STATE BANK OF VIETNAM THE MINISTRY OF EDUCATION TRAINING HO CHI MINH UNIVERSITY OF BANKING PHAN NGUYEN NGOC BICH STUDENT CODE 030805170203 FACTORS AFFECTING CONSUMER INTENTION TO PURCHASE ORGANIC FO.
THESIS INTRODUCTION AND OVERVIEW
The necessity of the topic
Against the backdrop of ongoing economic integration and robust growth, living standards have steadily improved As households gain purchasing power, expectations for food quality and safety have risen, making rigorous standards and transparency in food sourcing essential The direct link between what people buy and consume and their health underscores the impact of food choices on overall quality of life Therefore, ensuring high-quality, safe, and nutritious food is a priority for sustainable economic development and consumer welfare.
Along with the development of science and technology, the application of technical achievements to the food production and processing process is inevitable However, this has led to certain hazards to the environment and people The use of chemical fertilizers, pesticides, veterinary drugs growth stimulants, etc have been causing serious effects to the living environment and human health In general, from
2010 to 2019, there were 1,556 cases of food poisoning nationwide From that, it can be seen that focusing on the issue of buying and using food is indispensable and necessary to protect consumers and their families
Organic food has become the preferred choice for many consumers, as products raised, handled, and prepared without artificial chemicals, growth hormones, or antibiotics are produced in ways that are environmentally safe According to AC Nielsen's Trends in Organic Food Consumption Report (2017), up to 86% of Vietnamese consumers choose organic products for daily meals because they are safe, nutritious, and delicious In addition, organic food may help reduce the risk of cancer, heart disease, and high blood sugar Vietnam Report (2019) indicates that the prospects for the Vietnamese organic food market are on par with the international market, with sales projected to reach 320.5 billion USD by 2025.
Consumers in Ho Chi Minh City (HCMC) tend to buy and use higher-quality foods than those in other areas, driven by heightened awareness of unsafe food risks and the importance of food quality to protect the health of themselves and their families This prioritization of safety influences shopping choices, encouraging demand for reliable labeling, traceability, and higher standards from retailers and producers operating in HCMC Consequently, the emphasis on safe, high-quality food supports healthier communities and underscores why clear, trustworthy food safety information matters to consumers in Ho Chi Minh City.
Ho Chi Minh City consumers who can afford organic food that meets their quality and safety standards form a viable market By identifying the factors that influence their purchase intention for organic food, businesses can optimize organic food distribution channels and supply chains, enabling more efficient operations and sustainable growth in Ho Chi Minh City's OF market.
Although numerous studies have examined the factors influencing consumer intention to buy organic food, in-depth research within Ho Chi Minh City (HCM) remains limited This gap motivates a focused investigation into the Factors Affecting Consumer Intention To Buy Organic Food in Ho Chi Minh City, aiming to identify how health consciousness, perceived quality and safety, price sensitivity, environmental concern, trust in certification, social influences, product availability, and marketing communications shape consumers’ willingness to purchase organic products in this urban market The findings will offer actionable guidance for retailers, certification bodies, and policymakers seeking to promote sustainable and healthier consumption patterns in Ho Chi Minh City and enhance organic food adoption in Vietnam.
Purchasing organic food in Ho Chi Minh City is an essential focus of this study, which seeks to identify the factors that influence consumers’ intention to buy organic food By uncovering these drivers, the research will offer actionable recommendations for manufacturers and businesses to better meet consumer needs and to drive growth and development of the organic food market in Ho Chi Minh City.
Research Objectives
Determining factors affecting consumer intention to purchase Organic Food in Ho Chi Minh City, thereby proposing useful management implications to improve consumer intention to buy OF in this area
Objective 1: Determining factors affecting consumer intention to purchase Organic Food in HCMC.
Objective 2: Evaluate the impact of the above factors.
Objective 3: Propose several implications for producers and enterprises in the OF sector to reach more consumers and improve their businesses.
Research Questions
Which factors affect consumer intention to buy organic food in HCMC?
How is the influence level of these factors?
Is there a difference between demographic factors on intention to buy organic food?
Subject and scope of the research
Research Subjects: Factors affecting consumer intention to purchase Organic Food in HCMC.
Survey Subject: Consumers living in HCMC intend to buy Organic Food to use for themselves and their families.
Research Duration: The study was conducted during the period from March 2022 to May 2022 Specifically: o Collecting research materials: from 01/03/2022 to 31/03/2022 o Collecting primary data for research: from 01/04/2022 to 08/05/2022
Research Location: Ho Chi Minh City.
Research Method
The research combining qualitative and quantitative methods is carried out in 2 phases, particularly:
Qualitative research was conducted after the proposed research model was built from the theoretical basis and previous studies The implementation method used group discussions with experts in marketing and food to adjust and supplement the independent variables affecting the dependent variable, customer intention to buy OF in HCMC Subsequently, the scales were calibrated and the questionnaire was completed.
This study employs quantitative research, collecting data through direct and indirect interviews using a questionnaire Non-probability sampling was used to select an adequate sample size from March to April 2022 Data were processed with SPSS 20.0 and analyzed through a sequence of steps: descriptive statistics; scale reliability testing with Cronbach's Alpha; Exploratory Factor Analysis (EFA) to reshape the scale structure; and correlation analysis along with linear regression to test the hypotheses and validate the model Finally, the study identified the factors affecting customer intention and built a complete research model.
Research Contribution
This study adds to academic research by complementing and strengthening existing studies on the factors influencing purchase intention and by constructing a robust research model to explain customer intention to buy organic food in Ho Chi Minh City Consequently, it expands the theoretical framework of food and organic food, contributing new data to the body of knowledge and providing a reference source for future related studies.
This thesis identifies the key factors influencing the intention to buy OF among consumers in Ho Chi Minh City and measures the relative impact of each factor on OF purchase intention It ranks determinants of OF buying behavior, clarifying which factors most strongly drive consumer decisions Based on these results, the author proposes governance implications to help producers and traders reach more customers and design effective policies that enhance OF purchase intention, thereby supporting targeted market growth in Ho Chi Minh City.
Thesis Structure
In addition to the introduction, references and appendices, the topic: "Factors Affecting Consumer Intention To Purchase Organic Food In Ho Chi Minh City" includes 5 chapters:
Chapter 1: Thesis Introduction and Overview
Chapter 1 sets out the rationale for selecting the topic and clearly defines the study’s objectives, research questions, and scope It specifies the research object, explains the chosen methodology, and describes data collection and analysis procedures The chapter also outlines the expected contributions to the field and provides a concise roadmap of the thesis structure to orient readers for the remaining chapters.
Chapter 2: Theoretical Foundation and Research Model
In Chapter 2, the author introduces the core concepts related to the topic and provides a concise review of the theoretical foundations, prior research, and the key research models associated with the thesis Building on this discussion, the author formulates research hypotheses and proposes a suitable research model to guide the study.
Chapter 3 outlines the research process and the methodologies used for both qualitative and quantitative research It explains that qualitative research is employed to develop research scales and questionnaires, while quantitative research relies on surveys administered to suitable subjects.
Chapter 4: Research results and discussion
Chapter 4 outlines and analyzes the research results, beginning with a description of the sample and an assessment of scale reliability It proceeds with exploratory factor analysis (EFA), followed by correlation analysis and regression analysis, and culminates in hypothesis testing Building on these analyses, the chapter discusses and evaluates the impact of the examined factors on the study outcomes, offering interpretation and implications for the research field.
Chapter 5 presents the thesis conclusion and practical management implications, highlights the study’s contributions, acknowledges the limitations encountered during the research process, and concludes with proposed directions for future research.
Chapter 1 offers a concise overview of the research topic, explains the rationale for topic selection, and clearly defines the research objective, questions, object, and scope It justifies a mixed-methods approach by combining qualitative and quantitative methods to produce comprehensive insights The chapter also highlights the study’s scientific and practical significance and outlines the expected structure of the thesis, laying the groundwork for a clear, systematic investigation.
In the next chapter, the author will refer to the theoretical bases related to the topic, briefly review related studies, how to establish hypotheses and build a proposed research model.
THEORETICAL BASIS AND RESEARCH MODEL
Concepts
Organic food (OF) refers to foods grown or processed for sale without synthetic chemicals, growth hormones, or antibiotics, according to Jones et al (2001) Building on this, Teng & Wang (2015) argue that OF is generally more nutritious, healthier, safer, and environmentally friendlier than conventional foods Brown (2021) provides the latest definition, describing OF as foods produced without artificial chemicals, hormones, antibiotics, or genetically modified organisms (GMOs) Collectively, these definitions depict organic food as produced without synthetic inputs or GMOs, with potential benefits in nutrition, safety, and environmental sustainability.
In order to be labeled organic, a food product must be free of artificial additives
In Vietnam, Decree No 109/2018/ND-CP on Organic Agriculture requires that organic foods be certified and properly labeled, with conformity assessments based on national standards for organic agriculture or applicable international, regional, or foreign standards used in the production of organic products.
