RESEARCH OVERVIEW
Reason to choose the research topic
In today's digital landscape, effective communication is crucial for developing successful marketing strategies The shift from traditional marketing to modern techniques has led many businesses, both domestically and internationally, to embrace influencer marketing This approach is particularly prevalent in the cosmetics industry, where beauty brands frequently utilize brand ambassadors to enhance their marketing efforts and connect with consumers.
Consumers encounter numerous beauty product advertisements daily but often lack knowledge about these products As a result, many prefer to rely on "reviews" from celebrities, family, and friends who have firsthand experience This trend has led to the widespread use of celebrities as brand ambassadors in the beauty industry Research indicates that celebrity endorsements significantly impact customer attention, reviews, and purchase intentions.
The rising demand for online shopping has led customers to favor products promoted by brand ambassadors, particularly those with positive evaluations These influencers significantly boost online orders through livestreaming on platforms like Facebook, TikTok, and Instagram, demonstrating their strong impact on consumer perceptions and purchasing intentions Recognizing this trend, our team has initiated a study titled "Research on The Effect of Brand Ambassador on Cosmetic Purchase Intention in Ho Chi Minh City" to explore this phenomenon further.
Research objective
Our research aims to explore how brand ambassadors influence customer buying intentions while identifying the limitations of this marketing strategy Ultimately, we seek to provide actionable recommendations for companies that utilize brand ambassadors as a central component of their marketing efforts.
Specific objective: In order to achieve the general objective that are confirmed as above, the study is generated to access these specific goals:
To determine the factors of brand ambassadors that affect consumers' intention to buy cosmetics in Ho Chi Minh City.
To measure the extent to which the brand ambassador's factors affect the purchase intention
Research question
There are two questions that will be answered at the end of this project:
Which factors of brand ambassadors influence the purchase intention of customers in Ho Chi Minh City?
How do these factors: reliability, attractiveness, specialization, familiarity, suitability and negative information of a brand ambassador influence the purchase intention of customers in Ho Chi Minh City?
Research object and research subject
Research object: the brand ambassador's factors affecting customers' intention to buy cosmetics
Research subject: cosmetic consumers in Ho Chi Minh City
Research method
This study used both qualitative method (preliminary research) and quantitative method (formal study).
Qualitative research method: refering to research works, related theories to synthesize into theoretical basis and analyze it.
Quantitative research method: from theoretical foundation and also refering to the research model to propose and build scales and questionnaires for survey and collecting research data.
Research structure
Along with tables, figures, references, and appendixes, this study report is conducted with five chapters, as follow:
LITERATURE REVIEW
Theoretical foundation
Brand ambassadors play a crucial role in attracting customers by representing their interests and needs, which fosters consumer acceptance These ambassadors can be mascots, industry experts, or even religious figures (Hafilah, Chaer, and Usman, 2019) Influencer opinion leadership and social interactions are vital in shaping followers' purchase intentions (Farivar et al., 2021) Today, celebrities such as singers, actresses, and models are commonly used as brand ambassadors due to their widespread popularity and influence Businesses leverage their image to enhance communication efforts, including attending press conferences and participating in promotional events, while also utilizing personal social media platforms to share brand-related content Notably, celebrities are featured in approximately one-sixth of all global advertisements (Shimp and Andrews, 2012) The strategic use of brand ambassadors not only boosts brand interest and positioning but also effectively conveys marketing messages to a broad audience.
Behavioral intention plays a crucial role in business, as it predicts the likelihood of repeated actions (Giao and Tra, 2020; Ajzen, 1991) Purchase intention reflects customers' motivations to choose specific brands, influenced by their behavior, perception, and attitudes Key factors such as price, quality, and perceived value significantly impact this intention Kotler et al (2016) outline six stages leading to a purchase: awareness, information search, product interest, product preference, persuasion, and the final purchase decision Additionally, both internal and external motives shape consumer behavior, with low-cost and simple-packaging products often perceived as high-risk due to quality concerns (Gogoi, 2013) Understanding these factors is essential for businesses aiming to maintain and expand their customer base.
Rational Action Theory (Fishbein and Ajzen, 1975) and the Theory of Planned Behavior (Ajzen, 1991) are influential frameworks in social psychology, frequently applied in modern marketing research, especially in the analysis of consumer behavior These theories suggest that an individual's actions are primarily influenced by their intentions and desires to engage in specific activities, as highlighted by Nguyen and others.
According to Le (2021), consumer behavior is influenced by two main factors: attitude toward the behavior and subjective norm Nguyen (2021) defines consumer behavior as the totality of actions related to the acquisition, use, and discontinuation of products and services, encompassing consumers' emotional, mental, and behavioral responses Furthermore, socio-cultural factors are identified as the primary influences on consumer behavior, as varying cultures and social classes result in distinct consumer behaviors across different regions (Nguyen, 2021).
2.1.4 Theory on the impact of brand ambassadors on consumer behavior
In today's marketing landscape, brands are increasingly utilizing social media influencers as brand ambassadors to promote their products and services, fostering meaningful interactions with consumers This dynamic enhances connections and helps brands influence consumer purchasing decisions more naturally (Antunes, 2021) Research by Utami et al (2020) indicates that brand ambassadors significantly boost customer purchase intentions through effective electronic word of mouth (e-WOM) Huynh (2020) highlights that social media influencers play a crucial role in the initial phases of the customer decision-making process in Vietnam, as they establish credibility by sharing valuable and relevant content However, as these influencers grow in popularity and receive more sponsorships, the authenticity of their product endorsements may come into question.
Related Studies Matrix
Research hypotheses
Trust is built on the essential qualities of reliability, honesty, and dependability Research indicates that young people often seek guidance from prominent figures, relying on their advice more than older generations (Pandey, 2011) Reference groups, particularly celebrities, significantly influence consumer attitudes, with celebrities being the most trusted sources (Escalas and Bettman, 2005) Notably, women tend to trust female celebrities, while men are more inclined to trust their male counterparts.
Companies can enhance their brand image by collaborating with reputable spokespeople who possess a strong public presence and high trustworthiness Consumers often perceive celebrities as reliable sources of information, making them valuable assets for marketing strategies.
Research indicates that the trustworthiness of brand ambassadors significantly enhances consumers' purchase intentions (Osei-Frimpong et al., 2019; Weismueller et al., 2020) By evaluating reliable online reviews from these ambassadors, consumers are more likely to develop confidence in their buying decisions (Utami et al.).
