Một số tiền tố của hành vi tiêu dùng thân thiện với môi trường một nghiên cứu về tiêu dùng xanh tại đà lạt

NHIỆM VỤ VÀ NỘI DUNG: - Develop a theoretical model that demonstrates the structural relationships among value orientations, environmental identity, self-efficacy, personal outcome expe

-LE HAI YEN

SOME ANTECEDENTS OF CONSUMERS'

PRO-ENVIRONMENTAL BEHAVIOR: A STUDY OF GREEN

CONSUMPTION IN DA LAT, VIETNAM

MỘT SỐ TIỀN TỐ CỦA HÀNH VI TIÊU DÙNG THÂN THIỆN VỚI MÔI TRƯỜNG: MỘT NGHIÊN CỨU VỀ TIÊU

ĐẠI HỌC QUỐC GIA TP HỒ CHÍ MINH

Cán bộ hướng dẫn khoa học: PGS.TS Nguyễn Mạnh Tuân Chữ ký:

2 Thư ký: PGS TS Vương Đức Hoàng Quân

3 Phản biện 1: TS Lê Hoành Sử

4 Phản biện 2: TS Nguyễn Vũ Quang

5 Ủy viên: PGS TS Nguyễn Mạnh Tuân

Xác nhận của Chủ tịch Hội đồng đánh giá luận văn và Trưởng khoa quản lý

chuyên ngành sau khi luận văn đã được sửa chữa (nếu có)

TS Phạm Quốc Trung

NHIỆM VỤ LUẬN VĂN THẠC SĨ

Ngày, tháng, năm sinh: 21-11-1990 Nơi sinh: Lâm Đồng

Chuyên ngành: Quản trị kinh doanh Mã số: 60340102

I TÊN ĐỀ TÀI: Some antecedents of consumers' pro-environmental

behavior: A study of green consumption in Da Lat, Vietnam

II NHIỆM VỤ VÀ NỘI DUNG:

- Develop a theoretical model that demonstrates the structural relationships

among value orientations, environmental identity, self-efficacy, personal outcome

expectation and pro-environmental behavior of individual consumers

- Validate the model in the context of green consumption in Da Lat, Vietnam

III NGÀY GIAO NHIỆM VỤ: Ngày 11 tháng 02 năm 2019

IV NGÀY HOÀN THÀNH NHIỆM VỤ: Ngày 05 tháng 07 năm 2019

V CÁN BỘ HƯỚNG DẪN: PGS.TS NGUYỄN MẠNH TUÂN

This thesis is a mark for the completion of my Master program in Ho Chi Minh University of Technology, opens a new page for my career During the time working in this thesis, I have received a lot of support to finish my study First of all, I would like to express my sincere gratitude to my advisor, Assoc Prof Dr Nguyen Manh Tuan He gave me not only appropriate advice and advance knowledge but also very good encouragement for me to overcome a lot of difficulties during such a time I would like to thank Dean of School of Industrial Management, Ho Chi Minh University of Technology, Faculty staffs for their facility and lectures support Finally, I would like to give a big thank to my family, all teachers, and friends for their encouragement and support during my time of studying My thesis work was finally completed with all these love and help, again, sincerely, thanks to all

Lam Dong, 5th July 2019

Le Hai Yen

Chủ nghĩa tiêu dùng xanh đang nhanh chóng lớn mạnh và trở thành một khuynh

hướng tiêu dùng được khách hàng ủng hộ Tựa trên nền tảng của định hướng giá trị

(gồm giá trị cá nhân, giá trị xã hội và giá trị môi trường lấy từ lý thuyết giá trị),

đồng nhất với môi trường (từ lý thuyết đồng nhất) và sự tự lập cũng như kỳ vọng kết quả cá nhân (từ lý thuyết tự điều tiết), bài viết này đề xuất và kiểm định một mô

hình lý thuyết để giải thích hành vi thân thiện với môi trường của khách hàng Nghiên cứu thực hiện phân tích PLS-SEM trên 179 mẫu khảo sát từ người tiêu dùng tại Đà Lạt, Việt Nam đã cho thấy có đến 14/15 giả thuyết nghiên cứu được ủng hộ Kết quả nghiên cứu đã khẳng định tác động trực tiếp của định hướng giá trị lên