For this thesis, organic food is defined as food that is grown and processed without artificial chemicals, growth hormones, or antibiotics, and that is certified and properly labeled in accordance with applicable regulations This definition emphasizes adherence to established organic standards and certification schemes to ensure product integrity and consumer trust.
Intention is a key predictor of future behavior and encompasses diverse forms, including purchase intentions (Blackwell et al., 2001) According to Ajzen (1991), intention motivates individuals to perform a behavior, and in the Theory of Planned Behavior, it is the central determinant of actual behavior, shaped by an attitude toward the behavior Rashid (2009) further emphasizes the crucial role of intention in linking attitudes to action for consumers.
"intent to buy organic food is the degree to which consumers prefer organic food to other conventional foods"
2.1.3 Concept of Consumers in Ho Chi Minh City
Consumers are people who purchase and use goods and services for personal consumption and daily-life purposes, including individuals, families, or organizations, as defined in Clause 1 of Article 3 of the Law on Protection of Consumer Rights 2010.
According to the Ho Chi Minh City Office of the People's Committee (2010), Ho Chi Minh City is located at about 10°10′–10°38′ North latitude and 106°22′–106°54′ East longitude, covering a total area of more than 2,095 square kilometers It borders Binh Duong province to the north, Tay Ninh province to the northwest, Dong Nai province to the east and northeast, Ba Ria–Vung Tau province to the southeast, and Long An and Tien Giang provinces to the west and southwest.
Consumers in Ho Chi Minh City are individuals, households, and organizations who purchase and use goods and services to support daily living and activities within Ho Chi Minh City's 2,095 square-kilometer area.
Theoretical basis
2.2.1 Theory of Reasoned Action - TRA
The Theory of Reasoned Action (TRA), originally proposed by Fishbein in 1967 and later refined by Ajzen and Fishbein in the 1970s, posits that actual behavior is largely determined by behavioral intention Behavioral intention, in turn, is shaped by two key factors: attitude toward the product and subjective norm Attitude toward the product reflects a consumer’s positive or negative beliefs about the product and is measured by perceived product attributes; the more benefits a consumer believes are associated with certain attributes, the stronger those attributes will influence purchase intention By evaluating and categorizing the importance of each product attribute, marketers can better predict and influence consumer choices.
Subjective norm in consumer behavior is measured by the influence exerted by people who have a close relationship with the consumer The closer the relationship, the greater its impact on purchase intention, reflecting how intimacy levels create varying strengths of influence across individuals In practice, each person’s influence can be strong or weak depending on how close the relationship is, which shapes the buyer’s decision process.
Fig 2.1 Theory of Reasoned Action (TRA) Model
2.2.2 Theory of Planned Behavior - TPB
The Theory of Planned Behavior (TPB) was developed by Ajzen in 1991, extending the Theory of Reasoned Action (TRA) TPB identifies three determinants of buying behavior: attitude toward the behavior, subjective norm, and perceived behavioral control Perceived behavioral control reflects how much control an individual feels they have over performing the purchase, based on their knowledge, skills, and available resources, and describes how difficult the behavior is perceived to be.
Fig 2.2 Theory of Planned Behavior (TPB) Model
Empirical Studies
Determinant factors influencing organic food purchase intention and the moderating role of awareness: A comparative analysis (Asif et al 2018)
Asif et al (2018) conducted a cross-country study in Pakistan, Turkey, and Iran, collecting 271, 245, and 220 responses respectively The study's model preserved the three TPB factors—Attitude, Subjective Norm, and Perceived Behavioral Control—and added Environmental Concern and Health Awareness The results show that all five factors positively influence the intention to buy organic food, with Health Awareness and Attitude emerging as the two strongest predictors of this purchase intention.
Fig 2.3 Model of Determinant factors influencing organic food purchase intention
Consumers perception, purchase intention and actual purchase behavior of organic food products (Wee et al., 2014)
Wee et al (2014) conducted a study in the Malaysian market, collecting 288 valid survey responses They proposed a four-factor model—Health Awareness, Safety Concern, Environmental Concern, and Perceived Quality—that influences the intention to buy organic food The findings show that all four factors significantly affect organic food purchase intention, with Safety Concern delivering the strongest impact among them.
Fig 2.4 Model of Consumers perception, purchase intention and actual purchase behavior of organic food products in Malaysia
Factors driving Thai consumers' intention to purchase organic foods (Pomsanam et al., 2014)
The study was analyzed with 400 valid survey responses Accordingly, the proposed model of the study includes 5 factors that have impact on the intention to buy organic food: Convenience & Availability, Environmental Concern, Subjective Norm, Brand Trust and Perceived Quality The research results show that, Subjective Norm and Environmental Concern are the two factors that have the greatest influence on the dependent variable of the model
Fig 2.5 Model of Factors driving Thai consumer intention to purchase organic foods
Drivers of organic food consumption in Greece (Kulikovski et al., 2010)
Kulikovski et al (2010) conducted a study on Greek consumer intention to buy organic food in several regions The proposed model includes 7 independent variables: Health Awareness, Perceived Quality, Perceived Value, Ethical Concern, Food Safety Concern, Perceived Price and Brand Trust Analytical results from 190 valid survey responses were collected, showing that the proposed factors all have a certain impact on the intention to buy OF In which, Food Safety Concern has the largest impact, followed by Perceived Quality and Perceived Value
Fig 2.6 Model of Drivers of organic food consumption in Greece
An Investigation of Consumer Buying Behavior for Food Products (Mor & Sethia, 2015)
Mor & Sethia (2015) conducted an empirical study on “Consumer purchasing behavior for food products in rural and urban areas of Haryana state” with a research sample of 500 consumers in cities of Haryana state, northern of India The research results show that the factors affecting the food buying behavior of consumers in this region are: Perceived Quality, Brand Trust, Health Awareness and Other factors (Advertising, Packaging, Pricing, Availability, Variety )
Health Awareness Perceived Quality Perceived Value Ethical Concern Food Safety Concern
Fig 2.7 Model of Consumer Buying Behavior for Food Products in Haryana
Factors affecting the intention to buy organic food of office workers in HCMC (Lien, 2021)
Based TPB theory, Lien (2021) has identified the factors Subjective Norm and Attitude that affect the intention to buy OF By reviewing other related studies, the author added Environmental Concern, Health Awareness and Domestic Origin to the research model Research results with 206 valid survey responses collected by convenience sampling method at company offices in central districts of HCMC show that all 5 factors positively impact on the intention to buy organic food
Fig 2.8 Model of Factors affecting the intention to buy organic food of office workers in HCMC
Factors affecting the intention to buy organic food of urban Vietnamese consumers (Hung, 2019)
The author has built a proposed model consisting of 5 main independent variables
In which, the author keeps 3 variables from the TPB model, namely Attitude, Subjective Norm and Perceived Behavioral Control and added two new variables: Price Perception and Brand Trust & Certifications In addition to the above 5 independent variables, the author thinks that there is a difference between the groups under the Control Variables (Age, Education, Gender, Income) to the customer intention Besides, according to the author, the Attitude variable will be affected by 3 sub-variables: Environmental Concern, Self Perception and Health Awareness The research results show that Perceived Behavioral Control is the factor that has the greatest impact on the intention to buy OF
Attitude Subjective Norm Environmental Concern
Fig 2.9 Model of Factors affecting the intention to buy organic food of urban Vietnamese consumers
Factors affecting consumer intention to buy organic food in Ho Chi Minh City (Minh, 2016)
After synthesizing and analyzing previous relevant domestic and foreign studies, the author believes that the intention to buy organic food depends on 5 variables: Food Safety Concern, Health Awareness, Environmental Concern, Perceived Quality and Perceived Price Besides, the author also thinks that there is a difference between Demographic Variables (Gender, Age, Education, Income) on Intention to buy OF The analysis results from 200 survey responses that are people over 18 years old who intend to buy healthy products in HCMC show that all 5 variables are accepted because they all influence on the intention to buy OF In which, the variables Food Safety Concern, Health Awareness, Environmental Concern and Perceived Quality have positive impact on dependent variable In contrast, the Perceived Price variable has a negative effect At the same time, the author also found that there is a difference in Age and Income to consumer intention
(Age, Education, Gender and Income) Attitude
Subjective Norm Perceived Behavioral Control
Demographic Variables (Gender, Age, Education, Income) Food Safety Concern
Fig 2.10 Model of Factors affecting consumer intention to buy organic food in HCMC
Factors affecting the intention to buy safe vegetable products of urban residents in HCMC area (Trang, 2017)
The author has proposed a research model consisting of 6 factors affecting the intention to buy safe vegetables, including: Health Awareness, Safety Concern, Perceived Quality, Subjective Norm, Environmental Concern and Perceived Price After distributing 300 questionnaires, the author collected 235 valid survey responses Quantitative research results show that, except for the Environmental Concern variable is removed, the remaining 5 variables all have an impact on the dependent variable In which, the factor Safety Concern and Health Awareness have the greatest positive impact on the intention to buy safe vegetables and Perceived Price factor is the only variable with negative effect
Fig 2.11 Model of Factors affecting the intention to buy safe vegetable products of urban residents in HCMC area
Factors affecting consumer intention to buy safe food in HCMC (Hien,
The author has proposed 7 factors that affect the intention to buy safe food of consumers in HCMC, including: Health Awareness, Environmental Concern, Reference Group, Subjective Norm, Perceived Quality, Perceived Availability, and Perceived Price Research results show that Health Awareness has the greatest impact on the dependent variable with the coefficient β = 0.835
Gender and Income) Health Awareness
Safety Concern Perceived Quality Subjective Norm
Intention to buy safe vegetable products
Fig 2.12 Model of Factors affecting consumer intention to buy safe food in HCMC
Summary of relevant theoretical basis and research models
2.4.1 Summary of relevant theoretical basis
Based on the theories and previous studies related to the intention to buy OF presented, the author has summarize the theoretical and research models reviewed in Table 2.1 below., to select suitable research references for this study
