INTRODUCTION
Problem discussion
As the country develops, building a home transcends mere shelter, becoming a sanctuary for relaxation and a reflection of the owner's personality This shift emphasizes the need for aesthetic and diverse living space design and interior decoration With an average GDP growth rate of 6% annually, coupled with advancements in the construction and real estate sectors, rising household incomes are driving increased consumption of home furniture.
Vietnam is a leading player in the global furniture market, ranking first in Southeast Asia, second in Asia, and fourth worldwide in furniture exports, as reported by EVBN In 2015 alone, Vietnam's furniture exports to Europe reached $7.2 billion, with an additional $1.7 billion from home decoration items The furniture manufacturing industry in Vietnam is projected to grow steadily at a rate of 9.4% annually However, the current market is primarily export-focused, leaving the domestic market largely dominated by imported goods from countries like China, Malaysia, and Thailand.
The domestic furniture market presents significant opportunities for businesses, as the home furniture sector encompasses various segments catering to diverse income groups Understanding customer preferences is challenging due to the vast and varied consumer base, making it essential to research consumer behavior regarding home furnishings Currently, there is a lack of studies in Vietnam that analyze consumer behavior towards interior products Consequently, my thesis focuses on "Factors Affecting Consumer Behavior in the Home Furniture Market: A Case Study of Hanoi."
What are the factors which affects consumers purchasing intention when buying home furniture?
Research objective
This thesis aims to assist furniture manufacturers and retailers in Vietnam by analyzing consumer behavior and purchase intentions within the industry By examining the various positive and negative influences on customers' purchasing decisions, the research will provide valuable insights Understanding these consumer behaviors will enable manufacturers and retailers to tailor their strategies effectively, aligning them with market demands.
Research scope
This research explores consumer behavior theory and builds on previous studies, employing a two-phase approach that includes qualitative and quantitative research Initially, secondary data is synthesized to identify factors influencing consumer purchasing behavior Subsequently, a consumer opinion survey is conducted in Hanoi to gather primary data Based on the analysis of this data, the author proposes several strategies to assist businesses in expanding the home furniture market in Hanoi.
Object and scope of the study
- Object: Home furniture consumers in Hanoi
- Scope of the study: Research is carried out to find out the purchase intention of home furniture in the period of 2017 - 2020 and vision to 2025.
Thesis outline
This thesis is structured into six chapters: an introduction, a literature review, the development of hypotheses and framework, research methodology, results of hypothesis testing, and a conclusion with recommendations Following the main body, the thesis includes references and two appendices, featuring questionnaires in both English and Vietnamese.
LITERATURE REVIEW
Overview of consumer behavior
Consumer behavior, as defined by M Khan (2007), encompasses the decision-making process regarding what, how, where, and when to purchase products, including the quantity It also involves choosing the appropriate source for buying Additionally, consumer behavior examines both internal and external factors that influence purchasing decisions, such as self-concept, social and cultural backgrounds, age, family dynamics, attitudes, personality traits, and social class.
Consumer behavior encompasses the actions taken by individuals during the investigation, purchase, use, and evaluation of goods and services to fulfill their needs It reflects how consumers make decisions regarding the allocation of their resources—such as money, time, and effort—when acquiring products and services to meet personal requirements (Kotler, 2007).
Marketers study consumer behavior to understand their needs, preferences, and purchasing habits, including what products or services they desire, the reasons behind their choices, and their buying patterns This analysis helps in formulating effective marketing strategies that encourage consumers to make purchases.
Consumer behavior encompasses the thoughts, emotions, and actions involved in the decision-making process when purchasing goods or services, influenced by both external environmental stimuli and internal psychological factors.
Consumer intentions are a key indicator of their willingness to engage in specific behaviors, particularly in terms of purchase decisions (Ajzen, 1991) Purchase intention serves as the primary predictor of actual buying behavior (Montaủo and Kasprzyk, 2015) This research emphasizes the importance of purchase intentions as a crucial variable, occurring at the pre-purchase stage and reflecting the motivational elements that influence customer behavior To effectively predict consumer behavior, it is essential to understand the attitudes, evaluations, and internal factors that contribute to purchase intent (Fishbein and Ajzen, 1977) Consequently, this study posits that purchase intention directly influences purchase decisions, focusing on the factors that affect the intention to buy home furniture.