đồng nhất với môi trường và từ đó lên hành vi thân thiện với môi trường của khách

hàng thông qua sự tự lập và sự kỳ vọng kết quả cá nhân Kết quả cũng khẳng định ảnh hưởng của sự tự lập lên sự kỳ vọng kết quả cá nhân, và hơn nữa đã khám phá được quan hệ cấu trúc bên trong khái niệm định hướng giá trị và trong hành vi thân

thiện với môi trường Ngoài ra, bài viết cũng là một trong các nghiên cứu trước nhất

về quan hệ đồng nhất với môi trường-tự điều tiết và về vai trò trung gian toàn phần của sự tự lập trong quan hệ giữa đồng nhất với môi trường và hành vi mua của

khách hàng

Green consumerism is rapidly growing and becoming a familiar shopping option

that consumers adopt Based on value orientations (including egoistic, altruistic and biospheric values, from value theory), environmental identity (from identity theory), and both self-efficacy and personal outcome expectation (from self-regulation theory), this paper build and validate a theoretical framework to explain pro-

environmental behaviors (including consumer purchase behavior and willingness to pay) A survey study with PLS-SEM analysis of 179 consumers in green

consumption in Dalat, Vietnam showed 14 of 15 hypotheses empirically supported The findings affirmed the direct influence of value orientations on environmental identity, and then on pro-environmental behaviors through self-efficacy and personal outcome expectation The results also reinforced the impact of self-efficacy on personal outcome expectation, and moreover, explored the structural relationships within value orientations and within pro-environmental behaviors The paper is among the first studies to empirically identify the environmental identity-self-regulation link and the full mediation of self-efficacy between environmental identity and consumer purchase behavior The theoretical and managerial implications of the paper were derived

I hereby declare that I am the sole author of this Master thesis and that I have not used any sources other than those listed in the bibliography and identified as references I further declare that I have not submitted this thesis at any other institution in order to obtain a degree

Lam Dong, 5th July 2019 Thesis author

Le Hai Yen

NHIỆM VỤ LUẬN VĂN THẠC SĨ

ACKNOWLEDGEMENT

TÓM TẮT

ABSTRACT

STATEMENT OF AUTHORSHIP

LIST OF FIGURES

LIST OF TABLE

LIST OF ABBREVIATION

CHAPTER 1: INTRODUCTION 1

1 1 Research motivation 1

1.1 1 Real-world problems 1

1.1.2 Current state of research 2

1.2 Objective 5

1.3 Research scope 6

1.4 Research methodology 6

1.5 Managerial implications 7

1.6 Structure of the thesis 7

CHAPTER 2: LITTERATURE REVIEW AND RESEARCH MODEL 9

2.1 Concepts determination 9

2.2 Previous studies 11

2.2.1 Judith I M de Groot and Linda Steg, 2008 11

2.2.2 Joanne Dono, Janine Webb and Ben Richardson, 2010 12

2.2.3 Hsiu Yi Lin and Meng Hsiang Hsu, 2013 13

2.2.4 Birgitta Gatersleben, Niamh Murtagha and Wokje Abrahamse, 2014 14

2.3 Conceptial model, proposed research model and hypothesis 14

CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY 21

3.1 Research process 21

3.2 Preliminary research 22

3.2.1 Preliminary measurement scales development 22

3.3 Official research 26

3.3.1 Sample size detection 26

3.3.2 Data collection design ……… 26

3.4 Data analysis method 28

3.4.1 Data collection and assessment (Hair et al., 2014) ……… 28

3.4.2 Structural model PLS – SEM 28

3.4.3 Measurement model assessment in PLS – SEM ……… 29

3.4.4 PLS – SEM structural mode assessment 30

CHAPTER 4: RESULTS 35

4.1 Preliminary quantitative research 35

4.2 Preliminary quantitative results 36

4.3 Official quantitative research 40

4.3.1 Sample statistics 40

4.3.2 Proposed research model assessment 42

4.3.2.1 Measurement model 42

4.3.2.2 Structural model assessment of multicollinearity 45

4.4 Results and discussion 49

4.4.1 Proposed research model 49

4.4.2 Discussion of proposed research model 54

CHAPTER 5: CONCLUSION AND RECOMMENDATION 58

REFERENCES 61

Appendix 1: Preliminary qualitative research 70

Appendix 2: Official questionnaires 79

CURICULUM VITAE 85

Figure 2.1 Corelations among 3 value orientations, environmental concern and

environmental beliefs 12

Figure 2.2 Research model constructed by Joanne et al., 2010 12

Figure 2.3 Conceptual research model constructed by Lin & Hsu, 2013 14

Figure 2.4 Conceptual framework 15

Figure 2.5 Proposed research model 20

Figure 3.1 Research process (Nguyen Dinh Tho, 2007) 21

Figure 4.1: Research model assessment results 50

Table 3.1 Draft measurement scales 23

Table 4.1 Internal consistent reliability 37

Table 4.2 Convergent validity 38

Table 4.3 Discriminant validity (Fornell Larcker standard) 38

Table 4.4 Loading and cross-loading factor 39

.Table 4.5 Sample analysis according to green consumption method 40

Table 4.6 General information of survey 41

Table 4.7 Personal information analysis 41

Table 4.8 Internal consistency reliability 43

Table 4.9 Convergent validity (AVE) 44

Table 4.10 Discriminant validity (Fornell-Larcker standard) 44

Table 4.11 Loading and cross-loading factor 45

Table 4.12 Multicollinearity 46

Table 4.13 Path coefficients β 46

Table 4.14 Coefficients of determination R2 48

Table 4.15 Effect size f2 49

Table 4.16 Results of research model’s hypotheses 50

AVE : Average variance extract CFA : Confirmatory factor analysis

In local context, Da Lat city is considered one of the green center of Vietnam, which is city of agriculture and forestry sectors At the AMME conference and the 15th ASEAN

