Final report ACKNOWLEDGEMENT

I am most grateful to Dr. Do Ba Khang, my Research Advisor and Chairman of Research Committee, for his invaluable guidance, constructive suggestions, and constant encouragement throughout the course of my research study.

A TWO-FACTOR MODEL FOR SERVICE QUALITY MANAGEMENT: THE CONCEPTUAL DEVELOPMENT AND AN APPLICATION TO

CO-OP MART NGUYEN DINH CHIEU, VIET NAM.

by

Ho Dac Nguyen Nga

A research study submitted in partial fulfillment of the requirements for the degree of Master of Business Administration

Examination Committee Dr Do Ba Khang (Chairman)

Ho Chi Minh City, Vietnam

Scholarship donor Government of Switzerland /SDC (SAV)

Asian Institute of TechnologySchool of ManagementBangkok, ThailandApril 2001

I am most grateful to Dr Do Ba Khang, my Research Advisor and Chairman of ResearchCommittee, for his invaluable guidance, constructive suggestions, and constantencouragement throughout the course of my research study His profound knowledge andboundless enthusiasm are a great inspiration for me to conduct this research

I would like to express sincere thanks to Dr Fredric W Swierczek and Dr ClemensBechter for their valuable contribution in serving as Research Committee members, as well

as for their constructive comments and critical suggestions on the research study

The sincerest thanks come to the Swiss-AIT-Vietnam Management DevelopmentProgramme (SAV) and Asian Institute of Technology (AIT), all the faculty members,staffs, and friends who contribute greatly to my learning progress

Thanks are also extended to Vietnam Asia Pacific Economic Center (VAPEC), managersand staffs of many supermarkets in Ho Chi Minh City, especially managers and staffs ofSaigon Co-op, for helping me in data collection

Last but not least, I would like to express my deep gratitude to my family and mygirlfriend, who provide continuous supports to me during my learning process and thecourse of this research study

The goal of this research is to investigate the quality level of Co-op Mart Nguyen DinhChieu in the perception of customers in order to help management improve the store’squality properly

For the purpose above, a new quality model, called “Two-factor model”, is developed.This model measures customer perception towards two factors -attributes performance andshopping preference of customers- and identifies the relationship of these two factors toconstruct the quality perception of customers The results of this research reveal that Two-factor model is more appropriate than Servqual in recommending the attributesimprovement priorities

Primary data of the research is collected through two surveys, one is for supermarketcustomers in Ho Chi Minh City and one is for customers of Co-op Mart Nguyen DinhChieu The former is used to collect general perception about supermarket qualityattributes and the latter is used to identify quality level of Co-op Mart Nguyen DinhChieu’s attributes Responses from these two surveys are analyzed by Two-factor model toderive characteristic curve of each attribute as well as current performance, contribution tocurrent shopping preference of customers, preference gap, and improvement efficiency ofeach attribute of the store

Based on the research findings, recommendations are made to improve the quality of thestore Quality attributes are classified into 3 main groups The first group includesattributes which should be improved The second group includes attributes which should

be improved after all the attributes of the first group reach their potential The attributes ofthe third group are recommended to be unchanged Moreover, priorities for improvement

of attributes in each group are also specified Among attributes which should be improved,quality of merchandise, agility of locker personnel, hospitality of personnel and variety ofproduct lines are four attributes which have the highest priorities for improvement Theyare the keys to improve quality of Co-op Mart Nguyen Dinh Chieu

TABLE OF CONTENTS

Acknowledgement i

Abstract ii

Table of Contents iii

List of Figures v

List of Tables vi

Chapter 1: Introduction 1

1.1 Rationale of The Research 1

1.2 Problem Statement 1

1.3 Research Objectives 2

1.4 Research Framework 2

1.5 Scope of The Research 3

1.6 Organization of The Research 3

Chapter 2: Co-op Mart Introduction 4

2.1 Background of Supermarket 4

2.2 Co-op Mart Introduction 6

Chapter 3: Development of Two-factor Model 13

3.1 Review of Multi-attribute Concept 13

3.2 Review of Servqual Model 13

3.3 Review of Kano Model 15

3.4 Proposed Two-factor Model 19

Chapter 4: Research Methodology 25

4.1 Information Needed 25

4.2 Target Populations 25

4.3 Questionnaires Development 25

4.4 Sampling Procedure 27

4.5 Sample Characteristics 28

Chapter 5: Findings Analysis and Discussion 31

5.1 Facility Attributes Analysis 31

5.2 Goods Attributes Analysis 34

5.3 Personnel Attributes Analysis 39

5.4 Policy Attributes Analysis 42

5.5 Two-factor Model vs Servqual 46

Chapter 6: Conclusion 48

6.1 Present Customer Perception 48

6.2 Improvement Strategy 49

6.3 Two-factor Model 52

Appendix A: Questionnaire A 53

Appendix B: Questionnaire B 56

Appendix C: Interview Schedule 59

Appendix D: Data Analysis 60

Appendix E: Quality Attributes of a Supermarket 74

References 76

LIST OF FIGURES

Figure 1.1: Research framework 2

Figure 2.1: Proportion of merchandises sold through supermarket 5

Figure 2.2: Saigon Co-op organization chart 7

Figure 2.3: Old procurement procedure 9

Figure 2.4: New procurement procedure 9

Figure 2.5: Co-op Mart Nguyen Dinh Chieu Organization chart 10

Figure 2.6: Co-op Mart Nguyen Dinh Chieu sales volume (2000) 10

Figure 3.1: The Kano diagram 16

Figure 3.2: Three basic shapes of characteristic curve 20

Figure 3.3: Structure of characteristic curve 21

Figure 3.4: Two-factor model’s coefficients 23

Figure 4.1: Respondents’ age of sample A 28

Figure 4.2: Respondents’ frequency of shopping in supermarket of sample A 28

Figure 4.3: Respondents’ average spending for each shopping of sample A 28

Figure 4.4: Respondents’ age of sample B 29

Figure 4.5: Respondents’ frequency of shopping in supermarket of sample B 29

Figure 4.6: Respondents’ average spending for each shopping of sample B 29

Figure 5.1: Convenience of parking 32

Figure 5.2: Air condition 33

Figure 5.3: Store size 33

Figure 5.4: Quality of merchandise 35

Figure 5.5: Variety of product lines, variety of brand names and varieties for each product line 36

Figure 5.6: Convenience of products arrangement for finding and attractiveness of products display 37

Figure 5.7: Introduction of new products 38

Figure 5.8: Unique products which customers cannot find somewhere else 38

Figure 5.9: Agility and hospitality of locker personnel 40

Figure 5.10: Hospitality, expertise in product information, and helpfulness in finding of salesperson 41