(2020) Hypothesis H1 is derived from the preceding statements:
Hypothesis H1: In Ho Chi Minh City, brand ambassadors’ reliability has a positive influence on customers' purchase intentions.
Research indicates that attractive appearances significantly influence consumer behavior during transactions (Reingen and Kernan, 1993) Effective advertising featuring appealing visuals can enhance desirability and foster positive attitudes, leading to increased purchase intentions (Clow et al., 2006) However, while physical attractiveness can play a role in celebrity effectiveness, it is not the sole factor (Winham and Hampl, 2008) The concept of attraction is multifaceted and cannot be solely defined by appearance, as non-physical traits like success, creativity, and relevance also contribute to celebrity appeal (Caballero and Solomon, 1984; Deshpandé and Stayman, 1994) Studies show that the engagement of brand ambassadors significantly impacts consumer purchasing intentions (Osei-Frimpong et al., 2019; Weismueller et al., 2020; Utami et al., 2020) Based on these findings, we propose hypothesis H2.
Hypothesis H2: The attractiveness of brand ambassadors has a positive outcome on the customers’ purchase intentions in Ho Chi Minh City.2.3.3 Specialization
The term "skill" refers to the perceived credibility of a communicator as a source of valuable information (Hovland et al., 1953) The expert's ability significantly influences the recipient's persuasion (Ohanian, 1990) Belch and Belch (199) suggest that information recipients tend to trust individuals with recognized expertise in the relevant field Experts with high knowledge and skills exert a stronger influence compared to those with lower expertise (Ohanian, 1990) The skill of celebrity endorsers also plays a crucial role in shaping consumer perceptions (Amos, Holmes, and Strutton, 2008) Consumers evaluate a brand representative's expertise based on the perceived value of the information provided Weismueller et al (2020) confirm that the expertise of brand ambassadors affects customers' purchasing intentions Furthermore, Fernández and Castillo (2021) find that perceived informational value can lead followers to express social expectations towards brands associated with influential figures Based on these insights, hypothesis H3 is proposed.
Hypothesis H3: The specialization of brand ambassadors has a positive effect on customers’ purchase intentions in Ho Chi Minh City.
Brand familiarity refers to the extent to which consumers directly and indirectly engage with a brand, reflecting their knowledge and perceptions stored in memory (Kent and Allen, 1994) This familiarity is closely linked to emotional connections with celebrities, where increased familiarity with a VIP can enhance positive consumer attitudes toward the brand (Fernández and Castillo, 2021) This phenomenon is known as the mere exposure effect (Wills, 2021), and when consumers have prolonged interactions with celebrities, it significantly influences their brand perceptions (Ha and Lam, 2016) Research by Utami et al (2020) confirms that familiarity with brand ambassadors positively impacts purchase intentions, while Osei-Frimpong et al (2019) highlight that celebrity endorsers possess traits such as attractiveness, trustworthiness, and familiarity, which shape consumer perceptions of quality, buying intentions, and brand loyalty Therefore, hypothesis H4 is proposed based on these insights.
Hypothesis H4: Brand ambassadors’ familiarity has a positive influence on the clients’ purchase intentions in Ho Chi Minh City.
The alignment between celebrities and brands plays a crucial role in modern advertising, as noted by Erdogan and Fessler (1999), who emphasize the need for a harmonious relationship between a brand's attributes and the characteristics of its celebrity representative The effectiveness of this pairing can be gauged by the connection between influencers and their followers Kalu (2019) highlights that brand ambassadors on social media significantly impact the purchasing intentions of millennial consumers When a product is endorsed by a celebrity whose reputation aligns closely with the brand, consumers are more likely to trust the advertisement compared to those endorsed by less relevant or lesser-known figures (Ha and Lam, 2016) Furthermore, Farivar et al (2021) confirm the positive influence of perceived credibility on consumers' buying intentions regarding brand ambassadors Based on these insights, we propose hypothesis H5.
Hypothesis H5: The brand ambassadors’ suitability with the brands/items has a positive effect on the purchase intentions of customers in Ho Chi Minh City.
Engaging brand ambassadors can be costly for organizations, not only due to the salaries of these high-profile individuals but also because of potential risks associated with negative publicity According to Knott and James (2004), employing celebrities with unfavorable information can adversely affect brand performance and recognition Furthermore, Reeves et al (2012) discovered that negative information surrounding brand ambassadors can diminish their appeal and credibility.
Research has shown mixed effects of celebrity endorsements on consumer purchasing intentions While a study by 2018 highlighted the negative impact of celebrity status and memorability on buying intentions, Osei-Frimpong et al (2019) found that unfavorable media coverage does not significantly affect consumer intentions towards the brand in question Additionally, Hafilah et al (2019) support the notion that brand representatives have minimal influence on consumer purchasing decisions Based on these insights, we propose hypothesis H6.
Hypothesis H6: The negative data of the brand ambassadors debilitates their effect on the buy purchase intentions of customers in Ho Chi Minh City.
From the above research outline, the proposed Research Model:
This research model consists of:
6 Independent variables: Reliability, Attraction, Specialization, Familiarity,
RESEARCH PROCESS
Research process
In order to complete the process of analyzing and evaluating the research topic, the team followed a sequence of 12 steps as follows:
+ Research theory and models Related studies
+ Build independent variables and research hypothesis
+ Perform multiple regression Check the suitability
+ Identify suitable and inappropriate variables
The process of building the scale, and designing the survey form
Step 1 - Define the content, identify and name the variable you want to measure
Step 2 - Make a list of statements or questions that have expressiveness based on theory, the study of previous research papers, group discussion
Step 3 – Determine the number and type of responses
Step 4 – Check all the questions that have been mined
Step 5 - Perform analysis of all items
Step 6 – Use the scale just built to re-analyze the questionnaire, preliminary assessment of the scale, Cronbach's alpha reliability coefficient, and EFA exploratory analysis to ensure the reliability of the scale.
3.2.1 The process of building the scale.
1.1 tc1 The brand ambassador mentioned above is a trusted celebrity.
1.2 tc2 The brand ambassador mentioned above is an honest celebrity.
1.3 tc3 The above brand ambassador is a celebrity who is honest when making statements about the brand/product in the advertisement.
1.4 tc4 I trust what the above brand ambassador says in the advertisement.
1.5 tc5 I trust the brands and products that the above brand ambassador endorse.