Factors affecting consumer behavior
Consumer behavior is shaped by various external factors and individual characteristics, which influence how consumers perceive and respond to stimuli The environment plays a significant role in shaping consumer preferences, while consumer behaviors also impact the surrounding environment (Blackwell et al., 2006) Additionally, purchasing decisions are heavily influenced by cultural, social, personal, and psychological factors (Kotler & Armstrong, 2004), as illustrated in Figure 2.2.
Figure 2.1: Characteristics Influencing Consumer Behavior
Cultural factors that exert intensely influence on consumer behavior consist of culture, subculture, and social class factor
Social factors such as reference groups, family, and social roles and status also impact consumer behavior
A purchaser’s decisions are also influenced by personal characteristics such as the purchaser’s age and life-cycle stage, occupation, economic situation, lifestyle, and personality
An individual’s purchasing choices are further influenced by four major psychological factors Four factors consist of motivation, perception, learning, and beliefs and attitudes
Economic scientists were pioneers in analyzing consumer behavior, offering insights into solutions for consumption issues They view humans as social and rational beings, emphasizing key economic factors such as personal income, family income, income expectations, liquid assets, and government policy in understanding consumer decisions.
Previous research about purchasing behavior of consumers
In this study, the author mentioned the consumer behavior model of Kotler
(2007) and the theory of reasoned action (TRA) and several researches related to consumer behavior in the furniture market
According to Kotler's model, consumer behavior is influenced by the four elements of the marketing mix, which interact to shape the consumer's decision-making process In a price-sensitive market, selecting the right pricing strategies is crucial for gaining a competitive advantage and effectively guiding product selection.
Consumers respond to various external stimuli, including the marketing mix and environmental factors within the market The marketing mix, known as the four Ps, consists of strategically planned stimuli developed by the company to influence consumer behavior.
The environmental stimuli are supplied by the economic, political, and cultural circumstances of a society Together these factors represent external circumstances that help shape consumer choices
The internal factors influencing consumer decisions are often referred to as the "black box," which encompasses various elements within an individual's mind This includes consumer characteristics such as beliefs, values, motivation, and lifestyle The decision-making process is integral to this "black box," as consumers identify problems that need resolution and evaluate how purchasing decisions can address these issues As consumers react to external stimuli, their internal "black box" processes these factors to guide their ultimate response—whether to make a purchase or not.
Similar to the economic man model, this framework posits that consumer responses stem from a deliberate and rational decision-making process, regardless of internal mental processes However, many marketers doubt this notion, believing that consumers frequently make impulsive or emotional purchasing choices In reality, marketers recognize that consumers' emotional and irrational behaviors often render them more receptive to marketing influences.
Consumer purchasing behavior is often viewed as a "black box," as individuals frequently lack insight into the factors influencing their choices This unpredictability complicates the exchange process, making it challenging for marketers to comprehend and anticipate consumer decisions.
The Theory of Reasoned Action (TRA), developed by Fishbein and Ajzen in 1975, is a prominent framework in social psychology that explores the relationship between beliefs, attitudes, intentions, and behaviors.
The Theory of Reasoned Action (TRA) explores how beliefs, norms, attitudes, intentions, and behaviors are interconnected It posits that an individual's behavioral intentions are the primary predictors of their actual behavior These intentions are shaped by the person's subjective norms and attitudes toward the behavior in question Additionally, the TRA indicates that modifying an individual's belief structure can influence their attitudes through external stimuli Furthermore, it acknowledges that external variables can indirectly affect behavior by altering subjective norms.
The TRA model focuses on the construction of a system of observation of two groups of variables, which are:
+ attitudes defined as a positive or negative feeling in relation to the achievement of an objective;
+subjective norms, which are the very representations of the individuals’ perception in relation to the ability of reaching those goals with the product
The author emphasizes the importance of purchase intention to actual purchase decision
When purchasing furniture, various important factors come into play This research highlights the key elements that significantly influence furniture buying decisions and examines their impacts on consumer choices.
Research “Factors Affecting Furniture Purchase in Pakistan”, M F Qureshi &
A Kamaran et al give 5 factors to consider including: (1) Price, (2) Customer service,
The study by M F Qureshi & A Kamaran et al (2020) reveals that price significantly influences furniture purchases, while product quality plays a crucial role in shaping customer attitudes Additionally, brand loyalty is identified as a key factor in influencing customer perceptions The research also confirms a strong relationship between customer reviews and customer attitudes, supporting the hypothesis that positive feedback impacts consumer behavior.
