Tài liệu Studying customer satisfaction’s impact on customer loyalty and shirt repeating purchase behavior of urban customers in vietnam under the influence of involvement and fashion style

Ministry of Education and Training NATIONAL ECONOMICS UNIVERSITY THE THESIS IS COMPLETED IN: NATIONAL ECONOMICS UNIVERSITY
Supervisors: 1. PHAM VAN TUAN PROF. TRAN MINH DAO 2. dR. NGUYEN NGOC QUANG STUTYing CUSTOMER SATISFACTION’S IMPACT ON CUSTOMER LOYALTY AND SHIRT REPeating Purchase OF URBAN CUSTOMERS IN VIET NAM UNDER THE INFLUENCE OF INVOLVEMENT AND FASHION STYLE Reviewer 1: ASS. PROF. NGUYÔN THÞ TUYÕT MAI Reviewer2: ASS. PROF. NGUYÔN HOµNG LONG Reviewer3: MAJOR:: BUSINESS ADMINISTRATION (marketing) dR. §µO TïNG CODE: CODE: 62340 62340102 The thesis will be protected against National termination Council at National Economics University in Hanoi On…….h……date…….month……year 2014 SUMMARY OF DOCTORAL THESIS The thesis can be found at 1. National Library 2. National Economics University’s Library HA NOI, 2014 1 2 CHAPTER 1. INTRODUCTION many impact evaluation models of satisfaction on loyalty and repurchase 1.1. Introduction of thesis Satisfaction, loyalty and repurchase behavior are among the most basic concepts, but always appear to be complex and not well understood in marketing science. Particularly, these concepts will need deeper investigation and clarification in specific situations and study contexts. This thesis, in addition to directly reaching those concepts as mentioned above, mainly does not study the concepts singly, independently and discretely, but focuses on the impact of satisfaction on loyalty and repurchase behavior. This relationship is put in the context of Vietnam urban customers and one typical product of fashionable shirt. Moreover, this research of the relationship is continuously and deeply discovered under the impact of two adjustment variables (factors) of involvement and fashion style. That is the official way in which the thesis will be developed. 1.2. Rationale Customer loyalty appears to be concerned by many managers as well as makerters. Frederick J.Reichheld and W. Earl Sasser stated that the maintenance of high number of loyal customers would make more profit for the enterprise, including: (1) Profit from high price strategy, (2) Profit from new customer introduced through Word of Mouth; (3) Profit from marketing and advertising cost reducing; (4) Profit from increasing product using/purchasing number and frequency. That’s why in recent decades, there have been a lot of researchers investigating in both theory and practice to find effective methods to build and retain customer loyalty. To make customers satisfied is one of the most important missions that marketing managers have to do. However, that’s just the first step. The core thing that firms concern is the satisfaction level that makes customers become loyal and the loyalty level that makes customers have repurchase behavior. To discover more deeply the impact of satisfaction on loyalty and repurchase behavior, the author studies this impact under the moderation of involvement and fashion style. Despite extensive researches that have built behavior in the world, it’s totally new in Vietnam, particularly when it is measured under the moderation of involvement and fashion style. In addition, when it is done related to fashionable shirt, this is a totally new topic. With this purpose, the author chose the topic:” Investigating the impact of satisfaction on loyalty and repurchase behavior for shirt of Vietnamese urban customers under the impact of involvement and fashion style”. 1.2 Research objectives, questions and missions 1.2.1. Research objectives This thesis focuses on evaluating, testing the impact of satisfaction on loyalty and repurchase behavior for shirt of Vietnam urban customers, under the impact of involvement and fashion style. Based on the results, this thesis gives some recommendations to enterprises producing and dealing in fashionable shirts to increase satisfaction, retain loyalty towards repurchase behavior. 1.2.2. Research questions With these above-mentioned objectives, this thesis follows these questions: - How does satisfaction affect the expressions of loyalty and repurchase behavior? - Which expression of loyalty (commitment, trust, word of mouth) has the strongest impact on repurchase behavior? - Moderating variables: How do involvement and fashion styles affect the impact of satisfaction on loyalty and repurchase behavior? 1.2.3. Research missions - Overviewing domestic and international research results about impact of satisfaction on loyalty and repurchase behavior under the moderation of involvement and fashion style. - Building model test and hypotheses 1.3 Research object and scope - Research object The impact of Satisfaction on loyalty and repurchase behavior for shirt under the moderation of involvement and fashion style. 3 - Research scope Urban Customers in 2 cities: Hanoi (North) and Ho Chi Minh city (South) 1.4. Research methodology - Secondary data: applying those methods: (1) Collecting statistical and analytical data; (2) Analyzing and collecting theories; (3) Classifying and systematizing theories; (4) Modelling; (5) Scientific thinking in interpretation and inductive perspectives. - Qualitative data: Manual process (collecting ideas, counting the frequency of important words, noting important questions, etc.) - Quantitative data: Dealing by the method of EFA and CFA statistics and analysis. Using the method of liner structure analysis (SEM) to test the suitability of the model and hypotheses with SPSS, AMOS version 18.0 software. 