Table 2.1 Summary of relevant theoretical basis and research models
Behavioral Beliefs Outcomes Evaluation Attitude
Normative Beliefs Motivation to Comply Subjective Norms
Reference Group Subjective Norm Perceived Quality
Consumer intention to buy safe food
Attitude Subjective Norms Perceived Behavioral Control Ajzen (1991) Behavioral Intention
Determinant factors influencing organic food purchase intention
Attitude Subjective Norm Perceived Behavioral Control Environmental Concern Health Awareness
Consumers perception, purchase intention and actual purchase behavior of organic food products
Health Awareness Safety Concern Environmental Concern Perceived Quality
Factors driving Thai consumers' intention to purchase organic foods
Convenience & Availability Environmental Concern Subjective Norm Brand Trust Perceived Quality
6 Drivers of organic food consumption in Greece
Health Awareness Perceived Quality Perceived Value Ethical Concern Safety Concern Perceived Price Brand Trust
Consumer Buying Behavior for Food Products: An Empirical
Study of Rural and Urban
Perceived Quality Brand Trust Health Awareness Other factors
Factors affecting the intention to buy organic food of office workers in
Attitude Subjective Norm Environmental Concern
Factors affecting the intention to buy organic food of urban Vietnamese consumers
Environmental Concern Self Perception Health Awareness Attitude Subjective Norm Perceived Behavioral Control
Factors affecting consumer intention to buy organic food in Ho
Safety Concern Health Awareness Environmental Concern Perceived Quality Perceived Price
Factors affecting the intention to buy safe vegetable products of urban residents in HCMC area
Health Awareness Safety Concern Perceived Quality Subjective Norm Environmental Concern Perceived Price
Factors affecting consumer intention to buy safe food in HCMC
Health Awareness Environmental Concern Reference Group Subjective Norm Perceived Quality Perceived Availability Perceived Price
Source: Compiled by the author, 2022
In general, the studies that have been examined on the intention to buy OF mostly the TPB model as the theoretical foundation At the same time, each study developed and analyzed several other independent variables to adjust and suit the research context and object, the region reality, etc Therefore, the author will choose the TPB model as the foundation and analyze some additional suitable factors to include in the proposed model for this research “Factors Affecting Consumer Intention To Purchase Organic Food In Ho Chi Minh City”.
Developing hypotheses and proposed research model
2.5.1 Scientific basis to develop suitable hypotheses for the topic
Based on the the relevant research models that have been presented, the author finds that the TPB model has been used in many studies on purchase intention in general and the intention to buy OF in particular At the same time, TPB is a theory that is more inclined to predict behavioral intentions and real human behavior than the TAM theory and use UTAUT theory, so TPB will be suitable to study direct purchase rather than applied research in the use of technology or information systems Therefore, the author will use the TPB theory as the basis for the research According to this, the study will use TPB theory as the foundation and keep 3 factors namely "Attitude", "Subjective Norm" and "Perceived Behavioral Control" However, according to Ajzen (1991), it is necessary to modify and add other factors to suit each topic and research period Therefore, in order to build a suitable model, the author needs to add other factors
From the relevant studies have been reviewed, the factor "Environmental Concern" appeared in 8/12 studies, the factor "Health Awareness" appeared in 9/12 studies., the factor "Perceived Quality" appeared in 7/12 studies These are the three factors with the highest frequency in the investigated studies (over 50%) This shows that, through different economic periods, these factors have a certain influence on consumer intention to buy OF Therefore, the author decided to add the those factors to the research model
Besides, the factor "Perceived Price" appeared in 5/12 studies Although the occurrence frequency of this factor is lower than the 3 selected factors, it still has a higher frequency than the other factors At the same time, price is always a factor that consumers care about, especially in the current economy This shows that this is also a necessary factor to be included in the research model for further study and analysis Therefore, the author will also add “Perceived Price" to the research hypothesis for the topic
Based on Table 2.1, the author summarizes the occurrence frequency of the factors selected through Table 2.2 to see that these are the necessary factors to bring in the research model
Table 2.2 Summary of the factors occurrence frequency
Source: Compiled by the author, 2022
From the scientific bases analyzed above, the author has selected 7 factors as independent variables that have an impact on the dependent variable Thereby, proceed to build research hypothesis based on selected factors
Attitude is defined by Fishbein & Ajzen (1977) as an individual's positive or negative feelings towards performing a particular behavior Ajzen (1991) also said that attitude is a factor that directly affects the intention to perform the behavior Thus, the attitude towards the OF purchase intention is the consumer's positive or negative perception towards the behavior of purchasing OF The related studies have all proven that Attitude has a positive impact on OF purchase intention Therefore, in this study, Attitude is defined as the level of positive assessment towards consumers' intention to buy OF in HCMC From there, the following hypothesis is built:
Hypothesis H1: Attitude (ATT) has a positive effect (+) on consumer intention to buy OF in HCMC
Subjective Norm are measured by those who have close relationship with consumers (Fishbein & Ajzen, 1977) Consumers often tend to perform desired behavior from those who are close to them All related show that Subjective Norm have a positive impact on the intention to buy OF From there, the following hypothesis is built:
Hypothesis H2: Subjective norm (SN) has a positive effect (+) on consumer intention to buy OF in HCMC
Perceived behavioral control is defined by Ajzen (1991) as an individual's perception of the ease or difficulty of performing a particular behavior and it has a lot to do with an individual's knowledge and skills Accordingly, perceived behavioral control has a direct impact on the intention to perform the behavior (Ajzen, 1991) Through the results of a review of related studies, perceived behavioral control is the factor that has the greatest impact on the intention to buy
OF in the study Accordingly, when consumers feel they have enough ability and resources to buy OF, their intention to buy OF will increase (Hung, 2019) Therefore, Perceived Behavioral Control has a positive effect on intention to buy
OF From there, the following hypothesis is built:
Hypothesis H3: Perceived behavioral control (PBC) has a positive effect (+) on consumer intention to buy OF in HCMC
Environmental Concern is the degree to which an individual is aware of environmental problems and is willing to contribute his or her own efforts to the implementation of solutions to those problems (Dunlap & Jones, 2002) According to Santa (2007), environmental concern is considered an individual's attitude towards the environment This attitude includes beliefs and attitudes about the environment, through which behaviors towards the environment are carried out Environmental concerns are analyzed quite a lot in related studies such as the study of Asif et al (2018), Wee et al (2014), Lien (2021), Trang (2017) The results of these studies all show that Environmental Concern has a positive impact on intention to buy OF From there, the following hypothesis is built:
Hypothesis H4: Environmental Concern (EC) has a positive effect (+) on consumer intention to buy OF in HCMC
Health Awareness is defined as the concern about one's own health status, while focusing on improving health and maintaining the best quality of life (Kraft & Goodell, 1993) Suh et al (2012) suggest that using OF will help improve consumers' health more than other common foods Therefore, this factor appears quite a lot in relevant researches such as the studies of Asif et al (2018), Kulikovski et al (2010), Lien (2021), Minh (2016) All these studies have shown that Health Awareness has a positive impact on intention to buy OF From there, the following hypothesis is built:
Hypothesis H5: Health Awareness (HA) has a positive effect (+) on consumer intention to buy OF in HCMC
Perceived quality is the way in which consumers perceive a product through quality indicators (Olson, 1977) These indicators include both internal (features, shape, size ) and external (brand, price, origin ) Accordingly, the perception of quality for OF related to the attributes of organic products such as taste, freshness, experience and enjoyment (Kulikovski et al 2010) The studies on the intention to buy OF included the Perceived quality such as the studies of Wee et al (2014), Kulikovski et al 2010), Minh (2016), Trang (2017)… These studies have all proved that Perceived quality has a positive impact on the intention to buy OF From there, the following hypothesis is built:
Hypothesis H6: Perceived quality (PQ) has a positive effect (+) on consumer intention to buy OF in HCMC
Price perception is the perception of the cost that consumers must pay to obtain the
OF product In the studies related to OF that have been examined, especially in the domestic studies, price perception is considered as a "barrier", the main obstacle in not buying OF Specifically, researches by Kulikovski et al (2010), Minh (2016), Trang (2017) all show that Perceived Price has a negative impact on the OF purchase intention From there, the following hypothesis is built:
Hypothesis H7: Perceived price (PP) has a negative effect (-) on consumer intention to buy OF in HCMC
Demographic factors such as gender, age are considered by Venkatesh et al
(2003) to have an impact on the factors affecting the intention to perform the behavior Therefore, in addition to the gender and age factors, the author adds 3 factors such as education level, occupation and income in the group of Demographic factors to check if there is any differences between the 5 factors on consumer intention to buy OF Therefore, the Demographic variable will also be added to the proposed research model
The proposed research model of the topic is based on the theoretical foundation of the TPB theory and the Model of Determinant factors influencing organic food purchase intention (Asif et al., 2018) Thereby, maintaining the 5 factors: Attitude, Subjective Norm, Perceived Behavioral Control, Environmental Concern and Health Awareness Besides, after reviewing other studies, the author decided to add
2 more independent variables that are Perceived Quality and Perceived Price In
Demographic Variables (Gender, Age, Education,
Perceived Price addition, Demographic Variables are also included in the model to test the difference between these groups of factors on consumer intention to purchase OF
Chapter 2 has presented the concepts, the theoretical basis of the consumer's intention to buy OF At the same time, the author also reviewed previous studies related to the intention to buy OF Since then, the author has presented the scientific basis for building appropriate hypotheses and the process of building research hypotheses Thereby, the author has chosen the TPB model as a theoretical foundation and added variables suitable to the topic to build the research model consisting of 7 independent variables that impact on the dependent variable.