A study by Parmana et al (2019) examined the factors influencing purchasing decisions for wooden furniture at Furnimart Bogor The research concluded that consumer perceptions significantly affect furniture buying behavior, with key influences including product design, completeness, quality, and discount rates of up to 70% Additionally, factors such as convenient payment methods, accessible store locations, fast delivery, store convenience, attractive promotions, quality sales assistance, and social media marketing also play crucial roles The SEM analysis identified price and promotion as the most impactful factors in the purchasing decision process.
Research “Consumer behavior in purchasing home furnishing products in Thailand”, Thanyamon consider 7 factors that have influences on furniture purchasing decision process including: (1) Product quality, (2) Product design, (3) Brand loyalty,
When selecting furniture, key factors include product quality, design, and price, with after-sales services, store location, and delivery speed also playing significant roles According to Thanyamon Sakpichaisakul (2012), these elements collectively influence consumer choices in the furniture market.
In their 2009 study, "Understanding Furniture Decision Making Process and Design Preference using Web-Based VR Technology," So-Yeon Yoon and Ji Young Cho highlighted the critical role of product design in influencing consumer purchasing decisions Their research underscores how web-based virtual reality technology can enhance the furniture selection process by allowing consumers to visualize and interact with designs before making a purchase.
The research on the "Consumer Behavior Model in the Furniture Market" identifies five key factors influencing purchasing decisions: product quality, price, customer service, social media, and customer reviews Among these, product quality emerges as the most critical criterion for consumers when considering furniture purchases.
Table 2.1 shows some of the research works on this topic and the factors investigated in those prior studies
Table 2.1 Factors influence purchasing intention home furniture
1 Attitude 1 M F Qureshi & A Kamaran et al (2020)
2 Product quality 4 M F Qureshi & A Kamaran et al (2020)
3 Product design 1 Parmana et al (2019)
3 So-Yeon Yoon & Yi Young Cho (2009)
4 Price 1 M F Qureshi & A Kamaran et al (2020) h
5 Store location 1 Parmana et al (2019)
6 Social media 1 Parmana et al (2019)
7 After-sales service 1 Parmana et al (2019)
Literature gap
In the past five years, Vietnam has seen the emergence of approximately 400,000 to 500,000 townhouses and luxury apartments, according to Mr Phan Dang Chuong, Deputy General Director of Ernst & Young Vietnam Limited Each apartment typically incurs interior costs ranging from 100 to 200 million VND, resulting in a total expenditure exceeding 100 trillion VND for this segment Understanding the factors influencing consumer buying behavior is crucial, as these factors can vary significantly across individuals, cultures, and regions This research will primarily focus on the Vietnamese market.
This study also aims to examine how the impact of factors changes when furniture consumption trends and furniture demand are changing.
Hypotheses development
2.5.1 Relationship between customer attitude and purchase intention
Various factors influence consumers' purchase intentions, including personal preferences, social and monetary standings (Mirabi et al., 2015) Key elements such as perceived quality, value, and price play a significant role, alongside both internal and external motivations Researchers outline six stages leading to the decision-making process: knowledge, awareness, interest, preference, persuasion, and purchase (Kawa et al., 2013) The concept of purchase intention is extensively studied in marketing literature, with research conducted globally to uncover its determinants (Kar et al., 2018) Additionally, brand-related elements like product quality, involvement, image, loyalty, knowledge, and attributes significantly impact buyers' purchase intentions (Abbasi et al., 2015).
Attitude plays a crucial role in shaping individual predispositions and is positively linked to behavior, as noted by Allport (1935) Defined by Fishbein and Ajzen (1977), attitude reflects the extent to which a person evaluates behavior positively or negatively In this study, consumer attitude is specifically examined in relation to purchasing home furniture, building on the findings of Andrews and Bianchi (2013).
According to the Theory of Reasoned Action (TRA), a positive attitude towards a behavior leads to a higher intention to engage in that behavior (Amaro and Duarte, 2015) Therefore, if consumers have a favorable assessment of purchasing home furniture, their intention to buy will likely increase This leads to the first research hypothesis.
H1: Attitude has a significant positive effect on purchase intention
2.5.2 Relationship between product quality and customer attitude
Product quality encompasses the characteristics and features that determine a product's ability to meet specific requirements Brands influence perceived quality by positioning themselves favorably against competitors Higher product quality significantly impacts customer attitudes, as evidenced by research indicating a strong correlation between the two To achieve quality, producers must ensure their products fulfill consumers' expectations for durability.
