MINISTRY OF EDUCATION AND TRAINING THE STATE BANK OF VIET NAM BANKING UNIVERSITY OF HO CHI MINH CITY NGUYEN HO TRAM ANH FACTORS AFFECTING CUSTOMERS’ DECISION TO BUY FRESH FOOD ONLINE A
INTRODUCTION
Research Statement
In recent years, consumer shopping behavior has changed dramatically due to the Covid-19 pandemic, marking a turning point in how people shop and accelerating the reshaping of shopping habits Online stores offer numerous advantages over physical shops, including convenience, time savings, no need to travel or queue, 24/7 access, and the ability to shop from anywhere They provide free, rich product information and online tools that help consumers compare products and make informed purchase decisions In Vietnam, isolation measures implemented to protect public health have further boosted online shopping by limiting exposure to infection To persuade and retain online buyers, sellers must understand the factors that buyers use to decide on purchases As Ahmad Reza Asadollahi et al note, analyzing the factors affecting consumer online shopping behavior is one of the most important issues in e-commerce and marketing.
Fresh food items are essential and among the most affected during lockdowns The Covid-19 pandemic caused a socio-economic disruption that led governments to issue measures to limit the virus's spread, including social distancing, physical distancing, large-scale social restrictions, and regional quarantines These regulations reshaped shopping behavior, changing where people shop for groceries, how they access fresh produce, and how often they buy, while retailers adjusted supply chains to meet shifting demand.
Vietnam's fresh food market is estimated at 948 trillion VND (about 31,919 thousand tons) in 2020 and is projected to reach 1,307 trillion VND by 2025, with an average annual growth of 6.6%, signaling rapid expansion in online and offline channels; iPrice data show increasing consumer attention to fresh food, meat, fish, beverages, packaged foods, and vegetables in online shopping since Q1 2021, underscoring shifting online shopping behavior Although prior research identifies numerous factors influencing online consumer behavior, a single model cannot comprehensively cover all potential influences due to diverse prior product and service experiences; this study adopts the Theory of Reasoned Action and the Theory of Planned Behavior to survey, review, and evaluate the factors affecting online shopping for fresh food products among Ho Chi Minh City consumers, with a focus on purchasing on supplier websites and platform-specific shopping behavior.
Therefore, I have chosen the topic “Factors Affecting Customers’ Decisions to Buy Fresh Food Online at Supermarkets’ Websites During the Covid-19 Pandemic in Ho Chi Minh City” as the focus of my graduate thesis This study seeks to identify the key drivers and barriers shaping consumer decision-making for online fresh-food purchases on supermarket websites during the Covid-19 period in Ho Chi Minh City, providing insights for retailers aiming to optimize their online grocery platforms.
Aims and objectives of the research
This study aims to identify the factors influencing online shopping behavior on fresh-food supermarket websites in Ho Chi Minh City and to quantify how strongly each factor affects consumer choices during the COVID-19 pandemic To achieve this, the research sets out specific objectives: to determine which factors drive online purchases of fresh groceries on Ho Chi Minh City’s supermarket sites, to measure the relative degree of influence of these factors, and to provide actionable insights for optimizing e-commerce strategies in the pandemic context.
In order to achieve this general goal, this study sets out the following specific research objectives:
- To determine and evaluate factors affecting the customers’ decision to buy fresh food online at supermarket's websites during the Covid-19 Pandemic in Ho Chi Minh City
- To propose recommendations and implications to managers have solutions for their supermarket's websites
To achieve this goal, the research questions are posed as follows:
Q1: What factors of online fresh food buying behavior affect the customer's decision to buy fresh food online at supermarket's websites during the Covid-19 Pandemic in Ho Chi Minh City?
Q2: How do these factors affect the customer's decision to buy fresh food online at supermarket's websites during the Covid-19 Pandemic in Ho Chi Minh City?
Q3: What administrative implications are proposed to attract customers to buy online at supermarket's websites?
Subject and scope of the study
The subject of the study analyzes factors affecting the customer's decision to buy fresh food online at supermarket’s websites during the Covid-19 pandemic in Ho Chi Minh City
The scope of the study includes consumers who bought fresh food online at supermarket’s websites during the Covid-19 pandemic in Ho Chi Minh City
Time of the research: Carry out from 02/2021 to 03/2021.
Research method
This qualitative study identifies the research problem by drawing on related research about consumer decision-making It relies on documents and prior studies to propose a model of the factors influencing consumer decisions Subsequently, a questionnaire was administered to participants to collect data.
4 lived in Ho Chi Minh City and used to buy fresh food at supermarkets' websites during social distancing
Quantitative approach: The data collected from the questionnaire include (attitude, subjective norms, perceived behavioral control, risk awareness and purchase decision)
The data will be analyzed to describe the sample and eliminate inconsistently observed variables in a factor by Cronbach's Alpha Besides, this research uses EFA to explore factors and evaluate the internal reliability of a scale Multivariate regression aims to examine factors that contribute to the change of independent variables ANOVA test uses to compare data sets and evaluate potential differences in a dependent scale variable.