Proposed model and research hypotheses
Expert group discussion Descriptive statistics
In-depth consumer interviews Exploratory factor analysis
Results discussion and governance implications
RESEARCH METHODOLOGY
Research process
The research process is designed by the author as Figure 3.1 below
Qualitative research
Qualitative research was carried out in order to discover, adjust, add or reduce the elements in the research model and observed variables of the scale The qualitative research results are the basis to build the questionnaire for the quantitative research The qualitative research is carried out through the following steps:
- Step 2: Team discussion with experts to make 1st scale adjustment
- Step 3: In - depth interview with consumers use with adjusted scale for the second time (if necessary)
- Step 4: Complete the final scale
The construction of a preliminary scale is referenced based on the scale of previous relevant research models in Chapter 2 Accordingly, the preliminary scale in this study is inherited mainly from 3 studies of Asif et al (2018); Pomsanam et al (2014) and Kulikovski et al (2010)
The original scales used for reference to build the preliminary scale are detailed in
From there, the author built the preliminary scale shown in Table 3.1 below
No Encrypted Scale Contents Source
1 ATT1 I think that purchasing organic food is a good idea
2 ATT2 I think that purchasing organic food is interesting
3 ATT3 I think that purchasing organic food is important
4 ATT4 I think that purchasing organic food is beneficial
5 ATT5 I think that purchasing organic food is wise
6 ATT6 I think that purchasing organic food is favorable
My family think that I should buy organic food rather than non-organic food
Most friends whose opinions regarding diet are important to me, think that I should buy organic food
People important to me (other than family, friends), such as teachers, doctors, well-known people think I should buy organic food
10 SN4 I intend to buy organic food because I follow my family
11 SN5 I intend to buy organic food because I follow my friends
12 SN6 I intend to buy organic food because I follow people important to me
13 SN7 I intend to buy organic food because society says it as a good choice
14 PBC1 If I wanted to, I could buy organic food instead of non-organic food
15 PBC2 I think it is easy for me to buy organic food
16 PBC3 It is mostly up to me whether or not to buy organic food
17 EC1 Buying organic foods reduces pollution problems
18 EC2 Buying organic foods affects plants and animals
19 EC3 Buying organic foods helps to save the environment
Organic foods is environmentally friendly
21 HA1 I consider myself as health conscious consumer Asif et al (2018)
22 HA2 I think often about health related issues
23 HA3 I consume organic food because it is healthier than the conventional one
24 HA4 In the long-run, people who take care of themselves stay healthy
25 HA5 I think that I protect my health by consuming organic food
26 PQ1 I think that organic food is of high quality Kulikovski et al
27 PQ2 I think organic food is of higher quality than conventional food
28 PQ3 Organic foods look fresh and clean Pomsanam et al
29 PQ4 Organic foods is safe to eat
30 PQ5 Organic food is less associated with health risks Kulikovski et al
31 PQ6 I feel I am getting higher quality food with organic products
32 PP1 The price for organic food is high
33 PP2 Organic food is expensive
34 PP3 I don't mind paying more for the organic food
35 PP4 It is important me to get the best price when I buy foods
Organic Food Purchase Intention (OFPI)
36 OFPI1 I intend to buy organic foods if they are available
37 OFPI2 I am willing to purchase organic foods if they are available
38 OFPI3 I plan to consume organic foods if they are available for purchase
39 OFPI4 I will try to consume organic foods if they are available for purchase
Group discussion with experts to discover, add or reduce factors in general or observed variables in particular that affect consumers' intention to buy OF in
HCMC Thereby, adjust the model and hypothesis as well as develop the observed variables to match the actual topic
The author conducts group discussions with 5 experts who are experienced, knowledgeable about the market as well as being consumers in the field of organic food for at least 2 years The list of members participating in the group discussion is in Appendix 2, including 5 experts
Discussion content: Discuss the hypotheses and the observed variables in each hypothesis that have an impact on consumers' intention to buy OF Details of the discussion are presented in Appendix 4: Group discussion outline
Conduct a group discussion between the study author and a group of 5 experts who have been selected to participate in qualitative research to collect relevant data
Based on the collected data, check and correct the model, hypothesis and observed variable scales
Corrected data will be discussed with the expert group again The group discussion ends when no new changes are found.
The experts agreed with all 7 independent factors that the author gave in the preliminary scale do affect the dependent factor However, experts also added and reduced some observed variables, edited some statements to make it easier for the respondents to understand
After discussing again with the revised changes, the experts agreed with the results of the group discussion Based on the group discussion results, the preliminary scale has the following changes:
In this factor, the preliminary scale has 6 observed variables The author edited ATT6 variable:
Editing ATT6 “I think that buying OF is favorable” to “I think that I will be satisfied with buying OF”.
In this factor, the preliminary scale has 7 observed variables The author edited 2 variables as follow SN4 and SN5
Editing SN4 “I intend to buy OF because I follow people who are important to me” to “I intend to buy OF because I follow my family, friends and people who are important to me
Editing SN5 “I intend to buy OF because society says it as a good choice” to “ I intend to buy OF because society (press, Internet,…) says it as a good choice
Perceived Behavioral Control (PBC) scale:
In this factor, the preliminary scale has 3 observed variables The author edited PBC3 and PBC4:
Editing PBC3 “It is mostly up to me whether or not to buy OF” to “The purchase of OF is under my control”
Editing PBC4 “I have enough awareness and ability to buy OF” to “I have enough awareness (information about products, services, stores,…) and ability (finance, time, means,…) to buy organic food
The preliminary scale has 4 observed variables The author edits the wording of the EC2 and EC5 variable Therefore, the adjusted scale will have 5 observed variables Specifically:
Editing EC2 “Buying OF affects plants and animals” to “Buying OF has positive environmental impacts”
Editing EC5 "I think environmental issues are needed to draw people’s attention" to “I think environmental issues in Vietnam are needed to draw people’s attention”
In this factor, the preliminary scale has 5 observed variables The author edited the wording of HA4 and HA5:
Editing HA4 “In the long-run, people who take care of themselves stay healthy” to “I believe that using OF can reduce health risks”
Editing HA5 “I think that OF can protect my health” to “I have intention to buy
OF in order to improve my health” to make statements clearer
The preliminary scale has 6 observed variables The author edited words in variables as follow PQ4, PQ5 and PQ6 Specifically:
- Editing PQ4 “OF is safe to eat.” to “OF is safe and delicious to eat”
Editing PQ5 “OF is less associated with health risks” to “The quality of OF is directly related to health”
Editing PQ6 “I believe consuming OF can better my life” to “I believe that consuming OF can improve my quality of life”
The preliminary scale has 4 observed variables The author edited the wording of PP2, PP3 and PP4 Specifically:
Editing PP2 “OF is not as cheap as CF” to “OF is more expensive than CF”
Editing PP3 “I am not willing to pay for OF” to “I am afraid of paying more to get OF”
Editing PP4 "It is important me to get the best price when buying OF" to
"Reasonable price is important to me when buying OF"
Organic Food Purchase Intention (OFPI) scale:
In this dependent variable, the preliminary scale has 4 observed variables The author edited words in 2 variables as follow OFPI1 and OFPI3 Specifically:
Editing OFPI1 “I intend to buy OF if they are available” to “I intend to buy OF in the near future”
Editing OFPI3 “I am planning to buy OF because they are beneficial to me and the environment” to “I am planning to buy organic food due to its benefits brought to me and the environment
3.2.3 In-depth interviews with consumers
In-depth interview with consumers is a process of trial interviewing with consumers using re-calibrated scales and observed variables after group discussion With the goal of editing and correcting the statements again with the customers who will be officially surveyed so that the statements are clearer and easier to understand Finally, build the official scale and the questionnaire
The in-depth interview subjects were selected according to the convenient method but still showed the characteristics of the sample Therefore, the author conducts in-depth interviews with 10 consumers who have had experience in purchasing OF at least 1 year The list of members participating in the in-depth interview is in Appendix 3, including 10 consumers
3.2.3.3 Contents of the in-depth interview (Appendix 4: Outline of the in- depth interview)
Contents of in-depth interviews: Discussing the statements of each observed variable that have an impact on consumer intention to buy OF Details of the content of the in-depth interview are presented in Appendix 5: Outline of the in- depth interview
Using the calibrated scale after group discussion with 5 experts for in- depth interviews with 10 consumers
Record comments and correct statements
Corrected data will be exchanged with the in-depth interview subjects again
