VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY TRINH NGOC HONG ANH FACTORS AFFECTING CUSTOMER PURCHASE INTENTION OF CLOTH BAGS FOR SHOPPING MASTER’S THESIS MASTER OF BUSINESS ADMINISTRATI[.]
VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY TRINH NGOC HONG ANH FACTORS AFFECTING CUSTOMER PURCHASE INTENTION OF CLOTH BAGS FOR SHOPPING MASTER’S THESIS MASTER OF BUSINESS ADMINISTRATION Hanoi, 2019 H i VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY TRINH NGOC HONG ANH FACTORS AFFECTING CUSTOMER PURCHASE INTENTION OF CLOTH BAGS FOR SHOPPING MAJOR: BUSINESS ADMINISTRATION CODE: 60340102 RESEARCH SUPERVISORS: ASSOC.PROF.DR VU ANH DUNG PROF.DR TOHRU INOUE Hanoi, 2019 ACKNOWLEDGMENT I would like to express my gratitude to all those who gave me the possibility to complete this thesis I have received extensive help, suggestions and encouragements at each step along the way, and I am very grateful to my beloved and dedicated professors, family and friends for their time and support First and foremost, I would like to thank my two supervisors, who are not only my guide throughout the completion of this thesis but also my lecturers, Associate Professor Vu Anh Dung and Professor Tohru Inoue, for their wholehearted guidance and help as well as invaluable feedbacks, which enable me to complete this paper Furthermore, during this work I have received remarkable help and encouragement from my beloved professor in Business Administration Department – Professor Tran Thi Bich Hanh for her thoughtful support and advice during my time of generating ideas and finalizing the topic Besides, I would like to say thank-you to all of the participants who helped me answer my questionnaire when I did the survey, without you there will not be the results for this research Last but not least, I am greatly indebted to Vietnam Japan University and Yokohama National University for creating a wonderful study environment for me to develop myself and study which encourage me to finish my thesis Student, Trinh Ngoc Hong Anh i ABSTRACT Plastic shopping bags are a daily use product of everyone, which is not only harmful to environment but also to people’s health Environmental terms such as global warming, green house effects are becoming more and more popular in the media channel People seem to be more aware of the negative effects of using plastic shopping bags, which leads to a need for an alternative Cloth bags are a typical option for changing in everyday life of customers, which is recyclable and is made with more eco-friendly materials compared to plastic bags This study explores the factors affecting consumer purchase intention of cloth bags for shopping by using theory of planned behavior The constructs of theory of planned behavior: attitude, subjective norms, perceived availability and perceived consumer effectiveness are tested to check if they have a positive significant relationship with consumer purchase intention of cloth bags for shopping or not Moreover, since media exposure seems to make a certain influence on the relationships between constructs of theory of planned behavior and purchase intention, it is used as a proposed moderator The study uses convenient sampling and distributes questionnaire adopted from previous researches to 120 respondents both offline and online 113 results sent back to analyzed by SPSS 20 The results found that attitude and subjective norm have a significant positive impact on purchase intention while perceived availability and perceived consumer effectiveness not Media exposure to environmental message also does not show a moderating effect to the relations ii TABLE OF CONTENTS ACKNOWLEDGMENT i ABSTRACT ii LIST OF TABLES vi LIST OF FIGURES viii CHAPTER 1: INTRODUCTION .1 1.1 Background and necessity of the research 1.1.1 Why is environment degrading gradually? 1.1.2 Plastic bags versus cloth bags .3 1.1.3 Green marketing and customer purchase intention 1.2 Statement of the problems 1.3 Research objectives 1.4 Research questions 10 1.5 Research scope 10 1.6 Structure of the research 10 CHAPTER 2: LITERATURE REVIEW 12 2.1 Green consumptions and related definitions 12 2.1.1 Sustainable consumption .12 2.1.2 Green consumption 13 2.1.3 Green products and related researches 16 iii 2.1.4 Eco-friendly shopping bags and cloth bags 17 2.2 Consumer behavior and purchase intention .22 2.3 Theory of Planned behavior .25 2.3.1 Attitudes towards behavior 26 2.3.2 Perceived behavioral control .27 2.3.3 Subjective norms 28 2.4 Media exposure to environmental messages: 28 2.5 