Figure 5.11: Hospitality, accuracy, and agility of cashier 41

Figure 5.12: Price level compared with other supermarkets 43

Figure 5.13: Returning unqualified products policy 44

Figure 5.14: Attractiveness of promotion campaigns 44

Figure 5.15: Flexibility of paying method 45

Figure 5.16: Convenience of open hours 46

Figure 5.17: Convenience of open hours and quality of merchandise 47

LIST OF TABLES

Table 2.1: Co-op Mart stores 8

Table 2.2: List of Co-op Mart’s direct competitors 11

Table 3.1: Kano questionnaire combination table 17

Table 4.1: List of studied attributes 26

Table 4.2: Attributes classification 27

Table 5.1: Facility attributes 31

Table 5.2: Goods attributes 34

Table 5.3: Personnel attributes 39

Table 5.4: Policy attributes 42

Table 6.1: Present customer perception 48

Table 6.2: Improvement priorities of the first group 51

Table 6.3: Improvement priorities of the second group 51

Table C.1: Schedule of survey A 59

Table C.2: Schedule of survey B 59

Table D.1: Variables coding 60

Table D.2: Shopping preference contributed by very good performance 61

Table D.3: Shopping preference contributed by very bad performance 62

Table D.4: Current attributes' performance 63

Table D.5: Shopping preference contributed by current attributes' performance 64

Table D.6: Difference between current attributes' performance and very good performance .65

Table D.7: Difference between current attributes' performance and very bad performance .66

Table D.8: Difference between shopping preference contributed by very good performance and shopping preference contributed by current performance 67

Table D.9: Difference between shopping preference contributed by current performance and shopping preference contributed by very bad performance 68

Table D.10: Attributes' characteristics 69

Table D.11: List of attributes in the order of preference gap 71

Table D.12: List of attributes in the order of improvement efficiency 72

Table D.13: Pair test for difference in improvement efficiency 73

CHAPTER 1: INTRODUCTION

After more than 20 years of planned economy, Vietnam entered to the market economywith a lot of new kinds of business imported to the country Among them, the concept ofsupermarket seems to be the one which has most impact on the lifestyle of Vietnamese.While the knowledge of supermarket quality receives many attentions from researchers inother countries, the specific knowledge on Vietnam’s situation is scarce Saigon Co-opMart is chosen for the research because it has the largest coverage in Ho Chi Minh City,the most dynamic city in Vietnam

Saigon Co-op is a co-operative which used to operate in many businesses frommanufacturing to trading, exporting and retailing In the recent years, it has a strategicmove from a diversified portfolio investment to focusing on retailing especially thesupermarket chain Many business units were divested for freeing resources to develop thesupermarket chain From the beginning, Co-op Mart gains a significant market share thank

to be one of the first movers However, it could not enjoy this advantage forever Recently,the competition on this kind of business increases rapidly in term of quantity as well asquality Many supermarkets have been opened with modern facilities and larger area in thelast year In the near future, the competition will be more intensive as the result of thepenetration of foreign supermarket chain and the co-operation of other chains Forsustaining the market leader position, Saigon Co-op must improve its competitiveness andthe basic for this competitiveness is quality Quality leads to customer satisfaction andloyalty; satisfied and loyal customers are a source of increasing future purchasing then thefirm long-term profitability and market share (Anderson, Fornell, and Lehmann, 1994)

On the other hand, for capturing the market quickly, Saigon Co-op Mart spreads its storesall over the city with very high speed As the result, the operation and quality level of thestores are not unified It creates a problem with customer perception of the brand.Moreover, it makes the co-operation among stores having problems and the operation ofthe whole chain is not effective To overcome these problems, the management of Co-opMart set up Co-op Mart Nguyen Dinh Chieu as the model for their next generation stores.After that, they standardize other stores based on this model

The result of this research will give the company management a better understanding of therelationship between store’s performance and shopping preference of customers and alsothe perception of customer about Co-op Mart Nguyen Dinh Chieu Consequently, they canhave more precise and systematic decisions to improve the quality of Co-op Mart NguyenDinh Chieu The result is not only the improvement of Co-op Mart Nguyen Dinh Chieu butalso the improvement of the whole chain when they apply this model to other stores

The goal of this research is to provide the management of Saigon Co-op a better insight ofCo-op Mart Nguyen Dinh Chieu quality in the perception of customer in order to improveits quality

 Identifying the attributes of a supermarket that affect perceived quality of customerand finding out how these attributes’ performance influence the shopping preferencelevel of customers in an incremental manner.

 Measuring customer’s perception about attributes performance of Co-op MartNguyen Dinh Chieu and how much these attributes contribute to shoppingpreference of customers

 Providing recommendations to improve Co-op Mart Nguyen Dinh Chieu’s quality

in an efficient way: improving the right attributes in the right priorities based on theincremental contribution of these attributes performance to shopping preference ofcustomers

Analysis

Quality attributes

ManagementinterviewPrevious research Focus group ofcustomer

Survey BSurvey A

Shopping preference

Managementinterview Observation

Recommendation

Quality attributes of a supermarket are discovered through previous research and focusgroup of customers Then an interview with the company management to decide whichattributes will be studied After that, two surveys will be conducted to collect primary data.The result of these two surveys will be used to characterize the attributes On the otherhand, the second survey will be used to measure Co-op Mart Nguyen Dinh Chieu’sattributes performance and how much this performance level contributes to shoppingpreference of customers Based on these findings, some recommendations will be made tohelp company management improve their store’s quality

Quality attributes are characterized, based on their impacts to shopping preference ofcustomers The concept underlies this method is twofold Firstly, when an attribute isimproved or declined, it does not affect shopping preference of customers by the sameamount Secondly, this difference is not the same for every attribute but each attribute hasits own property Some will have more impact on increase side and others will have moreimpact on decrease side

The research does not take into consideration the opinions of customers who visitsupermarket less than 4 times in the year since they are assumed not to be familiar withthis kind of shopping In addition, customers who are under 15 are also excluded

Chapter 1(Introduction) provides an introduction including the rationale of the research,problem identification, objectives, research framework, scope of the research, andorganization of the research report

Chapter 2 (Co-op Mart introduction) presents the literatures of supermarket backgroundand supermarket in Viet Nam and introduces the development, organization, and operation

of Saigon Co-op and the Co-op Mart chain, especially Co-op Mart Nguyen Dinh Chieustore

Chapter 3 (Development of Two-factor model) provides the literatures review of attribute concept and quality measurement, especially the Servqual and Kano models Thischapter also presents the development of Two-factor model, which is to be used in thisresearch to examine service quality of Co-op Mart Nguyen Dinh Chieu and set up prioritiesfor quality improvement

multi-Chapter 4 (Research methodology) presents the information needed, target populations,questionnaire development, sampling procedure, and sample characteristics of the research.Chapter 5 (Findings analysis and discussion) contains the discussion and analysis offindings from customer surveys as well as management interview and observation Somerecommendations are also discussed in this chapter

Chapter 6 (Conclusion) draws out the conclusions for current perceived quality of SaigonCo-op Mart Nguyen Dinh Chieu, quality improvement strategy, and Two-factor model

CHAPTER 2: CO-OP MART INTRODUCTION

7.1 History and concept of supermarket

The supermarket was born in the US and was one of the most interesting innovationswhich has changed the retailing environment Exact place and time it was born has beendebated for many years According to Peak H.S (1977), most authors cite December 8,

1932 as the birth date of the supermarket when a new store called the Big Bear was opened

in New Jersey This store, along with other early supermarkets, was a child of the GreatDepression Economic conditions were such that the price was all-important, and everypenny counted

The lower price in supermarkets is the result of low-margin operations that depend on veryhigh stock turnover rates to sustain profits Operation out of clean, modern facilities, thesupermarket is basically a self-service operation supported by such services as parking,check cashing, fast check out, and bagging Cash and carry is the preferred method ofdoing business (Lewison, 1994)

