Factors influencing domestic consumption in vietnams market

The model investigates the effect of someindependent variables on the domestic consumption such as import product qualityperception, consumer ethnocentrism, cosmopolitanism and patriotis

CAO QUỐC VIỆT

FACTORS INFLUENCING DOMESTIC CONSUMPTION IN VIETNAM’S MARKET

MASTER’S THESIS IN BUSINESS ADMINISTRATION

Ho Chi Minh City – 2012

MINISTRY OF EDUCATION AND TRAININGUNIVERSITY OF ECONOMICS HOCHIMINH CITY



CAO QUỐC VIỆT

FACTORS INFLUENCING DOMESTIC CONSUMPTION IN VIETNAM’S MARKET

MASTER’S THESIS

InBusiness AdministrationCode: 60340102

Supervisor

Dr Nguyễn Thị Mai Trang

Ho Chi Minh City – 2012

When I write these statements, it reminds me to turn back the time of two years ago;

as I read the book “Nghiên cứu khoa học Marketing - Ứng dụng mô hình cấu trúc tuyến tính SEM”, I could not understand anything It is a mysterious world that I

need to discover It was also the start point for my study road My sincere thankssend to the authors who have composed this wonderful book

I would like to thank my supervisor Dr Nguyen Thi Mai Trang for her supports,instructions, and valuable comments throughout the work leading to this thesis

I would like to thank all respondents who gave me priceless time during the datacollection phase of this thesis

I never forget my thanks to Mr Nguyen Khanh Duy with his first SEM Vietnamesebook

My deep thanks to my wife, Tsằn Schếng Mỹ, and our son; their love is themotivation behind my work

Thanks to my family, my Mother, Father, brothers and sister

Thanks God bless and lead me

I declare that the entirety of the work contained therein (except the citations) is myown work I declare that I am the only author and neither portion in the thesis hasbeen submitted in support of another degree nor published by another person

HoChiMinh City, 10/10/2012

Cao Quoc Viet

The purpose of this thesis is to investigate domestic consumption The aim is toexamine factors which influence the domestic consumption among consumers inVietnam Furthermore, the research focuses on finding out determinant that influencesdomestic consumption the most Based on Vida and Reardon’s Model and literaturereview a theoretical model was proposed The model investigates the effect of someindependent variables on the domestic consumption such as import product qualityperception, consumer ethnocentrism, cosmopolitanism and patriotism

A survey of 396 consumers was carried out Some techniques were used to analyze themodel, such as descriptive statistics, reliability assessment, exploratory factor analysisand confirmatory factor analysis The proposed theoretical model was also empiricallytested by means of the structural equation modeling technique The results show thatonly cosmopolitanism was insignificant factor The findings indicate that importproduct quality perception, consumer ethnocentrism, patriotism are significantpredictors of domestic consumption

Key words: domestic consumption, consumer ethnocentrism, patriotism,cosmopolitanism, import product quality perception

