Research paper factors of online marketing influences online shopping behavior of huflit student

MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY UNIVERSITY OF FOREIGN LANGUAGES - INFORMATION TECHNOLOGY FACULTY OF BUSINESS ADMINISTRATION SUBJECT OF SPECIALIZED RESEARCH PROJECT w

MINISTRY OF EDUCATION AND TRAINING

HO CHI MINH CITY UNIVERSITY OF FOREIGN LANGUAGES - INFORMATION TECHNOLOGY

FACULTY OF BUSINESS ADMINISTRATION SUBJECT OF SPECIALIZED RESEARCH PROJECT

whe

RESEARCH PAPER FACTORS OF ONLINE MARKETING INFLUENCES ONLINE SHOPPING BEHAVIOR OF HUFLIT STUDENT

Advisor : Nguyen Thi Huyen Tran Group name : NhŸÿÄm §

Group members : Nguyen Thanh Dat - 19DH120833

Luong The Van - I9DH480670 Phan Nguyen Tuong Vy - 19DH480436

HCMC - May, 2023

MINISTRY OF EDUCATION AND TRAINING

HO CHI MINH CITY UNIVERSITY OF FOREIGN LANGUAGES - INFORMATION TECHNOLOGY

FACULTY OF BUSINESS ADMINISTRATION SUBJECT OF SPECIALIZED RESEARCH PROJECT

whe

RESEARCH PAPER FACTORS OF ONLINE MARKETING INFLUENCES ONLINE SHOPPING BEHAVIOR OF HUFLIT STUDENT

Advisor : Nguyen Thi Huyen Tran Group name : NhŸÿÄm §

Group members : Nguyen Thanh Dat - 19DH120833

Luong The Van - I9DH480670 Phan Nguyen Tuong Vy - 19DH480436

HCMC - May, 2023

ABSTRACT

1.1 Introduction to online marketing 1 1.1.1 The concept of online marketing 1 1.1.2 Features of online marketing 2 1.1.3 Tools of online marketing 3 1.2 Influence of online marketing on customer behaVÏOT - 5< s<s «se << see 5 1.2.1 Influence of online marketing on decision makÌng << 5 << 5< << 5 1.2.2 Online buying involves risk 5 CHAPTER 2: LITERATURE REVIEW 7 2.1 Background of research 7 2.2 Research model and hypothesis 8 CHAPTER 3: RESEARCH METHODOLOGY 10

3.1.1 Research process construction 10 3.1.2 Research Subject 10 3.1.3 Purposes 10 3.1.4 Objective 11 3.1.5 Process 11 3.1.6 Methods and tools for collecting informafÏOTI .- <5 5< 5 << se s<es 11 3.2 Data processing method 12 3.2.1 Methods of data analysis 12 3.2.2 Factor analysis 13

List of charts

Ð Di n NFhG 200/2) 0 16 Chart 4, 2 Current academic year of students in school - -+-c<-c+ee+ses 17 Chart 4, 3 The frequency of online purchases by Huflit students .000000000000000 18 Chart 4 4 Huflit students' online purchase channeÌs . - - 55 55555 <s<<s+ 19

CHAPTER 1: INTRODUCTION 1.1 Introduction to online marketing

1.1.1 The concept of online marketing

Marketing refers to a broad system of organizational operations that are planned, promoted, and distributed to target audiences in order to fulfill their requirements and forward corporate goals the American Marketing Association (AMA) claims of the company According to Philip Kotler's definition of marketing, "marketing is a form of human activity aimed at satisfying their needs and wants through various forms of exchange." Original definition: "The science and art of exploring, creating, and delivering value to satisfy the needs of a target market while making a profit."

Both social marketing and production marketing are included in this definition

The promotion mixes direct marketing, which consists of marketing activities that leverage

IT and the internet to meet (measurable) consumer demands and/or interact with customers, includes online marketing merchandise all over the place to enter, safeguard, and build the market Online sales, electronic public relations, and online advertising are all part of online marketing This viewpoint is in agreement with that of Prof Dr Nguyen Bach Khoa, who stated: "Online marketing is a type of media channel of direct marketing tool (mode)

- one of 8 integrated marketing communication tools to refer to a comparable system A marketing firm that uses online communication methods to reach a consumer or potential customer anywhere in the world and give a quantifiable reaction or transaction with the aim of infiltrating, safeguarding, and expanding commercial markets Any communication activity that makes use of the internet for advertising, sales promotion, public relations, and online consumer sales is referred to as online marketing

We may infer from the aforementioned definitions of internet marketing that it is the use

of electronic methods to market and promote goods and services

Online marketing is a type of online marketing that makes use of the Internet to inform potential clients about a company's brand, products, or services Websites, emails, social media, display ads, search engine optimization (SEO), Google AdWords, and other channels are used for online marketing Online marketing aims to connect with potential consumers through the websites and apps they use to read, search, shop, and socialize 1.1.2 Features of online marketing