The supermarket was generally accepted as the food retailer until the years 1970s By theend of 70s, the changing lifestyle of the customers focused on convenience had led someoperators to a “one-stop-shopping-for-household-goods” supermarket The emphasis onone-stop shopping and the adding of new departments and sections has continued up tillnow (Peak, 1977) Today, a supermarket’s upgraded and upscale operations include suchnon-food lines as prescription medicines, electronic appliances, auto accessories, books,magazines, clothing, flowers, and house ware Moreover, many supermarkets have addednumerous services including dry cleaning, postal, banking, tailoring, medical, dental,insurance, and legal services

No one definition of a supermarket exists because of the wide range of business formulaused in this industry There are three criteria used to define a supermarket: the minimumsize requirement, the characteristics and extent of self-service, and the stock requirements

In the US, a supermarket was firstly defined by the Supermarket Institute as a self-servicedepartmentalized food store having a minimum sales volume of $250,000 per year In

1954, the sales volume was revised to $500,000 (Charvar, 1961 and McCleland, 1963).Prior to 1981, supermarket included food stores with annual sales of at least $1 million.Recently, the minimum annual sales requirement is $2 million (Lewison, 1994) In the UK,

a supermarket is defined as a self-service food store with centralized checkouts and a salesarea of over 2,000 square feet (Khalifa Othman, 1990)

In the US, supermarkets, using 80% of retail food industry’s personnel, account for about90% of all retail food sales (Peak, 1977) In the UK and Canada, supermarkets and largesuperstores account for about 85% of grocery trade Similarly in other European countries,supermarkets dominate the food retailing system (Khalifa Othman, 1990)

7.2 Supermarket in Viet Nam

In general, Vietnamese perceives that the store offering self-service, having fixed pricelabels and electronic checkout as a “supermarket” (Thu Huong, 1997, p.33) In herresearch, Thu Huong (1997) reached the conclusion that most of the people (94.9%)consider a store as a “supermarket” because of its self-service characteristic, which is

pretty new for Vietnamese shoppers Many customers cite self-service as a positive side ofthe store, because shoppers can enjoy free choice and comparison among brands, and canavoid the high pressure selling which is normal practice in other traditional stores Fixedpricing is also thought as a sign of a “supermarket” by nearly 85% of customers More thanhalf of the customers point out that a “supermarket” should have electronic checkout Inaddition, nice and convenience product displays are also considered as one attribute of a

“supermarket” by 40% of customers

In contrast with supermarket concept used in developed countries which is characterized

by a large volume, multi-line store that serves mass market at competitive prices, thesupermarkets in Viet Nam are more expensive than traditional outlets, and tend to cater themiddle and upper class customers The supermarkets are still fairly basic in terms ofdesign, range and service, but they are a far cry from the small and congested roadsidestalls Different with traditional markets, supermarkets have well designed interior andshopping comforts such as air conditioning, cold storage, and electronic checkouts Theyoffer self-service, high quality goods at fixed prices The merchandise mix is mainlyhousehold amenities and personal care products, ready-to-eat foods, confectionary andcooking materials, as well as footwear, clothing and some durable products (Thu Huong,

Figure 2.1: Proportion of merchandises sold through supermarket 1

Demand for supermarket is more and more increasing The number of customers who go toone supermarket for shopping increases from hundreds per day to 5 thousands per day inaverage Especially, Mien Dong Supermarket and Maximart received 20,000 customers perday in opening days after off days for extent In addition, the average value of one billincreases from 50,000VND to 200,000VND The number of above 1 million VND bill alsoincreases (Chanh Khai, 2001) Moreover, the traffic condition in big cities of Viet Nam isworse and worse As a result, one-stop shopping will be preferred more and more becausecustomer will avoid traveling from store to store for shopping For conclusion, thedevelopment of supermarket is the indispensable trend in big cities of Viet Nam.

From 1989, under the influence of social and economic reform process (Doi Moi), manymember co-operatives were disintegrated, merged, or reformed because of the fiercecompetition of other forms of business organization At that time, there were only 7member co-operatives under the management of Management Board To reinforce the co-operative segment, The Union of Trading and Service Co-operatives of Ho Chi Minh City(Saigon Co-op) was founded in 05/1989 based on resources of Management Board andmember co-operatives

Recently, Saigon Co-op operates in many areas as following:

 Supermarket: Co-op Mart chain is the most well-known supermarket chain in Ho ChiMinh City with 6 stores It will be discussed more deeply in next sections

 Mini mart: Saigon Co-op has just started to develop a mini mart system (mCo-op Mart)

in the form of joint ventures with other local co-operatives Up to now, 2 stores wereopened in Ho Chi Minh City; mCo-op Mart Cao Thang is a joint venture with Wards 2-4-5Co-operative and mCo-op Mart Cau Kinh is a joint venture with Cau Kinh Co-operative.These stores supply daily necessities to customer such as food-grains, foodstuffs, healthcare products, and family used items, etc The competitive advantage of this kind of mart isthe convenience for customer because the mart is located in residential area

 Distribution agent for other companies: distributing and selling domestic products andimported products all over the country

 Manufacturing: Nam Duong sauces with brand name “Con Meo Den” was elected asViet Nam high quality product during 4 successive years from 1997 to 2000 Nam Duongsauces are distributed and sold all over the country and exported to European, American,and Russian markets

 Export: agricultural products, aquatic products, seafoods, processed foods, textile,garment, leather products, and handicraft

 Import: raw and working materials (such as plastic, textile, and chemical), foods,machine, motorcycle, auto, and consumer’s goods.

 Services: besides tourism services of Saigon Co-op tourism center, Saigon Co-op alsoprovides other services such as advertisement, foreign exchange but not well developed tillnow

 Dong Khoi computer center: trading computer, computer parts, maintenance, andinstalling software

Figure 2.2: Saigon Co-op organization chart

Committee of member co-operatives Board of management Board of directors

Inspection section

Co-op Mart chain

Co-op Mart Cong Quynh Co-op Mart Tran Hung Dao

Co-op Mart Nguyen Dinh Chieu Co-op Mart Dinh Tien Hoang

Co-op Mart Dam Sen Co-op Mart Hau Giang

Distribution agent

for other companies

Nam Duong sauces

factory

Services

Nhat Nam trading

center

mCo-op Mart chain

Saigon Co-op tourism center Advertisement services Foreign exchange services

Ben Thanh trading

store

Dong Khoi

computer store

Joint-ventures

 Ben Thanh trading store: trading textile and garment with fixed price.

