition of a third and fourth variable would not increase the explanatory power of the model by a significant amount. Therefore, the increment in the R square (.03916) from the best two term model to the best four term model is not large enough to warrant using two addi- tional terms. The low F values and their insignificant R square values indicated that the six terms were not significant predictors of explaining Store One managers' overall impression of Store Two and Store Three. Store Two Managers The best one term model for predicting Store Two manager's over- all impression of Store One was identical to that of consumers' 61 So . v 9:; _.0. var... mo. v9... Fmoos mmooo. eeNmNmm.N o eeeeFem>eoo FeeoFoeeoo some teem omem Fmoos oooom. eeoNoNo.oF m eOFooeme emFoeeeoemz some mezzo ommm Fmoos ooNoo. eeooFom.oF N memeemoooq meeom some ego omem Fmoos NoFNo. eeoomom.FF F meoee tom meFe> some meo omem meeeom o o FFeee>o no eooeFooo< Fmooz oemF omen o 1 z mmeFoo< oomemoomm mszompm .Fmooz ooosomo woo oszzomm moo oo oowooooo< moo mooum mo oomemooEF FFoom>o .momooooz moo moopm .m.o mFooF 62 (see Table 4.6). With a significance level of p oonomme mmFoooooLmz zuFFooo Fmoos Eomu mmooa ammo Fmoos Eomp ozu ammo Fmoos Esme moo ammo moooom m o FFeee>o no moeeFeoo< Fmooz somh ammo mN u z mszFoo< oomemoomz mszomum ooo oszxomm moo op ooFooooo< moo mooum mo oomemooEF FFoom>o .momooooz ozF mooum .o.o mFooF .Fmooz ooosomo 64 So . v give 8. v9.5 mo. var FmooE mmm_e. eooomo.N m eoFoomFem mmooeeeoeez some e>Fc ommm Fmoos ooNoo. toFNom.m o economoooo eeoom oeeo teem ommo Fmoos oFFNm. 4NPN_m.o m NoFFeeo some mmezo omen Fmoos NNFem. eeoooFN.m N meoee tee m=Fe> some ozo ommm FmooE ooNFm. eemFNoN.oF F moF>emm mmFem emNoFoom some moo omen meeeom o a FFeem>o oo moeeFeooo Fmooz oemF memo mNuz mFmaFoo< oomemoomm mszomum .Fmooz ooosomo ooo oonzomm moo o» ooFooooo< ozF moopm mo oomemooEF FFocm>o .momomooz ozF mooum .N.o mFooe 65 addition of four terms does not significantly add to the explanatory power of Store Two managers' overall impression of Store Two. As was the case in the increase in the R square for the best terms which predict Store Two managers' overall impression of Store One, the increase in the R square for the best terms which predice Store Two managers' overall impression of Store Two were minimal. More specifically, the R square increased by .09547 from the best one term model to the best five term model for predicting Store Two managers' overall impression of their own store. The relatively small R squares indicate that additional salient terms for predicting Store Two managers' overall impression of Store One and Store Two were missing from the survey. Results of the Stepwise regression equation for predicting Store Two managers' overall impression of Store Three were identical to that of Store One managers' overall impression of Store Three. When uti- lizing the Beckwith and Lehmann multiple attribute model and employing a stepwise regression equation, none of the six attributes were sig- nificant predictors of Store One managers' and Store Two managers' overall impression of Store Three. Correlation CoeffiCients Consumers In order to determine the degree to which attributes were corre- lated, correlation coefficients were conducted. The strong positive 66 and negative relationship between the independent variables indicate multicollinearity (Lehmann, 1979). The results of the correlations revealed that consumers felt that the level of employee sales service was highly correlated (.58617) with the atmosphere offered by Store One (see summary Table 4.8). The quality offered by Store Two was viewed to be positively correlated (.44413) with its store atmosphere. Correlation coefficients were calculated using consumers' over- all impression of Store Three as the dependent variable. Although positive and negative correlations were viewed, none was highly signifi- cant. Table 4.8. Summary Table of Correlation Coefficients of Consumer Responses Correlation Coefficients for Store One N = 186 Store Atmosphere .58617 Employee Sales Service Correlation Coefficients for Store Two N = 186 Quality .44413 Store Atmosphere Store One Managers Overall, Store One managers had much stronger views concerning the relationship that store attributes can have on their overall impression of a store (see Table 4.9). 67 Table 4.9. Highly Correlated Attributes for Store One as Seen by Store One Managers Correlation Coefficients for Store One N = 9 Value for Price .78846 Store Atmosphere -.56403 Locational Convenience -.68528 Quality -.43590 .42712 .68202 Consumers' Employee Store Locational Overall Sales Atmosphere Convenience Impression Service Store One managers felt that a strong positive relationship existed between the value for price and their overall impression of their own store (.78846). However, their overall impression of Store One was negatively correlated with the store atmosphere (-.56403). Store One managers also felt that the level of employee sales service offered by Store One was negatively related to the perceived locational convenience (-.68528) and the quality of merchandise offered (-.43590). The quality of merchandise was, however, positively correlated with the perceived store atmosphere (.42712) and locational convenience of Store One (.68202). Store One managers felt that many attributes had extremely posi- tive and negative relationships with attributes offered by Store Two. 68 As depicted in Table 4.10, Store One managers felt that the quality of merchandise offered by Store Two was inversely related to four attributes at Store Two. That is, Store One managers felt that the quality of merchandise offered by Store Two was negatively related to: their overall impression of that store (-.56704), the merchandise selection (-.39496), the level of employee sales service (-.50342) and the perceived atmosphere of Store Two (-.45167). Store One mana- gers did consider, however, that the quality of merchandise offered by Store Two was positively correlated with the perceived value for price (.89042) and locational convenience (.77805). Table 4.10. Highly Correlated Attributes for Store Two as Viewed by Store One Managers Correlation Coefficients for Store Two N = 9 Quality -.56704 -.39496 .89042 -.50342 -.45167 .77805 Employee Sales Service -.44447 .60545 Value for Price -.47007 -.44447 Overall Merchan- Value Employee Store Location- Impres- dise for Sales Atmos- a1 Con- sion Selection Price Service phere venience Store One managers felt that the level of employee sales service offered by Store Two greatly enhanced their store atmosphere (.60545) but had a negative affect on the perceived value for price (-.44447). 69 Store One managers' overall impression of their store and the atmos- phere of their store was negatively correlated with the perceived value for price (-.47007 and -.44447 respectively). Store One managers felt that several Store Three attributes were positively correlated with each other (see Table 4.11). Store One Managers felt that the quality of merchandise offered by Store Three was positively correlated with value for price (.92500), level of employee sales service (.39483), store atmosphere (.90488) and the locational convenience (.76823) of Store Three. The locational con- venience of Store Three was also deemed to be positively related to the perceived merchandise selection (.31920), value for price (.86496) and atmosphere of Store Three (.52788). Table 4.11. Highly Correlated Attributes for Store Three as Seen by Store One Managers Correlation Coefficients for Store Three N = 8 Employee Sales Service .53457 Store Atmosphere .77539 .53457 Locational Convenience .86496 .52788 Quality .92500 .39483 .90488 .76823 Value for Employee Store Locational Price Sales Atmosphere Convenience Service 70 Store Two Managers A summary table of correlation coefficients results of Store Two managers' opinions concerning attributes offered by each discount store are presented in Table 4.12. An analysis of the results revealed that four attributes were viewed to be positively related to Store Two managers' overall impression of Store One. The level of employee sales service was positively related to the atmosphere offered by Store One (.71187). The quality offered by Store One was also viewed to be positively related to the value for price (.51003). Two attributes offered by Store One were viewed to be inversely related. The locational convenience was negatively correlated with the merchandise selection offered by Store One (-.72071). Store Two managers' overall impression of their store was posi- tively correlated with the level of employee sales service (.56434). Value for price was positively correlated with the level of employee sales service (.41141) and store atmosphere (.40299). Store Two mana- gers also felt that the level of employee sales service offered by their store was positively related to the perceived store atmosphere (.70805). The strongest correlation was between the quality of mer- chandise and atmosphere offered by Store Two (.79337). Using Store Two managers' overall impression of Store Three's level of employee sales service was positively correlated with store atmosphere (.50076). 71 moF>omm mmFom mmonoEN oNoom. meeeomoooo meoom mN u z mmoop moopm oom pomFoFFFmoo ooFooFmoooo momoomo5p< mooum mmmmm. oomemoosF FFoom>o momaomooo NoFFeeo momoomo2u< mooum moF>omm mmFom mmzoFoEN NwFFN. moF>omm moFoo mmFom monFoEm oom moFo> moNoN. ooNoo. FoFFo. moFoo Loo moFo> mooFm. omoom. mmFom mmaoFoEm mN u z ozF mooum tom mpomFonmmoo ooFqumoooo oonomme mmFoooooomz o... :25 FNoNN.1 moomFom>ooo FoooFuoooo momoomooo< moopm wN u z moo mooum tom moomFoFFFmoo oonoFmoooo momooooz o3» moopm op ooFooooo< mmuooFouu< omoeoeeeoo NFooFI to eFeeF Noeooem .NF.o mFeeF 72 Testing the Hypotheses The research objectives for this study were as follows: (1) Identify salient evaluative criteria for discount stores. Stepwise regression analyses were conducted in order to determine the best term models for expressing respondents impressions of Store One, Store Two and Store Three. (2) Measure consumers'and managers'professed overall impressions of each of the stores in the study. Analysis of variance and multiple t-tests were conducted in order to test the null hypotheses 1 and 2. (3) Identify consumers' and managers' perceived ideal amount of each store attribute. The third null hypothesis addresses this objective. (4) Compare consumers' and managers' perceived images of each store and the success with which the retailers offer the store attri- butes desired by the consumer. Multiple t-tests were conducted to analyze the fourth objective of the study. T-tests were conducted to compare significant differences between managers' and consumers' professed overall impressions of Store One, Store Two, Store Three. Many researchers have hypothesized that studying respondents' overall images or impressions of stores is insuf- ficient (Cohen, 1967). In order to obtain a more complete understand- ing of respondents' impressions, each store attribute must also be analyzed. Therefore, t-tests were also employed to compare the sig- nificant differences between managers' professed belief of the amount 73 of each attribute carried by each discount store in relation to their professed ideal store. H1: When comparing Store One, Store Two, Store Three with the ideal store, there will be no significant differences between managers' perceived image of their own stores. In order to simplify the reported results of the data, each hypothesis is broken down into sub-hypotheses. An analysis of respond- ents' overall impression of each store, the ideal amount of each attribute, and their professed belief of each attribute offered by each discount store constitute a sub-hypothesis. Hl-l: No significant differences exist between Store One managers' and Store Two managers' overall impression of Store One. A t-test was conducted to determine whether managers from Store One and Store Two held significantly different overall impressions of Store One. A summary of the results is presented in Table 4.13. An analysis of the data indicated that managers had significantly different overall impressions of Store One (p<:.001). Table 4.13. Managers' Overall Impression of Store One. T-test Overall Impression N Mean Mean Difference SD T-Value Store One Managers 9 87.556 50.413 7.585 lO.42*** Store Two Managers 28 37.143 21.819 ***p< .001 DF= 34.6. (Because it cannot be assumed that the two groups had the same variance, an F test of sample variance was performed. The F was less than or equal to .05, so the degrees of freedom were adjusted and t based on the separate variance estimate was used.) 74 Store One managers' overall impression of their own store was signifi- cantly higher than Store Two managers' overall impression for Store One. Due to the significant differences between managers' overall impression of Store One, the null sub-hypothesis 1-1 is rejected. H1-2: No significant differences exist between Store One and Store Two managers' overall impression of Store Two. When Store One and Store Two managers were asked about their overall impression of Store Two, significant differences resulted (see Table 4.14). Table 4.14. Store One and Store Two Managers' Overall Impression of Store Two. T-test Overall Impression N Mean Mean Difference SD T-Value Store One Managers 9 69.445 19.805 14.240 3.67*** Store Two Managers 28 89.250 14.031 DF = 35 ***p < . 