1.5. New contributions - Build the researching model on the impact of satisfaction on loyalty and repurchase behavior under the moderation of involvement and fashion style, for shirt in Vietnam - Discover the impact of satisfaction on loyalty expressions (commitment, trust, word of mouth) and impact of those expressions on repurchase behavior - Study deeply two moderating variables: (1) Fashion style for market segmentation and (2) Involvement to consider its domination over the impact of satisfaction on loyalty and repurchase behavior - Especially, investigate involvement in two aspects: Ego involvement and Purchase involvement. - Which loyalty expression (commitment, trust, word of mouth) has the strongest impact on repurchase behavior? Enterprises will focus their marketing resources on that expression. - Recommend, add more basis on market segmentation and positioning for enterprises dealing in shirts stemming from the behavior characteristics of customer groups with different fashionable styles in the business context of Vietnam 4 - Recommend suitable strategy orientation and propose suitable marketing solutions for enterprises dealing in fashionable shirts to increase satisfaction, maintain loyalty towards repurchase behavior 1.6. Layout This thesis is divided into 5 chapters with the following contents: Introduction, thesis overview, research methodology, research result, conclusion and recommendation with 150 pages including 50 tables and 29 figures CHAPTER 2: STUDY OVERVIEW 2.1 Theoretical Background and Study Overview Satisfaction is a multi-dimensional concept and is recognized under different aspects, including: satisfaction with the product itself, satisfaction with the sales process and satisfaction with after sales service. Brown (1992), Jamal and Kamal (2002) [69], (Oliver, 1996 [102]; E.W.Anderson, Fornell &Lehmann, 1994 [48]; E.W.Anderson & Mittal 2000 [44]) all had a conclusion: Customer satisfaction is an emotional status in which customer demand, expectation and hope about product values/benefits are lower, higher or equal to customer expectation. Satisfaction is indicated to be the antecedent of loyalty and repurchase behavior for the company’s product. 2.1.2 Loyalty 2.1.2.1 Theory and model Satisfaction is a multi-dimensional concept with different and complex expressions. The concept of satisfaction has been defined differently by researchers (Gretty and Thompson, [1994], Dick and Basu (1990), etc.) Overall, according to them, brand loyalty is a deep commitment leading to stable repurchase behavior for n interested service/product in the future despite different situation factors, such as marketing activities.  Common models Common models relating to satisfaction includes: model of Gounaris and Stathakopoulos (2004) [56], model of Ki-Joon Back and Sara C.Parks [73] 5 6 which argued that customers had attitudinal loyalty first, then emotional loyalty, 2.2 Impact of satisfaction on loyalty and repurchase behavior 2.2.1. Impact of satisfaction on loyalty Ruben Chumpitaz Caceres and Nicholas G. Paparoidamis (European Marketing Magazine, No. 41, 2007) [116] testing the impact of satisfaction on commitment and loyalty stated that: Higher customer satisfaction level for a product/servicewould generate higher commitment and trust level for that product/service. Morgan and Hunt (1994) [94], Leon G.Schiffma, Leslie Lazar Kanuh & Joseph [135], Gour C.Saha and Theingi [57] shared the same opinion. Overall, in the context of shirt consumption, this thesis assumes that satisfaction positively affects WOM, commitment and trust intention and behavioral loyalty. A number of researchers (Raruyen and Miller, 2007 [127]; Dick & Basu, 1994 [40]) accessed customer loyalty with concepts of Behavioral loyalty, Attitudinal loyalty and composite loyalty. 2.1.2.2 Loyalty expressions Commitment Some ideas on commitment are given by a number of researchers (Morgan & Hunt, 1994; Miller (2007), Abdelmajid Amine (1998), Burmkrant and Unnava (2000)). Accordingly, commitment could be identified as a motivation to stay with a partner and willing to lose an amount of profit. The above-mentioned researchers all believed that commitment was an aspect of loyalty 2.2.2. Impact of loyalty on repurchase behavior This thesis discovers loyalty and repurchase behavior as two specific concepts with 5 reasons: Customer attitude, customer inertia, demand change, Trust Trust is considered to be one of the critical factors for a successful relationship between parties and an expression of loyalty. Chaudhuri and Holbrook (2001) 31], Rauyreuen (2007) [127] defined trust as the trust in a multibrand loyalty and situational society factor Based on studies of Beth Davis-Sramek, John T.Mentzer and Theodore P.Stank [24] of Lousville and Tennessee University, US (operational management magazine, No. 26-2008; Wetzels et al (1998) [133]; Norizan party’s word or promise and that the party will fulfill his or her obligation in an exchange relationship. Word of month Word of mouth (WOM) refers to passing information orally about consumer personal experiences with a product or service from individual to individual. WOM plays an important part in shaping consumers’ behaviors and attitude, and forming loyalty. Butcher et al. (2001) indicated four variations of customer advocacy concept: Providing positive word-of-mouth, recommending the product/service to others, encouraging others to use the project, defending the product provider’s virtues Kassim and Nor Asiah Abdullah [99], the author indicated that: loyalty expressions including commitment, trust, wom affect repurchase behavior. 