Research significance
Against the Covid-19–shaped economic backdrop, this study aims to serve as a basis for further research on how the pandemic has influenced customers’ shopping decisions on online platforms It offers implications for supermarket managers to develop optimal strategies for product assortment, formats, and business models that boost sales of essential products through their websites, especially in Ho Chi Minh City and Vietnam, with the goal of reaching and retaining as many customers as possible By doing so, businesses can build resilient online solutions for black swan events and maintain a competitive advantage in the digital marketplace.
Expected contribution
This thesis provides clear empirical evidence on the factors that influence customers' purchasing decisions, forming a solid basis for proposing measures to improve the quality of products and services in order to influence purchase decisions For other researchers, these findings offer a robust platform for further study, including cross-context validation, refinement of measurement tools, and exploration of the interactions among drivers of consumer choice.
- As the reference for the other researcher who interested in customers’ decision
Thesis structure
The main content of the thesis, in addition to the two part: introduction and conclusion, the thesis includes 05 chapters as follows:
Chapter 4: Data analysis and discussion
Summary of Chapter 1
Chapter 1 presents why choosing this topic, gives an overview of the research objectives, research questions, research methods, and research scope In addition, it presents the significance of the research in terms of academics and practice, the expected contribution of research Finally, outline the presentation structure of the research topic to give readers an overall picture of the entire study This chapter is the basis for further research steps It is in-depth research on the theoretical basis, research methods, and data collection to produce research results and provide solution implications
LITERATURE REVIEW
The basic of concepts
According to the American Marketing Association, "Consumer behavior is the interaction between environmental stimuli and human cognition and behavior through which people change their lives " According to Philip Kotler,
"consumption behavior is the specific behavior of an individual when making decisions to purchase, use and dispose of a product or service" (Philip Kotler,
Solomon (2005) defines consumption behavior as a process that enables individuals or groups to choose, purchase, use, or discard a product, service, thoughts, experiences, or positives accumulated to satisfy a need or desire Accordingly, consumer behavior is the sequence by which people search for, select, purchase, use, and dispose of goods and services to satisfy their needs and wants Over the past 25 years, consumer behavior has changed substantially, though it remains an evolutionary process, with the seeds of change evident across generations (Kar, 2010).
(2007), buying behavior is influenced by disease outbreaks
2.1.2 The concept of customers’ buying decision
Consumer decision-making can be complex, with shoppers using information about specific items and brands, as well as their own experiences, to guide purchasing decisions (Jiang, Pingjun, Rosenbloom, Bert, 2005) After gathering information from various sources, the consumer decides what to purchase, evaluates and compares options, and then determines where to buy and what to buy In the end, consumers tend to buy the brand, product, or service that they rank highest in their evaluation.
7 the evaluation stage The surrounding environment also influences the purchase decision (Kotler at el, 2017)
2.1.3 Concept of customers’ buying decision process
Efthymios Constantinides (2004) identifies the most common approach to explaining consumer behavior as viewing the buying process as a learning, information-processing, and decision-making sequence This five-stage model shows the consumer moving through need recognition, information search, evaluation of alternatives, purchase decision, and post-purchase behavior, with learning and information processing shaping each step of the decision process and informing marketers about how information is perceived, compared, and acted upon.
(5) After-sale behavior (Bettman, 1979; Dibb et al., 2001; Jobber, 2001; Boyd et al., 2002; Kotler, 2003; Brassington and Pettitt, 2003)
This study assumes that consumers have already formed a purchase decision and are now in the stage of evaluating options and preparing to finalize a buying decision, with the process described according to Philip Kotler’s model (2002).
Figure 2.1 Steps to make a purchase decision of a customer
The buying decision process starts when consumers recognize a need and decide to satisfy it by purchasing a product They seek information through various channels—from mass media and online content to recommendations from friends, colleagues, and relatives who have used the product—then review and compare product features, price, and quality against similar options to determine the most suitable choice for their needs and budget Products that offer solid after-sales service tend to win high customer loyalty, with many buyers returning and referring others, making post-purchase behavior a crucial part of the decision process.
Philip Kotler's 2001 model of the purchasing decision/adoption process explains that a buying decision arises when consumers recognize a need for a product To satisfy this need, they seek information from diverse sources, including mass information channels and the experiences of friends, colleagues, and relatives who have used the product They then evaluate these inputs, read reviews, and compare the product with similar options to determine their final choice.
Alternatives assessment Intention to buy
Other people's point of view
When comparing nine products with similar quality or price, consumers should select the option that best matches their needs and budget to make the most optimal purchase decision Products that provide strong after-sales service are highly valued by customers, often winning loyalty, encouraging repeat purchases, and prompting referrals Thus, understanding post-purchase behavior is a crucial part of the buying decision process.
2.1.4 Behavioral characteristics of buying fresh food online
- The status of the fresh food market in the context of social distancing in Ho Chi Minh City
In Ho Chi Minh City, the average daily consumption demand is about 10,964 tons, with major categories including vegetables and fruits at 4,246 tons, rice products at 1,981 tons, dry processed foods (noodles, vermicelli, pho) at 660 tons, cattle meat at 755 tons, poultry meat at 660 tons, and processed foods at 236 tons Weekly demand averages 76,747 tons, and a 15‑day period totals about 164,460 tons According to baotintuc.vn, fresh food accounts for more than 50% of the city’s average daily consumption.