001 Based upon the analysis of data presented in Table 4.14, the sub- hypothesis of no difference between managers' overall impression of Store Two is rejected. Store Two managers did have a more favorable overall impression of their own store than Store Two managers. H1-3: No significant differences exist between Store One and Store Two managers' overall impression of Store Three. When managers were asked to indicate their overall impression of Store Three, very low scores were given. That is to say, managers from 75 Store One and Store Two held extremely low overall impressions of Store Three as a discount retailer (see Table 4.15). Table 4.15. Store One and Store Two Managers' Overall Impression of Store Three. T-test Overall Impression N Mean Mean Difference SD T-Value Store One Managers 8 28.250 .75 24.904 .09 Store Two Managers 28 27.500 19.491 The mean scores for Store One managers' and Store Two managers' overall impression of Store Three were 28.250 and 27.500 respectively. With a mean difference of .75. The two groups of managers' overall impression of Store Three were not significantly different. Due to the insignificant differences in managers' overall impression of Store Three, the null sub-hypothesis cannot be rejected. An understanding of respondents' professed belief of the ideal point for each attribute offered by a discount store is vital to store image research. Information received concerning consumers' and retailers' perceived ideal amount of the attributes offered by stores would give retailers a better understanding of what consumers desire and the success to which the retailer understands his target market. Hl-4: No significant differences exist between Store One and Store Two managers' professed belief of each attribute offered by an hypothetical ideal discount store. In order to determine whether managers held similar ideas concern- ing the "perfect" discount store and the amount of attributes carried 76 by this store, multiple t-tests were conducted. A summary of the results of the t-tests for managers' professed belief of the ideal amount of each attribute offered by a discount store is presented in Table 4.16. Analysis of the data showed that managers from Store One and Store Two had similar professed beliefs concerning the ideal amount of each attribute offered by a discount store. More specifically, no significant differences were seen between manager's professed belief of the ideal levels of merchandise selection, value for price, level of employee sales service, quality, locational convenience and atmos- phere offered by an hypothetical ideal discount store. Due to the lack of significant differences in managers' ideal level for each attribute, the null sub-hypothesis 1-4 cannot be rejected. Respondents were asked to evaluate the degree to which each store's attributes satisfied their perceived ideal amount for each attribute. Based upon a ratio scale of l to 100, respondents selected a number which indicated their professed belief of each attribute for Store One, Store Two, and Store Three. A number higher than their ideal amount meant that the store had a saturated amount of the attribute. A number lower than the ideal point meant the respondent felt the store did not offer enough of the attribute. H1-5: No significant differences exist between Store One and Store Two managers' professed belief of the level of each attribute offered by Store One. A t-test was conducted to determine whether managers had similar beliefs of attributes offered by Store One. An analysis of data re- vealed that managers had significantly different professed beliefs con- cerning the amount of five attributes offered by Store One (see Table 4.17). 77 .ommo mo: mpoEFpmm moooFoo> mooooomm mo» oo ommoo u ooo omumonoo mom: soommow Fo mmmoomo moo om .mo. om Fooom oo coop mmmF mm: o mob .omsoomomo mo: mooo 1Foo> mFoEom mo mmmu o oo .moooFoo> moom mom was mooooo oz“ mo» poop omEommo mo pooooo NF mmooomm F ooN.o nmo.oN oN momooooz o3» moopm Nm. oo. «MN.o ooN.F FFF.oF o momooooz moo mooom ooFuomme mmFoooooomz mmN.o ooo.oF NN momooooz ooh mooum Nmm. No. ooo.F moo. FFF.oF o momooooz moo mooum F moFoo coo moFo> Foo.m moo.NF oN meeoeeez 03F meeom om.NN oN.N ooo.m oNo.o Noo.oF o momooooz moo mooum momoomoso< mooom Neo.o omN.NF oN meooeeez ozF meoom om.oN Fm.F oFN.m Fom.N oom.oF o momooooz moo mooom moF>omm mmFom mmNoFoEm mo Fm>mo Fem.o mom.oF NN meeoeeez ozF eeeom No.oN oN. moo.m oom. moo.mF o momooooz moo moopm NoFFeeo Noo.m oNo.oF oN momooooz ozF mooum No.om No. mmm.oF oNo. oo.oF o momoooo: moo mooam F moomFom>ooo Fooonoooo mo moFo>1F om moomomFFFo oomz oomz z mpooFoup< ammu1h .mooom uoooomFo FomoF FooFomouooz: ee Ne omeeoeo eoeeFeooo seem too oeFee Femoo ego me meFFem oemmemoee.meeoeeez .oF.o eFeeF 78 .ommo mo: mpoeFomm moooFoo> mooooomm moo oo ommoo u now ompmonoo mom; EoommoF mo mmmoomo moo om .mo. ou Fooom oo coop mmmF mo: o mob .omEoomomo mo: mooo 1Foo> mFoEom Fo ammo o oo .moooFoo> mEom mop ooo mooooo oz“ mop poo» omsommo mo ooooou HF mmooommF :0. v93. Ibo. vat; ooN.o Nmo.ON NN memoeeez ozF eeeom oo.em emN.N moe.o mFo.m NNN.NF m meeoeeez meo meoom ooFuomme mmFoooooomz ooo.m omm.Fo NN memoeeez ozF eeoom eF.mm 1NN.N Foo.F NeN.eN FFF.NF o memoeeez meo meoom 00.7.5 Lou. m: Pm> moo.e omm.N oN memoeeez ozF meoom oo.mm eeeoe.e Fem.o oNN.o oNN.mF o memoeeez meo eeoom momoomoso< mooom omm.o FNN.N oN memoeeez 03F eeoom oo.mm 1No.N Fem.m ooN.m ooo.NF o memoeeez eeo meoom 9:5me mmFom mmaoFosu to Fm>mo omo.m Nmo.FF NN memoeeez 03F eeeom oo.em «oo.F oNo.m oNo.m oNN.mo m memoeeez meo meeom NoFFeeo FNN.FF NmN.oF NN memoeeez ozF meoom oo.Nm NF.F oNN.N omo.N Noo.mF a meeoeeez meo meeom F moomFom>ooo FoooFuoooo oo meFe>1F om eoememeeoo eeez eemz z eoeeFeoo< ommo1F .eeo eeoom Ne oeemeeo moeeeeooo eeeo to eeFme ommmmceee .meeoeeez .No.o mFeeF 79 Store One managers held a significantly higher professed belief concerning the value for price offered by their store than did Store Two managers. The second largest difference of opinion concerning attributes offered by Store One was the store atmosphere. The attrib- utes merchandise selection, quality of merchandise, and the level of employee sales service for Store One were also deemed significantly different between managers. These results indicate that the null sub- hypothesis 1-5 is rejected concerning the attributes quality, store atmosphere, employee sales service, merchandise selection, and value for price offered by Store One. No significant difference was found between managers' professed belief concerning the locational convenience of Store One. The null sub-hypothesis 1-5, therefore is not rejected for this Store One attribute. Hl-6: No significant differences exist between Store One and Store Two managers' professed belief concerning the amount of each attribute offered by Store Two. Results of the t-test comparing managers' professed belief of attributes offered by Store Two are summarized in Table 4.18. The level of employee sales service offered by Store Two was perceived signifi- cantly different by managers (p‘<.01). With a mean difference of 5.238, Store Two managers belief about the sales service offered by Store Two was significantly higher than Store One managers' belief. Due to the significantly different professed beliefs, the null sub- hypothesis Hl-6 is rejected for the level of employee sales service offered by Store Two. 8O .ommo mo: maoEFomm moooFoo> muoooomm mop oo ommoo p ooo omomonoo mom: Eoommow Fo mmmoomo moo om .mo. op Fooom oo ooou mmmF mm: o moh .omsoowomo mo: moom 1Foo> mFosom Fo ammo o oo .moooFoo> msom mop ooo mooooo o3» moo poop omsommo mo uooooo NF mmooomo F mo. van. 2b. v93 mom.FF FNN.FN NN meeoeeez 03F eeoom oo.mm 4N. oom.FF moo.F NNN.oN m meeoeeez eeo eeoom ooFuomme mmFoooooomz ooN.N oNo.mF NN memoeeez ezF eeoom mo.Nm FNm. NFm.m oom. mmm.o_ o mteoeeez meo meoom mot; 15.... m:..m> oNN.o omo.mo NN memoeeez ozF eeoom o_.Nm FNN. mom.m mo. ooo.mF o memoeeez eeo meoom momoomoEp< moopm oFF.N FNm.NF oN memoeeez ego meoom oo.mN FeeNo.N NNe.m omN.m mmm.NF o memoeeez eeo eeoom muggmm mmFom mmNoFoEm mo Fm>mo ooe.m oNo.oF NN memoeeez ezF meoom mN.mm NF._ NNN.N Nom.N FFF.oF o memoeeez eeo eeeom o . NoFFeeo Foo.F NmN.NF oN meeoeeez ezF eeoom oF.mm Foe.F moo.N oNo.e oNN.mF o memoeeez eeo meeom mucow=m>cou 3:05.951— oo meFe>1F om ooeeeeeLFo eemz eeez z moeeFeooo ammuuh .ozF eeoom Ne oeemmco moseFeooo eeem to cmFFmN ommmmceee.memoeeez .NF.o mFeeF 81 Managers from Store One and Store Two held similar professed beliefs concerning the locational convenience, quality, store atmosphere, merchandise selection, and value for price offered by Store Two. The lack of significant difference in professed beliefs indicate that the null sub-hypothesis 1-6 cannot be rejected for these five attributes offered. H1-7: No significant differences exist between Store One and Store Two managers' professed belief concerning the amount of each attribute offered by Store Three. When managers were asked about their professed belief of each attribute offered by Store Three, two of the six attributes were viewed significantly different between the managers (see Table 4.19). A t-test was conducted to determine whether managers had similar professed beliefs concerning the amount of each attribute offered by Store Three. An analysis of the results showed that Store One managers and Store Two managers held significantly different professed beliefs concerning two of the attributes offered by Store Three. More specifically, Store One managers' professed belief concerning the level of employee sales service and merchandise selection offered by Store Three was signifi- cantly higher than Store Two managers professed belief (p< .01). The sub-hypothesis that no differences exist between managers' professed beliefs relating to the level of employee sales service and merchandise selection offered by Store Three is rejected. Results of the t-test revealed that managers held similar pro- fessed beliefs concerning the value for price, locational convenience, quality and atmosphere offered by Store Three. Due to the similar professed beliefs for these four attributes offered by Store Three, .omm: mo: mpoEFpmm mpoooomm mop oo ommoo p ooo ompmoooo mom: soommop mo mmmoomo mop om .mo. op Fooom oo ooop mmmF mo: o mob .omsoomomo mo: moon 1Foo> mFosom mo pmmp o oo .moooFoo> moom mop coo mooooo o3p mop poop omsommo mo pooooo pF mmooommF 82 mo. v91 :0. vote moN.o moo.N NN meeoeeez exp meepm oo.mm .1FN.N ooN.o Nmm.m oom.mF o meeoeeez meo weepm oonumme mmpoooooomz ooo.N oNo.op NN meeoeeez esp mooom mo.Nm FNm. NFm.m mom. mmm.NF N meeoeeez eeo eeopm moFoo tom moFo> Foo.m mFF.o NN meeoeeez exp meoom mm.N oF.N NFo.N oN.o mNN.NF N meeoeeez eeo meepm F momoowosp< moopm NNN.m NNe.o NN meeoeeez esp meopm oo.Nm eeoo.N oFo.N NNm.N ooo.oF o meeoeeez meo meopm muF>omm mmFom mmonQEN mo Fm>mo mNN.m oom.o NN meeoeeez esp meepm oo.Nm om._ omo.o mNm.m mNo.NF o meeoeee: eeo meopm NpFFeeo omm.o peN.NF NN meeoeeez esp meopm No.om No. FNN.m omF. mNN.NF o meeoeeez meo eeopm F moomFom>ooo Fooonoooo oo meFe>1F om eoeeeeoopo eeez eat: 2 mpeepepp< pmmp1p .eeeep eeopm No oatmeco mpeepeppo seem to peppem ommmepoee .memoeeez .oF.o mFeeF 83 the null sub-hypothesis cannot be rejected. H2: No significant differences exist between consumers' pro- fessed image of Store One, Store Two, Store Three and the hypothetical ideal store as defined by the Beckwith and Lehmann multiple attribute model. H2-1: No significant differences existed between consumers' impressions of Store One, Store Two, and Store Three. An analysis of variance (ANOVA) was conducted to determine whether consumers' overall impressions of each store were significantly dif— ferent. The results of the ANOVA test are presented in Table 4.20. Table 4.20. Consumers' Overall Impressions Toward Store One, Store Two, and Store Three. One-way Analysis of Variance Dependent Variable = Consumers' Overall Impression 8-3-192 ................................................................ DF Sum of Squares Mean Sguares F Radio Between Groups 2 2333.1667 1166.5833 2.956 Within Groups 9 3551.7500 394.6389 Total 11 5884.9167 An analysis of the results indicated that consumers' overall impression toward Store One, Store Two, and Store Three were similar. Due to the lack of significant differences between consumers' overall impression of the three discount stores, the null sub-hypothesis 2-1 cannot be rejected. H2-2: No significant differences existed among consumers ideal point of each attribute and their professed beliefs of the amount offered by each discount store. 84 A one-way analysis of variance was conducted to determine whether the mean differences between consumers' ideal point of each attribute and their professed belief of the amount offered by each discount store differed significantly. An analysis of the data revealed that the beliefs were similar. Therefore, null sub-hypothesis 2-2 cannot be rejected concerning the value for price, quality, merchandise selection, locational convenience, level of employee sales service, and atmosphere offered by Store One, Store Two, and Store Three. H3: When comparing Store One, Store Two, Store Three and the ideal discount department store, there will be no sig- nificant difference between consumers' and managers' per- ceived image among stores as defined by the Beckwith and Lehmann model. As was the case with the analysis of the null hypotheses one and two, the null hypothesis three is broken down into sub-hypotheses. The first sub-hypothesis is related to an analysis of consumers' and managers' professed ideal amount for each attribute offered by a dis- count store. H3-1: No significant differences exist between consumers' and managers' ideal amount of each attribute offered by an hypothetical ideal discount store. A one-way analysis of variance was conducted to determine whether consumers and managers had similar views concerning the amount of attributes offered by an hypothetical ideal discount store. A summary table of results from this analysis is presented in Table 4.21. A re- view of the data revealed that consumers and managers had significantly different professed beliefs concerning the ideal level of three store sttributes. The ideal level of employee sales service, store atmosphere, 85 mo. v9. moo.NoNo NoN Fopep moo.NN omo.oomo ooN moeoeo epoppz om.N mom.mm moo.oFF N meeoeo eeezpem moF>omm mmFom mmNoFoEm mo Fm>mp mom.onFF FoN Fopop oom.mm mFm.omoFF ooF mesoeo eFesz No.N ooF.NmF ooN.oFm N meeoeo emmzpeo momoomoEp< moopm oNo.monF moN Fopop moF.eo moo.FooNF FoN meaoeo epzppz No.N NoF.FNF mom.Nom N meeoeo eeezpoo oonomme mmFoooooomz omo.oommF moN Fopop oNo.mN Noo.oonF FoN moeoeo epeppz NoN. mom.mm moo.FFF N moeeeo eoezpeo auto. Low mzpm> moo.NoFm_ moF Fopop Foo.oN mNo.omomF NNF meeeeo eoeppz oNN. oom.mm oNN.oFF N m ooeo eeeepem NpFFoeo NoN.oNNoF NoN Fopep oom.oo oNo.oFNNF ooN meeoeo epeppz tmNo. ooN.N on.o N meeoeo eeezpem mucmwcmZ—ou FocowumUnfi oppoz o mmoooom oomz mmoooom mo msom no mpoopopp< moooFoo> mo mFmNFoo< Ao31moo .mmoop moopm an omommmo mpooFopp< ooou Fo poooe< mop mo mFmFme ommmmwooo .mom53mooo ooo .momooooz .FN.o mFooe 86 and quality was viewed significantly different by consumers and managers. In order to determine the desire to which consumers and managers responses were different, a t-test was conducted (see Table 4.22). With mean differences of 5.805 and 4.762, consumers had significantly lower professed levels for the quality and level of employee sales service desired by their hypothetical ideal discount store than did managers. The level store atmosphere was also less for consumers than managers for the ideal store (p< .01). The value for price offered by the ideal store was the only attribute that consumers rated signifi- cantly higher than managers (p<:.05). Due to the significant differences between consumers' and managers' ideal levels for the quality, level of employee sales service, store atmosphere and value for price, the null sub-hypothesis 3-1 is rejected for these attributes. No significant differences in the ideal level of locational convenience and merchandise selection were seen, therefore the null sub-hypothesis 3-1 cannot be rejected for these attributes. H3-2: No significant differences exist between managers' and consumers' overall attitude toward Store One as defined by the Beckwith ahd Lehmann model. As previously discussed in Chapter III, the Beckwith and Lehmann model takes into consideration the ideal point of each attribute, the perceived amount of each attribute and the weight of each store attrib- ute. A one-way analysis of variance was conducted to determine whether consumers and managers had similar computed overall attitudes toward Store One when utilizing the Beckwith and Lehmann model. 87 .omm: mo: mpostmm moooFoo> mpoooomm mop oo ommoo p ooo ompmonoo mom: Eoommow mo mmmoomo mop om .mo. op Fooom oo ooop mmmF mo; o moF .omsoooomo mo: mooo 1Foo> mFoEom Fo pomp o oo .mooopoo> msom mop ooo mooooo ozp mop poop omsommo mo pooooo pp omooommF mo. 1... :o. vet. :8. var... oFo.o Noo.oF ooF momsomooo o.FNN co. on.o moN. Nmo.oN um momooooz oonomme mmpoooooomz moo.oF ooo.FN ooF momsomooo F.No FeFm.N Fmo.m NoF.m ooN.mF No mooooooz mopoo Lop moFo> NNo.o mmm.oF omF momsomooo o.FNN eeNo.m mom.m ooN.m NNo.mF No meeooeez momoomoEp< moopm oom.N oFo.FF owF momosmooo o.mo Feetom.o oom.m NoN.o oNo.oF mm momooooz moF>omm mmFom mmzoFoEm mo Fm>mo omo.m Nmo.FF ooF mmmwmmooo F.NoF NN.o FoN.o moo.m oNN.oF mm o oooz Fifi HF. PM: oom.~ NNm.mF omF momsomooo o.FNN Fm. oNo.o oNF.F moN.oF No momooooz ooomFom>ooo Foooppoooo mo moFo>1H om moomomFFFo com: com: 2 mpooFopp< pmop1p .eeopm poooomFo Fomoo mop zo omompwo mpoopopp< zoom mo poooe< FomoF .mom22mooo ooo .momoooo: .NN.¢ mFooF 88 A summary of results of consumers' and managers' computed atti- tudes for Store One is presented in Table 4.23. An analysis of the results indicate that consumers and managers have similar attitudes toward Store One. Due to the lack of significant differences in atti- tude toward Store One, the null sub-hypothesis 3-2 cannot be rejected. Table 4.23. Managers'and Consumers’Overall Attitude Toward Store One as Defined by the Beckwith and Lehmann Attitude Model. One-way Analysis of Variance Store One DF Sum of Squares Mean Squares F Ratio Between Groups 2 67.492 33.746 .703 Within Groups 220 10561.501 48.007 Total 222 10628.993 H3-3: No significant differences exist between managers' and consumers' overall attitudes toward Store Two as defined by the Beckwith and Lehmann model. A one-way analysis of variance was conducted to determine if con- sumers and managers held similar overall attitudes toward Store Two. Results of the ANOVA test are presented in Table 4.24, on the following page. An analysis of the data revealed that consumers and managers held similar overall computed attitudes toward Store Two. The lack of significant overall attitudes toward Store Two indicate that the null Sub-hypothesis 3-3 cannot be rejected. 89 Table 4.24. Managers' and Consumers' Overall Attitudes Toward Store Two as Defined by the Beckwith and Lehmann Model. One-way Analysis of Variance Store Two OF Sums of Squares Mean Squares F Ratio Between Groups 2 16242.803 8121.408 .465 Within Groups 220 3843866.108 17472.119 Total 222 3860108.911 H3-4: No significant differences exist between managers' and consumers' overall attitudes toward Store Three as defined by the Beckwith and Lehmann model. Results of the ANOVA test revealed that consumers and managers lield similar overall attitudes toward Store Three as well as Store One and Store Two (see Table 4.25). ‘Table 4.25. Managers' and Consumers' Overall Attitudes Toward Store Three as Defined by the Beckwith and Lehmann Model. One—way Analysis of Variance ‘ \ Store Three DF Sums of Squares Mean Squares F Ratio Between Groups 2 90.802. 45.401 .648 Within Groups 220 15405.415 70.027 Trotal 222 115496.217 ‘ «()ue to the lack of significant differences between managers' and con- ESIHners' overall attitudes toward Store Three, the null sub-hypothesis 3 -4 cannot be rejected. 9O H3-5: No significant differences exist between managers' and consumers' professed overall impressions of Store One. Respondents were asked to express their overall impression of Store One, on a ratio scale of l to 100. A score of one signified a very unfavorable overall impression of Store One whereas a score of 100 signified a very favorable overall impression of Store One. In order to determine whether differences existed between managers' and consumers' overall impressions of Store One, a one-way analysis of variance was conducted. Results of this test are summarized in Table 4.26. 'Table 4.26. Managers' and Consumers' Overall Impressions Toward Store One. One-way Analysis of Variance ‘ ¥ Store One DF Sums of Squares Mean Squares F Ratio [Between Groups 2 22272.099 11136.050 20.047** liithin Groups 217 120544.678 555.505 'Total 219 142816.777 **p< .0] ‘ lHanagers and consumers had significantly different professed overall impressions of Store One. The results of this test indicate that the riull sub-hypothesis 3—5 should be rejected. Before rejecting this sub- tlypothesis, however, an analysis of the t-test of managers' and con- Eiumers' overall impressions of Store One gave additional strength to the rejection. A summary of the t-test is presented in Table 4.27. 