Final proposed research model 30 CHAPTER 3: RESEARCH METHODOLOGY 31 3.1 Research process 31 3.2 Sampling and data collection .32 3.3 Designing questionnaire .33 3.4 Analyzing data plan 38 CHAPTER DATA PRESENTATIONS AND FINDINGS 47 4.1 Data description 47 4.2 Reliability analysis .51 4.2.1 Reliability of ATTITUDE-ATT scale 51 4.2.2 Reliability of SUBJECTIVE NORM – SN scale 52 4.2.3 Reliability of PERCEIVED AVAILABILITY - PA scale 53 4.2.4 Reliability of PERCEIVED CUSTOMER EFFECTIVENESS scale 54 4.2.5 Reliability of PURCHASE INTENTION scale 57 iv 4.2.6 Reliability of MEDIA EXPOSURE TO ENVIORNMENTAL MESSAGE scale .58 4.3 Exploratory factor analysis 59 4.3.1 Exploratory factor analysis of TPB scale 59 4.3.2 Exploratory factor analysis of PURCHASE INTENTION SCALE 62 4.3.3 Exploratory factor analysis of MEDIA EXPOSURE TO ENVIORNMENTAL MESSAGE Scale .64 4.4 Regression analysis 66 4.5 Moderation analysis via PROCESS model SPSS 72 4.5.1 Moderation analysis between MEDIA EXPOSURE to the relationship of ATTITUDE and PURCHASE INTENTION 73 4.5.2 Moderation analysis between MEDIA EXPOSURE to the relationship of SUBJECTIVE NORM and PURCHASE INTENTION .75 4.6 Hypothesis tested results .78 CHAPTER 5: RESULT DISCUSSION 80 5.1 Result discussion and implications 80 5.2 Contributions of the research .85 5.3 Limitations and future research direction 86 REFERENCES 88 APPENDIX QUESTIONNAIRE 94 v LIST OF TABLES Table 3.1 Measuring items for survey .35 Table 3.2 Likert scale of Agreement extent .37 Table 3.3 Likert scale of Frequency extent 38 Table 3.4 Encoded terms for data testing 38 Table 4.1 Descriptive Statistics of 113 participant 47 Table 4.2 Gender distributions among 113 participants 48 Table 4.3 Income distributions among 113 participants 50 Table 4.4 Reliability statistics of ATT scale 51 Table 4.5 Reliability statistics of SN scale 52 Table 4.6 Reliability statistics of PA scale 53 Table 4.7 Reliability statistics of PCE scale – 1st test .54 Table 4.8 Reliability statistics of PCE scale – 2nd test .55 Table 4.9 Reliability statistics of PCE scale – final test 56 Table 4.10 Reliability statistics of PI scale 57 Table 4.11 Reliability statistics of ME scale 58 Table 4.12 Exploratory factor analysis for TPB scale .59 Table 4.13 Exploratory factor analysis for PI scale 62 Table 4.14 Exploratory factor analysis for ME scale .64 Table 4.15 Pearson correlation statistic 66 Table 4.16: Collinearity Statistics .68 vi Table 4.17: Regression analysis summary 69 Table 4.18 Regression analysis summary 71 Table 4.19 Hypotheses tested results .78 vii LIST OF FIGURES Figure 2.1 Five stages of consumer buying process 23 Figure 2.2 Conceptual framework 25 Figure 2.3 Theory of reasoned action and theory of planned behavior 26 Figure 2.4 Proposed theoretical model 30 Figure 3.1 Research process proposed by the author .31 Figure 3.2 Conceptual Model of simple moderatio 44 Figure 3.3 Statistical Model of simple moderation 45 Figure 4.1 Age distribution among 113 participants 49 Figure 4.2 Education level distribution among 113 participants .50 Figure 4.3 Regression Standardized Residual 68 Figure 4.4 Proposed conceptual moderating effect of ME 73 Figure 4.5 Statistical diagram with ME moderates the relation .74 Figure 4.6 Moderation analysis statistic for ATT*ME 75 Figure 4.7 Proposed conceptual moderating effect of ME 76 Figure 4.8 Statistical diagram with ME moderates the relation .76 Figure 4.9 Moderation analysis statistic for SN*ME .77 viii products sold by socially and environmentally responsible companies Perceived availability I am familiar with the availability of cloth bags for shopping in my locality I can get cloth bags for shopping whenever I need them I have complete control over the number of cloth bags for shopping that I need to buy for personal use Subjective Norm My friends expect me to engage in cloth bags usage for shopping behavior My family expects me to engage in cloth bags usage for shopping behavior My society expects me to engage in cloth bags usage for shopping behavior People can rely on me to make a positive contribution to the society due to my cloth bags usage for shopping behavior Purchase intention I would like to use cloth bags for shopping I would buy cloth bags for shopping if I happen to see them in a store I would actively seek out cloth bags for shopping in a store in order to purchase it 36 I would patronize and recommend the use of cloth bags for shopping Media exposure to environmental messages During the past 30 days when you watch Reynaldo A TV, how often you see environmental