The group discussion ends when no new changes are found
After the in-depth interview process, in general, the opinions agree with the research hypotheses and the observed variables have been adjusted after group discussion with experts Therefore, there is no need to add, remove or modify any observed variables
3.2.4 Building the official scale and questionnaire
After conducting qualitative research, based on the results of calibration after group discussions with experts and in-depth interviews with appropriate consumers, the author has built the final scale as Table 3.2 below
No Symbol Preliminary scale Official scale
1 ATT1 I think that buying organic food is a good idea
I think that buying organic food is a good idea
2 ATT2 I think that buying organic food is interesting
I think that buying organic food is interesting
3 ATT3 I think that buying organic food is important
I think that buying organic food is important
4 ATT4 I think that buying organic food is beneficial
I think that buying organic food is beneficial
5 ATT5 I think that buying organic food is wise
I think that buying organic food is wise
6 ATT6 I think that buying organic food is favorable
I think that I will be satisfied with buying organic food
SN Subjective Norm Subjective Norm
My family think that I should buy organic food rather than non-organic food
My family think that I should buy organic food rather than non- organic food
Most friends whose opinions regarding diet are important to me, think that I should buy organic food
Most friends whose opinions regarding diet are important to me, think that I should buy organic food
People important to me (other than family, friends), such as teachers, doctors, well-known people think I should buy organic food
People important to me (other than family, friends), such as teachers, doctors, well-known people think I should buy organic food
I intend to buy organic food because I follow people who are important to me
I intend to buy organic food because I follow my family, my friends and people who are important to me
I intend to buy organic food because society says it as a good choice
I intend to buy organic food because society (press, Internet,…) says it as a good choice
PBC Perceived Behavioral Control Perceived Behavioral Control
If I wanted to, I could buy organic food instead of non- organic food
If I wanted to, I could buy organic food instead of non-organic food
13 PBC2 I think it is easy for me to buy organic food
I think it is easy for me to buy organic food
14 PBC3 It is mostly up to me whether or not to buy organic food
The purchase of organic food is under my control
15 PBC4 I have enough awareness and ability to buy organic food.
I have enough awareness (information about products, services, stores, ) and ability (finance, time, means,…) to buy organic food
EC Environmental Concern Environmental Concern
16 EC1 Buying organic foods reduces pollution problems
Buying organic food reduces pollution problems
17 EC2 Buying organic foods affects plants and animals
Buying organic food has positive environmental impacts
18 EC3 Buying organic food helps to save the environment
Buying organic foods helps to save the environment
19 EC4 Organic food is environmentally friendly
Organic foods is environmentally friendly
I think environmental issues are needed to draw people’s attention.
I think environmental issues in Vietnam are needed to draw people’s attention
HA Health Awareness Health Awareness
21 HA1 I consider myself as health conscious consumer
I consider myself as health conscious consumer
22 HA2 I am often worried about health issues
I am often worried about health issues
I consume organic food because it is healthier than the conventional one
I consume organic food because it is healthier than the conventional one
24 HA4 In the long-run, people who take care of themselves stay healthy
I believe that using organic food can reduce health risks
25 HA5 I think that organic food can protect my health
I have intention to buy organic food in order to improve my health
PQ Perceived Quality Perceived Quality
26 PQ1 I think that organic food is of high quality
I think that organic food is of high quality
27 PQ2 I think organic food is of higher quality than conventional food
I think organic food is of higher quality than conventional food
28 PQ3 Organic food looks fresh and clean Organic food looks fresh and clean
29 PQ4 Organic food is safe to eat Organic food is safe and delicious to eat
30 PQ5 Organic food is less associated with health risks
The quality of organic food is directly related to health
I believe that consuming organic food can better my life I believe that consuming organic food can improve my quality of life
PP Perceived Price Perceived Price
32 PP1 The price for organic food is high The price for organic food is high
33 PP2 Organic food is not as cheap as conventional food
Organic food is more expensive than conventional food
34 PP3 I am not willing to pay for organic food
I am afraid of paying more to get organic food
35 PP4 It is important for me to get the Reasonable price is important to best price when buying organic food me when buying organic food
Intention Organic Food Purchase Intention
36 OFPI1 I intend to buy organic food if they are available
I am planning to buy organic food in the near future
37 OFPI2 I am willing to buy organic food in the near future
I am willing to buy organic food in the near future
I am planning to buy organic food because they are beneficial to me and the environment
I am planning to buy organic food due to its benefits brought to me and the environment
39 OFPI4 I will buy organic food regularly
I will buy organic food regularly
40 OFPI5 I will introduce organic food to other people.
I will introduce organic food to other people
Based on the official scale has been built with a total of 40 observed variables In which, there are 35 observed variables of 7 independent variables and 5 observed variables of 1 dependent variable The author will design the final questionnaire to conduct quantitative research
Measuring tools: The author will use the interval scale to measure the observed variables because this scale brings high accuracy and is widely used in statistical analysis studies Therefore, the author uses a 5-level Likert scale to assess the degree of agreement on the statements of each observed variable and measures for the elements in the model The respective levels of measurement are as follows:
1 = Strongly disagree; 2 = Disagree; 3 = Normal; 4 = Agree; 5 = Strongly agree
Questionnaire Design: the questionnaire consists of 4 parts:
- Part 2: Refined questions and information questions about qualitative variables
- Part 3: The main body includes questions about the factors of analysis
Detailed questionnaire is presented in Appendix 6: Table of survey questions
Quantitative research
Quantitative research is the official research of the topic, done through data collection from the questionnaire at the qualitative research step The collected information will be processed and analyzed to test the research model, the scales as well as the hypotheses
3.3.1.1 Sampling scope and data type
The research sampling scope are consumers have bought OF or intending to buy OF in HCMC Therefore, the study uses primary data which is collected for the first time by the author
According to Hair et al (2006), the number of samples must be at least 5 times the number of observed variables, n ≥ 5k, where n is the number of samples to be surveyed, k is the number of observed variables Therefore, the minimum sample size for this study is 200 samples for 40 observed variables
Besides, in linear regression analysis, Tabachnick and Fidell (1991) suggested that the sample size is calculated according to the formula n ≥ 8m + 50, where: n is the number of samples to be surveyed, m is the number of independent variables Therefore, the minimum sample size with 7 independent variables of this study is 106 samples
Thus, from the above 2 sample calculation methods, it can be seen that the minimum number of samples for this study is 200 samples In order to increase objectivity and achieve higher accuracy, the author decided to choose a research sample size of 300 samples To achieve the set number of samples, the author will distribute 400 survey questionnaires
Sampling method: The sampling method of this study is convenience non- probability sampling method
Data collection method: The data is collected through interviews using the official questionnaire with the appropriate survey subjects who are consumers who have bought OF or are intending to buy OF in HCMC o For direct collection: Direct collection by face-to-face interview
Questionnaire is printed and broadcasted directly to the survey subjects to get comments by filling in and recording it (120 sheets were distributed) o For indirect collection: Indirect collection by online survey, questionnaire is created on Google Form, then sent to survey subjects via email address and social networking sites (280 forms were sent)
Survey period: From April 1, 2022 to May 8, 2022
Collected results: With a total of 400 survey tables sent, the author has collected 300 valid research samples, satisfying the designed sample size requirements
After collecting information and filtering all 300 research samples, the author processed and analyzed the data through SPSS 20.0 software according to the following steps:
Step 2: Test the reliability of Cronbach's Alpha to eliminate garbage variables with small-sum correlation coefficients.
Step 3: Anthropological analysis to discover EFA
Step 6: Test the research hypotheses and the difference of Demographic factors to the quantitative variables in the model.
Descriptive statistics are performed for the purpose of statistics on the frequency and percentage of the general information of the survey object to evaluate the characteristics of the consumer At the same time, test the difference in the level of evaluation through each feature of the Demographic groups for the quantitative variable of the model.