2013) Hence the second hypothesis is:
H2: Product Quality has a significant positive effect on customer attitude
2.5.3 Relationship between design product and customer attitude
Psychologist Carl Jung (1967) highlighted that the self-archetype can manifest through self-expression in physical structures The home serves as a tangible representation of one's self-image (Cooper, 1976) Furthermore, So-Yeon Yoon and Ji Young Cho (2009) underscored the critical role of product design in influencing purchasing decisions Therefore, the third hypothesis is established.
H3: Product Design has a significant positive effect on customer attitude
2.5.4 Relationship between price and customer attitude
Price is a critical factor influencing consumer purchasing decisions, prompting businesses to adopt various pricing strategies to meet diverse customer needs The relationship between price and buying behavior is complex; higher-priced items are not always deemed undesirable In fact, consumers may associate higher prices with greater perceived value and quality, leading them to view these products more favorably.
Price is a crucial determinant in consumer purchasing decisions across a wide range of products and services, with the exception of essential items such as lifesaving medications This leads to the formulation of the fourth hypothesis.
H4: Price has a significant positive effect on customer attitude
2.5.5 Relationship between store location and customer attitude
Location is crucial in retailing, significantly influencing consumer store choice An ideal location is perceived as convenient for shopping, with good public transport access, ample parking, and a consumer-friendly environment Store characteristics, such as layout, sales personnel, atmospherics, and type, play a vital role in shaping the shopping experience The design and presentation of merchandise are key elements of store characteristics that retailers utilize to encourage impulsive purchases Consumer shopping behavior can be categorized into planned and impulse purchases, with planned decisions made prior to store visits and impulse decisions triggered by in-store stimuli Factors like store layout, signage, feature areas, visual merchandising, and atmosphere—all influence impulsive buying Understanding these in-store characteristics is essential for retailers, as impulse purchases account for a significant portion of supermarket sales.
Thus, the fifth hypothesis is:
H5: Store location has a significant positive effect on customer attitude
2.5.6 Relationship between social media and customer attitude
Social media serves as a platform for individuals to express and share their ideas, fostering connections similar to those formed over centuries Concurrently, significant shifts in consumer behavior, driven by changes in values, lifestyles, and a desire for value for money, have emerged, resulting in a new generation of tourism consumers who are more informed, independent, and individualistic (Poon, 1993) Additionally, travel-related consumer behavior has become increasingly paradoxical, as consumers exhibit conflicting holiday preferences within short timeframes; they are willing to invest in luxury travel experiences while simultaneously seeking the best hotel rates online (Gretzel et al., 2006).
Hence the sixth hypothesis is:
H6: Social media has a significant positive effect on customer attitude
2.5.7 Relationship between after-sales service and customer attitude
After-sales service is essential for ensuring that customer needs are thoroughly addressed, aligning with their expectations Effective customer service plays a crucial role in meeting the requirements of every client The quality of customer service directly influences customer satisfaction and their overall perception of their purchase.
K, Khaksar SMS, 2011) Some qualities of good customer service include Promptness i.e., ensuring that things are always on time and no customer has to be kept waiting (Hsu
Politeness and professionalism are essential in enhancing customer experience Being sociable and sweet in conduct helps please the customer (CL, Chang KC, Chen MC, 2012; Jahanshahi AA et al., 2011) Additionally, interactions should be handled with professionalism, focusing on direct and relevant conversations to address the customer's needs effectively (Bashir F, Soroya SH et al.).
Personalization in customer service enhances the experience by addressing customers by their names, making them feel valued (Jasmand et al., 2012) The effectiveness of customer experience management depends on the nature of the product or service offered, the specific needs of customers, and whether the service is problem-focused or aimed at improving the overall consumer experience (Agnihotri et al., 2015).
Hence the seventh hypothesis is:
Research model
RESEARCH METHODOLOGY
Research process
The research was conducted by following steps which shown in the figure below:
Data collection
3.2.1 Data collection method: survey method
Therefore, in this study, the survey is used to be the main method due to the following reasons:
The researcher developed a user-friendly online survey that facilitates quick administration (Cherry, 2018) Various methods exist for creating these surveys, allowing users to design questions and share links with respondents Responses are collected in real-time, presented in charts or Excel format for immediate analysis.
Surveys are a cost-effective data collection method compared to focus groups and interviews, as noted by Cherry (2018) Unlike direct meetings that require time constraints for respondents, online surveys are free to create and allow participants to complete them at their convenience.