91 Table 4.27. Managers' and Consumers' Overall Impressions of Store One. T-test Dependent Variable = Overall Impression Toward Store Two N Mean Mean Difference SD T-Value Managers 37 49.405 12.735 29.166 2.80** Consumers 183 63.104 24.273 DF = 218 **p< .Ol With a mean difference of 12.735, consumers had a significantly higher overall impression of Store One than did managers. Since the analysis of the t-test and ANOVA showed that consumers and managers held sig- nificantly different professed overall impressions of Store One, the null sub-hypothesis 3-5 is rejected. H3-6: No significant differences exist between managers' and consumers' overall impressions of Store Two. When respondents were asked to express their overall impression (of Store Two, significantly different responses were given by managers iand consumers. A summary of a one-way analysis of variance between nnanagers' and consumers; professed overall impressions of Store Two ‘is presented in Table 4.28, on the following page. With a significance "level of p<:.OOl, the test results indicated that consumers and nnanagers did not hold similar overall impressions of Store Two. Due to 1:he significant differences between managers' and consumers' overall impressions of Store Two, a t-test was conducted to determine the mean 92 Table 4.28. One-way Analysis of Variance Between Managers' and Con- sumers' Overall Impressions of Store Two Dependent Variable = Overall Impression Toward Store Two Store Two DF Sums of Squares Mean Squares F-Ratio Between Groups 2 6815.924 3407.962 9.421*** Within Groups 220 79579.951 361.727 Total 222 86395.875 ***p < .001 differences in opinions. A summary of this test is presented in Table 4.29. Table 4.29. Managers' and Consumers' Overall Impressions of Store Two. T-test Dependent Variable = Overall Impression Toward Store Two N Mean Mean Difference SD T-Value Managers 37 84.432 11.588 2.686 3.34*** Consumers 186 72.844 1.453 DF = 221 ***p < .001 With a mean difference of 11.588, managers held a significantly higher overall impression of Store Two than did consumers. Due to the 93 significantly different professed overall impressions of Store Two (p<:.OOl), the null sub-hypothesis 3-6 is rejected. H3-7: No significant differences exist between managers' and consumers' overall impressions of Store Three. A one-way analysis of variance was conducted to determine whether managers and consumers held similar professed overall impressions of Store Three. A summary of the results from this test is presented in Table 4.30. As was the case in the analysis of respondents' overall impressions of Store One and Store Two, consumers and managers held significantly different professed overall impressions of Store Three. Table 4.30. Managers' and Consumers' Overall Impressions of Store Three. One-way Analysis of Variance Store Three DF Sum of Squares Mean Squares F Ratio Between Groups 2 4198.645 2099.323 3.868* Within Groups 206 111817.575 542.804 Total 208 116016.220 ' *p< .05 An analysis of the results indicate that the sub-hypothesis of no dif- ference between managers' and consumers' overall impressions of Store Three is rejected. A further analysis of the data through the employ- ment of_a t-test helped determine the exact differences in impressions (see Table 4.31). With a mean difference of 11.865, managers' overall impression of Store Three is significantly lower than that of 94 Table 4.31. Managers' and Consumers' Overall Impressions of Store Three. T-test Dependent Variable = Overall Impression Toward Store Two N Mean Mean Difference SD T-Value Managers 36 27.667 11.865 20.425 2.79** Consumers 173 39.532 23.775 DF = 207 **p< .Ol consumers'. The null sub-hypothesis 3-7 is therefore rejected due to respondents' significantly different overall impressions toward Store Three. Respondents were asked to evaluate the perceived level of attrib- utes offered by each store in relation to their hypothetical ideal discount store. One-way analyses of variances were conducted to deter- mine whether the six store attributes offered by Store One, Store Two and Store Three were perceived similarly by managers and consumers. H3-8: No significant differences exist between managers' and consumers' professed beliefs concerning the amount of each attribute offered by Store One. A summary of the results for the ANOVA test between managers' and consumers' professed beliefs of the amount of each attribute offered by Store One is presented in Table 4.32. Results of the test showed that managers and consumers held significantly different professed beliefs concerning one attribute offered by Store One. With a 95 mo. vat. mNo.ommoF mFN Fopop NoF.oo NNN.NoNoF NFN aoeoeo epeppz .o.F oNe.oo Nmo.oo N meeoeo eeezpeo mop>omm mmFom mmaoFoEu No Fm>mo Nom.eNmNF oFN Fopop ooN.om oNN.oooNF oFN moeoeo epeppz eFoN.e Foo.omN Foo.oNe N meeoeo eoezpeo momommosp< moopm Foo.NomFN oFN Fopop moo.No NNN.ooooN oFN mooeeo eposz mmm.N ooN.oNN oFo.omo N moeoeo eeezpeo oonomme mmFoooooomz ooo.eNomN NFN Fopep Noo.NoF ooN.mNomN mFN moeoeo epeppz NoF.N ooF.mNN ooN.omo N moeoeo eeezpeo mopoo to» moFo> Noo.ooNoN oFN Fopep mmF.oFF mNN.oNeoN eFN meeeeo epeppz on._ Foo.oNF oNN.omm N moeooo eoozpmm NpFFoeo oFN Fopoh oom.oFF NNN.NNmmN oFN moeoeo epeppz ooN. NFm.NN omo.oNF N moooeo eeozpoo mucm wcm>cou pmco H3001— oppom m mmoooom oomz mmoooom mo mEom no mpoopopp< moooFoo> mo mFmaFoo< No31moo .moo moopm on omommmo mpoopopp< zoom mo poooe< mop oo mmmFme ommmmoooo .mome=mooo ooo .momooooz .Nm.o mFooF 96 significance level of p<:.05, the atmosphere offered by Store One was perceived significantly different by managers and consumers. Due to the significant difference, the null sub-hypothesis 3-8 is rejected concerning the atmosphere offered by Store One. Managers and consumers had similar professed beliefs concerning the locational convenience, value for price, quality, merchandise selection and the level of employee sales service offered by Store One. Because no significant differences were viewed concerning these five attributes offered by Store One, the null sub-hypothesis 3-8 cannot be rejected. In order to determine the mean difference between managers' and consumers' professed beliefs of the atmosphere offered by Store One, a t-test was conducted (see Table 4.33). Unlike the results of the one- way analysis of variance, an analysis of the t-test revealed that managers and consumers held significantly different professed beliefs concerning two attributes offered by Store One; the ANOVA revealed that managers and consumers held significantly different professed beliefs of only one attribute offered by Store One. When respondents were asked to state their professed belief con- cerning the level of quality offered by Store One, consumers' belief was significantly higher than that of managers. Consumers also held a significantly higher professed belief relating to the amount of value for price offered by Store One. Due to the significant differences between managers' and consumers' professed beliefs of the quality and value for price offered by Store One, the null sub-hypothesis 3-8 is rejected for these two attributes. 97 ompmonoo mom: soommom wo mmmoomo mop om .mo. op Fooom oo ooop mmmF mo: o moF 1Foo> mFoEom mo pmmp o oo .moooFoo> meow mop ooo mooooo ozp mop poop omsommo mo pooooo pF mmooomo .ommo mo: mposppmm moooFoo> mpoooomm mop oo oomoo p ooo .omEooFomo mo: mooo F mo. vat 2b. v9.5. Nmo.o FooN.oF NNF moeoemeeo Foo. oFF.NF NFN. ooem.mF om meeoeeoz cowuumpwm mmwtcmsugmz FFo.FF NNo.oF oNF meeoemeoo o.oo peo.N mme.o NON.N NoN.mF om meoooeez moth. L0,... w:..m> oom.N ooN.N mNF meeoemeeo o.oo FoNo. moo.o NoN. Noo.o om meeooeoz momoomosp< moopm oom.N meo.o mNp meeoemeoo o.Fo Foeo. oFm.e FNo. oFm.o mm meoooeoz 9:?me mmFom mmzoFoEu mo Fo>mo omo.FF ooo.mF oNF memommeoo o._eF mN.N omm.m oom.m eNo.NF om wee eeez Fe. NoFFoeo omo.FF oom.mF NNF meeoemeoo o.mFN NN.F Noo.o mom.N moo.mF om meeoeeez moomFom>ooo Fooonoooo oo eeFo>1F om ooeeeeoeFo eoez com: 2 opeoFeppo pmmp1h .moo moopm an omomppo mpoopopp< ooou mo pooos< mop mo mFmFFmo ommmmpooo .mom53mooo ooo .momooooz .mm.¢ mFooh 98 The results of the t-test revealed that consumers and managers held similar professed beliefs concerning the locational convenience, store atmosphere, employee sales service, and merchandise selection offered by Store One. These similar professed beliefs indicate that the null sub-hypothesis 3-8 cannot be rejected for these four attrib- utes offered by Store One. ' H3-9: No significant differences exist between managers' and consumers' professed beliefs concerning the amount of attributes offered by Store Two. A one-way analysis of variance was conducted to determine whether consumers and managers held similar professed beliefs of the attributes offered by Store Two. When asked to state their professed beliefs con- cerning the amount of each attribute offered by Store Two, managers and consumers stated significantly different beliefs of three attributes offered by Store Two (see Table 4.34). The level of employee sales service and atmosphere offered by Store Two was perceived significantly different by managers and consumers (p<:.001). With a significance level of p<:.05, consumers and managers also held different professed beliefs concerning the value for price offered by Store Two. The results from the one-way analysis of variance revealed that consumers and managers held similar professed beliefs concerning the levels of merchandise selection, quality, and locational convenience offered by Store Two. Before any decisions were made concerning the rejection of this null sub-hypothesis, a t-test was conducted. The one-way analysis of variance revealed that the value for price offered by Store Two was viewed significantly different between 99 mo. v9: :0. v9.5. mSo. vat»; moo.omoNF FNN FopoF omo.Fm Noo.oomFF oFN meeeeo eFosz eteNNo.mF ooo.NNo Nom.memF N meeeeo eoezpem moF>oom mmFom mmNoFoEm mo Fm>mo NNF.NNFNF FNN FopoF Foo.mN ooN.FoooF oFN meeoeo eFosz teeoNo.N oNN.Nom NNo.mNFF N moeoeo eeezpem momoowosp< moopm ooo.mooom NNN FopoF Noo.NNF Fom.oNoNo oNN moeeeo eFosz ooo. oNo.oF Noo.oN N meeooo eoezpem oonumme mmFoooooomz ooo.NmoNN FNN FopoF moo.FoF NNN.oNMNN oFN moeoeo eFopF: eFNo. oFF.N NNN.o N moeoeo eeezpeo moFoo tom moFo> mFm.omFFm NNN FopoF mom.FoF NmF.FmFFo oNN moooeo eposz‘ NoN. Foo.No ooN.mo N moeoeo eoeepoo onFoeo oNF.omooN oNN FopoF NFm.oNF ooo.NFmoN oNN meeoeo eFosz mFo.F oNN.NFN oom.Nmo N moeoeo eeezpeo wucm wcm>cou Face .5000..— onom o mmoooom oomz mmoooom mo moom no mpoopopp< mooopoo> mo mpmaFoo< Aoz1moo .ozp moopm No omommmo mpoopopp< ooom mo pooos< mop Fo mFmFme ommmmmooo .mom53mooo ooo .momooooz .om.o mFooh 100 managers and consumers. Results of the t-test, however, indicate that the respondents held similar professed beliefs of the value for price offered by Store Two (see Table 4.35). The discrepancy of results lies in the fact that since it cannot be assumed that the two groups had the same variance, an F test of sample variance was performed. The F was less than or equal to .05, so the degrees of freedom were adjusted and t was based on the separate variance estimate. Managers' professed belief of the store atmosphere and level of employee sales service was significantly higher than consumers'. The null sub-hypothesis 3-9 is therefore rejected for these Store Two attributes. Managers and consumers had similar professed beliefs concerning the quality, locational convenience, value for price and merchandise selection offered by Store Two. Due to the lack of significant differ- ences, the null sub-hypothesis 3-9 cannot be rejected for these four Store Two attributes. H3-10: No significant differences exist between managers' and consumers' professed beliefs concerning the amount of each attribute offered by Store Three. A summary of results of a one-way analysis of variance between managers' and consumers' professed beliefs of the amount of each attribute offered by Store Three is presented in Table 4.36. An analy- sis of the results indicate that the two sub-samples held significantly different professed beliefs of a single attribute offered by Store Three. More specifically, managers and consumers viewed the locational convenience of Store Three significantly different (p<:.05). 