3.3.3.2 Check the scale reliability with Cronbach's Alpha
Cronbach's Alpha assesses the reliability of the scale to eliminate inappropriate variables According to Hair et al (2006), variables with item-correlation coefficients less than 0.3 will be excluded According to many researchers, Cronbach's Alpha of 0.8 or more is a good scale However, Trong & Ngoc (2008) added that in case the concept under study is new or new to respondents in the context of research, it is acceptable to accept Cronbach's Alpha from 0.6 above Many studies believe that Cronbach's Alpha from 0.8 or higher is a good scale, from 0.7 to 0.8 is usable, research with new contexts can use 0.6 or higher
Thereby, in this study, the author will keep the scale with Cronbach's Alpha from 0.7 or higher and the variable-total correlation greater than 0.3
Factor analysis is a statistical analysis method used to reduce a set of many observed variables into a group to make them more meaningful, but still does not contain most of the information content of the original variable Use Principle Component extraction, use Varimax rotation to group elements Then proceed to look at the following metrics:
Factor loading is an index showing the correlation between variables and factors, this coefficient indicates the close relationship between variables This coefficient must satisfy the condition that it is greater than 0.5 (Tho, 2011)
The KMO coefficient (Kaiser-Mayer-Olkin) meets the condition 0.55 ≤ KMO ≤ 1 to ensure the appropriate factor analysis with the research data
The Bartlett test considers the hypothesis H0 that the correlation between the observed variables is zero in the population If this test has statistical significance (sig < 0.05), then the hypothesis H0 is rejected, that is, the observed variables are correlated with each other This proves the data is suitable for factor analysis (Trong & Ngoc, 2008)
The breakpoint when extracting factors with Eigenvalue coefficients (representing the variation explained by each factor) must be greater than 1 to show that the extracted factor has good summary meaning (Gerbing & Anderson,1998)
The scale is accepted with the total variance extracted equal to or greater than 50% (Nunnally & Bernstein, 1994, extracted from Tho & Trang, 2004)
In addition, the difference between the factor loading coefficient of an observed variable among the factors is larger than or equal to 0.3 to create discriminant between factors (Jabnoun & Al-Tamimi, 2003)
Pearson's correlation analysis was performed to determine the statistically significant relationships between the independent and dependent variables Pearson's absolute value (r) indicates how tight the linear relationship is The closer the r value is to 1, the more closely the two variables are correlated and vice versa
In addition, correlation analysis helps to timely detect correlations between independent variables because it can greatly affect the results of regression analysis such as causing multicollinearity (Trong & Ngoc, 2008).
RESEARCH RESULTS AND DISCUSSION
Sample descriptive statistics
After collecting samples from consumers who have bought OF or are intending to buy OF in HCMC, with a total of 400 survey samples distributed, the author obtained 300 valid survey samples This is consistent with the sample size that has been designed with the appropriate survey object, so it is possible to continue to perform descriptive statistics and data analysis The author uses SPSS software to make descriptive statistics of the sample structure to get an overview of the information of the research sample This will show through statistics describing qualitative and quantitative factors
4.1.1 Descriptive statistics on qualitative variables
The results after descriptive statistics of the qualitative variables on Demographics shown in Table 4.1 show that out of 300 consumers participating in the survey, there are 176 women (58.7%) and 124 men (41.3%) The majority of respondents are aged between 18-22 years old (26.7%) and 30-35 years old (27.3%) Education level College - University accounted for the highest rate with 66.7% The occupations of the survey subjects are mainly Students (22.7%), Office workers (14.7%) and Public employees (13.7%) Income level from 5 to less than 10 million accounted for the highest rate with 35.3%
For details, see the descriptive demographic statistics in Appendix 7A
Table 4.1 Descriptive Statistics for Demographics
The results of descriptive statistics on the perception of buying OF in Table 4.2 show that out of 300 consumers participating in the survey, 206 have bought OF (accounting for 68.7%) and 94 are intending to buy (accounting for 31.3%) The distribution channels that are often preferred are Supermarkets (80.0%) and retail stores (78.7%) The issues that the survey respondents are most interested in when buying OF are Quality (98.3%) and Price (91.3%) Details are in Appendix 7B
Table 4.2 Descriptive statistics on the perception of buying OF
Experience Have bought OF or are planning to buy OF 300 100
Stores that specialize in OF 236 78.7%
Buy online through the website of OF stores 123 41.0%
Buy online via websites of supermarkets/convenience stores 220 73.3% Buy online through e-commerce sites 91 30.3%
4.1.2 Descriptive statistics on quantitative variables
The descriptive statistical results of the quantitative variables in the model show that the mean values of all observed variables are greater than 3 (above the mean), showing that the tendency to agree with the statements is high Therefore, the data is "valid" and the observed variables all play an important role In which, the PBC2 “I think it is easy for me to buy OF” has the highest mean value and the SN1
“My family thinks that I should buy OF rather than non-organic food” has the lowest mean value (Appendix 7C for details).
Cronbach's Alpha reliability test
Evaluation criteria: Based on the results of section 3.3.3.2 of chapter 3, the author will keep the scale with Cronbach's Alpha coefficient of 0.7 or higher and variable- total correlation greater than 0.3 to show that the scale and the variable under consideration have acceptable and further analysis
Table 4.3 Results of Cronbach's Alpha test
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted
Subjective Norm (SN): Cronbach's Alpha = 0.830
Perceived Behavioral Control (PBC): Cronbach's Alpha = 0.780
Environmental Concern (EC): Cronbach's Alpha = 0.827
Health Awareness (HA): Cronbach's Alpha = 0.820
Perceived Quality (PQ): Cronbach's Alpha = 0.787
Perceived Price (PP): Cronbach's Alpha = 0.792
Organic Food Purchase Intention (OFPI): Cronbach's Alpha = 0.808
Through Table 4.3, the results of Cronbach's Alpha test (detailed in
Appendix 8) for 35 observed variables of 7 independent variables and 5 observed variables of 1 dependent variable are shown:
- Subjective Norm, Perceived Behavioral Control, Environmental Concern, Health Awareness, Perceived Price all have Cronbach's Alpha coefficients greater than 0.7 At the same time, all the observed variables of these scales have the total correlation coefficient greater than 0.3 Therefore, these 5 independent variable scales are all good measurement scales and reliable variables for further analysis
- Attitude: Variable correlation coefficient - the sum of observed variable ATT5 is 0.199 < 0.3, so it is excluded The remaining 5 variables including ATT1, ATT2, ATT3, ATT4 and ATT6 all have variable correlation coefficients - the sum is greater than 0.3 The scale also has a better Cronbach's Alpha coefficient when removing the ATT5 variable because it increases from 0.786 to 0.829 > 0.6 Therefore, after removing the variable ATT5, the scale meets the requirements of reliability, so it can be included in factor analysis
- Perceived Quality: The variable correlation coefficient - the sum of the observed variable PQ5 is 0.204 < 0.3, so it is excluded The remaining 5 variables including PQ1, PQ2, PQ3, PQ4 and PQ6 all have variable correlation coefficients - the sum is greater than 0.3 The scale also has a better Cronbach's Alpha coefficient when removing the variable PQ5 because it increases from 0.787 to 0.831 > 0.6 Therefore, after removing the variable PQ5, the scale meets the requirements for reliability, so it can be included in factor analysis
The results of the reliability test for the dependent variable “Intent to buy OF - OFPI” show that the scale has Cronbach's Alpha coefficient = 0.808 > 0.7 and the variable correlation coefficient - the sum of the 5 observed variables have values higher than 0.3 Therefore, the dependent variable scale is satisfactory
After testing for reliability, all concepts have Cronbach's Alpha coefficient greater than 0.7 This shows that the observed variables are closely related to each other in the same scale of concept of composition However, from the initial 40 observed variables, the author removed the ATT5 and PQ5 variables because the variable- total correlation coefficient was not reached After removing the variable, the Perceived Quality scale with Cronbach's Alpha of 0.831 is the highest and the lowest is the Perceived Behavioral Control with a coefficient of 0.780 Therefore, the remaining 38 observed variables of 8 scales were included in the EFA analysis.