Vietnam relies on fresh ingredients for daily meals, and Ho Chi Minh City is the largest market for fresh food in the country Most of the city’s food is not produced locally but is supplied from provinces across the country, which makes the city vulnerable to supply disruptions As a result, the Covid‑19 lockdown in Ho Chi Minh City caused food shortages due to supply chain bottlenecks for more than three months, following Directive 16 issued in July.
In Ho Chi Minh City, after the government entailed for the population to avoid mass gatherings and personal interactions to help slow the transmission of COVID-
During the COVID-19 wave from August 23 to September 15, 2021, most citizens stayed at home in Vietnam Consequently, the majority of restaurants were ordered to suspend dine-in services, with only takeout, drive-thru, and delivery options permitted.
People have shifted toward stocking up with enough food for 1–2 weeks in a single shopping trip, while Ho Chi Minh City struggles with import difficulties caused by a citywide blockade that creates local shortages even as nearby provinces have excess supply, according to vnexpress.net Minh Tan, owner of the Foodmap Flagship Store on LazMall, says that since the start of 2021 there has been a rise in orders for fresh foods and agricultural products, with purchasing power during periods of social distancing increasing about 4.5 times compared with normal times.
An Industry and Trade Magazine article dated March 14, 2021 reports that in Ho Chi Minh City, the COVID-19 pandemic has driven a shift from daily shopping to centralized shopping, with weekends featuring bulk-buying and hoarding of essential items that can lead to local shortages.
- Behavioral characteristics of buying fresh food online
Through the concept of buying behavior and steps to make a buying decision, the online buying behavior of fresh food has the following characteristics:
Whenever you need fresh groceries, you don’t have to go to a store or market—you can stay at home or anywhere and shop online through sales websites and social networks to search for and choose products that can be delivered to your location This online shopping approach lets consumers gather extensive product information from many websites and suppliers, enabling them to compare options and make informed purchase decisions.
During the Covid-19 pandemic, the most significant benefit of online shopping is its convenience, enabling consumers to shop from home and limit contact with others This reduced social interaction helps lower the transmission of the virus while still delivering access to goods and services.
One major barrier to online fresh food shopping is the inability to touch or inspect the product to verify its freshness before purchase, making it hard for customers to assess quality from a digital description This creates a delivery risk that the food arriving at customers may not be as fresh as illustrated or described on the retailer’s site, potentially leading to dissatisfaction, higher return rates, and concerns about online grocery reliability.
Theoretical models
2.2.1 The theory of reasoned action model (TRA)
Developed by Ajzen and Fishbein in 1975, the Theory of Reasoned Action (TRA) is a pioneering model in social psychology that explains how attitudes and perceived social pressures shape both intention and behavior The TRA posits a strong, empirically verified link between intention and behavior across many fields, with two main determinants of behavioral choice: personal attitudes and subjective norms Attitude toward the behavior is assessed through behavioral beliefs and the evaluation of expected outcomes, while the subjective norm is assessed through normative beliefs and motivation to comply Ajzen (1991) defines the subjective norm as the perception that significant influencers think an individual should or should not perform the behavior This framework helps predict behavior by combining what a person intends to do with the social pressures they perceive, making TRA a foundational theory for research on behavioral intention.
Figure 2.2 The theory of reasoned action model (TRA)
The theory of reasoned action model (TRA) has three components:
Attitude: the individual's positive or negative feelings when performing a behavior, determined through personal beliefs and judgments
Subjective norms: Thoughts and opinions of individuals and people around them influence behavioral intentions
Behavioral intention: is a combination of attitudes and subjective norms that determine an individual's final behavior
The rational action theory model suggests that a person's behavioral intentions depend on their behavior and subjective norms If a person intends to perform an act, they will likely do it
2.2.2 The theory of planned behavior model (TPB)
The Theory of Planned Behavior (TPB) offers a robust framework for explaining how intention and action unfold across diverse behavioral domains by integrating attitudes, subjective norms, and perceived behavioral control (Fishbein & Ajzen, 2010) To address flaws in the Theory of Reasoned Action, Ajzen (1991) extended the model into the Theory of Planned Behavior, asserting that perceived behavioral control—driven by confidence in one’s capacity to undertake a behavior—strongly influences both intention and actual behavior As Sheeran (2002, p 1) summarizes, the fundamental assumption is that people do what they intend to do and do not do what they do not intend.
Figure 2.3 Theory of planned behavior
Attitude toward behavior is a key psychological factor that captures an individual’s positive or negative beliefs about a product or service, reflecting agreement or disagreement and guiding behavioral outcomes through overall beliefs and evaluations (Ajzen, 1991) In the context of online shopping, attitude denotes how customers assess their shopping experiences on a website, including evaluations of offers, perceived benefits for consumers, convenience, and alignment with their online service preferences (Lin, 2007).
Subjective norm is the perceived social pressure to perform or not perform a behavior, a concept introduced by Ajzen (1991) Social influence refers to the effects and impacts of important, close others—such as family and friends—whose opinions and expectations can shape an individual’s actions Together, these concepts explain how behavior is guided by the expectations of those who matter most, influencing intentions and actual conduct.