2.1.3 Repurchase behavior Some researchers assert that repurchase behavior is an expression of loyalty (behavioral loyalty). However, in this thesis, the author will investigate repurchase behavior as a specific concept led from loyalty. 2.2.3. Impact of satisfaction on repurchase behavior In 1969, Howard & Sheth [67], Taylor and Baker (1994) [130]; Olsen (2002) [117] argued that satisfaction had an impact on repurchase behavior although these were not enough consistent evidences about this. 2.3. Fashion market and the impact of satisfaction on loyalty and repurchase behavior under the influence of fashion style 2.3.1. Fashion, fashion market in Vietnam In this part, the author describes the fashion market in Vietnam. Fashion style has strong influence on customer behavior so the author chose the shirt market to conduct research in Vietnam. 2.3.2. Impact of satisfaction on loyalty and repurchase behavior under the impact of fashion style Fashion style 7 8 Fashion style is used as a criterion on customer segmentation in many involvement and purchase involvement. At the same time, the author will use previous studies such as studies of Shim and Bickle (1994) [123]; Sofiah Abd Rahman; Abdul Rahman Abdul Rahim, 2006 [126], etc. The author has successfully applied the research work of Rafaeli, Dutton, Harquail and Mackie-Lewis- 1997. Accordingly, the thesis will study the impact of satisfaction on loyalty and repurchase behavior of 4 groups of customers: Customer group choosing their clothes based on their individual perspective on job title; Customer group whose emotion determines their decisions on clothes choosing; Customer group using clothes as a tool to build social relations and other customers group. Accordingly: The impact of satisfaction on loyalty and repurchase behavior will be different among customer groups with different fashion styles. involvement as a moderating variable in the impact of satisfaction on loyalty and repurchase behavior. From that, the study assumes: Ego involvement and purchase involvement affect the impact of satisfaction on loyalty (commitment, trust, word of mouth) and repurchase behavior. 2.5. Model and research hypotheses 2.5.1. Model From overall research, the author proposes the research model presented in the thesis as follows: H8a H8b 2.4. Impact of satisfaction on loyalty and repurchase behavior under the impact of involvement 2.4.1. Opinions on involvement The term “involvement” is translated into Vietnamese as the following concepts: “sự tham gia”, “sự lôi cuốn”, “sự can dự”, “sự dính líu”, etc. In this thesis, the author will use the expression “sự dính líu” to ensure the suitability with the research object of the thesis. Almost all of previous studies consider involvement as a multidimensional construct. Some first researchers in this field are Lastovicka and Gardner (1979) [84], Traylor and Joseph (1984) [132], (Rothschild, 1979) [115], Hupfer và Gardner, 1971 [68]; Manuel (2009) divided involvement into two types: ego involvement (long-term) and purchase involvement (situational). Accordingly, ego involvement is customer individual impact due to individual experience, familiarity and professional towards the purchase and consumption of products/services obtained from time to time. Purchase involvement is a situational involvement attached with environmental stimulus during the process of approving customer purchase decisions. There are many different attitudes on explaining the nature and structure of involvement, in this thesis, the author agrees with the structure of involvement proposed by researcher Manuel (2009), which means that involvement includes: ego Involvement (ego and purchase) and fashion style (4) H11a H11b H12a H12b H9a H9b Loyalty (2) H4 H1 Commitment (2a) H2 H5 Trust (2b) Repurchase behavior Satisfaction (1) H6 H3 Word of mouth (2c) H7 H14a H14b H10a H10b Involvement (ego and purchase) and fashion style (4) H13a H13b Figure 2.12. Research model (3) 9 2.5.2. Research hypotheses The research model has the variables considered under different impacts:  In studying the impact of satisfaction on loyalty: - Independent variable: satisfaction - Dependent variables: commitment, trust, word of mouth  In studying the impact of loyalty on repurchase behavior: - Independent variables: commitment, trust, word of mouth - Dependent variable: repurchase behavior  In studying the impact of satisfaction on repurchase behavior: - Independent variable: satisfaction - Dependent variable: repurchase behavior • Moderating variables 1. Ego involvement 2. Purchase involvement 3. Fashion style Hypothesis groups Hypothesis group studying the impact of satisfaction on loyalty and repurchase behavior from H1 to H7 Hypothesis group related to the moderating variable of involvement from H8 to H14 Hypothesis group related to the moderating variable of fashion style: H15 10 Interviewing marketing experts and fashion experts in Vietnam to preliminarily evaluate and adjust the scales used for research in the thesis. • Quantitative method Collecting primary data, analyzing primary data by SPSS software, analyzing EFA exploration factor. Continuing using AMOS 18.0 software to analyze CFA confirmation factor to test the model and test the research hypotheses. 