EFA analysis
Evaluation criteria: Based on section 3.3.3.3 of chapter 3, the author will use the Principal Component factor extraction method with Varimax rotation for 33 observed variables of the independent variable and 5 observed variables of the dependent variable based on the criteria:
- 0.5 ≤ KMO ≤ 1; Bartlett test with sig < 0.05;
- Eigenvalues coefficient > 1; Total variance extracted 50%
4.3.1 EFA analysis results of independent variables
4.3.1.1 EFA analysis results of independent variables 1st time
Table 4.4 EFA analysis results of independent variables 1st time
The results of the KMO, Bartlett's test and the Eigenvalue coefficients of the independent variables for the first time are shown in Table 4.4 (see Appendix 9A for details), showing that the KMO coefficient = 0.827 > 0.5 and Bartlett's Test has Sig level = 0.000 < 0.05 Therefore, the observed variables are correlated with each other and this study data is appropriate The EFA Figure model gives 7 factors with Eigenvalue = 1.336 (greater than 1) and these 7 factors together explain 63.536% (greater than 50%) of the model Therefore, 7 constitutive factors are valid and coincide with the original hypothesis that the research model consists of 7 independent variables
Table 4.5 Rotated Component Matrix result of independent variables 1st time
The results of Table 4.5 (details in Appendix 9A) show that after rotating the factor by the "Varimax" method, the 33 observed variables were mixed together and then converged to their own group with most of the load coefficients All factors are above 0.5 The Perceived Price scale group has a negative load coefficient, so we evaluate it by the absolute value of the load factor and the absolute value of the load coefficients in this scale group is also above 0.5 Therefore, the factor loading condition is adopted
However, 5 observed variables SN5, PQ2, HA4, ATT4 and EC5 were separated to another group At the same time, each variable is uploaded at the same time in both different factors and has a load factor > 0.5, leading to the discriminant value (difference of factor loading factor) less than 0.3 Therefore, it is necessary to remove the 5 variables SN5, PQ2, HA4, ATT4 and EC5 from the scale Therefore, it is necessary to perform the second EFA analysis for the independent variables with 28 observed variables
4.3.1.2 EFA analysis results of independent variables 2nd time
Table 4.6 EFA analysis results of independent variables 2nd time
The results of the second EFA factor analysis are shown in Table 4.6 (Appendix 9B for details) showing that the KMO coefficient = 0.887 > 0.5 and Bartlett's Test has level of Sig = 0.000 < 0.05 Therefore, the observed variables are correlated with each other and EFA analysis is appropriate for the research data The EFA model gives 7 factors with Eigenvalue = 1,263 (greater than 1) and these 7 factors together explain 61.277% (greater than 50%) of the research model
Therefore, 7 constitutive factors are valid and coincide with the original hypothesis that the research model consists of 7 independent variables
After rotating the factors by the "Varimax" method, 28 observed variables were mixed together and then converged to their own group according to the original 7 hypotheses There is no pooling or splitting of variables to another group of factors At the same time, all the observed variables have the standard factor loading coefficient greater than 0.5 and the conditional discriminant value ≥ 0.3 Based on the above results, it can be concluded that after EFA analysis, all 7 independent variables of the hypothesis ensure the conditions to continue to be included in the analysis of correlation and regression analysis in the next step
4.3.2 EFA analysis results of dependent variables
The results of the EFA analysis of the dependent variable “Intent to buy OF - OFPI” (Appendix 9C for details) show that the KMO coefficient = 0.842 > 0.5 and Bartlett's Test has a Sig value = 0.000 < 0.05 Therefore, the observed variables of the dependent variable are correlated with each other and the EFA analysis is appropriate The results also show a factor with Eigenvalues of 2,830 > 1 and this factor explains 56.607% of the model (> 50%) Therefore, 1 constitutive factor is valid and coincides with the hypothesis that there is 1 dependent variable At the same time, 5 observed variables all converge to their group with a load factor greater than 0.5 From the above results, after factor analysis, the dependent variable meets all the conditions and is suitable for regression analysis
4.3.3 Adjustments of the research model after EFA analysis
After analyzing the data collected through Cronbach's Alpha and EFA analysis steps, the research model still retains 7 independent variables (Attitude, Subjective Norm, Perceived Behavior Control, Environmental Concern, Health Awareness, Perceived Quality, Perceived Price) have an impact on the dependent variable Organic Food Purchase Intention of consumers in HCMC However, the independent variable scales removed 7 observed variables So, the Figure model has a total of 33 observed variables instead of the original 40 variables
Therefore, the adjusted model has the following change:
- Attitude (ATT): ATT1, ATT2, ATT3, ATT6
- Subjective Norm (SN): SN1, SN2, SN3, SN4
- Perceived Behavior Control (PBC): PBC1, PBC2, PBC3, PBC4
- Environmental Concern (EC): EC1, EC2, EC3, EC4
- Health Awareness (HA): HA1, HA2, HA3, HA5
- Perceived Quality (PQ): PQ1, PQ3, PQ4, PQ6
- Perceived Price (PP): PP1, PP2, PP3, PP4
Dependent variables: Organic Food Purchase Intention (OFPI): OFPI1, OFPI2,
Correlation analysis
Correlation analysis was performed between the dependent variable Organic Food Purchase Intention (OFPI) and the independent variables Attitude (ATT), Subjective Norm (SN), Perceived Behavior Control (PBC), Environmental
Concern (EC), Health Awareness (HA), Perceived Quality (PQ) and Perceived Price (PP) It also analyzes the correlation between the independent variables Because such correlations can affect the results of regression analysis such as causing multicollinearity The results of the correlation analysis are shown in Table 4.7 below
OFPI ATT SN PBC EC HA PQ PP
** Correlation is significant at the 0.01 level (2-tailed) r : Pearson Correlation
The results from Table 4.7 (details in Appendix 10) show that all the independent variables have a strong linear correlation with the dependent variable of intention to buy OF, the correlation coefficients are statistically significant (Sig < 0.05) In which, the Attitude variable has the strongest correlation (r = 0.594) and the Environmental Concern variable has the lowest correlation (r = 0.415) In addition to the Perceived Price variable having a negative correlation, the remaining independent variables are positively correlated with the intention to buy OF Besides, the results of correlation analysis also show that the correlation coefficient between the independent variables is weakly correlated with each other when the majority r < 0.4 However, there are still two pairs of variables with correlation coefficients greater than 0.4 It is the Subjective Norm and Environmental Concern pair with the highest r-value of 0.432 and the Environmental Concern and Health Awareness pair with the next-highest r-value of 0.403 Therefore, the phenomenon of multicollinearity among these highly correlated variables should be taken into account when analyzing multiple regression From the results of the correlation analysis, all independent variables can be included in the linear regression analysis in the next step.
Regression analysis
Regression analysis was performed with 7 independent variables: Attitude (ATT), Subjective Norm (SN), Perceived Behavior Control (PBC), Environmental Concern (EC), Health Awareness (HA), Perceived Quality (PQ), Perceived Price (PP) and a dependent variable, Organic Food Purchase Intention (OFPI), were adjusted in section 4.3.4 with a total of 33 observed variables Regression analysis was conducted through the Enter method With this method, SPSS will process all input variables once to analyze and give relevant statistical parameters
The multivariable linear regression has the following form:
YDMS = β 0 + β 1 *ATT + β 2 *SN + β 3 *PBC + β 4 *EC + β 5 *HA + β 6 *PQ+ β 7 *PP + ε
The results of the regression analysis are presented in Table 4.8, Table 4.9 and Table 4.10 Details at Appendix 11
Table 4.8 Evaluate model fit Model Summary
Model R R 2 Adjusted R 2 Std Error of the
Model Sum of Squares df Mean Square F Sig
Table 4.10 Correlation coefficients Correlation coefficients
4.5.1 Evaluate the fit of the regression model
The results from Tables 4.8 and 4.9 (Appendix 11A and Appendix 11B for details) show that:
- R = 0.852 shows that the relationship between the variables in the model has a close relationship and has been shown to be a function that does not decrease with the number of independent variables included in the model
- Adjusted R2 = 0.719 (equivalent to 71.9%), this shows that the linear regression figure model is suitable for the data set to 71.9% Or 71.9% of the variation of the dependent variable is explained by the variation of the independent variables, the remaining 28.1% is due to variables outside the model and random error
- F test shows that F = 110,329 with sig value = 0.000 < 0.05 So, the linear regression model is built in accordance with the collected data and the input variables are statistically significant at the 5% level of significance
From the above results, it can be concluded that this regression model has a relevance of 71.9% and satisfies the test conditions for relevance, so it can be used for managerial implications
4.5.2 Regression coefficients and regression equation
Table 4.10 shows that there are 7 independent variables included in the regression analysis model In which, there are 6 variables that have a relationship with the dependent variable Organic Food Purchase Intention (OFPI) which are Attitude (ATT), Subjective Norm (SN), Perceived Behavior Control (PBC), Health Awareness (HA), Perceived Quality (PQ) and Perceived Price (PP) because all 6 variables have Sig values < 0.05
However, the Environmental Concern (EC) factor has a Sig value = 0.230 > 0.05, so it is necessary to remove this variable from the Figure model This shows that the hypothesis H4: Environmental Concern (EC) alone has a positive effect on the intention to buy OF is rejected The remaining 6 hypotheses are H1, H2, H3, H5, H6, H7 which are accepted and have statistical significance in the model with the significance level of 5%, the confidence level of 95% So the dependent variable Organic Food Purchase Intention is affected by 6 independent variables namely
Attitude, Subjective Norm, Perceived Behavior Control, Health Awareness, Perceived Quality and Perceived Price In which, 1 independent variable, Perceived Price (PP), has a negative impact on the dependent variable and the remaining 5 independent variables are Attitude (ATT), Subjective Norm (SN), Perceived Behavior Control (PBC), Health Awareness ( HA), Perceived Quality (PQ) has a positive impact on the dependent variable