Perceived Behavioral Control (PBC) refers to how easy or difficult it is to perform a particular behavior and whether its execution feels within one’s control or subject to restrictions Ajzen (1991) introduced PBC as a core component of the Theory of Planned Behavior, suggesting that the degree of perceived control influences both behavioral intentions and the likelihood of actually performing the behavior Factors such as available resources, opportunities, past experiences, and perceived barriers shape PBC, making it a key predictor of behavior alongside attitude and subjective norms In essence, perceived behavioral control captures the sense of controllability over actions, linking feasibility with confidence to determine whether a behavior will be performed.
14 directly affects behavioral intention and that if the consumer is correct in his perception, behavioral control also predicts the consumer's behavior.
Previous research
Descriptive norm Injunctive norm Attitude
Perceived control over behavior Self-efficacy
Perceived behavioral control over consumption was not related to the decision or to consumption frequency in the basic model, but it showed a significant negative association with both outcomes in the extended model Self-efficacy, defined as confidence in one's ability to regularly consume functional foods, emerged as the strongest predictor of the decision The decision was further predicted by both descriptive norms and injunctive norms, with each norm demonstrating a significant and reasonably strong predictive effect.
Study results show a dramatic shift in customer behavior toward e-shopping, with growing interest and decisive preference for online purchases Consumers now favor using multiple websites and online platforms that rely on online payment infrastructure, indicating a move toward seamless electronic commerce across channels This trend underscores the central role of online payment systems in shaping shopping decisions and overall customer engagement.
According to the findings, subjective criteria and perceived utility have a considerable favorable effect on online
Subjective norm Enjoyment Perceived usefulness
Findings indicate that utility and subjective norms influence users' decisions to continue, but enjoyment drives a higher level of commitment For online initiatives in Saudi Arabia, it is essential to address both the direct and indirect implications on continuation decisions to foster sustained engagement.
Nguyen Thu Ha and Hoang Dam
Perceived behavioral control Subjective norms Attitude
Analysis results indicate that three factors positively influence Vietnamese consumers' online purchase decisions and behavior during the COVID-19 pandemic, with Attitude having the largest Beta coefficient and, therefore, representing the most decisive impact on online purchasing decisions on Vietnamese business websites amid COVID-19.
Perceived risk Perceived usability Service quality Attitude
This study demonstrates the relative importance of the model's factors as indicated by the standardized regression coefficients Perceived usefulness has the strongest impact on online shopping decisions, followed by Service quality, while Perceived risk exerts the smallest negative influence.
In the context of mobile commerce, the findings reveal that perceived value
Perceived behavioral control, subjective norms, and perceived consumer values—comprising utilitarian value, hedonic value, and social value—collectively explain customer attitudes, subjective norms, and behavioral decisions The framework also reveals cross-effects among these subjective criteria, showing that utilitarian, hedonic, and social values interact to influence customer sentiments.
Research model and hypothesis
According to the hypotheses above, the thesis suggests research model including five factors: subjective norms, perceived behavioral, control attitude, perceived usefulness and risk awareness
The linear regression equation and conceptual model are below:
Y = PD = β 0 + β 1 *SN + β 2 *PBC + β 3 *ATT + β 4 * PU + β 5 * RA + ɛ i
2.4.2.1 The relationship between Subjective norm and customer’s purchase decision
Subjective norm is the third most influential factor after controlling cognitive behavior to influence purchase decision to shop online (Orapin, 2009) He et al
Since 2008, the idea that subjective norms significantly affect consumers' purchase decisions has been proposed A large body of findings indicates that subjective norms directly influence online shopping decisions, as demonstrated in studies by Nguyen Thu Ha (2020); Nguyen Huu Khoi (2017); Björn Tore Nystrand (2020); Yi Jin Lim (2016); and Al-maghrabi, T., Dennis, C., Halliday, S V (2011) In e-commerce, subjective norms reflect consumers' perceptions of the influence exerted by surrounding social factors on their purchasing behavior.
Reference groups and media strongly influence online purchasing decisions, a factor that becomes even more critical amid social distancing campaigns promoted by the government Messages such as #stayathome and #onhavanvui in the media, alongside distancing directives, shape consumer behavior toward e-commerce The reality that residents can only leave for essential needs during Ho Chi Minh City's lockdown has significantly driven the adoption of online platforms to buy fresh food Based on this discussion, the study has formulated the following hypothesis:
H 1 : Subjective norms has a positive impact on customers’ purchase decision
2.4.2.2 The relationship between Perceived behavioral control and customer’s purchase decision
Perceived behavioral control (PBC) combines external controllability and self-efficacy to shape a person’s desire to perform an activity and the actual performance of that behavior, extending the Theory of Reasoned Action (TRA) to explain actions when behavioral control is limited PBC refers to people’s perceptions of how easy or difficult it is to engage in a given act During the COVID-19 era in Ho Chi Minh City, everyday habits such as gathering, socializing in crowded areas, dining out, and especially shopping changed dramatically, with online orders rising as concerns about direct purchasing and on-site dining increased This shift reflects a broader move to improve shopping and delivery to make it easier and safer In this study, how will perceived behavioral control impact online shopping behavior of fresh food in Ho Chi Minh City? Based on the discussion above, the study has created the following hypothesis:
H 2 : Perceived behavioral control has a positive impact on the purchase decision
2.4.2.3 The relationship between subjective norms and customer’s purchase decision
The relationship between perceived usefulness and the decision to adopt a technology originated in an acceptance context (Bhattacherjee, 2001) In e-commerce, subjective norms reflect consumers’ perceptions of the influence of surrounding reference groups and media on online purchasing From a technology adoption perspective, perceived usefulness has been shown to be a strong determinant of usage decisions (Davis et al., 1989; Taylor and Todd, 1995) Gurito and Siringoringo (2013) found that perceived usefulness has a dominant influence on online shopping decisions, a view that is also supported by Singh, Keswani, Singh, and Sharma.