Finally, the survey method can be applied to use for collecting information on a broad range of sample which is includes personal fact, the attitudes, respondents’ behaviors, and opinions (Cherry, 2018)
In general, the survey is the most suitable method to gather the necessary data required for this analysis
According to Hair et al (2014), a minimum sample size should be at least five times the number of items in a questionnaire, meaning each variable requires a minimum of five respondents This study adopts this guideline for determining sample size, given its quantitative approach and a total of 32 variables Therefore, the calculated sample size should exceed 160 respondents (32 variables x 5) However, this research successfully collected 270 samples from participants, surpassing the minimum requirement.
Convenience sampling involves selecting participants from a target population based on practical criteria such as geographical proximity, accessibility, availability, and willingness to participate Often referred to as "accidental samples," these participants are chosen simply because they are nearby during the data collection process In the context of online surveys, respondents typically include individuals aged 18 and older, who have experience in purchasing furniture, and come from diverse backgrounds in terms of occupation, education level, and housing type.
Questionnaire design
This research article explores the factors influencing purchasing decisions for home furniture by utilizing a Likert scale to assess participants' opinions The evaluation comprises multiple items, allowing users to indicate their level of engagement effectively.
The study employs Likert scales, which demonstrate sufficient test-retest reliability, concurrent validity, and predictive validity (Jacoby & Mattel, 1971) The choice of the Likert scale is justified by its ability to capture primarily the directional component of total scores, with only a minor focus on intensity (Peabody, Cronbach, 1962) This approach is particularly beneficial for research centered on user actions.
In addition, the Likert scale offers a perfect way to assess attitudes, awareness, beliefs, values and behaviors
The questionnaire of this research was designed to be simple for the respondents easily to understand and answer This approach is in harmony with Easterby-Mith et al
In 2008, it was suggested that shorter questionnaires with simpler questions tend to yield higher response rates The questionnaire was initially crafted in English and subsequently translated into Vietnamese for local respondents The questions included were adapted and refined from those used in prior research studies.
The questionnaire is used in this study composed of two sections: h
• The first section is introduction and it contained 10 questions for demographic information
• The second section consisted 32 measuring items of 8 constructs including Purchase intention, Customer attitude, Product quality, Product design, Price, Store location, Social media, After-sales service
Table 3.1 Measuring items for questionnaire
1 Buying home furniture is something I would do
2 I feel comfortable in sharing my information about furniture product
3 I usually buy a few new products when I move to a new residence
4 Even when there is no need I still like to see furniture products
5 I often think about buying new products Thanyamon
6 I am interested in information of new furniture models
7 I believe it is necessary to buy new furniture regularly
8 My family influences my furniture purchasing a lot
9 A salesperson can influence my furniture selection process
10 I like products with a lifelong durability
11 I like products that integrate many functions
12 I usually read the product specifications
13 I like solid products no matter how heavy or thick they appear
14 I like new model rather than current model furniture
15 I usually choose products that match my home décor Éva & Judit
16 If there are two similar products, I would choose the nicer one even though the price might be higher
17 When I buy furniture, I choose the cheapest one
18 I usually buy on sale furniture Parmana et al
19 I like cheaper furniture only if it meets quality requirements
20 For furniture, higher price equals higher quality
21 Discounted furniture means it is left- over or obsolete item
22 I usually buy new furniture at the store near my house
23 I like furniture stores that are conveniently located
24 I prefer independent stores to stores located in shopping malls
25 I tend to buy a furniture that a celebrity I like endorsing it Éva & Judit
26 I tend to buy a furniture that I've seen advertised before
27 I often watch furniture ads on social media
28 I will look at the product review before buying
29 I will buy furniture if store provides delivery service
30 I feel more satisfied if speed delivery Ying Li et al
31 I like products with return policy Parmana et al
32 I will buy furniture if store provides installation service
(2012) Each item will be measured by 5 - Likert scale:
Analyzing data plan
This research utilized SPSS 22 software for data analysis, chosen for its flexibility in managing sample sizes and measurement models Following the collection of online questionnaire responses, each item was systematically coded, as detailed in Table 4.2.
Table 3.2 Encoded terms for data testing
INT1 Buying home furniture is something I would do
INT2 I feel comfortable in sharing my information about furniture product
When relocating to a new home, I typically purchase several new items to enhance my living space Even in the absence of necessity, I enjoy browsing furniture products The idea of acquiring new items frequently crosses my mind.
ATD1 I am interested in information of new furniture models.
ATD2 I believe it is necessary to buy new furniture regularly
ATD3 My family influences my furniture purchasing a lot.
ATD4 A salesperson can influence my furniture selection process.