 Nhat Nam trading center: this is a joint venture of Saigon Co-op and HasegawaCompany (Japan) In addition to a supermarket, it has many departments for rent

However, the strategy for long-term development of Saigon Co-op is to develop the Co-opMart chain It is the core business of Saigon Co-op in future

8.2 The development of Co-op Mart chain

In 1993, some supermarkets were established in Ho Chi Minh City such as Maxi Mart, CitiMart, etc These events created a new kind of shopping behavior of customer in Ho ChiMinh City Operating in retailer segment, Saigon Co-op realized the necessary ofdevelopment a supermarket chain It was also the opportunity to develop and diversify theoperation

Co-op Mart Cong Quynh is the first store of Co-op Mart Chain It was founded on01/06/1996 with operating area of 732 m2 and investment fund of 6,530 billion VND.Based on the very good result of the first year, Saigon Co-op decided to develop the Co-opMart chain by founding some more stores after many carefully done market researches.Recently, Co-op Mart Chain has 6 stores placed in Ho Chi Minh City As the commontrend, the operating area of the store is increasing

Table 2.1: Co-op Mart stores 2

date

AreaCo-op Mart Cong Quynh 189c Cong Quynh, dist 1 09/02/1996 760 m2Co-op Mart Tran Hung Dao 727 Tran Hung Dao, dist 5 30/04/1997 1200 m2Co-op Mart Hau Giang 188 Hau Giang, dist 6 05/01/1998 1272 m2Co-op Mart Dam Sen 3 Hoa Binh, dist 11 16/12/1998 2060 m2Co-op Mart Nguyen Dinh

 Standardizing the operation of the stores in Co-op Mart chain

 Opening new stores in many important areas in Ho Chi Minh City and other provinces

 Considering the market price and implementing direct procurement from foreign anddomestic suppliers to establish a reasonable and competitive pricing policy

2 Source: Ho Chi Minh City Commerce Service, List of Supermarkets & Trading Centers in Ho Chi Minh

City, 07/2000.

 Establishing the center warehouse and distribution center, standardizing theprocurement procedures, and selecting suppliers to standardize product quality, price, andproduct coding.

 Increasing promotion campaigns and advertisement

 Developing “Khach Hang Than Thiet” program (friendly customer) to increase thequantity and quality of loyal customer

 Computerizing all of the operation and management in Co-op Mart Chain

 Standardizing the labor force and preparing the labor force for future development

Figure 2.3: Old procurement procedure

Figure 2.4: New procurement procedure

8.3 Co-op Mart Nguyen Dinh Chieu

As the result of the fast development, the operation of Co-op Mart stores was differentfrom store to store Realizing the necessary of standardization in supermarket chain,management of Co-op Mart is trying to construct a suitable model for supermarket thenstandardize the supermarket chain based on this model Therefore, they establish the newstore based on the experience of old stores and learning from other domestic and foreignsupermarket then synchronize the whole chain with the new store Co-op Mart NguyenDinh Chieu is the one

Opened on 31/12/1999 with operating area of 2620 m2, Co-op Mart Nguyen Dinh Chieu isthe flagship of Co-op Mart chain Applying the newest techniques, arrangement, andprocedures, Co-op Mart management attempt to position this store as a model for the nextgeneration of supermarket in Viet Nam In case of success, other Co-op Mart stores will bereformed as this model

Figure 2.5: Co-op Mart Nguyen Dinh Chieu Organization chart

Director

Security supervisor

Marketing staff

Cashier supervisor

Office supervisor

Operating

manager

Operating manager

Sales supervisor

Warehouse keepers

Salespeople

Securities Vehicle keepers

Cashiers Administrator

Accountant Treasurer Office staffs

Figure 2.6: Co-op Mart Nguyen Dinh Chieu sales volume (2000) 3

3 1USD = 14500VND (03/2001)

8.4 Competition

The competitors of Co-op Mart chain are other supermarkets and other kinds of retailersuch as traditional market and grocery However, supermarket has some specialcharacteristics which are different from others, for example, fixed price, self-service,highly merchandise concentration, cleanliness, etc Therefore, the direct competitors of Co-

op Mart are other supermarkets

Table 2.2: List of Co-op Mart’s direct competitors

(m2) No ofitemsCora Dong Nai 11C4KP1 Long Binh Tan,

Bien Hoa, Dong Nai

20,000 20,000

Cora An Lac 354 Hung Vuong, dist Binh

Chanh

Underconstruction

dist.10

2,000 15,000

In Ho Chi Minh City (and Cora Dong Nai in Bien Hoa), there are 16 supermarkets of 11supermarket chains (excluding Co-op Mart) have considerable scale Among them, thereare three redoubtable competitors: Cora (Bourbon), Maximart, and Mien DongSupermarket All of the three have considerable larger operating area than Co-op Martstores

 Cora: Bourbon Corporation (France) Although Cora Dong Nai is the only operatingsupermarket, Cora will develop fast with Cora An Lac and another one in district 10 are inconstruction With a strong financial fund and the support from Bourbon Corporation, they

can spend a lot of money in marketing, promotion campaigns and lower the price to gainmore market share In addition, the experience in supermarket business and goodmanagement skills help them operate more efficient and professional.

 Maximart: there are many product lines and varieties as well as brand names,especially in imported products Other strengths of Maximart are good management skillsand low tax

 Mien Dong Supermarket: the operating area is large and having ability to extent Thereare many product lines and varieties as well as brand names The price is lower than othersupermarket

In addition, there is the threat of merger of other supermarket into a bigger supermarketchain Especially, the intention of the co-operation and might be merger of Maximart,Citimart, and Mien Dong Supermarket into the biggest supermarket chain in Ho Chi MinhCity

For survival and development in such a fierce competitive environment, Co-op Martpositions itself as the most coverage supermarket chain with standardized service and price

by establishing many supermarkets with average operating area cover a large region Thefirst step of standardization is to establish the flagship store then synchronize the wholechain Co-op Mart Nguyen Dinh Chieu is to be the one

CHAPTER 3: DEVELOPMENT OF TWO-FACTOR MODEL

In this research, a new model is developed to examine service quality of Co-op MartNguyen Dinh Chieu and set up priorities for quality improvement Although manytextbooks introduce Servqual as a prefer model to measure service quality, it faces a lot ofcriticisms from literature as well as empirical testing In addition, its managerialimplication is also challenged Therefore, based on the ideas of the relationship of twoquality constructed components and incremental effect of Kano model, an alternativemodel is designed As Servqual and Kano models, this model is also a multi-attributemodel, which provides more insight to managers and researchers than an overall effectmodel Before developing the new model, multi-attribute concept, Servqual model, andKano model are reviewed

According to Wilkie and Pessemier (1973), in multi-attribute models, a service (orproduct) is viewed as a bundle of benefits and costs The potential advantage of multi-attribute models over the simpler “overall effect” approach is that yields understanding andfeasible improvement direction A model which cannot help analysts gain these things isprone to be theoretically vacuous Attributes provide the basic dimensionality of the model.However, the weakness of the construct in this model is that attributes are perceptual ratherthan objective product construct, that reasonable candidate lists can be generated bycombining unstructured interviews Each attribute has two measures, importance weightand belief rating Both of them are presumed to add explanatory power; belief ratingcontribute product differences while importance weight provide differential stress onattributes

For applying this kind of models, a service (or product) will be divided into manyattributes Based on customer perception about each attribute, the measurement will beconducted By doing that way, researcher could understand deeper the construct of qualitythen know how to improve the quality perception of customer by improving each attributeperformance

Many multi-attribute models were designed to measure service quality up to this time such

as Servqual (Parasuraman, Zeithaml, and Berry, 1985, 1988, 1991), Servperf (Cronin andTaylor, 1992), Evaluated performance (EP) model and Normed quality model (Teas,1993) Among them, Servqual is the most well known one in textbooks In addition, Kanoquality model is also a multi-attribute model even though it is a quality identificationmodel rather than a quality measurement model The next sections will present Servqualand Kano models as the bases to develop Two-factor model

10.1 Quality definition

The concept of Servqual model is generally based on gap theory of Parasuraman, Zeithaml,and Berry (1985), which suggests that “the difference between customers’ assessment ofthe actual performance of a specific firm within a general class of service providers andtheir expectation about the performance of that class (P-E gap) drives the perception ofservice quality.”