Perceived usefulness can influence consumers’ online shopping decisions A 2016 study by S found that perceived usefulness positively influences consumers’ attitudes toward online shopping and their decision to make online purchases The study by Pham Thi Thuy Mien similarly supports the notion that perceived usefulness shapes both online shopping attitude and the decision to engage in online shopping.
Research across studies by Yi Jin Lim (2016), Tared Hashem (2020), and Al‑Maghrabi et al (2011) shows that perceived usefulness partially influences online purchase decisions In the COVID-19 era, mandatory e-commerce for fresh food and electronics has delivered time savings and improved decision support through website navigation, item classification, quality and price comparisons Consumers now benefit from diverse payment methods and flexible delivery options This study investigates how perceived usefulness affects online shopping behavior for fresh food in Ho Chi Minh City and posits a hypothesis that higher perceived usefulness positively influences online purchasing behavior in this category.
H 3 : Perceived usefulness has a positive impact on customers’ purchase decision 2.2.4.4 The relationship between Attitude and customer’s purchase decision
According to Fishbein & Ajzen (2010), attitude is "a latent disposition or disposition to react with a degree of favorability or unfavorableness towards a psychological object." Building on this idea, Hung et al (2016) demonstrated that attitude is a key factor shaping how individuals evaluate and respond to objects, underscoring its central role in attitude–behavior relationships and decision-making processes.
Attitudes play a crucial role in online shopping decisions for new functional meat products, a finding echoed across studies by Nguyen Thu Ha (2020), Nguyen Huu Khoi (2017), Nali and Ping Zhang (2002), Pham Thi Thuy Mien (2021), Tared Hashem (2020), and Björn Tore Nystrand (2020) In consumer theory, attitude is defined as the individual's own judgment that performing the behavior will be beneficial, meaning that after researching and targeting a product, the consumer may conclude this is the best choice for them (Giner-Sorolla, 1999) Consequently, researchers argue there is a positive relationship between attitude and purchasing decisions: the more positive the attitude, the higher the likelihood of making a purchase for the functional meat product.
This study investigates how attitude impacts customers' decisions to buy fresh food online on a supermarket website during the Covid-19 pandemic in Ho Chi Minh City Based on the discussion above, the study has created the hypothesis that attitude significantly influences online fresh-food purchase decisions in this pandemic context.
H 4 : Attitude has a positive impact on customers’ purchase decision
2.2.4.5 The relationship between risk awareness and customer’s purchase decision The perceived risk a consumer feels in a choice circumstance is often stated as being based on the ambiguity present in the information the customer has and the expected repercussions of purchase," according to Foxall, Goldsmirth, and Brown
Perceived hazards in consumer decision making are traditionally categorized into six types by Jacoby and Kaplan (1972): financial risk, product performance risk, physical risk, time risk (time/convenience), psychological risk, and social risk The current study focuses on four of these dimensions—financial risk, product performance risk, psychological risk, and time/convenience risk—exploring how concerns about costs, product failures, emotional discomfort, and wasted time or effort shape consumer choices Although the broader framework acknowledges physical and social risks, they are outside the scope of this research.
21 believe that there is a negative relationship between risk awareness and purchasing decisions The more negative the risk awareness, the lower the buying decision
In this study, how will risk awareness impact customers' decision to buy fresh food online at the supermarket's website during Covid-19 pandemic in Ho Chi Minh City? Based on the discussion above, the study has created the following hypothesis:
H 5 : Risk awareness has a negative impact on customers’ purchase decision
Table 2.2 Synthesize the hypotheses of the current study
The relationship between subjective norms and customers’ purchase decision
The relationship between perceived behavioral control and customers’ purchase decision
The relationship between perceived usefulness and customers’ purchase decision
The relationship between attitude and customers’ purchase decision
The relationship between risk awareness and customers’ purchase decision
Summary of Chapter 2
This chapter codify the rationale for customer decision and the factors that influence customer decision; synthesize and compare with previous studies so that the thesis has identified the factors leading to customers online shopping decisions at supermarket website; the changes in customers’ perceptions in the shopping process before and after Covid-19 pandemic
RESEACH METHODOLOGY
Design of research
An effective research process starts with clearly defining the research problem and objectives, followed by a literature review to understand the current state of knowledge in the field and to identify the theories guiding the study The next step is research design, which involves choosing the methodology, developing measurement scales or instruments, and planning data collection Finally, the collected data are analyzed to generate findings, insights, and implications that advance understanding in the domain.