+ 3 Environment Ministerial Conference, held in Brunei Darussalam from 11 to 14 September 2017, this city is the representative for the whole country to receive ASEAN sustainable city award (Lao Dong, 2018) This is an award to honor and promote the image of typical cities for environmental quality, while raising awareness

of the community about the importance of sustainable environmental protection This

is the motivation for Da Lat people take more involvement into green life As a citizen

of Da Lat city, I have observed Da Lat people, especially the young generation have awared of changing their consumption to sustainable way In accordance with this mementum, some researches are needed, so that producers, service suppliers, or even government can have appropriate information for there short and long term plan making for further development activities

For managerial meaning, many businesses and startups are also more concerned with the environment, so green consumption is a worth considering trend and sustainable developing way (VNN, 2018) Customer behavior analysis is a useful tool for producers to develop their strategies In such, a lot of green customer behavior studies have been conducted in order to adapt customer behavior research into plenty aspects

of green production It is, therefore, necessary to discuss pro-environmental behaviors

in consumer research in specific and business research in generic (Grimmer et al, 2015)

1.1.2 Current state of research

Some previously published results which are available for this research area are mentioned as follow:

One of the first basement studies for this research is Christina Tobler’s doctor of science work in year 2011 The topic was “green customer behavior: consumers’ knowledge and willingness to act pro-environmentally”, conducted in Zurich,

Switzerland and focused in food product consumption Since there had been no proper measurement for climate change knowledge, Christina developed an intensive scale in form of questionnaires to measure customer’s knowledge of climate change Following the model of ecological behavior, consumers' perceived benefits and consequences of ecological behaviors were examined in the second part Likerd scale questionnaires were also designed to study relationship among customers’ knowledge, attitude and their green consuming behavior The research findings showed the lack of climate change as well as environmental knowledge in most of consumers The willingness to act pro-environmental had relationship to cost of climate behavior and perceived climate-profit These results led to the need of climate education development in certain level for society (Tobler, 2011)

Later, in 2015, Anees Ahmad and K S Thyagaraj published their study about customers’ intention to purchase green brand, specifically, electronic products (Ahmad

& Thyagaraj, 2015) The results showed positive corelations among most of variables and there was no relationship between environmental knowledge and purchase intention However, the research also found environmental education/media communication has indirect effects on customers’ purchase intention of green brand products

Continue green consumer behavior research, in recent years, some groups of authors conducted their studies on green/sustainable packaged products in Sri Lanka (G Madushanka & V.R.Ragel, 2016) and Romania (Gheorghe Orzan, Anca Francisca Cruceru,Cristina Teodora Bălăceanu & Raluca-Giorgiana Chivu, 2018) with questionnaires survey method Both researchs showed the strong corelation between environmental concern and attitude to green packaging However, consumer’s attitude towards green packaging is varying based on the income level of the consumers (Madushanka & Ragel, 2016) For the case in Romania, 58.2% of the respondents considered green packaged products are more expensive, 65.6% of the respondents

lacked of information about green packaging and 11.5% respondents were not willing

to pay extra for packaging (Orzan et al., 2018)

After years studying on green customer behavior, researchers from different contries have clearly showed that customers’ knowledge of environment or climate change is limited, environmental concern and attitude relly affect to customer intention of purchasing green products (Tobler, 2011; Ahmad & Thyagaraj, 2015; Madushanka & Ragel, 2016; Orzan et al., 2018)

Besides, many of the past studies have focused on explaining and predicting

pro-environmental behaviors by exploiting value theory in which consumer values and

beliefs as the main antecedents as early framed in Stern et al (1995)’s often-cited model Another stream of research, which was much promoted by Stets & Biga (2003),

have adopted identity theory as the base to have effects on consumer attitude and then

on their buying behaviors It should be noted that, few studies (for example,

Gatersleben et al, 2014) have recently brought value theory and identity theory together

to understand pro-environmental behaviors

A more promising approach resides in Bandura’s (1977) theory of self-regulation

which is strongly advocated by Bagozzi (1992) as a powerful theoretical framework to discuss human behaviors, especially purposeful or intended behaviors Numerous empirical studies across many fields over the past decade (e.g Wang et al, 2015; Vries

et al, 1988; for a review, see Sheeran et al, 2016 ) have integrated the substantial constructs (e.g self-efficacy and personal outcome expectation) from self-regulation with individual psychological variables (like attitudes, norms, values and beliefs) from attitude theories such as theory of reasoning action or theory of planned behaviors into some sort of comprehensive models to better understand human behaviors