Bautisca, Jr et messages? al (2017) During the past 30 days, how often you see environmental messages in the newspapers or magazines? During the past 30 days, when you access to the internet, how often you see environmental messages? Details about likert scale for agreement extent and frequency extent are presented as the tables below: Table 3.2 Likert scale of Agreement extent Strongly disagree Strongly agree Strongly Disagree Neither agree Agree Strongly agree disagree nor disagree 37 Table 3.3 Likert scale of Frequency extent Weakest frequency Strongest frequency Never Rarely Occasionally Frequently Very frequently PILOT TEST The questionnaire was sent to 10 people (who are highly educated people with education level from university or above and have high awareness of environmental problems) to give advice on the content and vocabulary chosen for each statement Generally, all pilot respondents were able to answer the questions, any words that made them difficult to understand was sent back and modified afterwards before sending the questionnaires to the main sample of this research 3.4 Analyzing data plan Firstly, the data will be input and then screened to find out the invalid ones, which will be rejected After that, the data will be encoded as the following tables: Table 3.4 Encoded terms for data testing No Label Explanations ATTITUDE I believe that the use of cloth bags for shopping by me will help in ATT1 reducing pollution and also help in improving the environment 38 I believe that use of cloth bags for shopping by me will help in ATT2 reducing wasteful use of natural resources I believe that use of cloth bags for shopping by me will help in ATT3 conserving natural resources I feel good about myself when I ATT4 use cloth bags for shopping PERCEIVED CONSUMER EFFECTIVENESS It is worthless for the individual PCE1 consumer to anything about pollution When I buy cloth bags for shopping, I try to understand how PCE2 its use will affect the environment and other consumers Since one person cannot have any effect upon pollution and natural PCE3 resources problems, it doesn’t make any difference what I Each consumer’s behavior can have a positive effect on society by PCE4 buying products sold by socially and environmentally responsible companies 39 PERCEIVED AVAILABILITY I am familiar with the availability of cloth bags for shopping in my PA1 locality 10 I can get cloth bags for shopping PA2 whenever I need them I have complete control over the 11 PA3 number of cloth bags for shopping that I need to buy for personal use SUBJECTIVE NORM My friends expect me to engage in 12 cloth bags usage for shopping SN1 behavior My family expects me to engage in 13 cloth bags usage for shopping SN2 behavior My society expects me to engage 14 in cloth bags usage for shopping SN3 behavior People can rely on me to make a 15 positive contribution to the society SN4 due to my cloth bags usage for shopping behavior PERCHASE INTENTION 40 16 I would like to use cloth bags for PI1 shopping I would buy cloth bags for 17 shopping if I happen to see them in PI2 a store I would actively seek out cloth 18 bags for shopping in a store in PI3 order to purchase it 19 I would patronize and recommend PI4 the use of cloth bags for shopping MEDIA EXPOSURE TO ENVIRONMENTAL MESSAGES During the past 30 days when you 20 watch TV, how often you see ME1 environmental messages? During the past 30 days, how often 21 you see environmental ME2 messages in the newspapers or magazines? During the past 30 days, when you 22 access to the internet, how often ME3 you see environmental messages? The data will be cleaned and process by software SPSS 20 and Process Macro addon developed by Hayes for testing moderation effect The processing data method will be presented as below: 41 a Reliability analysis by Cronbach’s alpha Cronbach’s alpha measures the internal consistency (reliability) of a test or a scale When a study has multiple Likert question in questionnaire, it is necessary to determine if the scale is reliable In this case, it is common to use Cronbach’s alpha The value of alpha (α) ranges between [0,1] Theoretically, the higher the value, the better but also not always the case If cronbach’s Alpha coefficient is too large (about 0.95 or more), it shows that there are many variables in the scale that not differ at all, and this phenomenon is called duplication or redundancy Based on guidelines by George & Mallery (2010), alpha is more than 0.9 can be considered the best and lower than 0.5 is poor In order to calculate