101 1Foo> mFoEom wo pmmp o oo .moooFoo> msom mop ooo mooooo ozp mop poop omsommo mo poooou pp mmooomm .omm: mo: mpoEFpmm moooFoo> mpoooomm mop oo ommoo p ooo ompmonoo mom: Eoommop Fo mmmoomo mop om .mo. op Fooom oo ooop mmmF mo: o mob .omsoomomo no: mooo F poo . v at}, woo.mF ooo.oN ooF momoomooo o.FNN oo. ooo.FF oFo. mFo.FN om momooooz oonomme mmpooooocmz NNo.oF ooo.oF ooF momsomooo o.No FoN. Nmo.o on. NFN.NF om momooooz mopoo ooF moFo> Nom.o omm.o omF momsomooo o.oNN «teem.m me.o Nmm.m FFN.mF om momooooz F momoomosp< moopm Fom.N mmF.oF ooF momsomooo o.oNN «eeNo.o Noo.o moo.m Fmo.oF om momooooz moF>omm mmFom mmNmFoEm mo Fm>mo moo.NF mFo.oF ooF momsmmooo o.ou om. moo.o om. mNo.NF om mom oooz F NoFFooo com. omm.mF oNF memosmooo o.FNN oo.F moo.F NNo.m oNo.NF om momooooz moomFom>ooo Fooonoooo oo moFo>1F om moomommppo oomz com: 2 mpooFopp< pmmp1h .ozh moopm No oeeeopo mpooFopp< ooom Fo poooe< mop Fo mmmFme ommmmmooo .mome:mooo ooo .momooooz .mm.o mFooh 102 1Foo> mFoEom mo pmmp o oo .muooFoo> msom mop ooo mooooo ozp mop poop om52mmo mo pooooo pF mmooomm .omm: mo: mpoEFpmm mooopoo> mpooooom mop oo ommoo p ooo ompmonoo mom: soommop mo mmmoomo mop om .mo. op Fooom oo ooop mmmF mo: o mop .omsoomomo mo: mooo F 80. v ate. Nmo.o mNo.FF NoF momsomooo o.ou om.F Noo.m oFo.F FFo.o Nm momooooz F oonomme mmFoooooomz ooF.o oNo.NF NoF moosomooo N.NN FNo. mom.o oN. omo.FF No meoooeoz moFoo tom moFo> mNo.N oom.m ooF momsomooo o.oo Fom.F Nem.m mom.F NNo.N om meaoeeoo momoomosp< moopm Nmo.o omo.m NoF momsomooo o.FoN tNo.F mom.o omN.F ooo.N om momooooz moF>omm mmFom mmNoFmam mo Fm>mo NNN.o oom.FF moF momsmmooo oN.No mo.F moo.m mNN.F ooN.oF mm mom oooz F NoFFeeo Nom.oF oNN.NF ooF momsomooo oN.No om. NoN.N moo. NoF.mF mm meooeeoz F moomFom>ooo Foooppoooo oo moFo>1h om moomomoppo oomz oomz z mpoopopp< pomp1h .mmooe moopm mo omommmo mpoopopp< ooou mo pooos< mop mo mFmFme ommmmmooo .momsomooo ooo .momoooo: .om.o mFooH 103 Results of the one-way analysis of variance revealed that managers and consumers held similar professed beliefs concerning the value for price, quality, merchandise selection, level of employee sales service and atmosphere offered by Store Three. Managers and consumers held significantly different beliefs concerning the locational convenience of Store Three. Because of the significant differences between managers' and consumers' professed beliefs of the locational convenience offered by Store Three, a t-test was conducted for further analysis (see Table 4.36). Because it cannot be assumed that the two groups had the same variance, an F test of sample variance was performed. The F was less than or equal to .05, so the degrees of freedom were adjusted and t was based on the separate estimate variance. The results of the t-test did not coincide with the results of the ANOVA test. More specifically, an analysis of the t-test showed that managers and consumers had significantly different professed beliefs concerning the level of employee sales service offered by Store Three. The null sub-hypothesis 3-10 is rejected for the level of employee sales service offered by Store Three due to the significantly different professed beliefs revealed by the t-test. An analysis of the t-test between managers' and consumers' professed beliefs concerning the amount of attributes offered by Store Three revealed that managers and consumers did not have significantly different professed beliefs concerning the quality, locational conven- ience, value for price, merchandise selection and atmosphere offered by Store Three. The null sub-hypothesis 3-10 cannot be rejected for these Store Three attributes. CHAPTER V DISCUSSION A stepwise regression analysis was conducted to determine which store attributes were significant in predicting the respondents' pro- fessed image of each discount store. The regression analysis helped fulfill the first objective of this study. The second and third objectives were related to the first and second hypotheses. The third and fourth objectives were related to the third hypothesis. The results stated earlier in this study address each store attribute, sub-sample and discount store separately. In this discussion chapter, the discussion of the objectives and hypotheses will be of a general nature. The hypotheses were originally stated in the null form; this discussion will focus on the alternative hypotheses. Multiple Regression Analysis The multiple regression analysis of the Beckwith and Lehmann multiple attribute model was highly significant for predicting con- sumers' overall impression of each discount store. The model was also highly significant in identifying the total sub-sample of managers' overall impression of Store One, and Store Two managers' overall impres- sion of Store Two. A summary of significant best term models for 104 105 respondents' overall impression of Store One, Store Two and Store Three is presented in Table 5.1. The stepwise regression analysis showed that all attributes were significant predictors for identifying consumers' overall impression of each discount store. This finding supports research by Beckwith and Lehmann (1973), Fishbein (1967), Marks (1976) and Rosenberg (1957). These researchers hypothesized that only salient attributed should be used to analyze store image. Despite the significant results of the model as predictors, the R square for each best term model foreconsumers' overall impression ' of each store was extremely low. The R squares for the best six term model for Store One, Store Two and Store Three were .12319, .06491, and .27342 respectively. These low values suggest that salient store attributes were not included in this study. One possible explanation for this may be in the selection process of salient store attributes in professing an image. A small sample (n = 27) was selected for dis- cussing important attributes in a discount store. Three store attrib- utes identified as salient by five of the 27 respondents were brand name merchandise, store reputation and services offered such as credit and layaway. Had a larger sample been used in the selection of attributes, these store attributes may have been viewed very important in a discount store by a larger number of people, thus have been included in the survey. None of the six store attributes was deemed a significant pre- dictor of Store One managers' overall impression of Store Two and the total sub-sample of managers' overall impression of Store Three. 106 oz oz oz moomFom>ooo Fooonoooo oz mm» mm> moFoo ooF moFo> oz mm» oz momoomosp< moopm oz mm> oz moF>omm mmFom mmzoFosm Fo Fm>mo oz mm> mm> oonomme mmFoooooomz oz mm> mm> zpFFooo momooooz ooh moopm oz oz mm» moomFom>ooo Fooonoooo oz oz mm» mopoo tom moFo> oz oz mm> momoomoEp< moopm oz oz oz moF>omm mmFom mmzoFoEu Fo Fm>mo oz oz mm» oonomme mmFoooooomz oz oz oz prFooo momooooz moo moopm mm> mm> mm> moomFom>ooo Fooonoooo mm> mm> mm» mopoo com moFo> mm> mm» mm» momoomosp< moopm mm> mm> mm> mop>omm mmFom monFoEu wo Fo>mo mm» mm» mm> oonomme mmpoooooomz mm> mm» mm> apFFooo momsomooo mmoop moopm ozF moopm moo moopm Noopmmmoosp FFoom>o .mpomoooommo oopmmmooxm op poooFFFooFm Fmoos mop mo: mmeFoo< oopmmmoomm mmpzompm mo mFooH Noossom .F.m mFoo» 107 Managers play an important role in selecting which store attributes to emphasize, how much to offer and when they should be offered. By not including managers in the selection of salient store attributes, an important sub-sample of the target market was omitted. Additional salient store attributes may have been suggested if managers had been able to participate in the selection process; the omission of salient attributes associated with consumers' overall impression of each store may have also been included. This finding of differences between the saliency of attributes by the type of subject studied is in agreement with the results of the Hirschman study. Hirschman et a1. (1978) conducted a study to determine whether salient store attributes differed between target markets and objects studied. The researchers concluded that the saliency of attributes did vary between the target markets and the type of stores being investigated. Correlation Coefficients of Salient Store Attributes These correlations revealed that consumers' and managers' overall impressions of each store was strongly affected by the several attrib- utes offered and their interaction with one another (see Tables 5.2, 5.3 and 5.4). The strong interaction of several store attributes on consumers' and managers' overall impressions of each discount store ‘was expected due to the complexity of attitudes. Wilkie and Pessemier (1973) hypothesized that due to the complex- ity of attitudes, their analysis should be studied on the basis of many 108 moomFom>ooo momoomo5p< moF>omm mmFom mopoo Fooonoooo moopm mmaoFoEN Fo Fm>mo com moFo> mNooN. ooooo. oomNo. ApFFooo NNNNm. ooooo. moompom>ooo Fooonoooo Nmomm. ommNN. meeoomoopo eeepm mpoopopp< mmooh moopm op mmmooowmm .momooooz moo moopm momoo moF>omm mmFom moFoo oonmmme moompom>ooo 1mosp< mmaoFoEm top mmFo oomemooEF Fooonoooo moopm mo Fm>mo moFo> 1ooooomz FFoom>o moons. NoFmo.1 Nomom.1 Noooo. oooom.1 eonom.1 apFFooo mNNoN. oFoom.1 momom.1 moomFom>ooo Foooppoooo momoo. omFmo.1 momoomosp< moopm Noooo.1 moF>oom mmFom mmonoEN mo Fm>mo Noouo.1 mopoo com moFo> mmpoopopp< ozF moopm Fo mmmooommm .momooooz moo moopm moomFom>ooo momoomosp< moF>omm mmFom oopmmmoosF Fooonoooo moopm mmzoFosm.Fo Fm>mo FFoom>o .momooooz moo moopm Nono. NFNNo. oommo.1 prFooo onoo.1 moomFom>ooo Foooppoooo mooom.1 momoomosp< moopm memos. moFoo Low moFo> mmpoopopp< moo moopm mo mmmoomwmm .mommoooz moo moopm wepeoFepp< oopoFeeeeo »_oon to oFooF Nooeoom .N.m eFooF 109 momoomoEp< moF>omm mmFom oonomme moopm mmzoFoEm mo Fm>mo mmpoooooomz mFooo. onoN. NoooN. FoFoN. NmnmN.1 meoo. NoooN.1 mFmom.1 moF>omm mmFom moonoEm mo Fm>mo mmem. NpFFoeo moomFom>ooo Fooonoooo momoomoop< moopm moF>omm mmFom mmzoFoEu mo Fm>mo ozF moopm op mmmooommm .momEomooo momoomosp< moopm moo moopm op mmmooowmm .momsomooo mopaopeppo oepoFeeeeo NFooFI to oFooF NooEEem .o.m eFooF 110 moF>omm mmFom oonomme mmonoEN Fo Fm>mo mmFoooooomz oFooo.1 moomFom>ooo Fooonoooo oNoom. 2232.2 953 oFooo.1 moF>omm mmFom mmaoFosm po Fm>mo mmoop moopm momoomoEp< muF>omm mmFom mopoo oomeoooEF moopm mmonosu mo Fm>mo Lop moFo> FFoom>o nmmom. NpFFooo momoN. ooNoo. momoomosp< moopm FoFFo. oncom. moF>omm mmFom mmonoEm mo Fm>mo ozF moopm ooF>omm mmFom mopoo ooppomme mmzoFoEN mo Fm>oo Lop moFo> mmFoooooomz mooFm. zpFFooo FNoNN.1 moomFom>ooo Fooonoooo 3E. 2238.2 953 moo moopm mmmooommm .momooooz ozF moopm Fo mpomFoFmeoo oonoFmoooo .o.m mFooo 111 attributes and/or dimensions. Studying a person's overall impression of a store based on just one attribute would be misleading; the results of the data are oversimplified and erroneous conclusions could be made. As was discussed in the Review of Literature, attributes and experi- ences affect impressions and images. Due to the significance that store attributes can have on consumers' image of a store, researchers have emphasized the importance of understanding consumers' impression of each attribute offered by the store as well as their overall impres- sion of that store (Cohen, 1967; Pathak, 1971; Wyckham, 1969). For example, a person may have a favorable overall impression toward a store. However, this favorable overall impression does not necessarily mean that the store is "perfect" in the eyes of the consumer. The con- sumer may have felt that various attributes offered by the store are negatively correlated but do not greatly affect their overall impres- sion of that store. A summary table of the hypotheses and statistical analysis is presented in Table 5.5. A discussion of the results for each hypothesis and sub-hypothesis are given in the following section. Hypothesis 1 H1: No significant differences exist between Store One managers' and Store Two managers' professed image of Store One, Store Two, Store Three, and the hypothetical ideal store. Store One A t-test was conducted to determine whether differences exist between Store One managers' and Store Two managers' professed belief of 112 mpmep1p eFonFoz oee z<>oz oo mpmapoeo z<>oz co mFmNFoeo momep1p eFeppFez .Fmoos mooprpo mpooFoppo mFonFoe ooosomo ooo oppzoomm mop mo omoFFmo mo moopm Fomoo Foonmopooxo mop ooo mmooh moopm .oz» moopm .moo moopm mo mmoosp oommmmooo .momooooe ooo .mom52mooo ommzpmo mexm mmoomompmpo poooFFFooFm oz .moopm FomoF Foonmopooao mop ooo .mmoob moopm .o3F moopm .moo moopm Fo moosp ommmmmooo .momsomooo ommzpmo pmpxm mmoomommppo poooFmFooFm oz .moopm FomoF Foonmopooxo mop ooo .mmooF moopm .oz» moopm .moo moopm Fo mooEF ommmmpooo .momooooe ozo moopm ooo .momoooos moo moopm ommzpmo mexm mmoomompmpo poooFmFooFm oz ooNeFeom pmeF Feonmeopm mmeopooz: mpmaFoo< Foonmeopm ooo mmmmopooaz mo mFoop NgoEEom .m.m mFooH 112 momep1p eFerFez .Fmoos mooppppo mpoopoppo mFonFos ooosomo ooo oppzoomm mop an omopmmo mo moopm Fomoo FoonmopooNo mop ooo mmoop moopm .ozp moopm .moo moopm mo mmoosp ommmmpooo .momooooa ooo ooo F<>oz mo mFmaFoo< .momsomooo ommzpmo pmpxm mmoomomFFFo poooFFFooFm oz ”m: .moopm FomoF Fooppmopooao mop ooo .mmooF moopm .ozF moopm .moo moopm mo moosF ommmmmooo F<>oz mo mFmaFoo< .momsomooo ommzpmo mexm mmoomommwpo poooFFFooFm oz ”N: .oeopm Foooo Foonoopoozo mop ooo .omooe moopm .ozp moopm .moo moopm Fo mooep ommmmmooo .momoooos ozF moopm ooo .momooooe mpmmp1p mFonFoz moo moopm ommzpmo mexm mmoomomFmFo pooonFooFm oz "F: omonoEN pomp Foonmeopm mmeopooxz mFmaFoo< Foonmeopm ooo mmmmopoo»: Fo mFook Nooesom .m.m mFoo» 113 each attribute offered by three competing discount department stores. Results of the t-test revealed that Store One managers held a signifi- cantly higher professed belief concerning the quality of merchandise, level of employee sales service, store atmosphere and merchandise selection offered by their store than did Store Two managers. Store Two Managers held significantly different professed beliefs concern- ing one of the six attributes offered by Store Two. The level of employee sales service offered by Store Two was the only attribute to be perceived differently between managers. Store Three Two of the six attributes offered by Store Three were viewed significantly different between managers. Store One managers viewed the merchandise selection and employee sales service offered by Store Three significantly higher than did Store Two managers. These findings support research conducted by Isaacson (1964). In a study concerning store image, Isaacson noted that managers tended to hold a biased viewpoint in favor of their respective stores. More specifically, the managers surveyed rated their impressions and per- ceived level of attributes offered by their store significantly higher than did their competitor. Attitudes are formed through experiences and associations. The biasism in favor of the managers' respective stores may be due in part to their experiences with their employers (i.e., the company). 