10.2 Servqual dimensions

According to its developers, Servqual instrument consists of 22 attributes which can beclassified into 5 dimensions: tangible, reliability, responsiveness, assurance, and empathy(Parasuraman, Berry, and Zeithaml, 1988)

 Tangibility dimension: because of the absence of a physical product, customers oftenrely on the tangible evidence that surrounds the service in forming evaluation Thisdimension includes variety of objects such as desks, lightning, wall color, brochures,appearance of firm’s personnel, etc

 Reliability dimension: reflects the consistency and dependability of a firm’sperformance Does the firm provide the same level of service time after time, or doesquality dramatically vary with each encounter? Does the firm keep its promises, bill itscustomers accurately, keep accurate records, and perform the service correctly the firsttime?

 Responsiveness dimension: reflects the commitment of a firm to provide its service in atimely manner This dimension concerns the willingness and readiness of personnel toprovide a service It reflects the preparedness of the firm to provide the service

 Assurance dimension: addresses the competence of a firm, the courtesy it extends to itscustomers, and the security of the service This dimension refers to how a firm’s personnelinteract with customers and customers’ possession such as courtesy reflects politeness,friendliness, and consideration for the customers’ property

 Empathy dimension: is ability of the firm’s personnel to experience customers’ feeling

as their own Empathic firms understand their customer needs and make their serviceaccessible to customers

10.3 Servqual questionnaire

Servqual questionnaire includes 22 pairs of question designed to capture the perception ofcustomer about their expectation toward 22 attributes of a service and their perceivedperformance of these attributes These pairs of question have the same format according toParasuraman, Berry, and Zeithaml (1991):

Expectation: Excellent company will have “attribute 1” (attribute 2, 3, , 22)

Performance: XYZ company has “attribute 1” (attribute 2, 3, , 22)

10.4 Managerial implication

The bigger the gap between attribute performance and customer expectation of the attribute

is, the lower the quality is Therefore, there is more improvement needed for this attributerather than others As a result, the priorities for improvement are associated with themagnitude of P-E gaps The larger the P-E gap is, the more benefits can be gained byclosing the gap and the higher priority to improve the attribute has

10.5 Critiques for Servqual model

a) Reliability and dimensionality of 5-dimension structure

The reliability tests of Cronin and Taylor (1992) in four industries (banks, pest control, drycleaning, and fast food) prove that the 5-dimension structure of Servqual is not confirmed

in any of research samples The chi square statistic universally indicates a poor fit betweenthe theoretical and measurement model for 5-dimension structure The adjusted goodness-

of-fit indices are also not indicative of a good fit In addition, the dimensional test indicatesthat 22 attributes are unidimensional or they are considered as one composite of individualmeasures.

From the testing results above, Cronin and Taylor suggest that the dimensional structure ofquality measures and quality attributes should be constructed flexibly according to specificindustry Five-dimension structure of Servqual has a conceptual meaning rather than aframework to design a practical research

b) Validity of Servqual measures

The primary threat to validity of Servqual measures is construct validity Carmines andZeller (1979, p.23) state, “fundamentally, construct validity is concerned with the extent towhich a particular measure relates to other measures consistent with theoretically derivedhypotheses concerning the concepts (or constructs) that are being measure” The validitytest of Cronin and Taylor (1992) suggests that the performance-based measures provide amore construct-valid explication of service quality than Servqual measures because of theircontent validity In addition, the findings of Kenneth Teas’ empirical test (1993) prove thatevaluated performance measurement framework is characterized by higher validity than P-

E framework

Hence, the attributes performance is measured in this research rather than the P-E gaps

c) Priorities for improvement

The managerial implication of Servqual model implies that the improvement of an attributewhich has bigger P-E gap will provide more benefits than improvement of an attributewhich has smaller P-E gap so the former has higher priority to improve The results of thisresearch will prove that this implication is not always valid In case of some attributes, theimprovement is not worth even though the performance gap is large

11.1 Quality definition

In JIS (Japanese Industrial Standards) Z8101, quality is defined as “the totality of qualitycharacteristics, or level of performance, that determines whether a product or servicesatisfies the purpose of use” (Kano, 1996, p.114)

Quality concept is divided into 2 components, performance and satisfaction Having moremeaning than quality definition in Servqual, this definition does not stop at performancemeasure (performance is not a benefits related factor) but it goes further, this performancemust contribute to customer satisfaction in order to be defined as quality This definitionlinks quality concept with customer benefits

11.2 Kano diagram

It is clear that companies should provide service (or product) with high quality and it iseasy to understand that higher service (or product) performance can result in highercustomer satisfaction However, the relationship between customer satisfaction and service(or product) performance is more complicated than this For some attributes, customersatisfaction can be greatly improved only with a small improvement of performance;while, for some other attributes, customer satisfaction can only be improved a little evenwhen the performance of the service (or product) has been greatly improved Thus, the

deep understanding of this relationship is the prerequisite to achieving customersatisfaction.

For the above purpose, Kano have developed a very useful diagram to characterisecustomer needs This diagram is a tool in the field of R&D for development new productsand services, especially in the integration with Quality Function Deployment (QFD) (Shen,Tan, and Xie, 2000) It devides service (or product) features into three distinct categories,each of them affects customer satisfaction in a very different way:

 Must-be attributes: customer take them for granted when fulfilled However, if theservice (or product) does not meet this basic need sufficiently, the customer may becomevery dissatisfied

 One-dimensional attributes: these attributes result in customer satisfaction whenfulfilled and dissatisfied when not fulfilled The better the attributes are, the better thecustomer like them

 Attractive attributes: the absence of attractive attributes does not cause dissatisfactionbecause they are not expected by customers who may be unaware of such service (orproduct) features However, strong achievement in these attributes delights customer

Figure 3.1: The Kano diagram

In addition to three main categories above, Kano introduced two more categories (Kano,

1996, p.130) to cover all posible attributes:

 Indifference attributes: an increase or decrease in performance level does not make anydifference However, it is very hard to have an absolute indifference attribute Usually, anattribute has some effects, more or less, on some customer

 Adverse attributes: an increase in performance level results in customer disatisfaction,and a decrease results in satisfaction These attributes are nothing than the oppositions tothree main categories

Satisfaction

Must-beAttractive

One-dimensional

Performance

To conclude, in Kano model, the measurements of performance level and customersatisfaction are distinct Moreover, the relationship of these two measurements is notalways linear but may be convex.

 If there are many brands for one product in the supermarket, how do you feel?

1 I like it that way

2 It must be that way

3 I am neutral

4 I can live with it that way

5 I dislike it

 If there is only one brand for one product in the supermarket, how do you feel?