2.1 th1 The brand ambassador mentioned above has an attractive appearance.
& associates (2021) 2.2 th2 The brand ambassador mentioned above is attractive.
2.3 th3 Brand ambassador above is luxurious.
2.4 th4 Brand ambassador above is elegant.
2.5 th5 The brand ambassador mentioned above is charming.
3.1 cm1 The above brand ambassador is someone who has experiences in using the product/brand.
3.2 cm2 The above brand ambassador is knowledgeable about the product/brand.
3.3 cm3 The above brand ambassador is qualified to endorse the product/ brand.
3.4 cm4 The above brand ambassador is a person who is skilled in using the product/brand.
4.1 qt1 You think the brand ambassador mentioned above and you share the same culture.
4.2 qt2 You think that the brand ambassador mentioned above and you have the same lifestyle.
4.3 qt3 You think the brand ambassador mentioned above and you have the same interest.
4.4 qt4 You think the brand ambassador mentioned above and you have the same opinion.
5.1 ph1 You often see the above brand ambassador in the advertisements of this brand.
5.2 ph2 You think the above brand ambassador’s image is appropriate for this brand.
5.3 ph3 You think that this brand is completely suitable for the above brand ambassador to represent.
5.4 ph4 You believe that the above brand ambassador is trustworthy to represent this brand.
5.5 ph5 You believe that the above brand ambassador is using this brand’s products.
6.1 tttc1 You often come across a lot of bad news about the brand ambassador mentioned above.
6.2 tttc2 You do not like the information about the brand ambassador mentioned above.
6.3 tttc3 You find that the above brand ambassador’s information affects your purchasing behavior for this brand product.
7.1 ydm1 High chances that you will buy this brand product.
Pham Xuan Kien and Quach Nu Phuc Vuong (2020)
7.2 ydm2 You will actively learn about this brand product.
The fact that the brand ambassador is the representative of this brand is also a factor encouraging you to buy this brand’s products.
7.4 ydm4 You will try this brand’s products in the future.
Source The author based on previous studies.
3.2.2 Design of questionnaires and surveys
The survey was designed using theories, research studies, and scorecards, drawing from relevant literature by both domestic and international authors It focuses on developing key factors based on previous theoretical models, culminating in a main survey that encompasses seven distinct variables.
We designed the survey’s questionnaire using Google Forms, conducted the survey process in one month and collected data by:
+ Sending survey’s link to family, friends and acquaintances or even strangers who live in Ho Chi Minh City
+ Posting survey on Ho Chi Minh City’s student groups, forums, personal facebook. The survey ended with the number of responses up to 271 surveys, of which 214 were remained satisfactory.
3.3 Sampling process and data collection
The sample in the formal study was selected according to the conventional method The survey subjects are people living in Ho Chi Minh City, conducting the survey using a questionnaire.
The sample size in a study is influenced by the estimation method, the number of parameters, and the normal distribution of responses This report employs exploratory factor analysis (EFA), which mandates a minimum sample size calculated using the formula n = 5 * m, where m represents the number of questions excluding individual queries Given that the initial survey includes 30 observed variables, the minimum required sample size is n = 5 * 30, resulting in a total of 150 participants.
Survey by questionnaire survey over the Internet by creating a form on Google Form about the intention to buy cosmetics of people living in Ho Chi Minh City.
The study utilized a convenience sampling method, targeting an expected sample size of 400 participants, with a minimum of 150 samples calculated based on the research model Questionnaires were distributed through social media and directly to friends, family, and colleagues The survey comprised five main content areas with observed variables, utilizing closed questions on a 5-point scale Data analysis was conducted using SPSS 26 and Excel software to generate descriptive statistics and to identify and analyze factors influencing students' motivation to start a business, ultimately leading to the development of a regression model.
Quantitative analysis
3.4.1 Summary of the quantitative analysis process
A quantitative research study was conducted in Ho Chi Minh City, surveying 271 residents to gather data on three key areas Initially, general information was collected to refine the research scope The reliability of the variables was assessed using a 5-point Likert scale, ranging from "Totally Disagree" to "Totally Agree." Additionally, personal information was gathered to better understand the demographic of the participants Once the survey results were compiled, data cleaning and processing were performed, including reliability checks using Cronbach's Alpha and Exploratory Factor Analysis (EFA) Research hypotheses were tested through correlation and regression models utilizing SPSS version 26.
3.4.2 Check the difference between statistical groups (Descriptive Statistics)
Using a one-way T-test and ANOVA to test whether there is a difference in the assessment of purchase intention of customers living in Ho Chi Minh City.
However, before performing the ANOVA test, it is necessary to check the Levene’s Test for equality of population variances to consider the homogeneity of the observed data.
If Sig = 1 are retained in the exploratory factor analysis model because these factors have a better effect on synthesizing information than those with Eigenvalue less than 1.
The Total Variance Explained indicates the degree to which factors condense the data and the percentage of observed variables that may be lost, measured against a perfect score of 100% For an Exploratory Factor Analysis (EFA) model to be deemed appropriate, this value should be at least 50%.
+ If 0.3 10 then there is multicollinearity.
The Durbin-Watson value (D-W) is a statistical measure used to detect autocorrelation in the residuals of a regression time series Ranging from 0 to 4, a D-W value of 0 signifies a positive correlation, while a value of 4 indicates a negative correlation A D-W value of 2 suggests the absence of autocorrelation within the sample, making it a crucial tool for assessing the validity of regression models.
+ If 1 < D-W 2, there is a sign of multicollinearity, which is undesirable If VIF > 10 then there is multicollinearity.
Test the hypothesis about the fit of the model Hypothesis testing the significance of regression coefficients for each component.
Hypothesis testing of the normal distribution of residuals: based on the histogram of the normalized residuals; consider the mean to be 0 and the standard deviation to be 1.
RESEARCH CONCLUSIONS
Descriptive Statistics
4.1.1 Frequancy table and descriptive statistics
Dang tin cay Frequency Percent Valid
Cumulative Percent Valid Hoàn toàn không đồng ý 32 15.0 15.0 15.0
Source Conducted by authors’ data analysis results
- For the respondents who chose totally disagree, there are 32 answers, accounting for 15% of the total survey answers.
- For the ones who disagreed with the question, there are 19 answers, accounting for 8.9% of the total survey answers.
- For people who were neutral with the question, there are 22 answers, accounting for 10.3% of the total survey answers.