Cronbach’s alpha for a scale, the scale must have at least three measuring items Nunnally & Bernstein (1994) mentioned that item-total correlation also plays a vital role to perform the correlation of one variable with others in the same scale The low level of smaller than 0.3 for item-total correlations will be removed while more than 0.3 will be accepted b Exploratory factor analysis Exploratory factor analysis (EFA) is a quantitative analysis method used to reduce a set of many variables k into a smaller variable set F (F < k) to make them meaningful but still contain the content of the original set (Hair et al 2013) There are some criteria to consider when doing exploratory factor analysis: To use EFA, it is necessary to use a large sample size, but the problem of determining the appropriate sample size is complicated Researchers often rely on previous researches Here are a few ideas and suggestions from factor analysis experts , we can consider: The number of observations (sample size) must be at least to times the number of variables in factor analysis 42 Hair et al (2013) suggested that to use EFA, the minimum sample size should be 50, preferably 100 Hair suggested trying to maximize the observation rate on each measurement variable of 5: 1, which means that for every measurement, there is a minimum of observations Factor loading > 0.5 Kaiser-Meyer-Olkin to test the sampling adequacy of factor analysis The value of KMO must reach 0.5 or more (0.5 ≤ KMO ≤ 1) which is a sufficient condition for the factor analysis to be appropriate If this value is less than 0.5, then factor analysis is unlikely to be suitable for the study data set Bartlett’s test of sphericity: compares the correlation matrix to the identity matrix, which means it if there is a redundancy between variables that can be summarized with some factors The indicator is sig Bartlett’s Test < 0.05 c Regression analysis Regression analysis is a modeling technique for analyzing the relationship between a real-valued dependent variable Y and one or more independent variables X1, X2, X3,…., Xk (Ragsdale, 2007) In order to regression analysis, it is necessary to find out a regression function that represents the relationship between the independent and dependent variables, from that the influential impacts can be observed The regression function will be as the following in this study: Y = b0 + b1 X1 + b2 X2,……+ bk Xk + ε Where: - Y = Scores on the Customer Purchase Intention of Cloth bag for shopping - X1, X2,……, Xk = Scores on the constructs of TPB (Attitude; Subjective Norms; Perceived Availability; Perceived Consumer Effectiveness) - b1, b2 ……, bk = Regression co-efficient of independent variables - b0 = an intercept 43 - ε = an error term First step when doing the analysis is to test assumption that the relations between variables is linear Because there are multiple independent variables, multicollinearity checking is needed, which shown by VIF number (Variance inflation factor ) If the model has VIF < 2, it is accepted while >2, the variable which has multi-collinearity need removing from the model Other assumptions need checking are the error ε, mean value and constant variances for a model to be acceptable Goodness-of-fit is also important to check for a model to have a high fit for analyzing d Moderation Analysis According to Hayes (2013), the moderation effect can be illustrated in the following figure: M X Y Figure 3.2 Conceptual Model of simple moderation (Hayes, 2013) In which: X is the independent variable Y is the dependent variable M is the moderator The effect of the independent variable X on the dependent variable Y is regulated by variable M, which means variable M affects the direction or intensity of the relation between predictor X and dependent variable Y In this case, the variable M is called the moderator variable of the term X, which is affected by Y Note that M is not a predictor for Y in moderation model, but an interaction 44 According to Dardas and Ahmad (2015), a model with moderating variable can lead to the following results: (1) Increase the impact of predictor on the outcome (2) Decrease the impact of predictor on the outcome (3) Reverse the impact of predictor on the outcome Through some algorithms, the above moderator variable model can be reinterpreted as follows: X b1 M b2 Y b3 X*M Figure 3.3 Statistical Model of simple moderation (Hayes, 2013) The figure can be explained by the following equation: Y = i + b1X + b2M + b3XM + eY In which: i is the regression intercept eY is the error when