PQ1 I like products with a lifelong durability
PQ2 I usually read the product specifications
PQ3 I like products that integrate many functions
PQ4 I like solid products no matter how heavy or thick they appear
PD1 I like new model rather than current model furniture
PD2 I usually choose products that match my home décor
PD3 If there are two similar products I would choose the nicer one even though the price might be higher
P1 When I buy furniture, I choose the cheapest one
P2 I usually buy on sale furniture
P3 I like cheaper furniture only if it meets quality requirements P4 For furniture, higher price equals higher quality
P5 Discounted furniture means it is left-over or obsolete item
SL1 I usually buy new furniture at the store near my house
SL2 I like furniture stores that are conveniently located
SL3 I prefer independent stores to stores located in shopping malls h
SM1 I tend to buy a furniture that a celebrity I like endorsing it
SM2 I tend to buy a furniture that I've seen advertised before
SM3 I often watch furniture ads on social media
SM4 I will look at the product review before buying
ASS1 I will buy furniture if store provides delivery service
ASS2 I feel more satisfied if speed delivery
ASS3 I like products with return policy
ASS4 I will buy furniture if store provides installation service
With SPSS software, perform data analysis through tools such as descriptive statistics, frequency tables, reliability testing of scales, exploratory analysis, regression
To assess the reliability of a direct scale, the Cronbach Alpha coefficient is frequently employed as an internal consistency index, indicating whether the scale's questions share a common structure A higher Cronbach's Alpha value signifies greater internal consistency It is advisable to utilize the Cronbach's Alpha reliability coefficient prior to conducting exploratory factor analysis (EFA) to remove unsuitable variables, as these can lead to the emergence of spurious factors (Nguyen Dinh Tho and Nguyen Thi Mai Trang).
Cronbach's Alpha reliability coefficient assesses the relationship between measured variables but does not specify which variables to retain or exclude To enhance the measurement's validity, it is essential to utilize the variable-total correlation coefficient to identify and remove variables that contribute minimally to the intended concept (Hoang Trong and Chu Nguyen Mong Ngoc, 2008) Key criteria for evaluating scale reliability should be established to ensure accurate assessment.
The Cronbach Alpha reliability coefficient is a crucial metric for evaluating the consistency of a scale A coefficient greater than 0.8 indicates a good scale, while values ranging from 0.7 to 0.8 are considered usable Additionally, a coefficient of 0.6 and above may be acceptable if the research concept is novel or has not been extensively studied in the given context, as noted by Nunnally (1998), Peterson (1994), and Slater (1995).
Hoang Trong and Chu Nguyen Mong Ngoc, 2008) ) In this study, the author chose a scale with a reliability of 0.6 or more
The variable-to-sum correlation coefficient indicates that observed variables with a correlation of less than 0.3 are deemed ineffective and will be discarded A scale is considered valid when the reliability coefficient, Cronbach's Alpha, meets the necessary criteria.
Factor analysis is a statistical method employed to condense data by reducing a large set of observed variables into a smaller number of significant factors for enhanced analysis This process retains the essential information from the original data while providing clearer insights Exploratory factor analysis specifically assesses the conceptual validity of measurement scales, ensuring their reliability for further research.
Implementation method and evaluation criteria in EFA exploratory factor analysis:
The multi-directional scale utilizes Principal Axis Factoring with Promax rotation, extracting factors with EigenValues of 1 or greater for improved data representation, as supported by Nguyen Dinh Tho and Nguyen Thi Mai Trang (2007) In contrast, the unidirectional scale employs the Principal Components factor extraction method, which is deemed acceptable when the total variance extracted is 50% or higher.
In Exploratory Factor Analysis (EFA), it is essential for the factor loading to be at least 0.5 to ensure practical significance The levels of factor loading are categorized as follows: a minimum of 0.3 is acceptable, 0.4 is considered important, and a value above 0.5 is deemed practically significant When selecting factor loading thresholds, a sample size of 270 allows for a minimum loading of 0.3, while a smaller sample size of around 100 necessitates a higher loading value for meaningful results.
0.55; if the sample size is about 50 then the factor loading factor must be greater than 0.75
From the above theoretical basis, the model uses 35 observed variables for EFA factor analysis and the implementation follows the following steps:
To effectively measure the component concepts using unidirectional scales, it is essential to employ the Principal Components factor extraction method with Varimax rotation, stopping the extraction process when encountering EigenValues greater than 1 Following this, it is crucial to verify the related requirements to ensure the validity of the results.