List of Acronyms

AMOS - Analysis of Moment Structures

CET - Consumer ethnocentrism

CFA - Confirmatory Factor Analysis

CFI - Comparative Fit Index

CMIN - Chi square

CMIN/df - Chi square/degree of freedom

COS - Cosmopolitanism

DC - Domestic consumption

EFA - Exploratory Factor Analysis

FDI - Foreign Direct Investment

GFI - Goodness of Fit Index

IPQP - Import product quality perception

KMO - The Kaiser - Meyer - Olkin

Table of Contents

Acknowledgements ii

Declaration iii

Abstract iv

List of Acronyms v

Chapter 1: INTRODUCTION 1

1.1 Research background 1

1.2 Problem statement 1

1.3 Research objectives 2

1.4 Scope and Research methodology 3

1.4.1 Scope of the study 3

1.4.2 Research Method 3

1.5 Structure of the research 4

Chapter 2: LITERATURE REVIEW 5

2.1 Introduction 5

2.2 Domestic Consumption (DC) 5

2.3 Import Product Quality Perception (IPQP), Consumer Ethnocentrism (CET), Patriotism (PAT) and Cosmopolitanism (COS) 5

2.3.1 Import Product Quality Perception (IPQP) 5

2.3.2 Consumer Ethnocentrism (CET) 6

2.3.3 Patriotism (PAT) 6

2.3.4 Cosmopolitanism (COS) 7

2.4 Hypothesis development and Research Model 7

2.4.1 Import Product Quality Perception (IPQP), Consumer Ethnocentrism (CET) and Domestic Consumption (DC) 7

2.4.2 Cosmopolitanism (COS), Patriotism (PAT) and Consumer Ethnocentrism (CET) 8

2.4.3 Patriotism (PAT) and Domestic Consumption (DC) 9

2.4.4 Research Model 9

Chapter 3: METHODOLOGY 11

3.1 Introduction 11

3.2 Research design 11

3.2.1 Pilot study 13

3.2.1.1 Qualitative research 13

3.2.1.2 Quantitative research 13

3.2.1.3 Pilot study results 15

3.2.2 Official Quantitative research 18

3.2.2.1 Sample 18

3.2.2.2 Data analysis methods 19

3.3 Measurement 19

3.3.1 Scale to measure Consumer Ethnocentrism (CET) 20

3.3.2 Scale to measure Import Product Quality Perception (IPQP) 21

3.3.3 Scale to measure Patriotism 21

3.3.4 Scale to measure Cosmopolitanism 23

3.3.5 Scale to measure Domestic Consumption 23

3.4 Summary 24

Chapter 4: DATA ANALYSIS AND FINDINGS 25

4.1 Introduction 25

4.2 Descriptions of sample 25

4.3 Scale assessment 26

4.3.1 Normality assessment 26

4.3.2 Reliability testing 27

4.3.3 Exploratory Factor Analysis 28

4.4 Confirmatory Factor Analysis (CFA) 31

4.5 Testing Theoretical Model 41

4.6 Testing Hypotheses and Summary 44

4.6.1 Testing Hypotheses 44

4.6.2 Summary 46

Chapter 5: CONCLUSIONS AND IMPLICATIONS 47

5.1 Introduction 47

5.2 Discussion of findings 47

5.3 Managerial Implications 48

5.4 Limitations and Future research 50

List of References 52

Appendix 57

Appendix 1 Refrigerator statistics Vietnam 2005 - 2010 57

Appendix 3.1 Group discussion 59

Appendix 3.2 English Questionnaire 62

Appendix 3.3 Bảng câu hỏi khảo sát 66

Appendix 4.1 Descriptive Statistics 71

Appendix 4.2 Cronbach’s Alpha 72

Appendix 4.3 EFA 75

Appendix 4.4 CFA results 77

Appendix 4.5 Theoretical Model 85

LIST OF FIGURES

Figure 2.1: Theoretical Model 10

Figure 3.1: Research procedure 12

Figure 4.4: Saturated Model 34

Figure 4.5: Refined Saturated Model 35

Figure 4.6: Final Saturated Model 37

Figure 4.7: SEM’s Result - Theoretical Model (Standardized) 41

LIST OF TABLES

Table 3.1: Cronbach’s Alpha results 15

Table 4.1: Sample characteristics 26

Table 4.2: Assessment of Normality 27

Table 4.3: Cronbach’s Alpha results 28

Table 4.4: EFA results 30

Table 4.5: Model Fit Summary - Comparison 36

Table 4.6: Assessment Construct Validity Result 37

Table 4.7: Model Fit Summary - Final Saturated Model 38

Table 4.8: Assessment Construct Validity Result 39

Table 4.9: Regression weights and standardized regression weights summary40 Table 4.10: Testing the discriminant validity 40

Table 4.11: Regression Weights - Theoretical Model 42

Table 4.12: Standardized Regression Weights - Theoretical Model 43

Table 4.13: Squared Multiple Correlations 43

Table 4.14: Bootstrapping results (N = 1000) 44

Table 4.15: Hypotheses testing and results (based on the Regression Weights table in Theoretical Model) 45

Table 4.16: Standardized Regression Weights (Theoretical Model) 46

Chapter 1: INTRODUCTION 1.1 Research background

“Nowadays, globalization is taking its long stride and positioning as an

indispensable process of all countries in over the world, in which, every nation

is in need of deeply integrating into the process of globalization unless they leftbackwards from the others” (Đang 2009) For that reason, Vietnam became the

WTO’s 150th member on 11 January 2007 The increased competition among

domestic and foreign goods and services are the results of economy‘sglobalization (Netemeyer 1991cited from Ranjbarian and Barari 2011)

Clearly, when we have opened the door and entered the international market,local enterprises have been faced with more challenges such as the penetrationand competition from foreign companies The relaxation of trade policies hasprovided consumers with more foreign product choices than ever before (Wangand Cheng 2004) Thus, local firms must try, for taking efforts to market,increasing sales and considering about the question: Which factors do influencethe purchase of domestic products instead of buying imported goods of