As the Internet grows and there are more and more users, consumer behavior is progressively changing as well With access to a wealth of information, consumers are able

to purchase and sell a variety of goods online, which greatly aids marketing efforts The internet will become a more important medium for engagement and trade as technology advances, progressively weakening traditional marketing The key reason is that internet marketing is more successful in marketing activities and promoting companies, products, and services since it has many distinctive qualities compared to traditional marketing Online marketing contains the following fundamental traits:

e Unlimited space and time

Geographical location and implementation time are limitations of traditional marketing Customers simply need to utilize the internet in order to deploy online marketing anywhere and at any time, as everything happens swiftly on the internet This enables several buyers and sellers to cut the conventional middlemen

e High interactivity

The internet's interaction is well illustrated They provide two-way communication, offer several informational levels, and establish a direct supplier-customer connection Online marketing initiatives provide people with the information they've asked for and let them access it Customers have the option to comment on a product, ask for more information about it, or decline to receive it

e Customer feedback

Customers usually reply quickly to online marketing since information is delivered over the network more quickly, but traditional marketing must wait a long time for a response

e Product diversification

Customers may now shop just like they would in a physical store while at home, seated in front of a computer with an internet connection These online shops’ offerings of goods and services are becoming more varied and richer, which draws customers’ interest There are several ways for online marketing to get the proper customers Just as they may utilize databases as the foundation for direct marketing, they can target certain organizations, nations, or geographic areas Businesses may also use customer behavior and personal preferences to target the correct demographic

1.1.3 Tools of online marketing

e Website marketing

The advantages of having a website for advertising a business's goods and brand are becoming increasingly significant in the present information technology era as people grow more accustomed to using the internet and it progressively becomes a necessary tool in daily life therefore it's enormous The website offers various advantages depending on the business field of each company

Here are a few examples of business apps that employ an electronic office website that is open 24/7:

- Market analysis

- Public relations, commercial advertising, and corporate goods

- Sell

- Press relations; - Brand management

- Management of human resoutces

Hundreds of millions of people every day from all over the world use websites to find information and goods from companies that are active in the online marketplace

Customers can get information about items and services that are always available on a website

e Social network channels for communication

Websites that offer online connection services include social networks allowing users to sign up and register their information on that website, then connecting with one another through information exchange, photo sharing, friend-making, group- and association- forming, etc The interpersonal connections and interaction are excellent Online media concentrating on social networks have been drawn to and circulated by the huge and strong social network

YouTube, Facebook, and Zing Me are now regarded as the top 3 social networks in Vietnam

Social network types:

- Social networks: Facebook, Zing me, Go.vn

- Share videos: Youtube, Clip.vn, Viadeo

Because they enable users to rapidly connect with their friends, family, or other online users, social networks are regarded as the most effective online marketing tools for internet consumers rapid travel throughout the globe, update information and images, register for nearby activities, or just serve as a forum for interpersonal communication (Lan Huong, 2013)

e Additional internet advertising tools

Online advertising, mobile advertising (also known as mobile marketing), banner advertisements, sponsored ads, and other online marketing methods are available in addition to the ones mentioned above But I suggest five of the most common and widely used online marketing tools for more study and improvement as part of my research project

on the use of internet marketing in Vietnam

1.2 Influence of online marketing on customer behavior

The consumer takes the time to visit the fashion store if they want to purchase a bag Thanks

to the internet, everything has changed, and buyers may still purchase bags created overseas without physically leaving their homes demonstrates how customer behavior in person and online are very different Consumers find it challenging to alter their behavior and routines as well as to accept practical realities like the inability to touch or feel actual things while making grocery purchases (Digital Svensk Handel, 2014) A change in behavior, which results from a change in attitude, is required to alter how customers purchase (Fishbein & Ajzen, 1975) This emphasizes models and attitudinal theories that can help in explaining customer acceptance of e-commerce Grandén, Nasco, and Mykytyn (2011) further stress the significance of comprehending how consumer attitudes impact the adoption of IT, particularly e-commerce

1.2.1 Influence of online marketing on decision making

The classic process theory of purchasing behavior, which holds that attitudes toward purchases are impacted by both traditional and online channels, becomes more complex with the introduction of internet media line Online marketers have little control over the inputs offered by peer reviews, blogs, social networks, or other types of user-generated material These sources determine the choice and decision to mark prompts

Consider tying perception and purchase to the purchasing process so that all variables, including brand reputation, application, performance, and enablement, are taken into account In comparison to conventional media, internet media engages viewers and uses the term perception throughout the review, overcoming perception and favoring consideration for, cognitively speaking, motivating purchase