In the light of this integrative approach which recently has also been applied to understand pro-environmental behaviors (e.g Ertz et al, 2016), drawing upon value

theory (i.e value orientations), identity theory (identity), and self-regulation theory

(self-efficacy and personal outcome expectation), we build and validate a model of green consumption, a typical case of pro-environmental behavior (ETO, 2018) Moreover, as both value orientations and pro-environmental behaviors are made up of

several components, our second goal is to explore the structural relationships among these components in each construct mentioned

These findings are good primary information and theories for researches in similar area

1.2 Objective

The study was implemented to achieve the following objectives:

- Develop a theoretical model that demonstrate the structural relationships among value orientations, environmental identity, self-efficacy, personal outcome expectation and pro-environmental behavior of individual consumers

- Validate the model in the context of green consumption in Da Lat, Vietnam with listed green consumption items:

+ Purchasing Green packaging products

+ Zero-wasted shopping: customers use their own bag/cup/food container to go shopping

+ Purchasing recycled products: buying products made from recycled materials + Buying green food

Research findings will help producers or companies, based on understanding of green customer behavior in specific context, to set their strategies of marketing and sale which promote for green-packaged products

1.3 Research scope

The research object is individual consumer Due to limited time and resources, the scope of research is done in Da Lat city The survey was conducted mainly at supermarkets, food stores, markets, travel destinations, households, and also online survey by social network

Data collected time was from 30th March 2019 to 20th May 2019

1.4 Research methodology

Information was collected based on 2 sources: primary and secondary

- Secondary information was collected from various sources including available theories, related research topics and other sources of reference This information was used for qualitative research

- Primary information was collected through surveys from individuals (pre-designed questionnaires) This information was used for quantitative research

Qualitative research

Discussion face to face between the researcher and those who need to collect information was conducted This study aimed to explore, adjust and supplement the scales of antecedents of pro-environmental behavior based on behavioral theory and build a questionnaire to survey

Quantitative research

Quantitative research was done through survey questionnaires that had been sent to each object selected by sampling directly or online Collected data was used to assess the reliability and value of the scale, to verify the suitability of the proposed model Data collected was processed by SPSS 23.0 software and ADANCO 1.1 (Henseler & Dijkstra, 2014) After collecting a sufficient number requested, all data was encrypted and cleaned before being included in the analysis Data encryption is the process of

converting the answers in the questionnaire into the form of codes for input and processing in SPSS 23.0 software and ADANCO 1.1 Clean data for the purpose of detecting and handling errors that may occur in the answer set such as a blank box or

an unreasonable answer Data analysis methods including steps to assess data sets evaluate models according to PLS-SEM, evaluate structure models according to PLS-SEM

1.5 Managerial implications

The research was conducted to provide general information on pro- environmental behavior, the impact level of each element on behavior in the process of individuals implementing sustainable consumption Research results were reference materials for agencies and enterprises when implementing environmental protection policies through the correct impact of the right people and the right factors to bring about high efficiency

1.6 Structure of the thesis

This thesis includes 5 chapters:

Chapter 1: Introduction - This chapter gives an overview of the research, the reasons

for forming the topic, research objectives, research scope and methods, managerial implications of the research, the layout of the thesis

Chapter 2: Theoretical basis and research model - This chapter presents the

theoretical basis, proposed research models and research hypotheses

Chapter 3: research methodology and design – This chapter presents detailed

research methods, qualitative research results, adjusting scale, completing questionnaires for quantitative research, sample design and Expected data analysis

Chapter 4: Research findings – This chapter describes data, presents steps for data

analysis and data analysis results

Chapter 5: Conclusion and recommendation - This chapter summarizes the results of

this study, presents the contributions of the thesis, makes recommendations based on the research results, limitations and subsequent research directions

CHAPTER 2: LITTERATURE REVIEW AND RESEARCH MODEL

Chapter 2 airms to summary theories about value orientations, environmental identity, self-regulation, and pro-environmental identity, then present the proposed hypotheses and research model Theoretical basis in the analysis and formation of research hypotheses include: Value orientations (Schwartz, 1992); pro-environmental behavior (Kollmuss & Agyeman, 2002); identity theory (Pillsbury, 1934); Self- regulation (Bandura, 1977, 1997) This chapter includes 2 main contents: related theoretical basis and hypotheses and research model

2.1 Concepts determination

Pro-environmental behaviours

For this study, pro-environmental behavior is generally defined as intentionally reducing the negative impact that a human action can have on the environment (Kollmuss & Agyeman, 2002) By this definition, pro-environmental behaviors can refer to some related activities such as sustainable consumption behaviors (Vermeir & Verbeke, 2006), or conservation behaviors (Monroe, 2003), to name a few In many consumption settings, pro-environmental behaviors can be operationalized as two dimensions including consumer purchase behavior and willingness to pay (Stern et al, 1995) From consumer’s view, while the two different types share the same feature of environmentally responsible consumption, they are distinct in terms of the psychological nature of human behaviors Drawing on the affective-behavior-cognition model of attitude formation mentioned above, given a typical consumption setting, whilst consumer purchase behavior could be considered as a sort of behavioral response, willingness to pay turns out a type of affective response (Dono, 2010)