Vietnam’s consumers? This thesis proposes a study containing four factors

affecting the domestic consumption of Vietnamese in Vietnam’s market

1.2 Problem statement

This problem has been studied by many academic researchers for many yearsboth in countries having developed and transition economies Research on thefactors affecting the choice of consumers between domestic and foreign

products has attracted several researchers in the past several years (Nguyen et

al 2008) Consumer ethnocentrism is one of factors affecting directly to the

purchase intention and is studied by some researchers such as Herche 1994;Klein and Ettenson 1999; Nguyen and Nguyen 2008; Shimp and Sharma 1987;Sumner 1906; Vida and Reardon 2008 Patriotism is studied by Bannister and

Saunders 1978; Balabanis et al 2001; Han 1988; Sharma et al 1995; Vida and

Reardon 2008 Research conducted on domestic/imports product qualityperceptions (e.g., Han 1988; Klein 2002; Nguyen and Nguyen 2008; Wang andCheng 2004; Vida and Reardon 2008) Cosmopolitanism construct (e.g.,Javalgi et al 2004; Suh and Kwon 2002; Vida and Reardon 2008) In Vietnam,Nguyen and Nguyen are the priority authors who had the research for thisproblem done in 2008 From that point, according to my best knowledge, Icould not find any research conducted the factors affecting the choice eitherdomestic products or foreign products of Vietnamese consumers by researchers

On the practice side, Vietnam had the programs like “Vietnam High Quality

Products” in 2004 On 31/07/2009, the Politburo issued Document No

264-TB/TW on the organization of the campaign “Vietnamese use Vietnamese’s

products” launched nationwide It proves that Vietnam’s government hasidentified the important role of Vietnamese’s buying and using domestic

products Thus, exploring determinants influence the domestic consumption in

Vietnam’s market is a valuable contribution for domestic enterprises within the

field of marketing

1.3 Research objectives

This thesis aims to apply the model of domestic consumption developed byVida and Reardon (2008) and test this model in Vietnam as an emergingeconomy From those results, the managerial implications will help localenterprises build the marketing and business strategy In order to confirm theapplicability of the model, this study will determine the contribution of each ofthe factors to the domestic consumption To serve this task, two questions need to

be answered:

Q1 Which factors do influence the domestic consumption in Vietnam?

Q2 How do factors influence the domestic consumption in Vietnam?

For the purposes of this study, imports or foreign products mean thoseimported from countries of higher origin (i.e Technologically/ economicallyadvanced countries) to Vietnam and domestic goods mean the goods are eitherproduced or manufactured by Vietnam companies/ FDI companies in Vietnam.

1.4 Scope and Research methodology

1.4.1 Scope of the study

This research concentrates on the Vietnam market and the behavior ofVietnamese in their consumption Refrigerator is chosen as a proxy forthe product in this study The reason is that, refrigerator is a necessaryproduct for every Vietnamese family According to EuromonitorInternational1(2010), “Due to busier lifestyle, especially in urban areas,Vietnamese consumers increasingly prefer to make one – stop groceryshopping trips just once a week or even once a fortnight During thereview period, this trend led many consumers to trade up torefrigeration appliances with a larger volume capacity, so they couldstore more food at home and spend less time on grocery shopping”.Appendix 1.1 presents the statistic data of refrigerator’s sales from

2005 to 2010

1.4.2 Research Method

The research is conducted in two main stages; the first stage: “focusgroup” and pilot test are used to build a clear completed questionnaire;the second stage: quantitative research with 396 samples (detailed inchapter 3) The reliability of the measurement scales is re-assessed byusing Cronbach alpha coefficient and Exploratory Factor Analysis(EFA) The next step uses CFA (Confirmatory Factor Analysis) and

1

Euromonitor International is the world leader in strategy research for consumer markets, with offices around the world, analysts in 80 countries and market research on every key trend and driver.

tests the theoretical model by using SEM technique All thesetechniques are used in the thesis under the operation of PASW Statistics

18 and Amos 18 software

1.5 Structure of the research

The thesis contains five chapters Chapter 1 provides the introduction of thethesis including research background, problem statement and researchobjectives This chapter also mentions to the scope of the study and researchmethod Chapter 2 focuses on literature review relating to factors affectingdomestic consumption A list of hypotheses and research model are presented

in this chapter Next, chapter 3 addresses the methodology of the research.Chapter 4 presents data analysis and findings Finally, chapter 5 discusses thefindings, managerial implications and limitation as well as future research