+ Bartlett test: observed variables are correlated with each other in the population
+ Consider the KMO value: if the KMO is in the range from 0.5 to 1, then factor analysis is appropriate for the data (Hoang Trong and Chu Nguyen Mong Ngoc, 2008)
+ To analyze EFA with practical value, remove observed variables with factor loading coefficient less than 0.5
+ Review the parameter Eigen Values (representing the variation explained by each factor) with a value greater than 1
+ Consider the total variance extracted (requires greater than or equal to 50%): indicate the extracted factors explaining the % variation of the observed variables
Scales that meet satisfactory evaluation criteria are included in the Pearson correlation analysis, which assesses the linear relationship between dependent and independent variables This analysis confirms the appropriateness of using linear regression The Pearson correlation coefficient (r) ranges from -1 to +1, with values closer to 1 indicating a strong correlation between the two variables A value of r = 0 signifies no linear relationship exists between the variables (Hoang Trong and Chu Nguyen Mong Ngoc).
After concluding that two variables have a linear relationship with each other, this causal relationship can be modeled by linear regression (Hoang Trong and Chu Nguyen Mong Ngoc, 2008)
The study performed multivariate regression by Enter method: all variables were included once and related statistical results were considered
The process of hypothesis testing is carried out according to the following steps:
+ Evaluate the fit of the multivariable regression model through R2 and R2 adjusted
Test hypothesis about the fit of the model
+ Test the hypothesis about the significance of the regression coefficients for each component
+ Hypothesis testing of the normal distribution of residuals: according to the histogram of the normalized residuals; see mean as 0 and standard deviation as 1
+ Check the assumption of multicollinearity through the value of tolerance (Tolerance) or the variance magnification factor VIF (Variance Inflation Factor) If VIF
> 10, there is multicollinearity (Hoang Trong and Chu Nguyen Mong Ngoc, 2008)
In assessing the impact of various influencing factors, the beta coefficient serves as a key indicator; a larger beta coefficient signifies a greater level of influence for that factor when compared to others within the research model.
DATA ANALYSIS & HYPOTHESIS TESTING
Descriptive data
A total of 270 valid survey responses were collected using a convenient sampling method, following the removal of invalid entries that lacked essential information or did not meet age criteria The results of this quantitative analysis are summarized in the accompanying table.
Table 4.1 Description of respondents’ profile
Eligible answers were chosen from 312 surveys that collected from 270 people
+ People aged 21 – 40 accounted for the most with 33.7% and 31.5% Next is the group 41-50 years old, accounting for 18.9% The last group over 50 years old, accounting for 15.9%
+ According to sex, the relative sample is not too big difference between male and female, in which female accounted for 59.3% and male 40.7%
+ In terms of education level, 15.1% of the surveyed sample have a degree High school; University accounted for 55.6% and Postgraduate accounted 29.3%
+ By occupation, the group of Employee accounted for the most with 53.7%; Self-employed accounted for 12.9%; Housewife accounted for 11.5%; Other specific occupations: 51.4%
+ By personal income, the group Under 6 million accounts for 13.7%; 5 – less than 10 million accounts for 14.1%; 10 - 15 million, accounting for 28.1%; 15 – 25 million accounted for 21.5%; 25 – 40 million accounted for 11.1%; group Over 40 million accounted for 11.5% h
+ By Total income, the group Under 15 million accounted for 11.5%; 15-25 million counted for 15.2%; 26-40 million accounted for 26.3%; 41-60 million accounted for 14.4%; group over 100 million accounted for 12.2%
+ According to residential type, the group of Tube house accounted for 20%; Highrise building accounted for 30.4%; Condominium accounted for 37.4%; group of Villa accounted 12.2%
The survey sample is considered representative of the population, as each group's individual characteristics are sufficiently large for statistical analysis, with all samples exceeding 30 participants.