Value orientations

Schwartz (1992) defines a value as: “a desirable trans-situational goal varying in importance, which serves as a guiding principle in the life of a person or other social entity” This definition includes most of the agreed on key features of values First, a value reflects a belief on the desirability of a certain end-state Second, values are rather abstract and therefore transcend specific situations Third, values serve as a guiding principle for selecting or evaluating behavior, people, and events And finally, values are ordered in a system of value priorities This feature implies that when different competing values are activated in a specific situation, choices are based on values that are considered to be most relevant to act on The characteristics of values illustrate at least two reasons why it is important to study values First, it has been theoretically reasoned and empirically validated that values play a significant role in explaining specific beliefs and behavior and can therefore be used as predictors for various variables such as attitudes and behavioral intentions (Stern, 2000; Stern & Dietz,1994) Second, the total number of values that people may consider is relatively small Therefore, relative to other antecedents of behavior (e.g., specific beliefs, attitudes), values provide an economically efficient instrument for describing and explaining similarities and differences between persons, groups, nations, and cultures (Rokeach, 1973)

Self-regulation

According to Bandura’s framework of regulation, human action is a regulatory system and self-regulation is defined as self-directed change influenced by the environment Bandura (1977, 1997) stated that behavior and self-regulation process are best predicted by the combined influence of one’s belief in the ability to

self-perform a task, also called self-efficacy, and personal outcome expectancy which is

results one anticipates from doing these tasks Both self-efficacy and outcome expectancy are part of cognitive process prior to one’s action (Bandura, 2003)

Self-identity

Self-identity refers to how an individual sees him/herself, and can encompass all aspects of the self such as physical attributes, preferences, values, personal goals, habitual behaviour, personality traits and personal narratives (Mc Adams, 1995) Individuals tend to present themselves in ways that are congruent with their self-identity (Burke & Reitzes,1991), and this extends to behaviour (Sparks & Shepherd, 1992) including consumption (Dittmar, 2010) Although identity represents an individual’ ssubjective perspective on the self, identities are formed through social interaction Theorists in the symbolic interaction ist tradition proposed the development of the self through reflection from others in social exchanges (Breakwell, 1986) and Stets and Burke (2000) proposed that identities develop through processes

of self-categorisation and identification People thus develop multiple identities Multiple identities are proposed as being managed in a ‘hierarchy of salience’ (Stryker,1984): identities vary in salience, and particular identities, such as gender, arel ikely to be chronically salient Identities can form barriers to pro-environmental behaviours Identities can also motivate ‘green’ behaviour An environmental identity reflects the extent to which people indicate that environmentalism is a central part of who they are, and a number of studies have shown that an environmental identity increases engagement in pro-environmental actions

2.2 Previous studies

2.2.1 Judith I M de Groot and Linda Steg, 2008

Publication of Judith and Linda in 2008 is about the method of measuring egoistic, altruistic, and biospheric value orientations with specific scales Also, the study mentioned about corelation between the 3 value orientations and value orientations with environmental conver and significant evironmental beliefs

Figure 2.1 Corelations among 3 value orientations, environmental concern and

environmental beliefs

2.2.2 Joanne Dono, Janine Webb and Ben Richardson, 2010

The study is about the relationship between environmental activism, environmental behaviour and social identity The results revealed that there was a significant relationship between social identity and environmental behaviour, only the citizenship component of environmental behaviour significantly predicted environmental activism These research findings add to the body of evidence regarding the importance of inter-group processes to collective action and suggest the potential usefulness of social identity theory for making predictions about when environmental activism will occur, why certain environmental activist strategies are more common than others, and suggest that there is potential to achieve a much more integrated approach to under-standing environmental activism and the inter-group processes that underlie such action

pro-Figure 2.2 Research model constructed by Joanne et al., 2010

2.2.3 Hsiu Yi Lin and Meng Hsiang Hsu, 2013

In this study, Lin and Hsu investigated green customer behaviour using social cognitive theory They constructed the conceptual research model to find the corelations among environment, person, outcome expectation and behaviour

Figure 2.3 Conceptual research model constructed by Lin & Hsu, 2013

2.2.4 Birgitta Gatersleben, Niamh Murtagha and Wokje Abrahamse, 2014

Before this research, there is significant evidence that values and identities play a role

in explaining and predicting pro-environmental behaviour Birgitta et al looked at values and identities simultaneously and studied the relative importance of each of these constructs in understanding pro-environmental behaviour The results showed the relationship between values and identity They also found that when identities were included, the relationship between environmental values and behaviours was weaker, suggesting that identities mediate the link between values and behaviours