Chapter 2: LITERATURE REVIEW 2.1 Introduction

The previous chapter introduces an overview of the thesis The aim of thischapter is to search and review relevant theories in the literature A researchmodel is introduced Chapter 2 includes main contents such as a literaturereview of domestic consumption and imports product quality perception,consumer ethnocentrism, patriotism, cosmopolitanism as well as presents theresearch model

2.2 Domestic Consumption (DC)

According to the Oxford Advanced Learner’s Dictionary (OALD), domesticconsumption (DC) means the act of buying and using domestic products Inmarketing, when scholars research on buying and using products,they mentionthe consumer behavior and scholars study on the attitudes and perceptions ofconsumers Consumers have diverse perceptions about products; theseperceptions affect consumer attitudes, purchasing intentions and behaviors(Vida and Reardon 2008)

2.3 Import Product Quality Perception (IPQP), Consumer Ethnocentrism (CET), Patriotism (PAT) and Cosmopolitanism (COS)

2.3.1 Import Product Quality Perception (IPQP)

Consumers’ intention to buy either domestic or import products will beinfluenced by perceived quality (Wang and Cheng 2004) In theliterature, quality perception is often treated as a multidimensionalmeasure including workmanship, appearance, design, durability,functionality, prestige, reliability, technical advancement and value formoney (Vida and Reardon 2008) Many researches show that consumers

in developed countries tend to perceive domestic products as being of

higher quality than imported products whereas the reverse is true for

consumers in developing countries (Nguyen et al 2008; Wang and

Cheng 2004)

2.3.2 Consumer Ethnocentrism (CET)

Sumner is the first author introduces the term ethnocentrism in 1906 Hedefines ethnocentrism as “the view of things in which one’s own group

is the center of everything and all others are scaled and rated with

reference to it” (Sharma et al 1995) Consumer ethnocentrism has been

defined as the “unique proclivity for people to view their own group asthe center of the universe, to interpret other social units from theperspective of their own group, and to reject persons who are culturallydissimilar while blindly accepting those who are culturally likethemselves,” (Shimp and Sharma 1987) Consumer ethnocentrism canpredict (with varying precision across product – categories) consumers’preferences to buy or own domestic as opposed to foreign products

(Herche 1992 cited from Balabanis et al 2001)

Consumer ethnocentrism represents beliefs that purchasing importedproducts is wrong and immoral because it can hurt the domesticeconomy and can cause the loss of jobs (Shimp and Sharma 1987)

2.3.3 Patriotism (PAT)

Patriotism means the love of your country and willingness to defend it(OALD 2010) Patriotism refers to the strong feeling of attachment andloyalty to one’s own country without the corresponding hostility towards

other nations (Balabanis et al 2001) Patriotic attitudes develop early in

the socialization process and as a result is very persistent, with the deep– rooted needs served by these attitudes (i.e Security, feeling ofbelonging, self – enhancement) making them very resistant to change(Druckman 1994) Patriots, on the whole, display a willingness to

sacrifice for their country and subordinate their personal interests to

national interests (Feshbach 1990 cited from Balabanis et al 2001).

2.3.4 Cosmopolitanism (COS)

Cosmopolitanism (from the Greek kosmopolitês, meaning citizen of the

world) is a concept drawn from the fields of anthropology and socialpsychology, with growing applications in marketing and management

(Cleveland et al 2010) Cosmopolitans tend to consume international

media, foreign books, and films while in their home countries (Hannerz

1990 cited from Riefler and Diamantopoulos 2009) Researchers nowhave the same opinion that cosmopolitanism, having been looselyapplied to describe people that move about in the world, should instead

be used to refer to a specific set of attitudes, beliefs and traits, most of all,

“an ethos of cultural openness” (Kurasawa 2004 cited from Cleveland et

al 2010).