The research topic investigates the factors that affect customers’ decisions to buy fresh food online, exploring how attitudes toward online grocery shopping influence purchasing behavior The study uses two types of survey questions: open-ended questions, which allow respondents to freely express their feelings and perspectives about the factors influencing online fresh-food purchases, and closed-ended questions, which present prepared statements for respondents to rate their level of agreement on a five-point scale (strongly agree, agree, neutral, disagree, strongly disagree) This questionnaire design captures both qualitative insights and quantitative attitudes, enabling a comprehensive analysis of consumer decision-making in the fresh-food market.
With a closed-ended question format with statements, the researcher will receive an answer sheet that clearly shows the consumer's assessment of the problem to be researched
Open-ended questions generate a wide range of responses from consumers and experts, making it difficult to control, quantify, or draw a single general conclusion from the data Despite this complexity, the format offers an objective view of the factors that matter most to consumers, helping researchers uncover the true drivers of consumer choices This insight supports building an effective measurement scale for robust data analysis and actionable insights.
There fore, his study is carried out through three phases: (1) Preliminary research,
This formal research begins with a preliminary study that uses a qualitative method, employing group discussions with customers who have shopped for fresh food through supermarket websites during the Covid-19 pandemic The author elicits in-depth information by posing a set of targeted questions within a focused group outline, enabling rich insights into online grocery shopping experiences and expectations during the health crisis.
1 In your opinion, what factors will affect the decision to buy fresh food online during the Covid-19 pandemic in Ho Chi Minh City?
2 Of the factors that you have listed, which factor do you feel will have the strongest impact on the decision to buy fresh food online during the Covid-19 pandemic in Ho Chi Minh City? Ho Chi Minh City?
3 In terms of Attitude, what aspects of Attitude factors do you think will affect customers' decision to buy food at Co.op Mart supermarket in Ho Chi Minh City?
4 In terms of Cognitive Behavioral Control, what aspects of the group of Cognitive Behavioral Control factors do you think will affect the decision to buy fresh food online during the pandemic period? The covid-19 pandemic occurred in Ho Chi Minh City?
5 In terms of Subjective Standards, what aspects do you think belong to the group of Subjective Norms that will affect the decision to buy fresh food online during the Covid-19 pandemic? happened in Ho Chi Minh City?
6 In terms of Perceived usefulness, what aspects of the group of factors Perceived usefulness do you think will affect the decision to buy fresh food online during the pandemic period? Covid-19 occurred in Ho Chi Minh City?
7 In terms of Risk awareness, what aspects of the group of factors Risk awareness do you think will affect the decision to buy fresh food online during the pandemic period? Covid-19 occurred in Ho Chi Minh City?
8 Do you have any other comments?
Through group discussions and in-depth interviews, the study refined the scale for suitability Subsequently, a pilot survey involving 10 consumers was conducted to test the questionnaires, and the results indicated that all participants understood and agreed with the questionnaires (Appendix 1).
Table 3.1 Measurement scale of observed variables
People who influence my behaviour would think that I should buy fresh food online at supermarket’s website
I chose to buy fresh food online at supermarket’s website because of government regulations to limit community spread
Ha and Hoang Dam Luong Thuy (2020)
People whose opinions I value would approve of me buying fresh food online at supermarket’s website
SN4 I choose this online form of shopping after seeing many people using it Ajzen (1991)
People who are important to me would agree if I buy fresh food online at supermarket’s website
PBC1 If I wanted to, I could buy fresh food online at supermarket’s website
I think this form of shopping is in line with the Government's directive on social distancing
Ha and Hoang Dam Luong Thuy (2020)
I have all the information I need to make a decision to buy fresh food at supermarket websites
I have enough resources, knowledge, and competence to buy fresh food online at supermarket’s website
Taylor and Todd (1995), George (2004) Khalifa and Limayem
PBC5 I know how to buy fresh food at supermarket websites
Buying fresh food on the supermarket's website is safe because it helps me limit in contact with people during Covid-19 pandemic
Taylor and Todd (1995) George, 2004 Khalifa and Limayem
Buying fresh food on the supermarket's website is convenient because supermarket's websites have many flexible payment methods and delivery of goods
Taylor and Todd (1995) George, 2004 Khalifa and Limayem
Buying fresh food on the supermarket's website is easy because the supermarket's website offers a wide variety of goods
Taylor and Todd (1995) George, 2004 Khalifa and Limayem
AT1 It's worth the money I spent
AT2 I trust this form of online shopping
AT3 I think this form of shopping is the best choice for myself
Taylor and Todd (1995), de Matos and
I think this form of shopping is a good idea in the context of the Covid-19 pandemic
Others may know my information when I buy fresh food on the supermarket's website in the context of
RA2 I am always at risk when online shopping Pavlou, 2001
Using the online shopping on the supermarket's website services may cost you money in your account
Online shopping on the supermarket's website seems to be a safe and reliable service
I will continue to buy fresh food on the supermarket's website in the context of
Recommended by the author PD2
I always give priority to buying fresh food on the supermarket's website when social distancing
PD3 I bought fresh food on the supermarket's website more times when I had a demand
I will introduce friends and relatives to buy fresh food on the supermarket's website in the context of Covid-19
Research method
The qualitative research was conducted as follows:
For qualitative research, the study focuses on a customer group of eight individuals who have purchased fresh food from the Bach Hoa Xanh online platform The author requested data directly from the Bach Hoa Xanh website (https://www.bachhoaxanh.com/) to support the analysis and ensure accurate insights into online grocery behavior.