2.3 Conceptial model, proposed research model and hypothesis

A conceptual framework is generated to learn the relationship between value orientations, environmental identity, self-regulation and pro-environmental behavior

Figure 2.4 Conceptual framework

The hypotheses were developed:

Value orientations

Values are “desirable, trans-situational goals, varying in importance, that serve as guiding principles in people’s lives” (Schwartz, 1996, p 2) According to Schwartz’s, there are many types of value, mainly being classified into two groups, self-transcendence versus self-enhancement and openness to change versus conservation (De Groot, 2008) From psychological view, there are the three distinct value orientations to shape human beliefs and behaviors, including egoistic, altruistic, and biospheric values (Stern, 2000) Egoistic values indicate individual ultimate outcomes, altruistic values involve concern for the good of others; and biospheric values, a more recent category, mention concern for the environment and the earth’s biological system (De Groot, 2008) Although it is widely recognized that the three value orientations exist (e.g Stern et al, 1995), the structural relationships among them seem

to be unclear

According to the affective-behavior-cognition model of attitude formation (Breckler, 1984; Van den Berg et al, 2006), attitude comprise three components of response - cognitive mentioning more to belief and evaluation, affective more to feeling and emotion, and behavior more to response and action – and further, cognitive response

may influence affective one, which in turn may impact behavioral response Thereby,

we argue that egoistic value refers to the cognitive dimension of value orientation, altruistic to affective dimension and biosphere to behavioral dimension of value orientation From this, the following hypotheses are recommended:

H1a: Egoistic value orientation significantly influences altruistic value orientation H1b: Altruistic value orientation significantly influences biospheric value orientation

Identity Theory

Self-identity refers to the way a person see one’s self, encompassing physical attributes, preferences, values, personal goals, habitual behavior, personality traits and personal narratives (Pillsbury, 1934) A person tends to present the ways that are congruent with their self-identity (Burke & Reitzes, 1991), and this extends to behavior (Sparks & Shepherd, 1992) including consumption (Oyserman et al, 2007; Dittmar, 2010) Although identity represents an individual’s subjective perspective on the self, identities are formed through social interaction (Pillsbury, 1934) For example, theorists in the symbolic interactionist tradition proposed the development of the self through reflection from others in social exchanges (Mead, 1934; Breakwell, 1986) and Stets and Burke (2000) also proposed that identities develop through processes of self-categorization and identification In this study, we consider person’s

identity in environment protection context as environmental identity

Value orientations and environmental identity

Using the narrative, MacAdam (1995) suggested that identity is a person’s long story resulting from his/her life-long development that is in turn based on his/her value or beliefs Recently, Gatersleben et al (2014), in a sustainable consumption, empirically found a relationship between value orientations and environmental identity

On these bases, we propose the following hypotheses for the context of green consumption with an individual environmental identity and the three value orientations:

H2a: Egoistic value orientation significantly influences environmental identity

H2b: Altruistic value orientation significantly influences environmental identity H2c: Biospheric value orientation significantly influences environmental identity

Environmental identity and pro-environmental behaviors

The theoretical contribution of social Dono et al (2010) demonstrated that social identity played prediction role toward pro-environmental behavior Meanwhile, environmental identity is developed through social interaction, action and recognition (Sarah, 2015) These two findings led us to assume that environmental identity may also predict pro-environmental behavior On the other hand, an environmental identity reflects the extent to which people indicate that environmentalism is a central part of who they are, and moreover, a number of studies have also shown that an environmental identity increases person engagement in pro-environmental actions For example, Whitmarsh and O’Neill (2010) found that people with a green identity more often act in an environmentally responsible manner Similarly, Van der Werff et al (2011) indicated that an energy saving identity is positively related to intentions to conserve energy Thus, we suggest the following hypotheses:

H3a: Environmental identity significantly influences consumer purchase behavior H3b: Environmental identity significantly influences willingness to pay

Self-regulation

According to Bandura’s framework of regulation, human action is a regulatory system and self-regulation is defined as self-directed change influenced by the environment Bandura (1977, 1997) stated that behavior and self-regulation process are best predicted by the combined influence of one’s belief in the ability to

self-perform a task, also called self-efficacy, and personal outcome expectancy which is

results one anticipates from doing these tasks Both self-efficacy and outcome expectancy are part of cognitive process prior to one’s action (Bandura, 2003)

Environmental identity and self-regulation

Meanwhile, given a general notion of identity as a set of meanings attached to the self that serves as a standard that shapes people cognition, attitude and behavior (Stets and Biga, 2003), we believe that environmental identity is a reliable reference to guide a variety of responses of environment related issues such as self-evaluation of own capability to do some relevant activities

On this basis, we propose the following hypotheses:

H4a: Environmental identity significantly influences self-efficacy

H4b: Environmental identity significantly influences personal outcome expectation

Self-efficacy and personal outcome expectation

Working on Bandura’s self-regulation framework, some scholars have recently raised

a challenge, or emphatically an unresolved contradiction (Williams, 2010) That is a question about which direction of the causal link between self-efficacy and outcome expectation While Bandura (1997) affirmed the influence of self-efficacy on outcome expectation but not vice versa, some works have suggested the reverse link between the two Therefore, it is an opportunity for us to propose and validate the following hypothesis in the context of green consumption:

H5: Self-efficacy significantly influences personal outcome expectation

Self-regulation and pro-environmental behaviors

In his conceptual frame of self-regulation, Bandura (1997) generally indicated on how self-efficacy can initialize and make changes on human behaviors In this paper, with the definition of green consumption self-efficacy as ones’ confidence in their own capabilities to follow green consumption via either their purchase behavior or willingness to pay more to gain benefits from their participation of environment protection Thus, based on Bandura’s seminal work, we naturally come to the two following hypotheses:

H7a: Self-efficacy significantly influences consumer purchase behavior

H7b: Self-efficacy significantly influences willingness to pay

The other key concept in Bandura’s frame is personal outcome expectation which is understood as individual’s estimate that a given behavior will lead to a certain outcome According to motivation theory (Deci & Ryan, 1985), when an individual perceives potentially positive benefit, one possibly increase incentive motivation and willingness to practice the relevant behaviors that, in the context of green consumption, turn to purchase behavior and wiliness to pay more in green products and services This leads us to provide the following hypotheses:

H6a: Personal outcome expectation significantly influences consumer purchase

behavior

H6b: Personal outcome expectation significantly influences willingness to pay

Finally, referring to the affective-behavior-cognition model of attitude formation mentioned above, we recognize that within the pro-environmental behavior, the dimension of willingness to pay expresses affective response while the one of purchase behavior centers on behavioral response Thus, we propose the following hypothesis:

H8: Willingness to pay significantly influences consumer purchase behavior

All hypotheses of this study are showed in the research model (Figure 2.5)

Figure 2.5 Proposed research model

CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY

Chapter 2 presented the theoretical basis related to this research, development of hypotheses and research models Chapter 3 will present research methodology in details, qualitative research results, scale adjustment, completion of questionnaires for quantitative research, and data analysis methods

3.1 Research process

The study consisted of two stages: the preliminary stage including preliminary qualitative research and preliminary quantitative research, and the official phase including official quantitative research, which are demonstrated Figure 3.1

Figure 3.1 Research process (Nguyen Dinh Tho, 2007)

The research started off with a preliminary qualitative step to assess the scale of concepts qualitatively The draft scale used in this step included the concepts in proposed research model and the scale of these concepts was translated into Vietnamese from the original scales of previous studies to suit the research context in Vietnam Preliminary qualitative research with group discussion techniques and face to face discussion was conducted to adjust the scale to suit the context of the study After the preliminary qualitative step, the study continued with preliminary scales for preliminary quantitative study with a sample of size 60 Preliminary scale was then calibrated based on internal consistency (CR convergent reliability), convergent validity (average variance extracted AVE, loading factor), discriminant validity (Fornell Larcker standard, cross-loading coefficient) After the preliminary quantitative step, official scale was finalized for official quantitative research

Next, the PLS-SEM method was applied to analyze and evaluate the research model and hypotheses for the purpose of the paper to predict and explore the theory First of all, the study evaluated the measurement model, including the following steps: assessing the internal consistency reliability (CR composite reliability), convergent validity (Average variance extracts AVE, load factor), discriminant validity (Fornell Larcker standard, cross-loading coefficient) Next, the research evaluated the structural model, including the steps: assessing the additive (variance inflation factor - VIF) of the concepts, path coefficient (β), coefficients of determination (R2) and the effect size (f2)

3.2 Preliminary research

3.2.1 Preliminary measurement scales development

The scale of the concepts in this research model was based on the scales of previous studies and the correction of words for easy understanding, in accordance with the context of the study The draft scales (Table 3.1) of the study model consists of 25

observed variables measuring 8 concepts

Preliminary measurement scales were inherited and modified from different studies are

Observed variables from 1 to 25 measured by Likert scale of 5 points, as follows:

• Selection "1" corresponds to "Very disagree"

• Selection "2" corresponds to "Disagree"

• Selection "3" corresponds to "No opinion"

• Selection "4" corresponds to "Agree"

• Selection "5" corresponds to "Very agree"