“Cosmopolitan consumers are much more open to new ideas, they tend

to address functional needs with the kind of products or services thatbest deliver the desired function, regardless of tradition or social

influence” (Cannon et al 1994 quoted from Riefler and Diamantopoulos 2009) Cannon et al reason that if cosmopolitan consumers base their

choices on functional needs rather than local customs or traditions, thenconsumer cosmopolitanism represents a potential transnationalsegmentation base (Riefler and Diamantopoulos 2009)

2.4 Hypothesis development and Research Model

2.4.1 Import Product Quality Perception (IPQP), Consumer

Ethnocentrism (CET) and Domestic Consumption (DC)

Wang and Cheng (2004) state that a consumer will tend to purchase aproduct made in a technologically-advanced country if he or she judges

its quality as better than that of merchandise produced in a developed country Moreover, a product’s country-of-origin often serves

less-as a cue activating a consumer’s ethnocentric tendency (Huddleston et al2001) Thus, domestic consumption is influenced by both ethnocentrismand quality judgment In a developed country, consumers in general tend

to have a higher quality perception of domestic than foreign products(Ahmed and d’Astous 2001; Bilkey and Nes 1982; Dickerson 1982;Morganosky and Lazarde 1987; Samiee 1994 cited from Wang andCheng 2004) However, it is not true in developing country such asNigeria, Romania, Turkey, India and China In these countries,consumers typically perceive foreign products, particularly those made

in higher origin countries, as being of higher quality than domesticproducts (Vida and Reardon 2008) In Vietnam, Nguyen and Nguyen(2008) find that there is a negative relationship between imports

products quality perception and domestic purchase intention Rybina et

al (2010) conclude that ethnocentric attitudes of consumers have a

positive effect on domestic purchase behavior in Kazakhstan By this Imean, there are some hypotheses offered in this study:

H1: Import Product Quality Perception is negatively related to Domestic Consumption.

H2: Consumer Ethnocentrism has a positive effect on Domestic Consumption.

2.4.2 Cosmopolitanism (COS), Patriotism (PAT) and Consumer

Ethnocentrism (CET)

Balabanis et al (2001) investigate the patriotism, nationalism, and

internationalism as antecedents of consumer ethnocentrism in Turkeyand the Czech Republic They find a significant effect of patriotism andnationalism These results are produced in the previous researches (e.g

Han 1988; Sharma et al 1995; Klein and Ettenson 1999) In 2008 a

research has the same similarly result composed by Vida and Reardon

Rybina et al (2010) find that patriotism has a significant positive effect

on ethnocentric attitudes of consumers and cosmopolitanism has asignificant negative effect on ethnocentric attitudes of consumers inKazakhstan Other studies produce confusing results on the impact ofcultural openness or related constructs used in the literature (e.g.,cosmopolitanism, internationalism, global mindedness) onethnocentrism (Suh and Kwon 2002) Vida and Reardon (2008) observethe result of cosmopolitanism, as a direct antecedent of consumerethnocentrism, is negatively related to CET in their research inSlovenia’s market Hence, there are two hypotheses offered in this study:

H3a Patriotism is positively related to Consumer Ethnocentrism H3b Cosmopolitanism is negatively related to Consumer Ethnocentrism.

2.4.3 Patriotism (PAT) and Domestic Consumption (DC)

A study by Han (1988) shows patriotism has a positiveinfluence ontheconsumers’ choice of domestic over foreign products (Granzin andOlsen 1998) Vida and Reardon (2008) test the hypothesis: Domesticconsumption is determined by the affective mechanism of patriotism inSlovenia This result is confirmed Thus, in this study, the fourthhypothesis is recommended:

H4: Patriotism is positively affected to Domestic Consumption.

2.4.4 Research Model

Based on presented hypotheses above, the research model in this study:

Figure 2.1: Theoretical Model

H4(+)

Cosmopolitanism

(COS)

Import product quality perception (IPQP)

Domestic consumption (DC) Consumer

ethnocentrism (CET)

Patriotism

(PAT)

H1(-)H3b(-)

H3a(+)

H2(+)

Chapter 3: METHODOLOGY 3.1 Introduction

This chapter will go into details the problems related to research methods,research procedure and scales for the purposes of measuring constructs andtesting the research model

3.2 Research design

This research was divided into two stages The first stage was pilot study Thequalitative research and quantitative research, in this stage, supported thebuilding of the questionnaire In the second stage, the official quantitativeresearch was conducted with the collected data, testing the research hypothesesand research model as well Figure 3.1 presents the research procedure

Figure 3.1: Research procedure

The first draft questionnaire

Literature reviews(DC scale, IPQP scale, CEscale, PAT scale, COS scale)

Pilot study(1) Qualitative research(Group discussion)

n = 9

The second draft questionnaire

(2) DraftQuantitativeresearch

3.2.1 Pilot study

3.2.1.1 Qualitative research

The main purpose of this stage was building the measuringscales suitable for Vietnam’s market In order to do that theoriginal scales were translated from English into Vietnamesebased on the procedure of Craig and Douglas (2003) whoproposed a process of forward and back translation for across-culture adoption In the forward translation process, theoriginal scales were translated into Vietnamese (except forCET scale taken from a local study in 2008 by Nguyen andNguyen) under the reference and instruction from a specialistwho has been lived and studied in American for a long time