This article begins by inviting members to share brief personal information to foster a smoother discussion, then engages customers with exploratory, open-ended prompts to reveal the factors they consider when buying fresh food online It then outlines the factors that influence purchasing decisions at Bach Hoa Xanh, analyzed within the author’s proposed discussion model that encourages member input and expression of opinions Finally, it presents a draft scale for customers to discuss and provide feedback, offering a structured tool for evaluating their preferences and decision criteria when shopping for fresh food on the Bach Hoa Xanh website.
Qualitative research confirms that the factors influencing consumers’ decisions to buy fresh food on Bach Hoa Xanh’s customer website remain consistent with the model originally proposed by the author The study reiterates that key drivers—such as product freshness, perceived quality, price transparency, site usability, trust, and reliable delivery—shape online purchase decisions for fresh groceries By preserving the original theoretical framework, these findings show how the Bach Hoa Xanh online platform translates expectations for freshness and reliability into actual consumer behavior, providing actionable insights for optimizing user experience and information clarity to boost conversions in online grocery shopping.
Specifically, the research model includes:
- Dependent variable: Purchase decision (PD)
-Independent variables include: (1) Subjective norm, (2) Perceived behavioral control, (3) Attitude, (4) Perceiver usefulness, and (5) Risk Awareness
A preliminary quantitative study was conducted through a pilot survey of 10 customers who purchase fresh food via a supermarket's website, using a pre-designed preliminary questionnaire (Appendix 2) Participants were selected by random sampling The results of this preliminary research will be used by the author as data to test the study's instruments and to inform subsequent stages of the research.
According to Hoang Trong and Chu Nguyen Mong Ngoc (2008), the reliability of the preliminary scale for the factors affecting the customer's decision to buy food is evaluated using Cronbach's Alpha Variables whose item-total correlation coefficients are below 0.3 are rejected The scale is accepted for analysis in the next steps only when its Cronbach's Alpha reliability is 0.6 or higher.
After collecting preliminary research data, the author conducts a preliminary assessment of the scale of independent variables using Cronbach's Alpha reliability coefficient.
Formal research
According to Hair et al (1998), exploratory factor analysis requires a minimum sample size of at least five observations per observed variable, and to ensure sample quality and a robust distribution for generalization, researchers should aim for adequate sampling For multivariate regression analysis, the minimum sample size is given by the formula n = 50 + 8m, where m is the number of independent variables (Tabachnick & Fidell, 1996) Consequently, a sample size of 300 was chosen to meet the required size and support reliable analysis.
To conduct this study, a non-probability sampling method is employed The advantages of this approach include convenience, ease of implementation, time savings, low cost, and high objectivity This exploratory study aims to gain an in-depth understanding of the problem, with a survey questionnaire distributed to respondents to collect the necessary data.
To investigate the factors shaping customers' decisions to purchase fresh food on supermarket websites, this study adopts closed-ended questions as its primary research method In addition, with the goal of determining the key determinants of online grocery purchasing behavior, the research design emphasizes quantifiable responses that can be analyzed to identify which factors most strongly influence buying decisions on online fresh-food platforms.
To measure the impact of each factor on customers’ decision to buy food on supermarket websites, this study uses a closed-ended questionnaire with a 5-point Likert scale, allowing respondents to rate items across five levels The Likert responses are treated as an interval scale, enabling quantitative processing to identify linear correlations and other relationships between factors and online purchasing decisions Guided by the proposed research model, established theories, relevant prior studies, and the local context, the author develops a draft scale to evaluate results The instrument assesses the impact of factors on customers’ decisions to buy fresh food on supermarket websites during COVID-19 by presenting statements that define each factor and measure its influence.
In this study, the variables are applied according to the 5-poinnt Likert scale, which is conventional to the scale of the scale by score as follow:
1 Absolutely disagree 2 Disagree 3 Normal 4 Agree 5 Absolutely agree
To ensure objectivity in the research, the measured variables are coded into symbols, specifically as follows:
Table 3.2 Coding the variables of the scale
People who influence my behaviour would think that
I should buy fresh food online at supermarket’s website
I chose to buy fresh food online at supermarket’s website because of government regulations to limit community spread
SN3 People whose opinions I value would approve of me buying fresh food online at supermarket’s website
SN4 I choose this online form of shopping after seeing many people using it
SN5 People who are important to me would agree if I buy fresh food online at supermarket’s website
PBC1 If I wanted to, I could buy fresh food online at supermarket’s website
PBC2 I think this form of shopping is in line with the
Government's directive on social distancing
PBC3 I have all the information I need to make a decision to buy fresh food at supermarket websites
I have enough resources, knowledge, and competence to buy fresh food online at supermarket’s website
PBC5 I know how to buy fresh food at supermarket websites
Buying fresh food on the supermarket's website is safe because it helps me limit in contact with people during Covid-19 pandemic
Buying fresh food on the supermarket's website is convenient because supermarket's websites have many flexible payment methods and delivery of goods
Buying fresh food on the supermarket's website is easy because the supermarket's website offers a wide variety of goods
AT1 It's worth the money I spent AT2 I trust this form of online shopping AT3 I think this form of shopping is the best choice for myself
AT4 I think this form of shopping is a good idea in the context of the Covid-19 pandemic
Others may know my information when I buy fresh food on the supermarket's website in the context of Covid-19 RA2 I am always at risk when online shopping
Using the online shopping on the supermarket's website services may cost you money in your account
RA4 Online shopping on the supermarket's website seems to be a safe and reliable service
PD1 I will continue to buy fresh food on the supermarket's website in the context of Covid-19
PD2 I always give priority to buying fresh food on the supermarket's website when social distancing
PD3 I bought fresh food on the supermarket's website more times when I had a demand
I will introduce friends and relatives to buy fresh food on the supermarket's website in the context of
The survey questionnaire is designed to collect data regarding each of the research objectives
Developed by Sir Francis Galton (1822-1911), the questionnaire is a standardized research tool designed to efficiently collect respondent responses A typical survey questionnaire comprises two parts: personal information and Likert-scale items Personal information is gathered with closed-ended questions, providing structured data, while questions about influencing factors or attitudes are measured on the Likert scale to capture varying levels of agreement or frequency This approach ensures consistent data collection and simplifies subsequent statistical analysis in social science research.