3.1.2. Research process To ensure the implementation of research objectives, the thesis will be developed and deployed under the following research process: Theoretical background Drafting scale 1 In-depth interviews with experts, focus group Amendment, supplement Drafting scale 2 Quantitative preliminary research N = 200 Cronbach’s Alpha EFA Final scale Removing observation variables with small Alpha coefficient and relation coefficent Removing observation variables with small EFA. Checking extracted variance CHAPTER 3: RESEARCH METHODOLOGY 3.1. Overview of research methodology and process 3.1.1. Data collection and analysis • Secondary data collection and analysis Collecting, analyzing, comparing and evaluating some studies on the impact of satisfaction on loyalty and repurchase behavior under the impact of moderating variables of involvement and fashion style of customers from available domestic and international sources on the contents related to the research. • Qualitative method (in-depth interview) Final research N= 1000 CFA SEM Removing variables with small CFA; Checking the suitability of the model; Calculating the general reliability coefficient, Calculating extracted variance Checking the suitability of the model. Testing the hypotheses Figure 3.1: Research process 3.1.3. Research plan From the above mentioned research process, the research plan will be implemented following an order of time and represented in the following table: 11 12 Table 3.1: Research plan image. Besides, Vietnamese people always pay much attention to clothes at the Research Method Technique Preliminary Qualitative In-depth interview with experts Focus group 01/1115/11/2013 Quantitative (N=200) Direct interview 11-12/2013 Quantitative (N=1000) Direct interview Final Time Location Hanoi 01/2014 Hanoi Hanoi, Ho Chi Minh City According to the plan, quantitative research will be organized in the 2 biggest cities in Vietnam: Hanoi and Ho Chi Minh City. To ensure the fairness and objectiveness of the research, the author will conduct the research with the similar number of samples in two cities. The interviews are also organized in some areas with many urban citizens with the total of 29 urban districts of two cities. For each district, the author lists the areas suitable for the criteria on selecting research samples of the thesis. 3.2. Designing preliminary research and developing scale 3.2.1. Qualitative preliminary research Qualitative preliminary research is to supplement and amend observation variables used to measure the research concepts used in the theoretical model suitable for the context of Vietnamese market. For the scale of involvement, the results of interviewing experts add two more observation variables into the original scale. Specifically, they are SDL_4 (The brand of my shirt helps to show my job position) and SDL_5 (I choose shirt brands which can help me build and maintain my own image). When developing the scale on ego involvement, the researchers state that in Vietnamese market in particular, with the characteristic of formality respect, with clear division among different job positions, fashionable shirts used in the workplace can help their owners represent their job positions as well as their first meeting to create good impression for partners. The experts also agree with the decision of applying the research work of the postgraduates including professionals and students in Michigan and Virginia university in US ((Rafaeli, Dutton, Harquail. and Mackie-Lewis- 1997) [111]) on the behavior of consuming and choosing business attire. Accordingly, the thesis will study 3 customer groups with different fashion styles at Vietnamese market, as well as add 01 selection option as “other customers group” to ensure the tightness of the answers. In addition to interviewing the experts, the author also interviewing focus groups to adjust the language used in observation variables suitable and well understood for interviewees. 3.2.2. Quantitative preliminary research Quantitative preliminary research was implemented from November 20, 2013 to December 8, 2013 in Hanoi by the method of convenience sampling. (N=146) As mentioned in chapter 2, there are 6 concepts studied in this thesis including: (1) satisfaction or STM with the measurement criteria from STM_1 to STM_8, (2) commitment or CK with the measurement criteria from CK_1 to CK_14, (3) Trust or TT with the measurement criteria from TT_1 to TT_4, (4) Work of mouth or TM with the measurement criteria from TM_1 to TM_4, (5) Repurchase behavior or MLL with the measurement criteria from MLL_1 to MLL_7, (6) Involvement or SDL with the measurement criteria from SDL_1 to SDL_11. The following table summarizes the steps on developing and evaluating the scales of the final research model. When evaluating the scales, the author uses the standards of Hair and ctg. Accordingly, the scales are considered to be fully reliable when Cronbach Alpha coefficient reaches ≥0.6 and extracted variance is more than 50%, and the Factor Loading must be more than 0.3. 