In back translation, another specialist who has experience inteaching English at the University of Economics Ho Chi MinhCity helped translate the first draft Vietnamese questionnaireback into English to validate this version The two versionswere then compared in the source language to check for errorsand the quality of the translation (Craig and Douglas 2003).After finishing this stage, the next step had been carried out in

- depth interview technique to ensure the adaptablequestionnaire to Vietnam’s market Nine customers wereasked around The questionnaire for group discussion waspresented in Appendix 3.1

3.2.1.2 Quantitative research

The second draft questionnaire was built on GoogleDocuments and sent to 400 MBA students; 129/400 feedbacks

within two weeks was coded, filtered and run by PASWStatistics 18 The purpose of this step was to compose theofficial completed questionnaire (through testing theCronbach’s alpha, EFA and adjusting some items in eachconstruct) used for quantitative research.

Reliability analysis was first used to remove items with lowItem-Total Correlations (<0.3) (Nunnally and Bernstein 1994

- cited from Nguyen 2011) Scales with a Cronbach’s alphacoefficient either equal to or greater than 0.6 are acceptable insome cases (Nunnally and Bernstein 1994 - quoted in Nguyen2011) Items those passed the test then were analyzed byusing Exploratory Factor Analysis (EFA) method todetermine the actual dimensions of each construct

In that working, items with factor loadings less than 0.5 weredeleted (Nunnally and Bernstein 1994 - quoted in Nguyen andNguyen 2008); variables should generally have communalities

of greater than 0.5 to be retained in the analysis (Hair et al

2010); Total Variance Cumulative is either equal or greaterthan 50% (Gerbing and Anderson 1988 cited from Nguyen andNguyen 2008) Furthermore, in EFA, researchers use Bartlett’stest of sphericity and The Kaiser - Meyer - Olkin (KMO)Measure of Sampling Adequacy

Bartlett’s test of sphericity is a statistical test for the presence

of correlation among variables It provides the statisticalsignificance that the correlation matrix has significantcorrelations among at least some of the variables (Hair et al2010) This means if the Bartlett’s test of sphericity to besignificant (p < 0.05), then we can reject the hypothesis that the

correlations in a correlation matrix are zero.

The Kaiser - Meyer - Olkin (KMO) Measure of SamplingAdequacy tests whether the partial correlations amongvariables are small (Norusis 2003) The overall KMO scorecan range from 0 to 1; KMO score should be above 0.6 toproceed with factor analysis (Norusis 2003) Kaiser (1974 ascited from Nguyen 2011) suggests that KMO >= 0.9:marvelous; >= 0.8: meritorious; >= 0.7: middling; >= 0.6:mediocre; >= 0.5: miserable and KMO < 0.5: unacceptable.Appendix 3.2 shows the official completed questionnaire

3.2.1.3 Pilot study results

a Reliability of scales:

All figures in Table 3.1 showed the results with Item TotalCorrelation were greater than 0.3 and Cronbach’s Alpha Coefficientswere greater than 0.6

Table 3.1: Cronbach’s Alpha results

AlphaScale Mean

if Item

Deleted

ScaleVariance ifItem Deleted

CorrectedItem-TotalCorrelation

Cronbach'sAlpha ifItem Deleted

Item-Total Statistics Cronbach's

Rotated Component Matrix a

Based on EFA’s result above: KMO> 0.5; Sig (Bartlett’s test) <0.05; Total Variance Explained > 50% (68.75% and 73.45%).These results showed that the draft questionnaire could be used inmain survey.