- For the respondents who agreed with the question, there are 74 answers, accounting for 34.6% of the total survey answers
- For the ones who chose totally agree, there are 67 answers, accounting for 31.3% of the total survey answers
Conclusion: With the analysis of do tin cay frequency, the average of the respondents’ chosen answers fluctuates between “agree” (34.6%).
N Minimum Maximum Mean Std Deviation
Source Conducted by authors’ data analysis results
- The standard deviation “Dang tin cay” is 1.397
Conclusion: Lowest Dang tin cay 1 "tc1", highest Dang tin cay 5"tc5", median at 3.58 4.1.2 CROSS TABULATION TABLES
Table 4.1.2.1 - Thu nhap*Hoc van Crosstabulation
N Percent N Percent N Percent thu nhap*hoc van 214 100.0% 0 0.0% 214 100.0%
Thu nhap*Hoc van Crosstabulation
Hoc van 12/12 Trung cấp/ cao đẳng Đại học Khác Total
Source Conducted by authors’ data analysis results
Table 4.1.2.2 - Hoc van*Nghe nghiep*Gioi tinh Crosstabulation
Hoc van*Nghe nghiep*Gioi tinh
Hoc van*Nghe nghiep*Gioi tinh Crosstabulation
Source Conducted by authors’ data analysis results
A comprehensive analysis of students' career outcomes reveals that 87 respondents possess a university education, while 8 hold a college degree Additionally, 26 individuals reported having completed 12/12 education, and only 1 respondent indicated a different academic level These findings highlight the significance of educational attainment in shaping career paths for both male and female students.
In a survey of 214 respondents regarding career paths for officers, it was found that 20 individuals with a university education, 8 with a college degree, and 3 with a 12/12 degree provided insights Notably, there were no responses from those with different educational levels, and the combined total of these four educational categories accounted for 31 responses.
In a study analyzing the career perspectives of housewives, both male and female respondents with university and college degrees provided five answers each Notably, those at the 12/12 education level contributed two responses, while individuals at varying educational levels did not provide any answers Overall, these educational backgrounds represented 12 out of the total 214 respondents.
In analyzing the career outcomes for workers with varying levels of education, data reveals that individuals with a university education provided 12 responses, while those with a college degree contributed 10 Notably, respondents holding a degree of 12/12 accounted for 22 answers However, there were no responses from individuals at different educational levels, indicating that these educational categories collectively represent 24 of the total responses.
In the realm of self-employment, both men and women with a university education provide 11 responses, while those with a college degree contribute 7 answers Notably, individuals at various educational levels, including those with 12/12 qualifications, also have no responses Collectively, these four educational categories represent a total of 18 responses in the self-employment career landscape.
Among the 214 respondents, those with a university education provided 6 answers, while individuals holding a college degree and those at the 12/12 level recorded 0 answers each Additionally, various educational levels contributed 1 answer, resulting in a total of 7 answers from the surveyed educational backgrounds.
Conclusion: The correlation of income between "Male" and "Female" genders in the occupation range "students" shows that the ratio between all variables is interconnected in terms of value.
4.1.2.3 For categorical variable and numerical variables
Source Conducted by authors’ data analysis results
Among respondents with incomes under 5 million, the age distribution reveals that 10 individuals are aged 18, 71 are between 18 and 22, 4 fall within the 22-30 age range, and 3 are over 30 Collectively, these age groups represent 88 out of 214 total respondents.
+ For incomes below 5-10 million: with the age group of 18 years old there are no answers, with groups of 18-22 years old, there are 18 answers, from 22-30 years old has
12 answers, the group over 30 years old has 3 answers All age groups account for 33 of
+ For incomes from 10 to 15 million: with the age group of 18 years old there are 3 answers, with groups of 18-20 years old, there are 11 answers, from 22-30 years old has
21 answers, the group over 30 years old has 5 answers All age groups account for 40 of
Among the 214 respondents, a total of 29 individuals reported incomes exceeding 15 million Specifically, 3 respondents were aged 18, 11 were between 18 and 20, 8 belonged to the 22-30 age group, and 7 were over 30 years old.
In the analysis of non-answers across different age groups, individuals aged 18 provided 15 responses, while those aged 18-20 contributed 8 answers Notably, the 22-30 age bracket recorded no responses, and only 1 answer came from individuals over 30 Overall, these age groups collectively accounted for a total of 24 responses.
Conclusion: With the age of accounting for the majority (from 18-22 years old), the main correlation of income and education fluctuates at the range of "from 10 to 15 million".
Figure 4.1.3 - Pie Chart Percent of tuoi
Source Conducted by authors’ data analysis results
Note The linked table describes the ages of the referenced participants
-The age of “under 18 years old” accounted for 14.49 percent (3rd highest number) shows that high school students who are interested in cosmetics are talking up a large number
- The age range from "18-22 years old" accounted for 55.61 percent (the highest number) shows that most of the survey respondents are in high school or university.
- The age range of "22-30years old" accounted for 21.03 percent (2nd highest number) shows that most of the survey participants have graduated from college and have financial income.
- The age of “over 30 years old” accounted for 8.88 percent (the lowest number) shows that middle-aged survey participants are the lowest.
Conclusion: The largest group of survey participants are university students.
TESTING RELIABILITY OF MEASUREMENT SCALES (CRONBACH’S ALPHA)
Cronbach’s Alpha equals 0.935, the measured measurement scale is good
Table 4.2.1 - Results of assessment of Reliability Scale
Scale Mean if item deleted Scale Variance if item deleted Corrected Item-Total Correlation Cronbach’s Alpha if item deleted Dang tin cay
Source Conducted by authors’ data analysis results
Keep these five items (the Corrected item – Total correlation ≥ 0.3)
Cronbach’s Alpha equals 0.923, the measured measurement scale is good
Table 4.2.2 - Results of assessment of Attraction Scale
Scale Mean if item deleted Scale Variance if item deleted Corrected Item-Total Correlation Cronbach’s Alpha if item deleted Hap dan
Source Conducted by authors’ data analysis results
Keep these five items (the Corrected item – Total correlation ≥ 0.3)
Cronbach’s Alpha equal 0.914, the measured measurement scale is good
Table 4.2.3 - Results of assessment of Specialization Scale
Scale Mean if item deleted Scale Variance if item deleted
Corrected Item-Total Correlation Cronbach’s Alpha if item deleted kinh nghiem
Source Conducted by authors’ data analysis results
Keep these four items (the Corrected item – Total correlation ≥ 0.3)
Cronbach’s Alpha equal 0.795, the measured measurement scale is good
Table 4.2.4 - Results of assessment of Specialization Scale
Scale Mean if item deleted Scale Variance if item deleted Corrected Item-Total Correlation Cronbach’s Alpha if item deleted chung nen van hoa
Source Conducted by authors’ data analysis results
Keep these four items (the Corrected item – Total correlation ≥ 0.3)
Cronbach’s Alpha equal 0.846, the measured measurement scale is good
Table 4.2.5 - Results of assessment of Specialization Scale
Scale Mean if item deleted Scale Variance if item deleted Corrected Item-Total Correlation Cronbach’s Alpha if item deleted thuong thay
Source Conducted by authors’ data analysis results
Keep these five items (the Corrected item – Total correlation ≥ 0.3)
Cronbach’s Alpha equal 0.725, the measured measurement scale is good.