estimating Y b1,b2, and b3 represents the coefficient of the effect from X to Y, M to Y, and XM to Y respectively In order to assess the moderation effect, PROCESS MACRO v3.3 developed by Hayes will be used to test regression With the equation, when running multivariate 45 regression, the dependent variable is Y, the three independent variables are X, M, XM, then sig of XM variable is brought into consideration If sig 0, and X> 0, then when the variable M increases, the relation between variables X and Y increases However, multiplying or turning X * M leads to some disadvantages, because it can be multi-collinear between variables, so for the test to be valid, mean centering technique is used in SPSS to help 46 CHAPTER DATA PRESENTATIONS AND FINDINGS 4.1 Data description Table 4.1 Descriptive Statistics of 113 participant N Minimum Maximum Mean Std Deviation ATT1 113 4.39 891 ATT2 113 4.14 1.025 ATT3 113 4.10 935 ATT4 113 4.03 1.004 SN1 113 3.19 854 SN2 113 3.08 1.019 SN3 113 3.69 983 SN4 113 3.67 881 PA1 113 2.65 1.148 PA2 113 2.81 1.122 PA3 113 3.00 1.134 PCE1 113 2.50 1.218 PCE2 113 3.15 1.054 PCE3 113 2.44 1.187 PCE4 113 3.94 909 PI1 113 3.75 1.014 47 PI2 113 3.47 964 PI3 113 3.26 1.059 PI4 113 3.64 1.070 ME1 113 3.19 1.125 ME2 113 3.16 1.192 ME3 113 3.47 1.218 Valid N (listwise) 113 Tải FULL (115 trang): https://bit.ly/3U3wpG0 Dự phòng: fb.com/TaiHo123doc.net A summary of the descriptive statistics is presented in table above For each item the minimum, maximum, mean and standard deviation are taken into consideration Among 120 questionnaires were distributed and collected both online and offline The received answered are screened to find out the invalid, then among 120 questionnaires, 113 samples are valid as in the table above and missing samples Data analysis excluded missing samples because participants did not provide sufficient information or they chose the same option for all questions Table 4.2 Gender distributions among 113 participants Frequency Percent Valid Percent Cumulative Percent Female 73 64.6 64.6 64.6 Valid Male 40 35.4 35.4 100.0 Total 113 100.0 100.0 48 From Table 4.2 above, it can be seen that 113 participants who answered questionnaires are not distributed equally about gender The number of females is bigger than the number of males There are 73 females and 40 males sent their responses back which correspond to 64.6% and 35.4%, respectively Tải FULL (115 trang): https://bit.ly/3U3wpG0 Dự phòng: fb.com/TaiHo123doc.net Figure 4.1 Age distribution among 113 participants Among 113 people answered the questionnaire, the main participants are 26 to 35 years old They account for 59 people or 52.2% The second largest group is in age of under 25 The quantity of this group is 50 people which represent 44.2% of total The group of 36 years old and older only accounts for 3.5% of the total with participants It can be seen that, most of participants are young working people with age range is from 26 to 35 years old 49 Figure 4.2 Education level distribution among 113 participants As the pie chart shows, a great deal of participants is in University and Graduate School level The largest number of participants belongs to University level, with the ratio is 64.6% with 73 people Participants who have Graduate School stand for a smaller volume of 29 participants The ratio number corresponds to 25.7% Next is the number of participants who are high school students (indicated in “Others” label) with people – 5.3% Participants who have Vocational Training and College level are the smallest amount, 3% (with participants) and 1.8% (with participants) respectively It can be seen that education level of survey participants is relatively homogenous Table 4.3 Income distributions among 113 participants Frequency Percent Valid Percent Cumulative Percent Under 15M 87 77.0 77.0 77.0 15M - 35M 24 21.2 21.2 98.2 1.8 1.8 100.0 113 100.0 100.0 Valid 36M and higher Total 50 6795212 ... consumer purchase intention of cloth bags for shopping That means the higher/lower consumers perceive the availability of cloth bags for shopping, the higher/lower level of customer forming intention. .. consumer purchase intention of cloth bags for shopping That means the higher/lower consumers perceive the effectiveness of cloth bags for shopping, the higher/lower level of consumers forming intention. .. consumer purchase intention of cloth bags for shopping That means the higher/lower consumers form an attitude toward purchasing cloth bags for shopping, the higher/lower level of consumers forming intention