3.2.2 Official Quantitative research

3.2.2.1 Sample

The basic idea of sampling is that by selecting some of theelements in a population we may draw conclusions about theentire population There are several compelling reasons forsampling, including: lower cost, the accuracy of results, thespeed of data collection and the availability of populationelements (Donald and Pamela 2003) The population for thisstudy was limited to customers who are living and working inVietnam; however, it was impossible to identify all thecustomers who make up the whole population and toestablish a sampling frame that includes a large proportion ofthe population Due to the tremendous limitations of time,budget and knowledge, this study used a non-probabilitysampling technique, specifically, convenience sampling This

is one of the least reliable sampling techniques, but it is thecheapest and easiest, and is the most feasible for this study

As a result, students who are studying a second – degree(They have graduated from a University and kept learning inanother school that they want) and MBA students at theUniversity of Economics of Ho Chi Minh City were surveyed.There is no consensus in the literature on how large the sample

size should be to represent a population According to Hair et al

(2010), preferably the sample size should be 100 or larger; as ageneral rule, the minimum is to have at least five times as manyobservations as the number of variables to be analyzed, and themore acceptable sample size would have a 10:1 ratio.Tabachnick and Fidell (2007) suppose that the number that isconsidered “comfortable” when it is higher than 300 cases Thisstudy is doing SEM; according to Bentler and Chou (1987) aratio of ten responses per free parameters is required to obtaintrustworthy estimates If data is found to violate multivariatenormality assumptions, the number of respondents perestimated parameter increases to 15 (Bentler and Chou 1987).The minimum sample size is five respondents per estimateparameter (Bollen 1989 cited from Nguyen and Nguyen 2008).The sample size for this study was intended to be 378 (equal 7times of parameter - 7*54) To obtain the desired sample size,total of 700 questionnaires (greater than 50% sample size) weredistributed to the respondents.

3.2.2.2 Data analysis methods

The feedback questionnaires were coded, cleaned as well asfiltered by PASW Statistics 18 It was also used in analysis,testing the scales, reliability testing, exploratory factoranalysis AMOS 18 was used for confirmatory factor analysisand testing the theoretical model with structural equationmodeling method

3.3 Measurement

This study used 7 - point Likert scale which ranges from 1 (strongly disagree)

to 7 (strongly agree) There were five measurement scales in this studyincluding CET scale, IPQP scale, PAT scale, COS scale and DC scale

3.3.1 Scale to measure Consumer Ethnocentrism (CET)

As shown in chapter 2, consumer ethnocentrism represents beliefs thatpurchasing imported products is wrong and immoral because it can hurtthe domestic economy and can cause the loss of jobs Hence, CET scaleincludes items which demonstrate this view

Shimp and Sharma (1987) built the CET’s scale with 17 items Fromthat point, lots of researchers had used them in their studies in developed

and emerging countries such as Balabanis et al (2001), Ettenson et al (1998), Good and Hudleston (1995), Klein et al (1998), Sharma et al

(1995) Nguyen and Nguyen (2008), Wang and Cheng (2004), Vida and

Reardon (2008); Rybina et al (2010)… The study used the CET scale

adapted from a local previous research of Nguyen and Nguyen inVietnam (2008) They had rebuilt the scale with five items:

3 A real Vietnamese should always buyVietnamese - made products

4 We should purchase productsmanufactured in Vietnam instead ofletting other countries get rich off us

5 Vietnamese should not buy foreignproducts, because this hurts Vietnamesebusiness and causes unemployment

3.3.2 Scale to measure Import Product Quality Perception (IPQP)

Product quality perception is measured through items showworkmanship, appearance, design, durability, functionality, prestige,reliability, technical advancement and value for money

To measure Import Product Quality Perception, this study adapted thescale developed and rebuilt by Vida and Reardon (2008) in Slovenia.Furthermore, the English version of this scale was translated intoVietnamese under the reference from the rebuilt scale of Nguyen andNguyen (2008) in Vietnam It included 4 items:

3.3.3 Scale to measure Patriotism

As shown in literature review; Patriots, on the whole, display awillingness to sacrifice for their country and subordinate their personalinterests to national interests Hence, the measurement scale for thisconstruct must demonstrate these meaning

Patriotism scale was Kosterman and Feshbach built in 1989 with 12items (Coryn et al 2004) It was used and rebuilt by some authors such

as Balabanis et al (2004), Klein and Ettenson (1999), Rybina et al

(2010), Vida and Reardon (2008) This study adapted the scale

developed by Rybina et al (2010) with 4 items:

2 I am proud to be a Vietnamese citizen

3 When a foreign person praises Vietnam,

it feels like a personal compliment

4 I feel strong ties with Vietnam

PAT1PAT2

PAT3PAT4This scale was translated based on the forward and back translatedprocedure of Craig and Douglas The qualitative research was applied toadjust some items Item 4 (PAT4) got some ideas from specialists InVietnamese, PAT4 has the meaning of either “tôi cảm nhận có mối liên

hệ chặt chẽ với đất nước Việt Nam” or “tôi cảm nhận có sự ràng buộcchặt chẽ với đất nước Việt Nam”

Furthermore, the first item in original scale was used in this study (Item5: I love my country) Explaining for this matter was the content validity

of the scale Group discussion results are supposed that when someonehas the love with their country, they should confirm to themselves thatthey love their country in their mind In summary, there were five items

in the new PAT scale including:

2 I am proud to be a Vietnamese citizen

3 When a foreign person praises Vietnam,

it feels like a personal compliment

PAT1PAT2

PAT3

modifying 4 I feel strong ties with Vietnam.