Table 3.1 Draft measurement scales

1 EVO1 For me, social position and power is important

2 EVO2 For me, having an impact on people and events is important

3 EVO3 For me, ambitious is important

Altruistic value

orientation

4 AVO1 For me, everyone should have equal opportunity

5 AVO2 For me, a world at peace is important

6 AVO3 For me, social justice is important

7 AVO4 For me, being helpful (working for the welfare of others) is

important

Biospheric value

orientation

8 BVO1 For me, protecting natural resources is important

9 BVO2 For me, respecting the earth (harmony with other species) is

12 EI2 I think of myself as a part of nature, not separate from it

13 EI3 Behaving responsibly toward the Earth - living a sustainable life

style is part of my moral code

Green

consumption

self- efficacy

14 SE1 I am confident in my ability to engage in waste utilization

15 SE2 I am confident in my ability to show preference for products

with Greenmark

16 SE3 I am confident in my ability to implement the concept that all

organisms are equal and not to damage the natural environment

21 CPB2 I often make a special effort to buy paper and plastic products

(small things like note books, tissue papers or bigger stuffs like furniture ) made from recycled materials

22 CPB3 I often do avoid product from accompany that I know maybe

harming the environment

Willingness to

pay

23 WP1 I would be willing to pay much higher taxes in order to protect

the environment

24 WP2 I would be willing to accept cuts in my standard of living to

protect the environment

25 WP3 I would be willing to pay much higher prices in order to protect

the environment

3.2.2 Preliminary qualitative research

Preliminary research was done through two steps: preliminary qualitative and preliminary quantitative analysis Preliminary qualitative research is a step to adjust the content, determine appropriate terminology and supplement/ remove observed variables used to measure research concepts For this step, group and face to face discussion techniques had been used based on the Vietnamese translation of draft English scales (Table 3.1) The content of the discussion will be recorded and synthesized as a basis for the construction of a preliminary scale used in preliminary quantitative research This preliminary qualitative step is expected to be carried out with 7 individuals participating in green consumption in Da Lat - Vietnam

3.2.3 Preliminary qualitative research

Preliminary quantitative research was carried out to evaluate the preliminary scale quantitatively to see if the scales were good enough to be used as official scales in official quantitative research step This preliminary quantitative research was done by directly survey using paper questionnaires Convenient sampling method was done with size of 60 samples (Hair & et al, 2014)

3.3 Official research

3.3.1 Sample size detection

Official quantitative research methods initiated with data collection by interviewing

according to predefined questionnaires and using convenient sampling methods (estimated 170 samples) The theoretical model of the topic consisted of 8 concepts Based on this model, many hypotheses were stated verified through collected primary data The thesis used PLS-SEM method to test the model and research hypotheses, from the results obtained, conclusions and recommendations were drawn The paper used PLS - SEM because PLS - SEM is mainly used to develop theory in discovery research by focusing on explaining the variance of the dependent variables when assessing models (Hair et al , 2014)

3.3.2 Data collection design

Subject of the survey: Green consumption in Da Lat recently is rapidly developing A

lot of citizens are aware of it and pay attention on it The research objects were individual consumers Due to limited time and resources, the scope of research was within Da Lat city The survey was conducted mainly at supermarkets, food stores, markets, households, gas station and also online survey by email and social network There are some green consumption items were surveyed in this study:

- Buying green food

- Zero-wasted shopping: customers use their own bag/cup/food container to go

shopping

- Purchasing Green packaging products

- Purchasing recycled products: buying products made from recycled materials

Sampling technique: Samples were selected by a convenient method, one of the

non-probability sampling patterns The size of the sample applied in the study was based on the requirements of exploratory factor analysis and correlation analysis Subjects participating in the study sample were selected so that the survey sample ensures the variance had been large enough to test the hypothesis if it was rejected

3.4 Data analysis method

Data collected was processed by SPSS 23.0 and ADANCO 1.1 software After collecting a sufficient number requested, all data was encrypted and cleaned before being included in the analysis Data encryption is the process of converting the answers

in the questionnaire into the form of codes for input and processing in SPSS 23.0 and ADANCO 1.1 Clean data for detecting and handling errors that may occur in the answer set such as a blank box or an unreasonable answer

3.4.1 Data collection and assessment (Hair et al., 2014)

At this step, data was collected and assessed based on following criteria

- The questionnaire lacked of data: If the data missed more than 15% for a

questionnaire, it should be removed If the data is missing a lot of focus on a certain concept, this questionnaire is also irrelevant even if the missing data is less than 15%

- Doubtful questionnaire: should be eliminated if the questionnaire had answers on the

same likert scale, or only select the smallest or largest value

- Standard data distribution: PLS - SEM is a non-parametric statistical method, does

not require data to be distributed normally However, it should be noted that the data should not exceed the standard deviation because it would cause an inflated standard error of the bootstrap method and lead to the possibility that some relations are not significantly evaluated

3.4.2 Structural model PLS – SEM

Structural equation modeling (SEM) is a statistical technique family that is currently very popular in social and business sciences, capable of evaluating hidden variables indirectly through measurement of observed variables Currently, there are two types of