Thirty-four factors influencing consumer purchasing decisions are assessed using a five-point Likert scale, ranging from 1 (completely disagree) to 5 (completely agree) The variables included in this thesis are selected based on a comprehensive literature review, with details of the selection presented in Appendix 1.
The detailed structure of the questionnaire is presented in the form of a survey form consisting of 3 main parts as follows (APPENDIX 2)
Part 1: Introduction to the survey and instructions for answering
Part 2: The questions revolve around the personal information of the surveyed people so that the author can make statistics and describe the survey sample
Part 3: The study's central questions are quantitative questions, using a 5-level Likert scale to survey the level of agreement for each observed variable of the study
This study adopts a non-probability sampling approach, using convenience sampling for its ease of implementation, time efficiency, low cost, and straightforward objectivity It is an exploratory study aimed at gaining a deep understanding of consumer behavior in online grocery shopping during the COVID-19 pandemic In this design, survey subjects who purchased fresh food on the supermarket's website will receive a survey questionnaire.
This study examines 300 adults aged 18 to over 55 who resided in Ho Chi Minh City during the COVID-19 pandemic, and data collection will be conducted using two methods: questionnaires and surveys.
Data analysis
Data were collected as primary data through a structured survey The author first screened and discarded invalid questionnaires to ensure data quality Valid responses were then entered into Excel for storage and organization, after which the data were imported into SPSS 22.0 for descriptive analysis.
35 statistics, evaluate the scale, determine the importance of the factors, and test the hypotheses
Reliability was assessed using Cronbach's Alpha to screen items before exploratory factor analysis (EFA) and to remove inappropriate variables that could introduce dummy factors (Nguyen Dinh Tho and Nguyen Thi Mai Trang, 2009) Cronbach's Alpha indicates whether the measurements are interrelated, but it does not identify which observed variables should be dropped or kept; therefore, the correlation between the total variables was calculated to exclude items that contribute little to describing the underlying concept When the correlation coefficient of the total score is small (less than 0.3), the item should be discarded Scale selection relies on Cronbach's Alpha values greater than 0.6, with higher alpha indicating higher internal consistency reliability (Nunnally, 1978).
Many researchers regard Cronbach's Alpha as a key measure of scale reliability and internal consistency: 0.8 or higher is considered good, 0.7–0.8 is usable, and 0.6 or higher can be acceptable when the concept being measured is new to respondents This guideline is often attributed to Nunnally.
1978) Therefore, items with Cronbach's Alpha coefficient less than 0.6 and corrected Total Item correlation less than 0.3 will be excluded from the model
After evaluating the scale's reliability using Cronbach's Alpha of the factors, the next step is to analyze the EFA exploratory factor
Exploratory Factor Analysis (EFA) is a statistical method used to reduce a large set of observed variables into a smaller set of general, interpretable latent factors based on the correlations among the variables By identifying these latent factors, EFA clarifies the data structure, enhances measurement efficiency, and supports scale development In this study, EFA analysis was employed to evaluate convergent validity—ensuring that items measuring the same construct correlate highly with their associated factor—and discriminant validity—demonstrating that factors representing different constructs are distinct from one another.
The EFA analysis must satisfy the following criteria:
1 To decide whether to keep or remove the observed variable to ensure that the scale reaches the convergent value, it must be based on factor loading If the loading factor is less than 0.5, it will be removed from the model and vice versa Factor loading ≥ 0.5 (Hair et al., 2006)
2 To determine the number of factors, we rely on the Eigenvalue index If the Eigenvalue index is less than 1, that factor will be excluded from the research model
3 Total Variance Explained to evaluate the scale: The scale is accepted when the extracted variance > 50% shows that the EFA model is suitable (Hair et al., 2006) Considering the variation as 100%, Total Variance Explained shows how the factors explain the variation of the variation
4 To determine the appropriateness of the EFA factor analysis, we use the Barlett test and the Kaiser - Meyer - Olkin coefficient (KMO) KMO coefficient must be greater than 0.5 (0,5 ≤ KMO ≤ 1) If the KMO coefficient is less than 0.5, the model is not accepted for EFA factor analysis
5 To test observed variables are correlated with each other in the population The Barlett test has statistical significance (Sig.