After removing the Environmental Concern (EC) variable from the model, the regression model is left with 7 variables with 6 independent variables and 1 dependent variable The multivariable linear regression equation is established based on the Beta coefficients of the variables that meet the conditions and is kept in the model In which, the unnormalized equation is built through the unnormalized Beta coefficient (β) and the normalized equation is built based on the normalized Beta coefficient (β') from the regression coefficient test results in Table 4.10
OFPI = 0.650 + 0.239*ATT + 0.217*SN + 0.185*PBC + 0.200*HA + 0.201*PQ - 0.202*PP + ε
OFPI = 0.244*ATT + 0.222*SN + 0.187*PBC + 0.205*HA + 0.209*PQ - 0.211*PP + ε
4.5.3 Testing the assumptions of the Regression model
Regression analysis is not only the description of the observed data, but also the need to infer conclusions for the relationship between the variables in the population from the observed results in the sample size Therefore, the results of the sample generalized to the value of the population must satisfy the necessary assumptions If the assumptions are violated, the estimation results of the model are no longer reliable (Trong & Ngoc, 2008) Assumptions that need to be tested are:
- No correlation between residuals assumption
- No violation of linear relationship assumption
- The variance of the residual is constant assumption
- The residuals have a normal distribution assumption
In the multivariable linear regression model, the assumption between the independent variables of the model is that there is no multicollinearity phenomenon The phenomenon of multicollinearity in the model is measured through the VIF coefficient In the experimental study, if VIF is less than 10, then the model is said to have no multicollinearity phenomenon The measurement results shown in Table 4.10 show that the variance magnification factor (VIF) has a value from 1,347 to 1,474 Since all components meet the requirements of VIF < 10, the regression model does not have multicollinearity Therefore, we accept the assumption that there is no multicollinearity
4.5.3.2 No correlation between the residuals assumption
This assumption is tested through Durbin-Watson statistics (d) The Durbin-Watson (DW) value of the Figure in Table 4.8 is d = 2,018 At the 5% level of significance, the model has an observation sample of n = 300, the number of independent variables k = 6, looking up Table DW we have dL = 1,770 and dU = 1,852 Attaching to the DW value bar, we see (dU) 1,852 < 2,018 < 2,148 (4 – dU) So accept the assumption that there is no correlation between the residuals
4.5.3.3 No violation of linear relationship assumption
The assumption of linear relationship is tested using a Scatter Plot plot for the normalized residuals The result of Figure 4.1 shows that the normalized residuals are randomly dispersed centered around the zero line without forming a particular Figure So the assumption of linear relationship is not violated (Details in
Source: SPSS analysis result, 2022 4.5.3.4 The variance of the residual is constant assumption
The results from Figure 4.1 also show that the percentiles fluctuate quite uniformly above and below the zero coordinate axis The percentiles are almost in the -2 to 2 range along the zero coordinate axis Therefore, assume determining the variance of the residuals unchanged is not violated
4.5.3.5 The residuals have a normal distribution assumption
Assumption that the residuals have a normal distribution is tested through the Histogram and P-P Plot (Appendix 11D for details) From the Histogram in Figure 4.2 below, we see a normal distribution curve superimposed on the histogram This curve has the form of a bell figure, which is consistent with the graph form of the normal distribution The values also show approximately normal distribution of residuals with mean Mean = 8.64*10-15 ≈ 0 and standard deviation Std.Dev 0.988 ≈ 1 so the residuals are normally distributed The P-P Plot Figure 4.3 histogram shows that the percentiles in the distribution of residuals are not scattered too far from the expected line and are concentrated along the expected line forming a diagonal So accept the assumption that the residuals are normally distributed.
Difference testing
Hypothesis H0: There is no mean difference between male and female gender groups for Intent to buy OF
Independent Samples T-test was performed Results of testing the difference by Gender (Appendix 13A for details) The Levene test shows the Sig value of all variables are > 0.05, which means that the variance between the two genders are the same Therefore, use the Sig value (2-tailed) T-test of Equal variances assumed (first line) The results show that Sig (OFPI) = 0.442 > 0.05 So accept the hypothesis H0 and conclude: There is no difference in intention to buy OF between
Hypothesis H0: There is no difference between age groups for Intent to buy OF One-Way ANOVA test was carried out The results of testing the difference by age (Appendix 13B for details) show that the Sig of Levene Statistic in Test of Homogeneity of Variances = 0.000 < 0.05, which means that the variance between age groups is heterogeneous Therefore, it is necessary to use the Sig result in Table Robust Tests for further review The results show that Sig (OFPI) = 0.000 < 0.05 So, reject hypothesis H0 and conclude: There is a difference in intention to buy OF among people of different age groups Specifically, the difference is that the consumer group over 40 years old has a higher intention to buy OF than the rest of the age groups
Difference testing between Education level
Hypothesis H0: There is no mean difference between groups of educational attainment for Intent to buy OF
One-Way ANOVA test was carried out The results of the test of differences by education (Appendix 13C for details) show that the Sig of Levene Statistic in Test of Homogeneity of Variances = 0.242 > 0.05, which means that the variance between the educational attainment groups is the same Therefore, it is necessary to use the Sig result in Table ANOVA for further review Sig results (OFPI) 0.006 < 0.05 So, reject hypothesis H0 and conclude: There is a difference in intention to buy OF between people of different education level groups Specifically, the difference is that the consumer group of Postgraduate level has a higher intention to buy OF than the rest of the group
Hypothesis H0: There is no mean difference between Occupation groups for Intent to buy OF
One-Way ANOVA test was carried out The results of testing the difference by occupation (Appendix 13D for details) show that the value Sig of Levene Statistic in Test of Homogeneity = 0.000 < 0.05, this means that the variance between occupational groups is not uniform Therefore, it is necessary to use the Sig result in Table Robust Tests for further review The results show that Sig (OFPI) = 0.000
< 0.05 So, reject hypothesis H0 and conclude: There is a difference in intention to buy OF between people of different occupation groups Specifically, the difference is that the consumer group of Management Level and above has a higher intention to buy OF than the rest of the occupational groups
Hypothesis H0: There is no mean difference between income groups for Intent to buy OF The results of the income difference test (details in Appendix 13E) show that the Sig of Levene Statistic in Test of Homogeneity = 0.000 < 0.05, which means that the variance between income groups is heterogeneous Therefore, it is necessary to use the Sig result in Table Robust Tests for further review The results show that Sig (OFPI) = 0.000 < 0.05 Therefore, we reject hypothesis H0 and conclude: There is a difference in intention to buy OF among people of different income groups Specifically, the difference is that the consumer group in the income group from 15 to 20 million has a higher intention to buy OF than the rest of the income groups.
Compare the results with previous domestic studies
Factors affecting the intention to buy organic food of office workers in HCMC (Lien, 2021)
Research results compared with research of Lien (2021) are similar in terms of the influence of 3 factors Attitude, Subjective Norms and Health Awareness All three variables have a positive impact on the intention to buy OF The difference is that the author has proven that the factor Environmental Concern has statistical significance in the research model At the same time, the author instead uses 3 other factors: Perceived behavioral control, Perceived quality and Perceived price into the model and the results show that all 3 variables impact on the dependent variable
Factors affecting consumer intention to buy organic food in HCMC (Minh, 2016).
The research results are similar to the study of Minh (2016) in terms of the impact of 3 factors: Health Awareness, Perceived quality and Perceived price In which, Health awareness, Perceived quality have positive impact and Perceived price has negative impact on the dependent variable The difference between these two studies is that this thesis does not prove the impact of Environmental Concern and does not propose the Safety Concern factor into the proposed model However, the author has analyzed and proved that other factors such as Attitude, Subjective Norms and Perceived behavioral control have a positive impact on consumer intention to buy OF
The thesis has analyzed 7 factors and proved that 6/7 factors have an impact on the intention to buy OF of consumers in TP.HCM These 6 factors have similarities and differences compared to related studies There are similar factors with this topic but other topics do not analyze, and vice versa At the same time, the level of impact of factors compared with previous studies is also not the same This shows that the research results on the intention to buy OF will be different according to each period, each subject and the scope of the study Therefore, based on the research results, the author gives appropriate management implications for managers in the
OF business field to have appropriate strategies for many consumers in HCMC
In Chapter 4, the author presented the results of quantitative research on the factors affecting the intention to buy OF of consumers in HCMC The author conducted a survey of consumers with different ages, education and occupations in HCMC from April 2022 to May 2022 There were 400 survey questionnaires sent and 300 valid samples were collected, so the sample size for analysis was 300 samples and were also analyzed by the author with descriptive statistics when analyzing the data
Next, the author performed Cronbach's Alpha test, analyzed EFA factors and removed 7/35 observed variables of independent variables (7/40 total observed variables) but did not affect the research hypotheses After performing Pearson correlation analysis and multivariable regression analysis, the results showed that 6/7 hypotheses were accepted and the hypothesis H4 about the independent variable Environmental concern should be removed Accordingly, in addition to 1 variable Perception of selling price has the opposite effect, the remaining 5 variables are Attitude, Subjective Norm, Perceived behavioral control, Health awareness and Perceived quality all have positive effect on the intention to buy OF At the same time, the author has also performed a difference test for Demographic factors through T-Test and ANOVA, the results show that there is a difference between age, education level, occupation and income groups on consumer intention.