Table 4.2.6 - Results of assessment of Negative Information Scale
Scale Mean if item deleted Scale Variance if item deleted
Corrected Item-Total Correlation Cronbach’s Alpha if item deleted
Source Conducted by authors’ data analysis results
Keep these three items (the Corrected item – Total correlation ≥ 0.3)
Cronbach’s Alpha equal 0.870, the measured measurement scale is good
Table 4.2.7 - Results of assessment of Purchase Intention Scale
Scale Mean if item deleted Scale Variance if item deleted Corrected Item-Total Correlation Cronbach’s Alpha if item deleted se mua
Source Conducted by authors’ data analysis results
Keep these four items (the Corrected item – Total correlation ≥ 0.3)
- General conclusion: After analyzing Cronbach’s Alpha of each scale, we find that
EXPLORATORY FACTOR ANALYSIS EFA
4.3.1 Exploratory Factor Analysis EFA with Independent variables
Factor analysis for 6 independent variables includes: Reliability, Specialization,Attraction, Similarities, Suitability, and Negative information This process has 2 parts:
Table 4.3.1.1 - KMO and Bartlett test results for independent variables
KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.933
Source Conducted by authors’ data analysis results
The factor analysis method has the standard in the KMO that must be greater than 05 and the Sig of the significance level must be less than 05
Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) value = 0.933(>0.5)
Acceptable - The data used for factor analysis are appropriate and the variables are correlated.
P-value (Sig.= 000 < 0.05) Acceptable - Measured variables are sufficiently intercorrelated to justify the factors
Total Variance Explained Intial Elgenvalues Extraction Sum of Squared Loadings Rotation Sums of Squared Loadings
Source Conducted by authors’ data analysis results
The analysis identifies four key factors based on an eigenvalue greater than 1, which effectively represent the information from 26 observed variables in the exploratory factor analysis (EFA) The first factor accounts for 41.784% of the total variance, while the second factor explains 12.237% Additionally, the third factor contributes 8.192% to the variance, and the fourth factor accounts for 7.488%.
- The total variance extracted is 69.701% (> 50%) The fourth factors are explained 69.701% of the data variation in 26 observed variables participating in EFA
Table 4.3.1.3 - The first exploratory factor analysis results
Source Conducted by authors’ data analysis results
After the first part, there are 2 variables will be dropped because factors loading of them
The variable "Often see" is not uploaded in factors Component 1, component 2, Component 3 and component 4 This variable will be dropped.
The variable “Same hobby” is uploaded in factors Component 2 and Component 4 with a factor loading of 0.546 and 0.503, the difference in load factor equal to 0.546- 0.503= 0.043< 0.3 This variable will be dropped.
Table 4.3.1.4 - KMO and Bartlett’s Test result for independent variables for the final
KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.935
The factor analysis method has the standard in the KMO that must be greater than 05 and the Sig of the significance level must be less than 05
Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) value = 0.935 (>0.5)
Acceptable - The data used for factor analysis are appropriate and the variables are correlated.
P-value (Sig.= 000 < 0.05) Acceptable - Measured variables are sufficiently intercorrelated to justify the factors
Intial Elgenvalues Extraction Sum of Squared Loadings Rotation Sums of Squared Loadings
Source Conducted by authors’ data analysis results
The analysis reveals that the first four variables should be retained, as they are supported by an eigenvalue greater than 1, indicating their significance in the exploratory factor analysis (EFA) of the 24 observed variables Specifically, the first factor accounts for 44.716% of the variance, the second factor explains 11.166% of the variance, the third factor captures 8.598% of the variance, and the fourth factor accounts for 6.728% of the variance, making them the most effective in representing the underlying data structure.
-The total variance extracted is 71.208% (> 50%) The fourth factors are explained
71.208% of the data variation in the observed variables participating in EFA
Table 4.3.1.6 - Final exploratory factor analysis results
Source Conducted by authors’ data analysis results
After eliminating the variables "often see" and "same hobby," the analysis reveals that the remaining observed variables can be categorized into four distinct factors, all exhibiting factor loading coefficients exceeding 0.5 This indicates that there are no problematic variables present, confirming the appropriateness of the results.
4.3.2 Exploratory Factor Analysis EFA with Dependent variable
Table 4.3.2.1 - KMO and Bartlett’s Test - scale of entrepreneurship intention of students
KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.807
Bartlett’s Test of Sphericity Approx Chi-Square 431.734
Note Factor analysis for dependent variable includes: Purchase Intension
The factor analysis method has the standard in the KMO that must be greater than 05 and the Sig of the significance level must be less than 05
Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) value = 0.807(>0.5) -
Accepted - The data used for factor analysis are appropriate.
P-value (Sig.= 000 < 0.05) Accepted - Measured variable is sufficiently intercorrelated to justify the factors
Total Variance Explained Intial Elgenvalues Extraction Sum of Squared Loadings
Source Conducted by authors’ data analysis results
A factor is based on the eigenvalue greater than 1 The total variance extracted is 72.123% (>50%) The factor is explained 72.123% of the data variation in the 4 observed variables participating in EFA.
Component a 1 se mua 831 chu dong tim hieu 816 yeu to khuyen khich 832 dung thu san pham 915
Source Conducted by authors’ data analysis results
Extraction Method: Principal Component Analysis a a 1 components extracted.
FINAL CONCLUSION: There are 2 variables that are dropped (thuong thay and cung so thich).