5 I love my country

PAT4PAT5

3.3.4 Scale to measure Cosmopolitanism

The construct of cosmopolitanism was conceptualized as an individual’sdesire to travel as she or he seeks new insights and experiences in othercultures (Cannon and Yapak 2002 cited from Vida and Reardon 2008)

This study adapted the Cosmopolitanism scale from Rybina et al (2010).

It includes 4 items:

Cosmopolitanism

(COS)(Adapted from

4 Getting information and news fromaround the world is important to me

COS1

COS2

COS3

COS4

3.3.5 Scale to measure Domestic Consumption

Domestic Consumption was measured by items which related toactivities associated with the purchase of domestic products/ brands andidentifying retail outlets that stock them

The original scale was developed by Granzin and Olsen (1998) in USwith 6 items Vida and Reardon (2008) rebuilt this scale in their researchwith 4 items This study used the 4 items in Vida and Reardon’s scale:

Domesticconsumption(DC)(Adapted fromVida andReardon 2008)

1 Mostly I try to buy brands of domesticcompanies

2 Whenever possible, I take time to look atlabels in order to knowingly buy morebrands of domestic companies

3 I shop first at retail outlets that make aspecial effort to offer a variety ofdomestic products

4 I shop first at retail outlets that make aspecial effort to offer brands of domesticproducts

Chapter 4: DATA ANALYSIS AND FINDINGS4.1. Introduction

The previous chapter provides the methodology that this study employsand the parameterization of the six research concepts contained in theresearch model This chapter presents the analysis of results from themain study Critical tasks of this chapter are testing the scales, reliabilitytesting and exploratory factor analysis Confirmatory factor analysis andstructural equation modeling analysis method were used to test thetheoretical model

4.2. Descriptions of sample

As shown in the previous chapter, 700 questionnaires were distributed to therespondents who have ever either used or intend to buy the refrigerator Thefinal sample consisted of 396 questionnaires The sample details are provided

in the Table 4.1 and Figure 4.1 There were more women (54%) in the samplethan men (46%) Approximately 34 percent of respondents got the income of

2 to under 6 million (VND), 38 percent of respondents had the income from 6

to 10 million and 29 percent of respondents had income larger than 10 million

Table 4.1: Sample characteristics

Sample Description Frequency Percent CumulativePercent

multi-Table 4.2: Assessment of Normality

2 This problem had presented in pilot test – chapter 3.

used in the EFA.

Table 4.3: Cronbach’s Alpha results

if itemdeleted

Corrected item-total correlation

Cronbach'sAlpha ifitem deleted

Cronbach 's Alpha

N ofitems

4.3.3 Exploratory Factor Analysis

Cronbach’s Alpha coefficient is used to evaluate the reliability of scale

But reliability is only a necessary - not a sufficient condition forvalidity (Thompson 2004) Consequently, factor analysis is used toinform evaluations of score validity All variables retained werethen put into EFA The purpose of this step was to continuouslypurify the instrument and to identify the components that explain forthe correlation between variables or sets of variables One of theunderlying objectives of EFA is to reduce a large number of variables

to a minimum number that can explain most of the characteristics of theoriginal variables Each component extract from the original set ofvariables represents a discriminated aspect of the construct beingstudied

This study used Principal Axis Factoring Method and Promax RotationMethod instead of Principal Components Analysis Method andVarimax Rotation Method The reason is that this method identifies thelatent dimensions represented in the original variable more accuratelyreflects the underlying structure of data than other methods (Gerbingand Anderson 1988 cited from Nguyen and Nguyen 2008)

EFA’s results in Table 4.3 show five factors extracted at Eigenvalue of1.61 (>=1) and Total Variance Explained of 59.91 (> 50%) All factorloadings were greater than the cut – off value of 0.4 and the result ofKMO was also quite high (0.82); Bartlett’s test of sphericity (sig < 0.05).These results indicated that EFA was suitable and all measures could beused for the next analysis