1 I like it that way

2 It must be that way

Table 3.1: Kano questionnaire combination table

Majority of the responses for a specific pair are in cell:

 “A”: attractive attribute

 “O”: one-dimensional attribute

11.4 Managerial implication

The priorities for improvement should be from must-be attributes to one-dimensionalattributes then attractive attributes at last (Rao, Carr, Dambolena, Kopp, Martin, Rafii,Schlesinger, 1996, p.402) In addition, there are some more specific managerialimplications:

 For must-be attributes: they must reach the threshold where the improvement ofperformance does not gain any increase in customer satisfaction or the increase is notconsiderable However, if they do not reach this threshold, the customer will terriblydissatisfy

 For one-dimensional attributes: the more we can improve them, the better the customersatisfaction

 For attractive attributes: a little improvement of these attributes might increase thecustomer satisfaction dramatically but the absence of them will not be recognized bycustomer

11.5 Critiques for Kano model

a) Qualitative vs quantitative analysis

The measurement in Kano model is qualitative rather than quantitative It is used toclassify quality attributes into five categories but it cannot distinguish quality attributeswithin each categories Therefore, all of attributes in one category are treated identically

As a result, the analyst cannot make the specific recommendation for each attribute.Moreover, the difference among attributes within one category may be very significant.For instance, in one-dimensional category, the steeper the attribute, the more meaningful toimprove the attribute In Kano model, all one-dimensional attributes are the same and thesteepness of attribute is ignored Similarity to other categories, the steepness and theconvexity of attributes are ignored in each category

b) Too much indifference attributes

A large proportion of attributes might fall into indifference category due to the largenumber of indifference cells in the Kano questionnaire combination table (9/25) It leadsthe analyst to ignore many attributes In fact, an absolute indifference attribute is scarce

An attribute usually has more or less impact into the perception of the customer Theattribute which is called indifference is the one that has less impact than others The widerthe range of indifference attributes in the model, the bigger the chance to ignore ameaningful attribute In addition, the central tendency bias in responses makes thisproblem more seriously

c) Quality identification vs quality measurement

The Kano model is used to identify the quality perception of customer about a service (orproduct), not to measure the quality of this service (or product) In the model, qualityattributes are classified into different categories but perception of customer about qualitylevel of a specific service (or product) is not measured This is the reason why Kano modelhas not been used in previous quality measurement research In fact, this model is usedwidely in service (or product) development in which the service’s (or product’s) attributesare designed according to their categories

However, the ideas of measuring the relationship between product performance level andcustomer satisfaction and incremental affect concept in Kano model are very interesting

The next section will be dedicated for developing the model which is to be used in thisresearch.

In this section, a new model is developed to examine a service quality and set up prioritiesfor quality improvement This model, called “Two-factor model”, will overcome someshortcoming of Servqual and Kano models as following discussion

 On one hand, Servqual model faces with some critical problems as discussion in thesecond section of this chapter:

 Firstly, the 5-dimension structure of attributes is not reliable and the dimensionalproperty of this structure is also weak (Cronin and Taylor, 1992) Other qualitymeasurement models such as Servperf model, Evaluated performance model, andNormed quality model also face this problem due to using the same dimensionalstructure Hence, it is not necessary to apply this structure into quality measurementmodel Using the interview or focus group to explore quality attributes of a specificproduct or service as in the basic of multi-attribute models (Wilkie and Pessemier,1973) may be a flexible alternative in which the characteristics of a product or serviceare reflected more specifically

 Secondly, using P-E gap to measure quality is worse than using performance invalidity tests (Cronin and Taylor, 1992, and Teas, 1993) It is suggested that the newmodel should be a performance measurement base rather than P-E gap measurementbase

 Thirdly, the implication of using P-E gap to determine priorities to improve ischallenged It is necessary to add one more component in the model to make sure thatthe improvement of performance will cause the improvement of firm benefit

 On the other hand, Kano model is not a quality measurement model but it is aqualitative model to determine the relationship of attribute performance and customersatisfaction in an incremental manner However, this idea opens a new approach tomeasure quality which overcomes the third weakness of Servqual

In conclusion, a new model is proposed to overcome the weaknesses of Servqual Based onthe concept of the relationship of two quality constructed components in an incrementalmanner of Kano model, Two-factor model is designed as a performance base quantitativemodel to examine quality by measuring attributes performance and a firm’s benefit relatedfactor and it also determines the relationship between them Moreover, it could be used toset up the priorities for quality improvement

12.1 Executive quality definition

The proposed Two-factor model is based on the similar quality definition of JIS (JapaneseIndustrial Standards) Z8101 as Kano model because it is also constructed by two qualitycomponents However, for practical purpose, this definition is somewhat modified.Satisfying the purpose of use is conceptually a customer benefit It is easy to agree thatcustomer satisfaction has positive relationships with customer loyalty, purchase intention,and shopping preference, which are the benefits of the firm (Anderson, Fornell, andLehmann, 1994) but “how strong are these relationships?” is a difficult question to answer.For associating quality concept directly with the firm benefits to bypass the questionabove, we should go further by replacing satisfaction factor with one of the firm’s benefit

related factors, shopping preference Therefore, the quality definition is modified as: “thetotality of quality characteristics, or level of performance, that determines the shoppingpreference of customers toward a product or service”.

Besides, using this executive definition could gain more interest of managers, who areinterested in the firm benefits, to improve quality The reason is that quality improvementhas direct effect on shopping preference of customers toward their product or service.According to the definition above, quality is constructed by the totality of product orservice’s attributes performance which contributes to shopping preference of customers.Therefore, a model for quality examination must satisfy four requirements:

 Total measuring

 Measuring the performance level (the first factor)

 Measuring the shopping preference of customers contributed by this performance level(the second factor)

 Identifying the relationship between two factors above

To ensure that total quality characteristics are measured, Two-factor model must be amulti-attribute model Idea of relationship of two factors in an incremental manner fromKano model is used to found the model Instead of the relationship of attributeperformance and customer satisfaction as in Kano model, the relationship in Two-factormodel is of attribute performance and shopping preference of customers Nevertheless, theconcept of the relationship is the same For some attributes, shopping preference ofcustomers can be greatly improved only with a small improvement of performance; while,for some other attributes, shopping preference of customers can only be improved a littleeven when the performance of the service or product has been greatly improved Inaddition, this relationship is not always linear but can be convex

Based on the quality concept in this section, an analytical framework is developed infollowing sections

12.2 Characteristic curve and managerial implication

Figure 3.2: Three basic shapes of characteristic curve

Shopping preference

Linear

Negative convexPositive convex

Performance

In Two-factor model, each attribute is presented as a curve called “characteristic curve” intwo-dimensional co-ordinates The horizontal axis represents performance level of anattribute and the vertical axis represents the shopping preference of customers contributed

by this performance level The characteristic curve has three basic shapes: negative convexcurve, straight line (linear), and positive convex curve In order to determine the shape ofthe curve, three points are needed Information of these three points comes from twosurveys:

 Survey A is used to measure shopping preference contributed by very goodperformance and by very bad performance of each attribute For this purpose, thequestionnaire of this survey consists of pairs of one extremely positive question and oneextremely negative question

 Survey B is used to measure performance level of each attribute of studied product orservice and shopping preference contributed by this performance level The questionnaire

of this survey is constructed of two sets of questions, one for performance level and one forshopping preference contributed by this performance level

From the response of survey A, two extreme points are determined Horizontal coordinates

of these two points are very good (positive point) and very bad (negative point)performance level of an attribute and vertical coordinates are respective shoppingpreference of customers contributed by these performance levels The third point (currentpoint) comes from the response of survey B Its coordinates are performance level ofstudied product or service in an attribute and shopping preference of customers contributed

by this performance level

Figure 3.3: Structure of characteristic curve

Characteristics of an attribute could be derived from characteristic curve as following:

 The more linear the curve is, the more symmetrically the attribute performance affectsshopping preference of customers This kind of attribute is called “linear attribute”

 The more positive convex the curve is, the more significantly the attribute performanceaffects shopping preference of customers on the decreasing direction than that on theincreasing direction This kind of attribute is called “positive convex attribute”

Shopping preference

Current point

Negative point

Positive point

Performance

 The more negative convex the curve is, the more significantly the attributeperformance affects shopping preference of customers on the increasing direction thanthat on the decreasing direction This kind of attribute is called “negative convexattribute”.