13 14 Table 3.2: Summarizing the results on preliminarily developing and evaluating the scales nd Scale Satisfaction (STM) Commitment (CK) Trust (TT) Word of mouth (TM) Ego involvement (SDL) Criteria 1st evaluation result Extracted variance Cronbach’s Alpha: Conclusion 27.745% 0.532 Recommending removing the following observation variables: STM_3; STM_4; STM_6 27.073% 0.664 Recommending removing the following observation variables: CK_3, CK_4, CK_5, CK_7, CK_9, CK_11, CK12 41.638% 0.483 Recommending removing the following observation variables: TT_1 57.021% 0.622 Put into official research 38.296 % 0.591 Recommending removing the following Extracted variance Cronbach’s Alpha: Conclusion Extracted variance Cronbach’s Alpha: Conclusion Extracted variance Cronbach’s Alpha: Conclusion Extracted variance Cronbach’s Alpha: Conclusion Scale Criteria 2 evaluation result (after removing some observation variables) 50.993% 0.609 Pass Purchase involvement (SDL) Extracted variance Cronbach’s Alpha: Conclusion 44.607% 0.786 Repurchase behavior (MLL) Extracted variance Cronbach’s Alpha: Conclusion Put into official research 52.127 % 0.529 Put into official research 45.949 0.605 Put into official research 1st evaluation result observation variables: SDL_ 5 35.866 0.601 Recommending removing the following observation variables: SDL_ 9 0.772 42.638% Put into official research 2nd evaluation result (after removing some observation variables) 48.391 0.629 Put into official research 3.3. Official research After implementing quantitative preliminary research, building and evaluating the scales, the author will implement official quantitative research with full sacles built at the previous step. 3.3.1. Sampling method Research object: all Vietnamese consumers over 18 years old; living and working in Hanoi and Ho Chi Minh City; There are many different attitudes on identifying sample size. However, the thesis uses the principle of Hair and ctg [6] which asserts that each measurement variable needs at least 5 observations. The thesis uses the method of factor analysis together with linear regression, the research model has 31 measurement variables. Therefore, the minimum sample size is 155 elements. The identification of sample size depends on research scope (Hanoi with 10 districts and Ho Chi Minh City with 19 districts); Population rate over 18 years old in the whole research; Population rate of district to be surveyed as compared to those of 29 districts to be studied. Therefore, the sample size is identified to include 1000 elements, the 15 16 sample scope for research is 7 times bigger than the required sample size as 1990)[23]; RMSEA ≤ 0.08. The standards evaluate the suitability of the model mentioned above. The author hopes that this will be an optimal sample scope to obtain a relatively accurate description. based on the aspects of content value including: The credibility of the scales; Being unidirectional/monad; Converge value; Difference value. 3.3.2 Describing research sample 4.1.2. CFA testing results Table 3.26: Describing research sample (N=615) Characteristics Male Sex Female South Region North <=23 24-30 Age 31- 45 > 45 High school Post graduate Qualification Under graduate Others Operational administrational field Student Current main job Enterprise Self-employment Others Employee Head manager Job position Department director Others and Frequency/Number 296 319 221 394 192 210 183 30 151 337 70 57 Rate (%) 48,1 51.9 36 64 31.2 34.1 29.8 4.9 24.6 54.8 11.4 9.3 74 12.0 248 206 56 31 366 143 39 67 40.3 33.5 9.1 5.0 59.5 23.3 6.3 10.9 GFI TLI Chi-square/df 4.523 0.931 0.944 0.95 0.076 Repurchase behavior 3.988 0.901 0.923 0.935 0.070 Satisfaction 0.91 0.928 0.94 0.071 Loyalty 4.052 CFI RMSEA The indicators at the above mentioned table all show that the scales are suitable with market data. However, the observation variables MLL_4, MLL_5, MLL_6, MLL_7 are all negative and lower than 0.5. Therefore, those observation variables will be removed from the research model to ensure the coverage value of the scales as well as of the research model. The following table 4.1 will summarize the results on retesting the general reliability coefficient and extracted variance of each scale after removing those observation variables of MLL scale. Table 4.1: Summarizing reliability and total extracted variance 1 Commitment (CK) General reliability Number of Total coefficient observation extracted Conclusion of variables variance Cronbach Alpha) 7 0.897 62.52% 2 Trust (TT) 3 0.904 84% 3 Word of mouth (TM) 4 0.852 69.37% 4 Satisfaction (STM) 5 0.805 56.42% 3 5 Repurchase behavior (MLL) 0.81 72.51% 6 Ego involvement (EGO) 4 0.832 66.71% 5 7 Purchase involvement (PUR) 0.802 56.63% No. CHAPTER 4: RESEARCH RESULTS 4.1 Testing the scales by the method of confirming factor analysis (CFA) 4.1.1. Testing standards following CFA method The standards used to check the suitability of the model: CMIN/df <5 (Kettinger and Lee 1995[81]); GFI, TLI, CFI ≥ 0.9 (Bentler & Bonett, Testing criteria Scale Scale All scales are reliable. 