Factor 1: Trustworthiness: dang tin cay, trung thuc, nhan dinh thanh that, tin tuong loi noi, tin tuong chung thuc, hap dan, thu hut, sang trong, thanh lich, quyen ru, kinh nghiem, hieu biet, kha nang, ki nang, chung nen van hoa, hinh anh phu hop, and hoan toan phu hop
Factor 2: Familiarity: chung loi song, cung quan diem
Factor 3: Negative information: thong tin khong tot, khong thich, anh huong
Factor 4: Suitability: dang tin cay, dang su dung
Dependent factor: Purchase Intension (se mua, chu dong tim hieu, yeu to khuyen khich, dung thu san pham)
TEST MODELS AND RESEARCH HYPOTHESES
DTC SQT TTTC SPH YDM
** Corelation is significant at the 0.01 level (2-tailed)
* Corelation is significant at the 0.05 level (2-tailed)
Source Conducted by authors’ data analysis results
H0: rab = 0 (there is no linear relationship between the two variables).
H1: rab ≠ 0 (there is a linear relationship between the two variables)
DTC and SQT have a positive relationship.
H0: rab = 0 (there is no linear relationship between the two variables).
H1: rab ≠ 0 (there is a linear relationship between the two variables)
DTC and SPH have a positive relationship.
H0: rab = 0 (there is no linear relationship between the two variables).
H1: rab ≠ 0 (there is a linear relationship between the two variables)
SPH and YDM have a positive relationship.
Thông tin tiêu cực, Sự quen thuộc, Độ tin cậy b
Enter a Dependent Variable: Ý định mua b All requested variables entered.
Source Conducted by authors’ data analysis results
YDM = 𝛽0 + 𝛽1*DTC + 𝛽2*SQT + 𝛽3*TTTC + 𝛽4*SPH
Table 4.4.2.2 - Explanatory Level of the Model
Std Error of the Estimate
1 0.771 a 0.595 0.587 0.63859 1.687 a Predictors: (Constant), Sự phù hợp, Thông tin tiêu cực, Sự quen thuộc, Độ tin cậy b Dependent Variable: Ý định mua
Source Conducted by authors’ data analysis results
R 2 = 0,595 (59,5%): The model is able to explain 59,5% of the observed data
This linear regression model fits the sample's data set.
Table 4.4.2.3 - Model Fit: Analysis of Variance ANOVA
Total 210.542 213 a Dependent Variable: Ý định mua b Predictors: (Constant), Sự phù hợp, Thông tin tiêu cực, Sự quen thuộc, Độ tin cậy
Source Conducted by authors’ data analysis results
H0: All regression coefficients are equal to zero - > the independent variables do not reliably predict the dependent variable
H1: The independent variables reliably predict the dependent variable.
From the ANOVA results: Significant = 0.000 < alpha = 0.05 From there reject H0,accept H1: The independent variables reliably predict the dependent variable.
Dependent Variable: Ý định mua
Source Conducted by authors’ data analysis results
CHECK THE VIOLATION OF ASSUMPTIONS
Source Conducted by authors’ data analysis results
From the four scatterplot graphs above, we can conclude the Linear relationships.
ASSUMPTION 2: RESIDUALS ARE NORMALLY DISTRIBUTED
Figure 4.4.2.5 - Histogram of the normalized residuals
Source Conducted by authors’ data analysis results
Source Conducted by authors’ data analysis results
Residuals are normally distributed Assumption 2: SATISFIED
Std Error of the Estimate
1 0.771 a 0.595 0.587 0.63859 1.687 a Predictors: (Constant), Sự phù hợp, Thông tin tiêu cực, Sự quen thuộc, Độ tin cậy b Dependent Variable: Ý định mua
Source Conducted by authors’ data analysis results
Durbin-Watson = 1.687 Weak positive autocorrelation: acceptable
ASSUMPTION 4: NO OR LITTLE MULTICOLLINEARITY: THE
INDEPENDENT VARIABLES ARE NOT HIGHLY CORRELATED WITH
EACH OTHER [COV(XI,XJ) =0]
6 - Statis tical analy sis of regre ssion coeffi cients
Dependent Variable: Ý định mua
Source Conducted by authors’ data analysis results
VIF =1: no multicollinearity: The independent variables are not highly correlated with each other [cov(xi,xj) =0]
YDM = 𝛽0 + 𝛽1*DTC + 𝛽2*SQT + 𝛽3*TTTC + 𝛽4*SPH
Figure 4.4.2.7 - Scatterplot: Regression Standardized Residual
Variance of residual (var(resid) are constant
Source Conducted by authors’ data analysis results
RESEARCH CONCLUSIONS
The study highlights the pressing nature of the issue and seeks to gain deeper insights into customer behavior towards brands It emphasizes the prevalence of celebrity endorsements in the cosmetics industry and outlines the positive contributions of these findings.
Identify the factors of famous people affecting the cosmetic purchase behavior of consumers in Ho Chi Minh City.
Measuring the influence of celebrity factors on consumer buying behavior of cosmetics in Ho Chi Minh City.
Thereby answering the questions posed before entering the study:
Which celebrity factors have the most profound influence on consumer buying behavior in Ho Chi Minh City?
How do the factors: reliability, specialization, attractiveness, familiarity, suitability, negative information of brand ambassadors affect the buying intention of consumers in Ho Chi Minh City?
The study demonstrates that the scale employed is reliable, with a Cronbach's Alpha greater than 0.6, indicating the significant influence of brand ambassadors on consumers' intention to purchase cosmetics in Ho Chi Minh City The research identifies four key factors impacting customers' purchase intentions: Reliability (DTC), Familiarity (SQT), Negative Information (TTTC), and Suitability (SPH).
YDM = 𝛽0 + 𝛽1*DTC + 𝛽2*SQT + 𝛽3*TTTC + 𝛽4*SPH
This research paper is the first of its kind, conducted within a limited timeframe and focusing on individuals residing in Ho Chi Minh City Traditionally, the study centers on females, who are more likely to use cosmetics, resulting in a significant gender imbalance in the sample population Consequently, this disparity may affect the overall representation of the findings.
Group 1 aims to broaden the scope of their research to provide a comprehensive overview, establishing a solid foundation for future studies They aspire to enhance understanding of the factors that influence brand ambassadors' purchasing intentions for cosmetics in Ho Chi Minh City.
Group 1 sincerely thanks you for reading our report!