 The steeper (positive or negative) the curve is, the more shopping preference issensitive (positive or negative) to performance improvement

 The higher the curve is, the more it contributes to shopping preference of customers.The current point represents the current situation of studied service (or product) at eachattribute Its horizontal coordinate indicates the current perceived performance of thisattribute and its vertical coordinate indicates the shopping preference of customerscontributed by this level of performance The negative point represents the worst case of anattribute when its performance is very bad In contrast, the positive point represents thebest case of an attribute when its performance is very good

Based on the relative position of current point compared with positive point on thecharacteristic curve and the shape of characteristic curve, we can answer the question “howwould we gain in shopping preference of customers if a specific attribute performance is to

be improved?” In general, the attribute which has steepest angular co-efficient at currentpoint will be in the first priority for improvement because shopping preference ofcustomers is most sensitive with this attribute

Similarly, the relative position of current point compared with negative point and the shape

of the characteristic curve indicates that “how would we lose in shopping preference ofcustomer if a specific attribute performance is to be decreased?”

For conclusion, characteristic curve determines the relationship between shoppingpreference of customers and a specific attribute performance in an incremental manner.Together with current point, negative point, and positive point, the shape of characteristiccurve uncovers the behavior of shopping preference of customers if a specific attributeperformance increases or decreases from current performance

12.3 Two-factor model’s coefficients and managerial implication

In addition to graphical evaluation, there are two measures which indicate the priorities forimprovement:

 Firstly, preference gap (difference between shopping preference level contributed byvery good performance and shopping preference level contributed by current performance)

of an attribute is used to measure the potential for shopping preference improvement whichcan be achieved by improving the attribute performance The larger the preference gap is,the more benefits the firm can gain in term of shopping preference of customer toward itsproduct or service by improving the attribute performance

e performanc current

e performanc good

preference shopping

gap e Performanc

gap Preference efficiency

Figure 3.4: Two-factor model’s coefficients

As a result, the higher the improvement efficiency is, the higher the priority forperformance improvement the attribute has Moreover, attributes which have lowimprovement efficiency but high preference gap are the reserve sources for improvement

in term of shopping preference when all of high efficient attributes reach their potential sothe firm cannot gain any amount of shopping preference of customer by improving them.However, the efforts and resources needed for performance improvement vary fromattribute to attribute For some attributes, it is very easy and cheap to improve theperformance In contrast, it is very difficult and expensive to improve some attributesperformance For detailed improvement program, this factor must be considered and anaction plan with cost/benefit analysis should be conducted This research model focusesmostly on benefit side (shopping preference of customers)

12.4 Critiques for Two-factor model

a) The determination of characteristic curve

By using three points in two dimensions co-ordinates as the method described above toconstruct characteristic curve, the shape of characteristic curve is determined rather thanthe exact curve However, it is good enough for the analysis in this research because theneeded information is the shape of characteristic curve and three critical points: negativepoint, current point, and positive point

b) The accuracy of improvement efficiency

The improvement efficiency is the average angular coefficient of characteristic curve fromcurrent point to positive point Therefore, it does not determine exactly the efficiency ofperformance improvement at current point but it determines the average efficiency of

Improvement efficiency

performance improvement from current performance to very good performance If theattribute performance does not improve thoroughly to very good performance, thiscoefficient will be less accurate In addition, the more convex the curve is (from currentpoint to positive point), the less accurate the improvement efficiency is because theconvexity leads to the variety of angular coefficient from current point to positive point Ifthe characteristic curve is degenerated to be a straight line, improvement efficiency will beabsolutely accurate.

For conclusion, improvement efficiency is an accurate coefficient when the attributeperformance improves thoroughly to very good performance or when the convexity is not

so significant Otherwise it should be considered as an approximate coefficient

c) The need of two surveys

Two surveys are needed for data analysis Survey A is used to measure shoppingpreference contributed by very good performance and by very bad performance of eachattribute and survey B is used to measure performance level of each attribute of studiedproduct or service and shopping preference contributed by this performance level It is theprice for the quality and quantity of information However, survey A could be reused forother research of the same kind of service (or product) For example, Co-op Mart can usethe same survey A to examine the quality of all stores in the chain

CHAPTER 4: RESEARCH METHODOLOGY

In order to achieve the objectives of the research study as discussed in Chapter 1, andbased on Two-factor model discussed in Chapter 3, the following information is needed forthe research:

 Attributes of a supermarket which construct the quality perception of customers

Source: previous research, focus group of customer, and management interview.

 Characteristics of these attributes which determine the relationship between attributesperformance and shopping preference of customers How does shopping preference ofcustomers behave when these attributes performance move up and down?

Source: survey A and survey B.

 Co-op Mart Nguyen Dinh Chieu’s attributes performance in the perception of customerand how much this performance level contributes to shopping preference of customers

Source: survey B.

 Recent operation of Co-op Mart Nguyen Dinh Chieu store

Source: management interview and observation.

Primary data of the research is collected through two kinds of survey called survey A andsurvey B These two surveys are conducted on two target populations

 Survey A: uses the questionnaire called “questionnaire A” (appendix A) to interview

customers of all supermarkets in Ho Chi Minh City

 Survey B: uses the questionnaire called “questionnaire B” (appendix B) to interview

Co-op Mart Nguyen Dinh Chieu’s customers, who have experiences of thissupermarket’s services

15.1 Studied attributes

As the discussion in chapter 3, quality attributes of supermarket is generated by combiningunstructured interviews as in other multi-attribute models to reflect the specific quality ofsupermarket My previous research with collaborators suggests the list of 60 qualityattributes for supermarket after many customer interviews and one focus group discussion.Starting with this list, convenience of open hours and location were added to make a list of

62 attributes (appendix E)

The list of 62 attributes was shortened by a focus group discussion There are 2 male and 3female customers in the group All of them were invited to the discussion after finishingtheir shopping at Co-op Mart Nguyen Dinh Chieu The moderator led the discussion frompreparation for shopping, went through self-shopping, then the billing process Allattributes which are not related to these three processes were wiped off Some remainedattributes were combined and others were discarded because of unimportance After the

discussion, a short list of quality attributes was established and this list has the agreement

of the whole group

The opinions of Co-op Mart management were also taken into account All attributes in thelist were discussed one by one The management decided to drop out some attributes whichthey cannot improve or modify by just basing on this research For example, the location ofsupermarket is decided as a result of a series of researches and the convenience of locationdepended on customer’s residence so including this attribute into this research is notrelevant In addition, there are many important attributes which are studied moreconveniently by other methods such as observation, incidence, etc They are also excludedfrom this research, for example, the availability of locker, the availability of shoppingtrolley, the convenience of entries, etc The management also added one more attributewhich was not considered by customer but the management wants to investigate: flexibility

of paying method After all, the list of studied attributes was completed with 23 items

Table 4.1: List of studied attributes

1 Convenience of parking

2 Hospitality of locker personnel

3 Agility of locker personnel

4 Quality of merchandise

5 Variety of product lines

6 Variety of brand names and varieties for each product line

7 Introduction of new products

8 Unique products which customers cannot find somewhere else

9 Store size

10 Convenience of products arrangement for finding

11 Attractiveness of products display

12 Air condition

13 Price level compared with other supermarkets

14 Hospitality of salesperson

15 Expertise of salesperson in product information

16 Helpfulness of salesperson in finding

17 Flexibility of paying method (credit card, check…)

18 Hospitality of cashier

19 Accuracy of cashier

20 Agility of cashier

21 Returning unqualified products policy

22 Attractiveness of promotion campaigns

23 Convenience of open hours

Based on this list of attributes, both two questionnaires A and B were developed Inaddition, these attributes could be classified by two ways, by process and by category, foranalysis There are three main processes: shopping preparation, self-shopping, and billing.The attributes which are not classified into any process are considered as general attributes.Attributes are also classified into four categories: facility, goods, personnel, and policy.