17 18 In the table: Estimate: average estimated value; SE: standard error, CR: 4.2. Testing research model and hypotheses 4.2.1. Testing research model critical value; P: value level, ***: p < 0.001. 4.2.2. Testing hypotheses 4.2.2.1. Testing H1 to H7 With P ≤ 0.05, the impact weighting has statistical meaning. In fact, the estimation result shown in Table 4.3 represents that **** or P<0.0001 and positive weighting. Satisfaction has positive effect to engage, trust, word of mouth and repurchase behavior at the same time engage, trust, word of mouth has effect on repurchase behavior. The model test results in table 4.3 show that: (1) satisfaction has positive effect on engagement, trust, word of mouth of customer with brand 1.196; 1.043; 1.309 respectively. (2) Engagement, trust, word of mouth have effect on repurchase behavior 0.240; 0.170; 0.36 respectively; (3) satisfaction has effect Figure 4.5. Analysis results (SEM) of research model (without the impact of moderating variables) The research model has the following coefficients Chi-square/df= 3.746; GFI= 0.900; TLI= 0.923; CFI= 0.934; RMSEA= 0.067, showing that the model on repurchase behavior of the brand: 0.33. As a result, hypotheses from H1 to H7 in the situation of ready-made shirt are accepted. 4.1.1.1 Testing H8 to H14 H8 to H14 defines effect of ego involvement and involvement from buying is suitable for market data. The result on model analysis will be represented in details in the following table 4.3: process on satisfaction, loyalty and repurchase behavior. This research use regression techniques of SPSS 18.0 Table 4.3: The result of model testing without the impact of moderating variables Estimation coefficient S.E. C.R. Table 4.4: Regression result to test H8a, H9a, H10a Commitment (CK) Trust (TT) Regression weighting P Model Unstan Differen dardize t from d standard Word of mouth (TM) Regression weighting Standardiz P value Unstandardi ed zed Different from standard Standardiz ed Regression weighting P value Differe P Unstanda nt from Standardiz value rdized standar ed d CK <--- STM 1.196 .081 14.706 *** Constant 1.048 .127 .000 2.625 .149 .000 .243 .144 TT <--- STM 1.043 .077 13.490 *** Satisfacti .565 on .027 .627 .000 .310 .041 .330 .000 .614 .031 .603 .000 TM <--- STM 1.390 .094 14.843 *** .006 .166 .000 .023 .010 .100 .023 .048 .007 .201 .000 MLL <--- TT .170 .039 20.853 *** STM*E GO MLL <--- TM .360 .037 16.984 *** MLL <--- CK .240 .039 19.383 *** MLL <--- STM .330 .038 17.572 *** .035 VIF 1.081 1.388 1.081 R2 0.476 0.151 0.469 .001 Dependent variables: Commitment/Trust/Word of mouth (respectively) The table above is summarized from Appendix 4.19 – 4.21 in Appendix 4 The relationship between satisfaction and ego involvement has statistical meaning (P < 0,05). This result shows that H1 is accepted, Ho is rejected or 19 20 H8a, H9a, H10a are accepted: ego involvement enhances effect of satisfaction on engagement, trust and word of mouth. Continued hypotheses are tested in similar way. The result, the thesis has some conclusions: involvement from buying process make the effect of satisfaction on engagement, trust and word of mouth: individual engagement enhance the effect of engagement, trust and word of mouth on repurchase behavior; involvement of buying process make effect of engagement, trust and word of mouth on repurchase behavior; individual engagement enhance the effect of satisfaction on repurchase behavior; involvement from buying process enhance the effect of satisfaction on repurchase behavior. That the result, all the hypotheses from H8 to H14 are accepted. 4.2.2.4 Testing H15 Table 4.12 the number of respondents who choose difference fashion style Description N People who choose shirt based on perception of job Group 1 209 position People who affected by emotional element when Group 2 255 choosing shirt People who take shirt as a means of building social Group 3 49 relationship Group 4 Others 102 The research has analyzed multi group to test level of effect of satisfaction on loyalty and repurchase behavior in 4 group of people who have different style. The result in following table. Table 4.13 Differences of variable and invariable model Chi-square df 4.2.2.3. Analysis Analysis in terms of sexes Table 4.10 shows that chosen model has differences between male and female in assessing the effect of satisfaction on loyalty and repurchase behavior with ready made Shirt. Variable model 331.24 156 Invariable model 375.80 169 Difference 44.56 13 Chidist(44.56,13)= 0.000 Table 4.10 Differences between criteria in the model Compared model χ2 df p Variable model 195.222 83 0.000 0.864 0.956 0.967 0.047 repurchase behavior among groups of people who have different fashion style” Invariable model 194.143 80 0.000 0.946 0.953 0.966 0.9048 Difference 1.079 3 0.00 4.2.3. Testing the difference between average value of latent variable in terms of group of customers who have different fashion style The result shows that: there is different in sample average between groups of people who have different fashion styles GFI TLI CFI RMSEA 0.082 0.003 0.001 0.001 χ2: chi-square, df: number of freedom levels, p: meaning level Analysis multi group structure in terms of living places Similarly, the result shows the difference between suitable criteria between