Nguyen, Q.C., Vo, T.X.Q., & Nguyen, A.P (2021, August 20) The Impacts of Brand Ambassador on Purchase Intention among Customers in Ho Chi Minh City Science and Technology Magazine, 50, 120-133
Rahmi, Y., Sekarasih, L., & Sjabadhyni, B (2017) The Influence of Beauty Vlog on Perceived Source Credibility and Purchase Intention Makara Human Behavior Studies in Asia, 21(1), 13-23
Adam, M.A., & Hussain, N (2017) Impact of Celebrity Endorsement on Consumer Buying Behavior British Journal of Marketing Studies, 5(3), 79-121
Putri, Y.I., Pradana, M., Utami, F.N., Nugraha, D.W., & Karnovi, R (2021) The Influence of Brand Ambassador on E-Commerce Purchase Intention IEOM Society
Nguyen, N.T (2020) The Influence of Celebrity Endorsement on Young Vietnamese Consumers’ Purchasing Intention Journal of Asian Finance, Economics and
Business, 8(1), 951-960.doi:10.13106/jafeb.2021.vol8.no1.951
Hafilah, N E., Chaer, V P., & Usman, O (2019) The effect of brand ambassador, brand image, product quality, and on purchase decisions Samsung Smartphones SSRN
Farivar, S., Wang, F., & Yuan, Y (2021) Opinion leadership vs para-social relationship: Key factors in influencer marketing Journal of Retailing and Consumer Services,
The decision-making process for purchasing airline tickets online among consumers in Ho Chi Minh City is influenced by various factors Research by Ha and Be (2020) highlights the significance of online platforms in shaping consumer behavior Additionally, the impact of celebrity endorsements on customer attitudes towards brands and their purchase intentions is emphasized in the work of Ha and Lam (2017), showcasing how influential figures can enhance brand perception and drive sales.
Ajzen, I (1991, December) The Theory of Planned Behavior Organization Behavior and Human Decision Processes, 50(2), 179-211 doi: 10.1016/0749(91)90020-T
Kotler, P., Keller, K.L., Brady, M., Goodman, M and Hansen, T (2016, December 28). E-Marketing and Its implementation on Developing Social Perceptions of Customers Through Effective Marketing Mix of Organic Textile Products.
American Journal of Industrial and Business Management, 6(12) doi:
Gogoi, B J (2013, June) Study of Antecedents of Purchase Intension and Its Effect on Brand Loyalty of Private Label Brand of Apparel International Journal of Sales &
Fishbein, M., & Ajzen, I (1976) Misconceptions about the Fishbein model: Reflections on a study by SongerNocks Journal of Experimental Social Psychology, 12(6), 579-584.
Nguyen, T M., & Le, G N H (2021) The influence of COVID-19 stress on psychological well-being among Vietnamese adults: The role of self-compassion and gratitude Traumatology, 27(1), 86-97 doi:10.1037
Lý thuyết hành vi người tiêu dùng đóng vai trò quan trọng trong việc phát triển hoạt động kinh doanh bảo hiểm phi nhân thọ tại Việt Nam hiện nay Việc hiểu rõ tâm lý và hành vi của người tiêu dùng giúp các doanh nghiệp bảo hiểm thiết kế sản phẩm phù hợp, tối ưu hóa chiến lược marketing và nâng cao trải nghiệm khách hàng Bài viết của Huy (2010) nhấn mạnh rằng việc áp dụng lý thuyết này không chỉ gia tăng tính cạnh tranh mà còn thúc đẩy sự phát triển bền vững của ngành bảo hiểm phi nhân thọ trong bối cảnh kinh tế đang biến đổi nhanh chóng.
Antunes, A C (2021) The Role of Social Media Influencers on the Consumer Decision- Making Process In Analyzing Global Social Media Consumption (pp 138-154). IGI Global.
Utami, S P., Setyowati, N & Mandasari, P (2020) Celebrity Brand Ambassador and e- WOM as Determinants of Purchase Intention: A Survey of Indonesian Celebrity Cake E3S Web of Conferences, 142 doi:10.1051/e3sconf/202014205001
The influence of social media personalities significantly shapes consumer choices, as highlighted by Vy (2020) These influencers create relatable content that resonates with their followers, impacting their purchasing behavior Additionally, Escalas and Bettman (2005) discuss how self-perception and social circles contribute to brand interpretation, further emphasizing the interconnectedness of social influences and consumer decision-making Together, these studies underscore the critical role of social media in modern marketing strategies.
Ronald, E G., Barbara, A L & Stephen, J N (2013, May 31) The Impact of Corporate Credibility and Celebrity Credibility on Consumer Reaction to Advertisements and
Brands Journal of Advertising, 29(3), 43-54. doi:10.1080/00913367.2000.10673616.
Osei-Frimpong, K., Owusu-Frimpong, N & Donkor, G (2019, January) The impact of Celebrity Endorsement on Consumer Purchase Intension: An Emerging Market Perspective The Journal of Marketing Theory and Practice, 27(1),103-121. doi:10.1080/106696679.2018.1534070
Weismueller, J., Harrigan, P & Wang, S (2020, November 1) Influencer Endorsements: How Advertising Disclosure and Source Credibility Affect Consumer Purchase Intention on Social Media Australasian Marketing Journal, doi:10.1016/2020.03.002
Reingen, P.H and Kernan, J.B (1993) The Impact of Attractiveness on Retail Managers’ Decision: Labor Market Discrimination When Hiring Salespeople.
Journal of Consumer Psychology, 2, 25-28 doi:10.1016/S1057-7408(08)80073-3
Clow, K.E., James, K., Kranenburg, K.E & Berry, C.T (2006, October) The relationship of the visual element of an advertisement to service quality expectations and source credibility Journal of Services Marketing, 20(6), 404-411. doi:10.1108/08876040691293
According to Winham and Hampl (2008), adolescents assert that television characters have no significant impact on their self-perception regarding body image, weight, clothing choices, or eating habits This finding suggests that marketers should consider the limited influence of media portrayals on young consumers' decisions and self-image when developing responsible marketing strategies.