Table 4.2: Attributes classification

Based on the pilot result, both questionnaires A and B were modified Wordings of somequestions and the question arrangement were changed to be more appropriate In addition,the numbers were added to check boxes for more transparency (-3 for strongly disagree, 3for strongly agree, -2, -1, 0, 1, 2 in between) Moreover, the survey method was changedfrom questionnaire distribution to structured interview with the support of questionnaire tofacilitate the accuracy of data

Since we cannot determine the whole population in detail (so we can not reach eachcustomer with the same probability), the probability sampling (random sampling) isinfeasible (Holbert and Speece, 1993, p.79, 81) Therefore, the convenience sampling wasused and the sample size was determined by the rule of thumb (minimum sample size is 30for each item), the bigger the sample size the more precise the statistics

The sample size of survey A is 320 (NA= 320) and this survey was conducted in 8randomly chosen supermarkets (appendix C) in Ho Chi Minh City (40 interviews persupermarket) to capture the opinions of customers in Ho Chi Minh City

The pre-determined sample size of survey B is 350 but only 302 (NB= 302) interviewswere actually conducted due to the difficulty at the end of open hours Based on theaverage distribution of bill quantity in a day, the survey B’s interviews were conducted(appendix C) but at the end of working hour (after 21:00), most of the customer refused theinterview because they are very hurry The reason for this arrangement is that theassimilation of the interview’s distribution with the bill quantity’s distribution mayincrease the representative property of the sample

A pre-determined procedure to choose interviewee was established to eliminate thepersonal preference of field interviewers which is the source of bias All the interviewswere conducted at the pre-determined time in the schedule (appendix C) The customers

who firstly finished billing process after these points of time in the schedule were chosenfor the interview In the cases they refused the interview or they are not in the scope of theresearch, the next customer would be the interviewee For excluding customers who visitsupermarket less than 4 times in the year or who are under 15, the interviewer must askeach customer before the interview.

less than 1per month

-sometimes notmuch, some timesvery big amountVND

Respondents in sample A are mostly (90.6%) in the age of 21 to 45 years old The range of26-30 is the biggest group, accounts for 27% of the sample 70% of the respondents are

female In average, more than 78% of respondents go to supermarket from 1 per month to

1 per week and more than 91% of them spend less than 500,000VND for each shopping.Particularly, more than 62% of them spend less than 200,000VND for each shopping(exchange rate: 14,500VND = 1USD) Therefore, the buying power of customer atsupermarket in Ho Chi Minh City is not much

less than 1per month

500,000 1,000,000

-more than 1,000,000

sometimes not much, some times very big amount

VND

Respondents in sample B are mostly (91.4%) in the age of 21 to 45 years old Nearly 70%

of the respondents are female In average, nearly 78% of respondents go to supermarket

from 1 per month to 1 per week and more than 92% of them spend less than 500,000VNDfor each shopping Particularly, more than 54% of them spend less than 200,000VND foreach shopping The pattern of sample B is similar with sample A Hence, the customer ofCo-op Mart Nguyen Dinh Chieu is not so different to supermarket’s customer in Ho ChiMinh City.

CHAPTER 5: FINDINGS ANALYSIS AND DISCUSSION

As discussion in chapter 4, all of the studied attributes are classified into 4 categories:facility, goods, personnel, and policy These attributes will be analyzed by Two-factormodel as discussion in chapter 3 Each attribute is presented in two-dimensional co-ordinates by characteristic curve The horizontal axis represents performance level of anattribute and the vertical axis represents the shopping preference of customers contributed

by this performance level There are three critical points on the curve, one on the left isnegative point, one on the right is positive point, and one another in between is currentpoint

Moreover, preference gap and improvement efficiency are also used to analyze qualityattributes On one hand, the larger the preference gap is, the more benefits the firm cangain in term of shopping preference of customer toward its product or service byimproving the attribute performance On the other hand, the higher the improvementefficiency is, the more shopping preference amount the firm gains by improving a certainamount of attribute performance

Based on the statistical analysis results (table D.2, D.3, D.4, D.5 in appendix D) and testing(table D.6, D.7, D.8, D.9 in appendix D), the characteristic curves’ co-ordinates, shoppingpreferences, and improvement efficiencies of studied attributes are summarized (table D.10

in appendix D) All characteristic curves’ information and coefficients in this chapter areextracted from this table

18 F ACILITY A TTRIBUTES A NALYSIS

There are 3 attributes which belong to facility attribute category: convenience of parking,store size, and air condition The characteristic curves of air condition attribute (figure 5.2)and store size attribute (figure 5.3) are similar They are nearly absolute linear andsymmetric The characteristic curve of convenience of parking attribute (figure 5.1) isslightly negative convex

Table 5.1: Facility attributes

Improvementefficiency

The improvement efficiency and preference gap of air condition attribute and store sizeattribute are not so different Therefore, they are similar in terms of benefits fromimprovement as well as room for improvement.

18.1 Convenience of parking

Customers’ response reveals that parking area of Co-op Mart Nguyen Dinh Chieu issomewhat convenient but this convenience contributes very little (just above neutral) tocurrent shopping preference of customers However, there are many rooms for increasingshopping preference of customers by improving this attribute thanks to high preference gapand negative convexity The investment in improving this attribute is also highly beneficialdue to the high improvement efficiency

The expansion of parking areas is especially expensive and the trade off with operatingareas restricts this option Rearrangement areas and way in, way out for more conveniencemay be a more feasible option For more convenient to customer, some entries should beopened to connect parking area with supermarket so customers can go from parking area tosupermarket and vice versa directly Recently, customer must go out of the parking areabefore enter to supermarket and vice versa

Figure 5.1: Convenience of parking

1.78

0.29

-1.49

-3-2-10123

preference gap reveals that there is room for shopping preference improvement byimproving this attribute in future.

For improving air condition significantly, the recent distributed 10 small air-conditionerssystem must be changed by a central air-conditioner This replacement project willconsume a lot of resources which could not be compensated by such an improvement ofshopping preference of customers Hence, the old system is recommended to continue inplace The other advantage of the old system is that it can be easy modified by addingsome small air-conditioners This incremental improvement is reasonable for such anattribute

Figure 5.2: Air condition

1.490.68

-1.67

-3-2-10123