variable model and invariable model is statistical meaning (p < 0.05). For this reason, variable model is chosen coming to conclusion that there is difference between customer living in northern cities and southern cities of Vietnam in assessing the effect of satisfaction on loyalty and repurchase behavior when buying ready made shirt. As the result shows, H5 is accepted: “Impact of satisfaction on loyalty and 4.2.4. Testing the different of average value of latent variables in the model in terms of sexes and living place of customer in the North and the South of Vietnam Group 1. People who choose fashion based on individual perception of job position  In terms of sexes Differences between male and female choosing clothes based on job position happen with some latent variable, however, it is relatively small due to 21 the maximum average value is d = 0.438  In terms of geographical area where customers are living (the North and the South of Vietnam) The difference in terms of geographical area of customers in group 1 who choose clothes based on perception on job position appear in all latent variables which is relatively clear because the minimum average value is d = 0.449, maximum is d = 0.819 Group 2: People who affected by emotional element when choosing shirt In terms of sexes The testing result on average value of two independent samples (Independent-sample T-test) (descriptive statistical analysis) is two customer categories of male and female [appendix 4.38] of the 2nd customer group showing that there is not any difference on value among the scales of Commitment (CK), Trust (TT), Word of Mouth (TM), Satisfaction (STM) and Repurchase behavior (MLL). In terms of geographical area where customers are living (the North and the South of Vietnam) The testing result on average value of two independent samples (Independent-sample T-test) (descriptive statistical analysis) is two customer categories in Hanoi and Ho Chi Minh City [appendix 4.39] of the 2nd customer group showing that there is difference on value among the four scales of Commitment (CK), Trust (TT), Word of Mouth (TM), Satisfaction (STM) with average difference of d=0.746, d=0.815, d=0.694, d=0.657 respectively. CHAPTER 5: CONCLUSION, RECOMMENDATION AND LIMITATION OF THE STUDY 5.1 Conclusion Based on the result of the study, we have the main findings: (1) Satisfaction have the positive impact on the expressions of loyalty; (2) Expressions of loyalty have the positive impact on repurchasing; (3) Satisfaction have positive impact on repurchasing; (5) Impact of the relationship of satisfaction, loyalty and repurchase behavior different among different fashion lifestyle groups. 22 5.2 Recommendation 5.2.1 Strategy: Result of the study give us some recommendation about strategy for garment Company in Vietnam: - Loyalty and repurchase behavior are asset of company - Segmentation, positioning based on fashion lifestyle. Specially focus on the segment of “people choose clothes based on job position”. The result also suggest company build up the marketing strategy for the urban residence should consider the geographic area (the north and the south) - Result of the study also suggests to marketing manager should pay attention on the level impact to the behavior that company expected. Marketing activity should focus the most on worth of mouth buzz, the secondly in commitment of customer and the trust shows the less impact than others. The involvement shows the quite important role in customer relationship. - Segmentation and marketing strategy for each fashion lifestyle segments. Result of the study shows that if company want to serve fashion lifestyle segments, they should have different marketing plan for each of them. For example: + Group 1: Choose based on job position + Group 2: Choose based on emotional + Group 3: Choose based on social status + Group 4: others 5.2.2 Solutions First of all, to ensure the satisfaction, loyalty of customer with brand, company should do the marketing research periodic to know the situation of company, competitors, catch new trend and also new demand of customer. Secondly, to improve the trust of customer, company have to improve the quality and service. Worth of mouth is the strongest impact on customer behavior and create the loyalty and repurchase behavior. Company should: - Worth of mouth through direct sales - Communication through mass media: Advertising, PR, event, fair… - Interactive media: Direct marketing such as Life and Fashion on VTV3… - Controlling the buzz and oriented public information. 23 1/ Carefully in pressing public information. Create social event such as washing event, advice customer in washing, using clothes… 2/ Differentiate position to creating the positive worth of mouth. Be sensitive on new trend in Vietnam, especial on group choose clothes based on job position and emotional. 3/ Focus on staff training about product, carefully in recruitments especial for sales force. If customer does not satisfy with the sales person, company will face with the risk of losing customer and further, they will spread the negative worth of mouth to others. 