Analysis and result
4.2.1 The reliability of the scale
The reliability of the scale was tested with the following results:
Table 4.2 Cronbach's Alpha analysis results table
Cronbach’s Alpha if Item deleted
Cronbach's alpha coefficient of the factor: 0.797
Cronbach's alpha coefficient of the factor: 0.841 h
Cronbach's alpha coefficient of the factor: 0.689
Cronbach's alpha coefficient of the factor: 0.611
Cronbach's alpha coefficient of the factor: 0.754
Cronbach's alpha coefficient of the factor: 0.896
Cronbach's alpha coefficient of the factor: 0.711
Cronbach's alpha coefficient of the factor: 0.796
The results show that all factors are statistically significant because Cronbach's Alpha coefficient is greater than 0.6, in which:
+ Product Quality with Cronbach's Alpha 0.797 and the total variable correlation coefficient from 0.588 - 0.629 so the variables will be kept
+ Product Design with Cronbach's Alpha 0.841 and the total variable correlation coefficient from 0.663 - 0.735 so the variables will be kept
+ Price with Cronbach's Alpha 0.689 and the total variable correlation coefficient from 0.355 - 0.531 so the variables will be kept
+ Store Location with Cronbach's Alpha 0.611 and the total variable correlation coefficient from 0.320 - 0.453 so the variables will be kept
+ Social Media with Cronbach's Alpha 0.754 and the total variable correlation coefficient from 0.420 - 0.662 so the variables will be kept
+ After-sales Service has the highest Cronbach's Alpha coefficient of 0.896 and the total correlation coefficient at the allowable level of 0.711 - 0.841, showing that the component variables have a close relationship h
+ Attitude with Cronbach's Alpha 0.711 and the total variable correlation coefficient from 0.441 - 0.542
+ Finally, Intention with Cronbach's Alpha 0.796 and the total variable correlation coefficient from 0.531 - 0.669
The evaluation of the scale's reliability identified eight key factors influencing consumer behavior: Product Quality, Product Design, Price, Store Location, Social Media, After-sales Service, Attitude, and Intention These factors will be analyzed through Exploratory Factor Analysis (EFA) to uncover their relationships and significance.
4.2.2.1 Exploratory factor analysis (EFA) for 6 independent variables
The model identifies six independent variables supported by 23 statistically significant observed variables These independent variables will be retained for further analysis in the exploratory factor analysis (EFA) scale test.
EFA analysis for 6 independent variables was performed with the hypothesis H0: The observed variables have no correlation in the population The results obtained from the analysis are summarized as follows:
▪ Barlett test: Sig = 0.000 < 5%: Rejecting hypothesis H0, observed variables in EFA analysis are correlated with each other in the population
▪ KMO coefficient = 0.922 > 0.5: factor analysis is required for the data
▪ There are 6 factors extracted from EFA analysis with:
EigenValues of all factors are > 1: satisfactory
The exploratory factor analysis demonstrated compliance with requirements, as the total value of variance extracted reached 65.903%, exceeding the 50% threshold This indicates that the six extracted factors collectively account for 65.903% of the data variation Additionally, the observed variables exhibited a difference in factor loading coefficients greater than 0.3, highlighting the high discriminatory value of the factors.
Table 4.3 Table of results of EFA analysis of independent variables
4.2.2.2 Exploratory factor analysis (EFA) for 2 dependent variables
Attitude scale used to measure customer’s attitude includes 4 observed variables The results of the EFA analysis showed:
▪ 5 observed variables are grouped into 1 factor Factor loading coefficients are all > 0.5, so they have practical significance
▪ Each observed variable has a difference in factor loading coefficient ≥ 0.3, so it ensures the distinction between factors
▪ KMO coefficient = 0.838 > 0.5 factor analysis is required for the data
▪ The Chi-square statistic of Bartlett Test reached the significance level value of 0.000 Therefore, the observed variables are correlated with each other on the overall scale
The extracted variance is 61.411%, showing that one factor can be explained 61.411% variation of the data, so the scale drawn is accepted Factor extraction with Eigenvalue = 3,071 met the requirements
Intention scale used to measure customer’s purchase intention includes 5 observed variables The results of the EFA analysis showed:
▪ 5 observed variables are grouped into 1 factor Factor loading coefficients are all > 0.5, so they have practical significance
▪ Each observed variable has a difference in factor loading coefficient ≥ 0.3, so it ensures the distinction between factors
▪ KMO coefficient = 0.812 > 0.5 factor analysis is required for the data h
▪ The Chi-square statistic of Bartlett Test reached the significance level value of 0.000 Therefore, the observed variables are correlated with each other on the overall scale
The extracted variance is 60.017%, showing that one factor can be explained 61.411% variation of the data, so the scale drawn is accepted Factor extraction with Eigenvalue = 2,611 met the requirements
Based on the results of EFA analysis, the extracted factors of the main research hypotheses are satisfactory.
Hypotheses testing
A validity test was conducted to ensure that the measurement accurately assessed its intended purpose This involved performing a Pearson correlation analysis to evaluate the relationship between the variables A correlation value between -1 and +1 indicates the strength of their relationship, with values closer to +1 signifying a stronger correlation.
Table 4.4 Results of Pearson correlation analysis
INT ATD PQ PD P SL SM ASS
According to the results, the independent variables all have a strong linear correlation with the dependent variable, the correlation coefficients are statistically significant (p