114 Managers' close involvement with the employees in particular and the company in general may have influenced their responses. Interpretations of these findings indicated that the store image is perceived differ- ently by various retailers. In addition, the amount of each attribute offered by the competitor may also be viewed differently. Hypothesis 2 H2: No significant differences exist between consumers' professed image of Store One, Store Two, Store Three, 1 and the hypothetical Ideal Store. 1 Results of an ANOVA test revealed that consumers did not hold significantly different overall impressions of Store One, Store Two, and Store Three. These implications are contradictory to many researchers' hypotheses. Furthermore, the difference between the ideal point for each attribute and the amount of that attribute perceived to be offered by each discount store was not significantly different. One possible reason for the lack of significant difference in consumers' overall impression of the three discount department stores and the mean differences between the ideal point and actual point may be caused by consumers' view of discount stores only as a discount store. The consumers have not refined the definition of the discount store. That is to say, they have not defined a "low image" discount store, a "medium image" discount store and/or a "high image" discount store. Future studies dealing with the perceived image of competing discount department stores are needed in order to assess whether consumers do in fact view the stores similarly. It is interesting to 115 note, however, that studies dealing with consumers' image of department stores revealed that consumers held significantly different images of the competing stores. In an image study conducted by Pathak (1972), the researcher concluded that consumers' impressions or images of various department stores varied. More specifically, consumers viewed three department stores to have a "high", "medium", and "low" image. Rich and Portis (1964) also came to conclusions similar to that of Pathak. Rich et al., analyzed consumers' image of nine department stores in New York and Cleveland. Although all nine stores were classified as the "typical department store" (i.e., the stores carried similar merchandise and offered similar services and by definition were classified as a depart- ment store), the researchers saw that the perceived images of the stores were separated into three categories. The categories included high-fashion appeal, price appeal, and broad appeal. The studies conducted by Pathak (1972) and Rich et a1. (1964), consisted of comparing department stores. Few image studies have dealt exclusively with discount department stores. Future research in this area may show that consumers define and form images of the dis- count store in the same fashion due to similar store attributes. Hypothesis 3 H3: No significant differences exist between consumers' and managers' professed images of Store One, Store Two, Store Three and the hypothetical Ideal Store as defined bydthe Beckwith and Lehmann multiple attribute attitude mo e . 116 Respondents were asked to state their overall impression of each discount store. Their impressions of each discount store were based on a ratio scale of l to 100, 1 being very unfavorable and 100 being very favorable. When analyzing respondents' overall impressions of Store One, Store Two and Store Three based upon this ratio scale, sig- nificant differences were seen. Table 5.6. Summary Table of Respondents' Overall Impression and Atti- tudes Toward Each Discount Store as Defined by the Beckwith and Lehmann Model Did a significant difference exist between managers' and consumers' responses toward each store? Store 1 Store 2 Store 3 Overall Impression Yes Yes Yes Attitude No No No These findings are similar to those of May (1971). May concluded that managers and consumers held significantly different impressions of their store image. Implications of these perceived differences indicate that retailers' marketing efforts are not accomplishing their goal; an image other than that desired by the corporate headquarters is being portrayed. Significantly different ideal levels, and professed beliefs con- cerning the amount of each attribute carried by the three discount stores were viewed by managers and consumer. The professed mean ideal 117 level for each attribute is presented in Table 5.7. Consumers' and managers' mean point for each attribute for Store One, Store Two and Store Three was compared with the mean ideal point for that attribute. Ideal Store Attributes Consumers and managers had significantly different professed ideal levels for three store attributes. Consumers placed a signifi- cantly higher value on the ideal level of quality of merchandise to be carried by a discount store than the managers. The sub-sample of managers however, held a higher ideal level of employee sales service and store atmosphere. These findings indicate that consumers would sacrifice store ambiance and additional sales service for better quality merchandise (see Table 5.7). A review of the current literature showed, however, that retailers, especially discounters are heavily investing in updating the atmosphere within the store. These retailers emphasize the importance of a differ- ential advantage of a positive store image and have tried to achieve this through additional service and store atmosphere; these attributes have not typically been offered by discounters. Goodman (1972) stated that an emphasis on store atmosphere can benefit the retailer in two ways: reorganization of the floor layout could result in increased productivity, traffic flow, merchandise visibility as well as make the store aesthetically pleasing. Despite this trend to trade-up, however, the discount retailer draws the majority of the traffic due to the price values offered. 118 mo. v9... m5. vote :oo. vat}. omzoo omoopz omzoo Noo.oF momsomooo omzoo omoopz omzoo Nmo.oN momooooz oonomme mmpoooooomz omzoo omzoo omzoo moo.FN momsomooo omzoo Fooom eomzoo eomN.mF momooooz mopoo to» moFo> omzoo omzoo oozoo mmm.oF momsomooo omzoo tteomoopz omzoo h.e..NNo.mF momooooz momoomoEp< moopm omzoo omzoo omzoo oFo.FF momsomooo agmzoo oerpozam Luzoo ottmmm.mp mgmmocoz moF>omm mmFom mmzoFoEN mo Fm>mo omzoo omzoo omzoo mNo.NN memosmooo omzoo omooF: to7833 ttooNNoF momooooz prFoom omzoo Fooom Fooou NNm.mF memosmooo omzoo omooF: omooF: moN.oF momooooz moomFom>ooo Fooonoooo mmooF moopm oz» moopm moo moopm moopm FomoF mpooFopp< moopm FomoH Foonmopooa: mop op oomFoooEoo oF moopm ooou zo ompoooo mmpoopopp< mop ooo poFoo FomoH mop ooFoomoooo mmmFFmo .momsomooo ooo .momooooz mo poooo zoosaom .N.m mFooF 119 In order to continue the values, these retailers must continue to operate at a smaller than average gross margin. The easiest way to do this is through self-service. As soon as the discounter offers addi- tional employee sales service, some other attribute offered must suffer (Goodman, 1972). Store One Results from a t-test showed that managers from Store One and Store Two held significantly different overall impressions of Store One. Store One managers had a much more favorable overall impression of their store than did Store Two managers (p<:.001). An analysis of the results from a t-test indicated that managers and consumers had significantly different professed beliefs concerning the quality and value for price offered by Store One. In addition to these significant differences, respondents felt that Store One carried less than its ideal level of five attributes. The low professed belief scores signified that consumers' "desires" are not totally fulfilled in any aspect other than the locational convenience. More specifically, Store One did not carry managers' and consumers' ideal level of merchan- dise selection, value for price, store atmosphere, employee sales service and quality. Store Two Managers and consumers expressed similar professed beliefs con- cerning the quality, locational convenience, value for price and 120 merchandise selection offered by Store Two. Consumers' professed beliefs of the level of employee sales service and store atmosphere however was deemed significantly lower than managers'. The results presented in Table 5.7 indicated that according to managers, Store Two offered an excessive level of quality, merchandise selection, atmosphere and locational convenience. The value for price and level of employee sales service, however, was seen to be approxi- mately equal to that of their professed ideal levels. Store Three The level of employee sales service offered by Store Three was the only attribute deemed significantly different between managers and consumers. Unlike the respondents' professed beliefs concerning attributes offered by Store One and Store Two, consumers and managers felt that every attribute offered by Store Three was lower than their ideal level. Summary of Interpretations An analysis of respondents' professed belief of six store attrib- utes offered by three specific discount department stores was made. The results showed that managers' and consumers' professed beliefs differed significantly. Interpretation of these findings indicate that the store attributes and perceived store image is seen differently by managers and consumers. 121 Studies conducted by McClure and Ryans (1968) and May (1971) had similar findings. These researchers concluded that consumers and managers had significantly different professed beliefs about the stores. Managers tended to rate their competition more negatively than they rated their own store. Managers also rated their competition more negatively than did consumers. In 1969, R. Wyckham conducted a study which dealt with the image of three traditional department stores, a discount store and an hypo- thetical ideal store. The researcher concluded that none of the stores analyzed was rated to be ideal, nor were any of the stores perceived to have too much of an attribute. In direct contradiction to the results of Wyckham's study, several store attributes were rated approximately equal to the ideal level by managers and consumers. In addition, managers consistently rated all the attributes offered by Store Two approximately equal to or hgiher than their professed ideal level. These results led the researcher to conclude that the discount retailers do not have a good comprehension of the attributes and weights of each attribute desired by the consumers. Although managers felt their respective stores were ideal, the consumers saw deficiencies in most of the attributes offered. CHAPTER VI SUMMARY AND RECOMMENDATIONS The purpose of this study was to investigate the perceived image of three discount department stores and the success with which these retailers offer store attributes desired by the consumer. The research objectives for this study included: (1) identify salient evaluative criteria for discount department stores; (2) measure consumers' and managers' professed overall impressions of each discount store in the survey; (3) identify consumers' and managers' perceptions of the ideal amount of each store attribute; and (4) compare consumers' and