4/ Build up the system to control and evaluate the buzz. For example: regularly check out the information and immediately have the solution for it. 5/ Build up system based on team work, educate the knowledge and improve the skills. Set up the CRM tools for control information better. Attention in handling the complaint. Company should find the solution to improve the involvement by improving the stick between brand and the motivation of customer. To do this, company should: 1/ Build the brand advantage by create point of different, unique positioning to reduce the impact of competitive and consolidation the involvement of customer. 2/ Secure the brand: Register the brand identity in law and protect the brand image of customer. 5.3 Limitation and suggestion for next study: First of all, this thesis just choose sample in 2 main cities: Hanoi (the north) and Ho Chi Minh city (the south) so the representative of result is not so high. Secondly, this studies just using 3 expressions of loyalty (commitment, trust, word of mouth). If we have more resource to give more expressions, the result could be more reliable and have more interesting findings. Thirdly, artist just considers “word of mouth” is positive and the satisfaction is the general. The negative word of mouth and the spread satisfaction have been not shown on this study. Fourthly, I just choose 2 moderator variables: fashion lifestyle and involvement. It will be more useful if the artist survey more impact of others variables. This is the suggestion for next study. 24 CONCLUSION Result of the thesis: • Overview the previous studies, systematized the basic theoretical about the main concepts of marketing: satisfaction, loyalty, repurchase behavior and the impact between them, detect gaps need further study in order to build the foundation for the research model in this thesis. • The thesis uses structural equation balance (SEM) to test the research model, check the suitable of the model to the market, evaluate the credibility and value of differentiated, value convergence of scales, testing the hypothesis. • Describe the level of impact (strong, weak) from satisfaction to expressions of loyalty and repurchase behavior of customers. • Analyze and evaluate the specific impact of the moderator, the fashion lifestyle and involvement in relation to each pair. • Provide some hints about the strategic direction and marketing solutions for business firms shirt fashion apparel available to enhance loyalty and motivate them towards repurchase behavior. New contributions on the theoretical level • The thesis explored in depth the impact of satisfaction to loyalty and repurchase behavior. While previous research usually only refers to the influence of factors unique to each site or just stop talking on the couple relationship. • An in-depth study of the impact satisfies the specific expression of loyalty and the expression of loyalty to repurchase behavior. • Research 02 moderator variables: involvement (ego involvement and purchase involvement) and fashion lifestyle. New contributions on the practical level • Proposal for the fashion apparel business marketing resource allocation to enhance loyalty and promote repurchase behavior. • The thesis also recommends, additional on the criteria of segmentation and market positioning for the business enterprise clothes comes from the study of behavioral characteristics of the customer group row with different fashion lifestyles. • The thesis also provides recommendations and strategic directions proposed marketing solutions suitable for businesses clothes fashion towards enhancing satisfaction, loyalty and promote repurchase behavior of customers. THE WORKS OF THE AUTHOR HAVE BEEN PUBLISHED RELATED TO THE THESIS 1. Pham Van Tuan &Nguyen Dinh Toan (2014), “Mối quan hệ giữa sự thoả mãn, lòng trung thành và hành vi mua lặp lại đối với sản phẩm áo sơ mi may sẵn”, Economics and Development Journal, No. 2000, February, National Economics University Publisher. 2. Pham Van Tuan và Tran Minh Đao (2014), “Mối quan hệ giữa sự thoả mãn, lòng trung thành và hành vi mua lặp lại dưới sự ảnh hưởng của phong cách thời trang tại Việt Nam”, International Scientific Conference: Những vấn đề về quản lý Kinh tế và Quản trị kinh doanh hiện đại, No.647-2014 CXB/03-TK, Statistics Publisher. 3. Pham Van Tuan (2014), “Lý thuyết lòng trung thành và thực tiễn mối quan hệ giữa các biểu hiện của lòng trung thành (cam kết, tin tưởng và truyền miệng) với sự thoả mãn và hành vi mua lặp lại”, Economics Managemet Journal, June, 2014,Institute of Economic Management Central - Ministry of Planning and Investment Publisher. 4. Pham Van Tuan (2014), “Sự dính líu từ quá trình mua của khách hàng”, Industry and Trade Journal, No.11, June, 2014, Ministry of Industry and Trade. 5. Pham Van Tuan, Nguyen Ngoc Quang & Vu Xuan Truong (2014), “Nghiên cứu ảnh hưởng của dính líu cá nhân trong tác động của sự thỏa mãn đến lòng trung thành và hành vi mua lặp lại: Ứng dụng trong xây dựng chiến lược marketing cho các doanh nghiệp nhỏ và siêu nhỏ tại Việt Nam”, National scientific conference:Relationship marketing of micro and small enterprises in Vietnam, National Economics University Publisher. 6. Pham Van Tuan and Pham Thi Giang (2014), “Relationship between satisfaction, loyalty and repurchase behavior of consumer under the impact of fashion style for shirt product market in Vietnam”, The 4rd Khon Kaen University National and International Conference, June 11 – 13, 2014, Khon Kaen University, Thai Lan.