Bower, A B & Grau, S L (2001, March) Is Beauty Best? High versus NormallyAttractive Models in Advertising Journal of Advertising, 30(1), 1-12.Doi:10.1080/00913367.2001.10673627
Cooper, M J & Solomon, P J (1984, March) Effects of Model Attractiveness on Sales Response Journal of Advertising, 13(1), 17-23 Doi:
Deshpande, R & Stayman, D M (1994, February 1) A Table of Two Cities: Distinctiveness Theory and Advertising Effectiveness Journal of Marketing
Hovland, C.I., Janis, I.L., & Kelley, H.H (1953) Communication and persuasion APA
PsycNet Retrieved from https://psycnet.apa.org/record/1953-15071-000
Ohanian, R (2013, May 29) Construction and Validation of a Scale to Measure Celebrity Endorsers' Perceived Expertise, Trustworthiness, and Attractiveness.
Belch, G E & Belch, M A (1994) Introduction to Advertising and Promotion: An
Integrated Marketing Communications Perspective (3 rd ) Burr Ridge, IL, United States: Irwin Professional Pub
Amos, C., Holmes, G., & Strutton, D (2008) Exploring the relationship between celebrity endorser effects and advertising effectiveness: A quantitative synthesis of effect size International Journal of Advertising: The Quarterly Review of
Allen, C & Kent, R J (1994, July) Competitive Interference Effects in Consumer Memory for Advertising: The Role of Brand Familiarity Journal of Marketing,
The doctoral dissertation by Kalu (2019) explores the influence of social media influencer marketing on the purchase intentions of Irish male millennials, specifically within the Irish fashion industry The study highlights how influencer marketing shapes consumer perceptions and decision-making processes, emphasizing its significance in contemporary marketing strategies By focusing on the unique perspectives of this demographic, the research provides valuable insights into the effectiveness of influencer collaborations in driving sales and engagement in the fashion sector.
Wongweeranonchai, P & McClelland, R J (2016, October) The Influence of Perceived Celebrity Endorser Credibility in Advertising on Purchase Intention of Thai
Consumers Research Gate Retrieved from https://www.researchgate.net/publication/308076331_The_Influence_of_Perceived_Celebrity_Endorser_Credibility_in_Advertising_on_Purchase_Intention_of_Thai_C onsumers
Reeves, R A., Thwaites, D., Lowe, B & Monkhouse, L (2012, September) The Impact of Negative Publicity on Celebrity Ad Endorsements Psychology and Marketing,
Prameswara, D H & Sjabadhyni, B (2018, July 1) The Effect of Negative Celebrity Publicity and Celebrity Identification Towards Purchase Intention Advances in
Social Science, Education and Humanities Research, 139 Doi:10.2991/UIPSUR-
Research on The Effect of Brand Ambassador on Cosmetic Purchase Intention in Ho Chi Minh City
We are a group of Industrial Management students from the Department of High Quality Training at HCMC University of Technology and Education Today, we conducted a survey titled "Research on The Effect of Brand Ambassador on Cosmetic Purchase Intention in Ho Chi Minh City." This research aims to explore how brand ambassadors influence purchasing decisions and factors affecting cosmetic buying intentions among consumers in Ho Chi Minh City.
We kindly ask you to take a moment to complete this survey honestly Please rest assured that all information you provide will remain confidential and will solely be used for research purposes.
How often do you see ads featuring brand ambassadors?
Do you know any celebrities who are brand ambassadors?
Which brand does that celebrity represent?
How often do you use that product?
The rating bar level is calculated as follows:
Instructions for answering the question: Please tick (x) into boxes which express your level of agreement from 1 to 5
Disagree Neutral Agree Totally agree
The brand ambassador mentioned above is a trusted celebrity.
The brand ambassador mentioned above is an honest celebrity.
The above brand ambassador is a celebrity who is honest when making statements about the brand/product in the advertisement.
I trust what the above brand ambassador says in the advertisement.
I trust the brands and products that the above brand ambassador endorse.
The brand ambassador mentioned above has an attractive appearance.
The brand ambassador mentioned above is attractive.
Brand ambassador above is luxurious.
Brand ambassador above is elegant.
The brand ambassador mentioned above is charming.
The above brand ambassador is someone who has experiences in using the product/brand.
The above brand ambassador is knowledgeable about the product/brand.
The above brand ambassador is qualified to endorse the product/brand.
The above brand ambassador is a person who is skilled in using the product/brand.
You think the brand ambassador mentioned above and you share the same culture.
You think that the brand ambassador mentioned above and you have the same lifestyle.
You think the brand ambassador mentioned above and you have the same interest.
You think the brand ambassador mentioned above and you have the same opinion.
You often see the above brand ambassador in the advertisements of this brand.
You think the above brand ambassador’s image is appropriate for this brand.
You think that this brand is completely suitable for the above brand ambassador to represent.
You believe that the above brand ambassador is trustworthy to represent this brand.
You believe that the above brand ambassador is using this brand’s products.
You often come across a lot of bad news about the brand ambassador mentioned above.
You do not like the information about the brand ambassador mentioned above.
You find that the above brand ambassador’s information affects your purchasing behavior for this brand product.
High chances that you will buy this brand product.
You will actively learn about this brand product.
The fact that the brand ambassador is the representative of this brand is also a factor encouraging you to buy this brand’s products.
You will try this brand’s products in the future.
under 18 years old 18 - 22 years old 22 - 30 years old Over 30
Your income is about: (unit VND)
Thank you very much for taking your time to participate in our survey!
FREQUENCY TABLE AND DESCRIPTIVE STATISTICS
For categorical variable and numerical variables:
Appendix 3 : TESTING RELIABILITY OF MEASUREMENT SCALES (CRONBACH’S ALPHA)
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted chung nen van hoa
10.58 7.334 530 780 chung loi song 11.22 7.111 605 744 cung so thich 10.86 6.528 684 702 cung quan diem 11.07 7.089 604 744
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted thuong thay 15.65 12.152 623 824 hinh anh phu hop
15.62 12.077 734 795 dang tin cay 15.83 12.357 675 810 dang su dung 15.84 12.463 555 843
Appendix 4: EXPLORATORY FACTOR ANALYSIS (EFA)
Exploratory Factor Analysis EFA with Independent variables.
Exploratory Factor Analysis EFA with Dependent variable.
Appendix 5 : TEST MODELS AND RESEARCH HYPOTHESES
CHECK THE VIOLATION OF ASSUMPTIONS
ASSUMPTION 2: RESIDUALS ARE NORMALLY DISTRIBUTED
ASSUMPTION 4: NO OR LITTLE MULTICOLLINEARITY: THE
INDEPENDENT VARIABLES ARE NOT HIGHLY CORRELATED WITH
EACH OTHER [COV(XI,XJ) =0]