managers' perceived images of each store and the success with which the retailers offer the amount of each attribute desired by the consumer. Random telephone interviews were conducted to determine store attributes to be used in the study. The questionnaire was then de- veloped around the six chosen salient attributes. Additional random telephone interviews were conducted in order to pretest and revise the questionnaire. The prefix telephone numbers for the Okemos, East Lansing, and Williamston, Michigan areas and a random numbers table were used to determine the survey of consumers. A total of 186 questionnaires were successfully completed by consumers. Management from Store One and 122 123 Store Two participated in the survey. Nine managers from Store One and 28 managers from Store Two completed the survey. A stepwise regression analysis was employed to analyze the inde- pendent contribution of each term in the Beckwith and Lehmann model. The results of the stepwise regression indicated that quality was the best, one term model for explaining consumers' professed beliefs about all three discount stores. Quality and merchandise selection was the best two term model for explaining consumers' overall impression of Store One and Store Three. Despite the highly significant levels for each term, the low R squares indicated that several potential attributes may be missing from the study. The terms used in the Beckwith and Lehmann model were significant predictors of consumers' overall impre- sion of the three discount department stores. Furthermore, the term quality, was the most salient predictor of consumers' overall impression of a discount store. The results of the stepwise regression of managers' overall impression of the three discount stores indicated that the best one term and best two term models varied between managers and stores. The terms used in the Beckwith and Lehmann model were significant predictors of managers' overall impression of Store One and Store Two but not Store Three. However, no generalization concerning the best model for studying managers' overall impression of discount stores can be made due to the varying best one term model for each store. Managers and consumers were asked to express their overall impres- sions of each discount store. An analysis of a t-test showed that managers and consumers had significantly different overall impressions 124 of Store One, Store Two and Store Three. The implication of_these significant differences in opinions led the researcher to conclude that overall image of a store is based at least partially on the perceived ideal level of each attribute and the success to which the store satisfies the consumers' desires. Consumers and managers displayed significantly different professed beliefs concerning the quality, level of employee sales service, value for price and atmosphere carried by an hypothetical ideal discount store. The differences between the ideal amount of each store could inevitably affect how consumers view the store image in relation to the emphasis placed on each attribute by the retailer. The importance of these differing viewpoints were emphasized by Brown (1970). Brown stated that customer satisfaction, which in turn help form overall impressions, is achieved through offering the amount (i.e., the "correct" weight) of each attribute. The weight or amount of each attribute offered by a store will positively or negatively affect the perceived image of that store. The significant differences in consumers' and managers' overall impressions and perceived beliefs of each attribUte offered by the three discount stores may have been caused by the differences in opinions of the ideal amount of each attribute and that offered by the stores. Limitations A major limitation of this study was the small sample and unequal number of participating store managers. Because of the unusual cell sizes the results of the managers' attitude toward the overall 125 impression of each store are questionable. Their ideal amount of each attribute also could not be generalized for all discount managers. Managers who participated in the survey did so out of convenience. Employees who were in the store at the time of the survey were asked to complete the questionnaire. The store manager as well as all other participating managers completed the survey at the same time, inter- acting with each other. This interaction and presence of the store manager may have influenced certain responses regarding their overall impression of each store. The omission of Store Three managers also limited the analyses between management teams. The exclusion of salient store attributes was a second major limitation of the study. Results of the data may have differed if all significant attributes were included. Because the consumers evaluated a particular branch store location rather than the company as a whole, generalizations concerning consumers' overall impression of the company are impossible. The questions to the consumers were read over the telephone. The accuracy of respondents answers rested solely on their ability to mentally comprehend questions and responses without actually seeing the questions. Depending upon the respondents ability, questions may have been misunderstood or not fully comprehended. Recommendations Based upon the results of this study, the use of the Beckwith and Lehmann multiple attribute model is recommended for studying store 126 image. Due to the fact that the importance of terms in the model vary in predicting managers' overall impressions of each store, additional branch locations, stores and larger sample sizes must be employed before any generalizations can be made. The findings that managers and consumers held different images of the ideal level of each attribute and different overall impressions of each store is significant. Additional research concentrated on branch locations of each discount store is needed. The results from the research would indicate whether the differences of images pertain only to one branch location or to the entire corporate organization. The inclusion of hypothetical "ideal" attributes allowed an analysis of deficiencies in each discount store according to managers and consumers. The significant differences between the ideal level of each attribute and the amount offered by each store indicate that consumers are not completely satisfied with the total offering of the discount stores. More importantly, however, is the fact that managers have different opinions concerning the ideal level of attributes; the attributes emphasized in each discount store therefore are not neces- sarily the attributes most desired by the consumers. Further investiga- tion into attributes offered by each store and those most desired by the target market would assist the retailer's marketing efforts. LIST OF REFERENCES LIST OF REFERENCES Alderson, Wroe, and Robert E. Sessions. "Basic Research on Consumer Behavior: Report on a Study of Shopping Behavior and Method for Its Investigation," in Quantitative Techniques in Marketing Anal sis by Ronald E. Fran, Alfred A. Kuehn, and William F. Massy (Homewood, 111.: Richard D. Irwin, 1962). Alpert, Mark 1. "Identification of Determinant Attributes: A Com- parison of Methods," Journal of Marketing Research Vol. VIII (May 1971): 184-191. Bass, Frank M. and William L. Wilkie. "A Comparative Analysis of Attitudinal Prediction of Brand Preference," Journal of Marketing Research Vol. X (August 1973): 262-269. Beckwith, Neil E. and Donald R. Lehmann. "The Importance of Differen- tial Weights in Multiple Attribute Models of Consumer Attitude," Journal of Marketinngesearch Vol. X (May 1973): 141-145. Bellenger, Danny N., Earle Steinberg and Wilbur W. Stanton. "The Congruence of Store Image and Self Image as it Related to Store Loyalty,"Journal of Retailing Vol. LII (Spring 1976): 17-32. Berkowitz, Eric N., Terry Deutscher and Robert A. Hansen. "Retail Image Research: A Case of Significant Unrealized Potential," Educators Proceedings, 1978. Berry, Leonard L. "The Components of Department Store Image: A Theo- retical and Empirical Analysis," Journal ongetailing Vol. XLV (Spring 1969): 3-20. Beyer, William. CRC Standard Mathematical Tables 25th edition (West Palm Beach, Fla.: CRC Press Inc., 1974). Bohr, Peter. "America's Quest for Quality," Money Vol. IX (December 1980): 44-47. Brown, F. E. "Price Perception and Store Patronage" eds. Steuart Henderson, Britt in Psychological Experiments in Consumer Behavior, New York: Wiley & Sons, 1970. Brown, F. E. and George Fisk. "Department Stores and Discount Houses: Who Dies Next?" Journal of Retailing Vol. XXXXI (Fall 1965): 5-27. 127 128 Bucklin, Louis P. "Retail Strategy and the Classification of Consumer Goods," Journal of Marketipg Vol. XXVII (January 1963): 50-55. Cohen, Joel B., Martin Fishbein, and 011i T. Ahtola. "The Nature and Uses of Expectancy-Value Models in Consumer Attitude Research," Journal of Marketing Research Vol. IV (November 1972): 456-460. Cohen, Louis. "The Differentiation Ratio in Corporate Image Research," Journal of Advertisipg Research Vol. VII (September 1967): 32-36. Collins, Kenneth. "Institutional Advertising," Journal of Retailing Vol. III (April, 1927): 10. "Color Schemes Can Help or Hinder Productivity," International Manage ment Vol. XXXII (September 1977): 5. Cook, Stuart W. and Claire Selltiz. "A Multiple-Indicator Approach to Attitude Measurement," Psychological Bulletin Vol. LXII (1964): 36-55. Crespi, Leo P. "Some Observations on the Concept of Image," Public Opinion Quarterly Vol. XXV (1961). Crissy, William J. E. "Image: What is it?" MSU Business Topics Vol. XIX (Winter 1971): 77-80. Dash, Joseph F., Leon G. Schiffman, and Conrad Berenson. "Information Search and Store Choice," Journal of Advertisinngesearch Vol. XVI (June 1976): 35-40. Dardis, Rachel and Marie Sandler. "Shopping Behavior of Discount Store Customers in a Small City," Journal of Retailing Vol. XLVII (Summer 1971): 60-72, 91. Day, George S. "Evaluating Models of Attitude Structure," Journal of Marketing Research Vol. IX (August 1972): 279-86. "Discounter Puts on New Face," Chain Store Age Executive Vol. LIV (September 1978): 43-64. "Doing More With Decor," Hardware Retailing Vol. CXLI (October 1981): 88-89. Dreyfus, Patricia A. "First-Class Secondhand Clothes," Money Vol. IX (December 1980): 63-66. Dulany, Don E. "Awareness, Rules, and Propositional Control: A Con- frontation with S-R Behavior Theory," in Theordore R. Dixon and David L. Horton, Eds. Verbal Behavior and General Behavior Theory (Englewood Cliffs, N.J.: Prentice-Hall 1968): 340-87. 129 Egan, Douglas. "Evaluative Bias, Consumer Attitudes and Department Store Image," Southern Journal of Business Vol. VI (November 1971): 73-81. Ellsworth, T., 0. Benjamin and H. Rudolf. "Customer Response of Trading Stamps," Journal of Retailing Vol. XXXIII (Winter 1957- 1958): 165-169. Enis, Ben M. "An Analytical Approach to the Concept of Image," California Management Review Vol. IX (Summer 1967): 51-58. Enis, Ben M. and Gordon W. Paul. "'Store Loyalty' as a Basis for Market Segmentation," Journal of Retailing Vol. XLVI (Fall 1970): 42-56. Fisk, George. "A Conceptual Model for Studying Customer Image," Journal of Retailing Vol. XXXVII (Winter 1961-1962): 1-8. Fishbein, M. "A Consideration of Beliefs and Their Role in Attitude Measurement," in M. Fishbein ed., Readings in Attitude Theory and Measurement. (New York: John Wiley and Sons, 1967). Gensch, Dennis H. "Image-Measurement Segmentation," Journal of Market- ing Research Vol. XV (August 1978): 384-94. Goodman, E. Lawrence. "Department Stores Versus Discounters: They Learn from Each Other," Stores Vol. LIV (1972): 36-37. Granzin, Kent L. and Louis K. Sharpe. "Brand AttributesThat Determine Purchase," Journal of Advertising Research Vol. XIV (April 1974): 39-42. Green, P. E. and Y. Wind. Multi-attribute Decisions in Marketing: A Meesurement Approach. (Hinsdale, Illinois: The Dreyden Press, 973 . Grubb, Edward L. and Harrison L. Grathwohl. "Consumer Selfeconcept Symbolism and Market Behavior: A Theoretical Approach," Journal ofeMarketing Vol. XXXI (October 1967): 22-27. Hammond, Kenneth R. Cognition and Conflict Program on Cognitive Processes, Report LOO, Institute of Behavioral Sciences, University of Colorado, August, 1967. Hansen, Fleming. "Consumer Choice Behavior: An Experimental Approach," Journal of Marketing Research Vol. VI (November 1969): 439-43. Hirschman, Elizabeth C. "A Descriptive Theory of Retail Market Structure," Journal of Retailipg Vol. LIV (Winter 1978): 29-48. 130 Hirschman, Elizabeth C., Barnett Greenberg and Dan H. Robertson. "The Intermarket Reliability of Retail Image Research: An Empirical Examination," Journal of Retailing Vol. VIV (Spring 1978): 3-11. James, Don L., Richard M. Durand and Robert A. Dreves. ”The Use of a Multi-attribute Attitude Model in a Store Image Study," Journal of Retailing Vol. LII (Summer 1976): 23-32. Kelly, Robert F. and Ronald Stephenson. "The Semantic Differential: An Information Source for Designing Retail Patronage Appeals," Journal of Marketing Vol. XXXI (October 1967): 43-47. Kunkel, John H. and Leonard L. Berry. "A Behavioral Conception of Retail Image," Journal of Marketing Vol. XXXII (October 1968): 21-27. Kusher, Richard. "An Individual Differences Multi-dimensional Scaling Approach Toward Consumer Decision-Making," in M. Venkatesan Ed. Association for Consumer Research: Proceedings 3rd Annual Con- ference 1972. (Chicago: Association for Consumer Research, 1972): 546-561. Lazarfeld, P. F. "Latent Structure Analysis," in S. Koch, Ed. Psychology: A Study of a Science Vol. III Formulations of the Person and Social Context (New York: McGraw-Hill, 1959). Lehmann, Donald. Market Research and Analysis. Homewood, Illinois: Richard D. Irwin, 1979. Lemon, Nisel. Attitudes and Their Measurement (USA: Halsted Press, 1973 . Lessig, V. Parker. "Consumer Store Images and Store Loyalties," Journal of Marketing Vol. XXXVII (October 1973): 72-74. Lindquist, Jay 0. "Meaning of Image: A Survey of Empirical and Hypothetical Evidence," Journal of Retailing Vol. L (Winter 1974-1975): 29-38. Lutz, Richard. "An Experimental Investigation of Causal Relations Among Cognitions, Affect, and Behavioral Intentions," Journal of Consumer Research Vol. III (March 1977): 197-208. McClure, Peter J. and John K. Ryans, Jr. "Differences Between Retailers' and Consumers' Perceptions," Journal of Marketing Research Vol. V (February 1968): 35-40. McDougall, G. H. G., and J. N. Fry. "Combining Two Methods of Image geaggrement," Journal of Retailing Vol. L (Winter 1974-1975): 131 Marks, Ronald B. "Operationalizing the Concept of Store Image," Journal of Retailing Vol. LII (Fall 1976): 37-46. Martineau, Pierre. Motivation in Advertising. (New York: McGraw-Hill Book Company, Inc., 1957). Martineau, Pierre. "Sharper Focus for the Corporate Image," Harvard Business Review Vol. XXXVI (November-December 1958): 49-58. Martineau, Pierre. "The Personality of the Retail Store," Harvard Business Review Vol. XXXVI (January-February 1958): 47-55. Mason, Joseph Barry and Morris L. Mayer. "The Problem of the Self- Concept in Store Image Studies," Journal of Marketing Vol. XXXIV (April 1970): 67-69. May, Eleanor G. "Image Evaluation of a Department Store," A Market- ing Science Institute Working Paper (October 1971). May, Eleanor G. "Management Applications of Retail Image Research," A Marketing Science Institute Working Paper (September 1973). May, Eleanor G. "Practical Applications of Recent Retail Image Research," Journal of Retailing Vol. L (Winter 1974-1975): 15-20, 116. Myers, John C. Consumer Image and Attitude. (California: Institute of Business and Economic Research Publications, 1981). Myers, John G. "Determinates of Private Brand Attitude," Journal of Marketing Research Vol. IV (February 1967): 73-81. Myers, Robet H. "Sharpenin Your Store Image," Journal of Retailing Vol. XXXVI (Fall 1960 : 129-137. Nelson, Barden H. "Seven Principles In Image Formation," Journal of Marketing Vol. XXVI (January 1962): 67-71. Osgood, Charles E., George J. Suci and Percy H. Tannembaum. "Attitude Measurement," in Gene F. Summers Ed., Attitude Measure- ment. (Chicago: Rand McNally & Co.). Pathak, Devendra S. "A Study of Department Stores' Images Held by Customers and Management." Unpublished Dissertation from Michigan State University, East Lansing, Michigan 1972. Pathak, Devendra 5., William J. E. Crissy, and Robert W. Sweitzer. "Customer Image Versus the Retailer's Anticipated Image," Journal of Retailing Vol. L (Winter 1974-1975): 22-28. Pekelman, Dov and Subrata Sen. "Mathematical Programming Models' for the Determination of Attribute Weights," Management Science Vol. XX (April 1974): 1217-1229. 132 Perry, Michael and Nancy J. Norton. "Dimensions of Store Image," Southern Journal of Business Vol. V (April 1970): 1-7. Pessemier, Edgar A. "Store Image and Positioning," Journal of Retailing Vol. LVI (Spring 1980): 94-106. Peters, William S. and Richard Kuhn. "An Exploration in Store Image Measurement," Decisions Sciences Vol. I (1970): 113-128. Rachman, David J. and Linda J. Kemp. "Profile of the Boston Discount House Customer," Journal of Retailing Vol. XXXIX (Summer 1963): 1-8. "Remodelling Gives New Look to Discounters," Chain Store Age Executive Vol. LIV (August 1978): 91. Rich, Stuart U. Shppping Behavior of Department Store Customers. (Boston, Mass.: Division of Research, Graduate School of Business Administration, Harvard University, 1963). Rich, Stuart U. and Bernard 0. Portis. "The 'Imageries' of Department Stores," Journal of Marketing Vol. XXVIII (April 1964): lO-15. Rosenberg, Milton J. "Cognitive Structure and Attitudinal Affect," Journal of Abnormal and Social Psycholpgy Vol. LIII (November 1956): 367-72. Sheth, Jagdish N. and W. Wayne Talarzyk. "Perceived Instrumentality and Value Importance as Determinants of Attitudes," Journal of Marketing Research Vol. IX (February 1972): 6-9. Singson, Ricardo. "Multi-dimensional Scaling Analysis of Store Image and Shopping Behavior," Journal of Retailing Vol. LII (Summer 1975): 38-52, 93. Snyder, James D. "In-Store Servicing Catches Fire," Sales Management Vol. CVIII (April 1972): 27-28. Stefflre, Volney. "Multidimensional Scaling as a Model for Individual and Aggregate Perception and Cognition," in Subhash C. Jain, Ed., Research Frontiers in Marketing: Dialogues and Eirectigns 1978 E ucators ProCeedin 5. Chicago: American Mar eting ssocia- tion, 1978): l7-2l. Tillman, Rollie. "Semantic Differential Measurement of Consumer Images of Retail Stores," Southern Journal of Business Vol. II (April 1967): 67-73. Tuhy, Carrie. "How and Where to Find the Finest," Money Vol. IX (December 1980): 53-60. 133 Tuhy, Carrie. "Rags Without Riches," Money Vol. IX (April 1980): 96-104. Tyler. William D. "The Image, the Brand, and the Consumer," Journal of Marketing Vol. XXII (October 1957): 162-165. Vaughn, Ronald, Joseph Pitlik, and Behram Honsotia. "Understanding University Choice: A Multi-attribute Approach," in Proceedinge: Advances in Consumer Research (1978): 26-31. Walters, C. Glenn. Consumer Behavior: Theory and Practice. (Homewood, Illinois: Richard D. Irwin, Inc., 1978). Weale, W. Bruce. "Measuring the Customer's Image of a Department Store," Journal of Retailing Vol. XXXVII (Summer 1961): 40-48. Werbel, Richard. "Measured Attribute Weights Can Make a Difference," in Proceedings from Advances in Consumer Research (1978): 180- 185. Wilkie, William L. and Edgar A. Pessemeir. "Issues in Marketing's Use of Multi-attribute Attitude Models," Journal of Marketing Research Vol. X (November 1973): 428-41. Will, R. Ted and Ronald W. Hasty. “Attitude Measurement Under Condi- tions of Multiple Stimuli," Journal of Marketing Vol. XXXV (January 1971): 66-70. Wingate, John H. "Developments in the Supermarket Field," New York Retailer (October 1958): 6. Wyckham, Robert George. Aggregate Department Store Images: Social and Experimental Factors. Unpublished Ph.D. dissertation from Michigan State University, East Lansing, Michigan, 1967. Yager, Ronald R. "On A Programming Model of Consumer Choice Among Multi-attributed Brands," Journal of Consumer Research Vol. VI (December 1979): 317-319. APPENDIX STUDY QUESTIONNAIRE APPENDIX sruov QUESTIONNAIRE This is a study conducted by Marianna Mahoney, a graduate student at Michigan State University. I would appreciate your assistance in completing this questionnaire. Consent Form--I have freely consented to take part in this study being conducted By Marianne Mahoney, under the supervision of Dr. Brenda Stemquist Witter, Assistant Professor. The study has been explained to *“me-and’l'finderstand the explanation that has been given and what my participation will involve. I understand that I am free to discontinue my participation in the study at any time without penalty. The results of the study will be treated in strict confidence and my answers and participation is confidential and anonymous. My participation in the study does not guarantee any beneficial results to me. In this section of the questionnaire, 1 would like you to grade the per- formance of Store One, Store Two, and Store Three and your IDEAL dis- count store based upon certain characteristics. Think about your view of an IDEAL discount store. This store does not necessarily exist, but is what you feel would be the perfect discount store. Based on a scale of 100 points, grade the following: IDEAL STORE 1. Merchandise Selection 2. Value for Price 3. Sales Personnel 4. Store Atmosphere 5. Locational Convenience 6. Quality Your IDEAL store has points for (characteristic 1 through 6). How many points WEETE you give Store One, Store Two, and Store Three for (characteristic 1 through 6). Store One Store Two Store Three 7. Merchandise Selection 8. Value for Price 9. Sales Personnel 10. Store AtmOSphere 11. Locational Convenience 12. Quality 134 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 135 How many.times, on an average, during this past year, have you shopped at: Store One Store Two Store Three Once a year Twice a year Once a month Once every other month Two or three times a month Weekly Never —h 9’ < O T On a scale of 1 to 100 (l = very unfavorable and 100 = very able) what is your overall impression of: Store One Store Two Store Three Please indicate your position in this store How long have you been employed with this store? Less than one year. More than one year, but less than three years. More than three years. On a scale of l to 100 (l = very unfavorable and 100 = very favor- able) what do you think other peoples impression is of: Store One Store Two Store Three When choosing a store, other people might give you advice. On a scale of l to 100 (1 = highly unlikely, and lOO.- extremely likely), do you think that other people who are important to you would recom- mend that you shop at: Store One ‘ Store Two Store Three 136 On a scale of l to 100 (1 = don't care at all and 100 = care a great deal), do you care if your choice of store is similar to what other people who are important to you would recommend? 25. 26. 27. 28. 29. 30. 31. 32. What is your marital status? ____ single .___ married ___ widowed ____ separated/divorced How many children do you have living at home (including any children supported by parents who are away at school)? ,_ Please state the position of the head of your family. ___ Professional or technical ____ Machine operator ____ Manager or administrator, except farm .___ Non-farm laborer ____Sa1es ____ Service worker , ____ Clerical ____ Farm worker ____ Craftsperson ____ Retired ___ Machine operator ____Unemployed _ Other Please specify the amount of school you have completed: ____Some elementary school ____Completed college ____ Completed elementary school (4 year degree) .___ Some high school ___ Some graduate work (Master's ____ Completed high school or Professional degree) Some college ____Completed graduate program Please state an approximate family income before taxes last year. ____Under $5,000 ___ $15,000 to $19,999 $5.000 to $7.499 ____$20,000 to $24,900 ____ $7,500 to $9,999 .___ $25,000 to $49,999 $10,000 to $14,999 ____$50,000 and over Please indicate your age at your last birthday. Under 18 years 18 to 24 years 25 to 34 years 35 to 44 years 45 to 54 years 55 to 64 years 65 years and over What is your sex? _ Male ____Female ‘11111111111111