1.2.2 Online buying involves risk

However, there are drawbacks to making purchases online as opposed to offline Customers who shop online are unable to understand the items they see, touch, taste, smell, and hear online when they search for and purchase them Customers who purchase online

could develop little trust and see lifting as high risk because to the absence of face-to-face interaction However, adopting certain software solutions, such as online dealer recommendations, might lessen this challenge Consumers who are aware of the risks may decide to switch to traditional retailers to purchase goods Online buying is encouraged by

a reduction in perceived risk (Tan, 1999) Both technology failure (such as system breaches) and human error (such as incorrect data entry) can contribute to perceived risk Financial risk (e.g., is my credit card information safe? ), product risk (e.g., is the product

of the same quality as what I can see on the screen? ), convenience risk (e.g., will I understand how to order and return goods? ), and risk of undeliverability (e.g., what if the product is not delivered? ) are the most frequently mentioned risks associated with online shopping Consumers’ perceptions of perceived risks are influenced by the level of ambiguity surrounding the online purchase process (Bhatnagar et al., 2000)

CHAPTER 2: LITERATURE REVIEW 2.1 Background of research

Sandra Forsythe, Chuanlan Liu, and David Shannon's "Establishing a Scale to Measure the Benefits and Risks of Online Shopping" study in 2006, Liu Chun Gardner published an article in the Journal of Interactive Marketing: This study constructed a scale for the benefits of online shopping, such as shopping convenience and product selection ability Ease/Comfort of Shopping, Hedonic/Enjoyment The risk structure includes financial risk, product risk, and time/convenience risk

In 2002, Na Li and Ping Zhang from Syracuse University published a review titled

"Research on Customer Attitudes and Behaviors in Online Shopping: This study showcases and explains customer attitudes and behavioral patterns in online shopping It also provides independent variables, which are typically used to study customer behavior and attitudes

in online shopping

Strauss, El Anssary&Frost, 2003: Introduction to the Internet and other technologies that have had a profound impact on business methods This transformation has led to new business technologies, increased customer value, established customer relationships, and increased company profits

Internet Marketing: A Combination Strategy of Online Advertising and Direct Advertising, Mary Low Roberts, 2002: This work involves developing marketing strategies The author introduces the concept of internal marketing from different perspectives and focuses on building communication strategies based on this online tool

The book "Ung dụng Marketing điện tử trong kinh doanh" written by Pham Thu Huong and Nguyen Van Thoan, 2009: This book points out the advantages and status of electronic marketing in e-commerce, and the advantages of electronic marketing compared with traditional marketing In addition, the author also introduces the application of classic marketing in international business and the practical application of e-marketing in Vietnamese import and export enterprises

The book “Giáo trình Marketing thương mại điện tử” by Nguyen Hoang Viet, 2011: The author gives an overview of e-commerce marketing and e-commerce marketing management In addition, the author also puts forward the theory of customer purchase behavior and mixed marketing management in e-commerce

2.2 Research model and hypothesis

Based on the research model of customer attitudes in online shopping by Nali and Ping Zhang (2002), as well as the online shopping return and risk measurement table developed

by Sandra Forsythe, Chuanlan Liu, David Shannon, and Liu Chun Gardner (2006) teams Simultaneously passing the author's test Since then, a research model consisting of independent variables such as websites, social networks, e-commerce exchanges, search engines, and email has been proposed Another dependent variable is accepting the following risks:

following assumptions were constructed during the research Based on previous research, the author's qualitative methods were combined with observational variables to construct

a risk tolerance component for online purchases:

- Risks of purchasing through websites

- The risks of purchasing through social networks

- The risk of purchasing goods through e-commerce exchanges

- The risk of purchasing through search engines

- The risk of purchasing through email

In order for customers to purchase products online, they must bear less risk Therefore, research shows that H hypothesis is as follows:

H: Customers purchase products with minimal risk, assuming the following factors: H1: The professional design of the website makes customers feel less risky

H2: User connections on social media make customers feel less risky

H3: The security guarantee of e-commerce transactions allows customers to bear less risk H4: Information support in search engines reduces customer risk

H5: The permission in the email places less risk on the customer

CHAPTER 3: RESEARCH METHODOLOGY

3.1 Research Process

3.1.1 Research process construction

- Choose the topic for the research: First of all, our group depended on our major in university to choose the main factor for this research that is Online Marketing Secondly, choosing the topic about online shopping behaviour that have a reason Because young people are purchasing a lot on online platforms, and there are many reasons which attract them to make purchasing decisions

- Collect documents: After choosing the topic, we started to collect documents that will help us to solve the problems in this research The first source where we take information

to serve the research process is our knowledge about Online Marketing field The most important source that helps us a lot in this research is books of Marketing and Market Research experts We collected documents which are verified by experts Some documents that we used are Books, E — books, research of experts

3.1.2 Research Subject

Researching about factors of Online Marketing and online shopping behaviour, our group chose the subject to do the research that is HUFLIT student With the age from 18 — 27 years old, online shopping demand of them is very high

3.1.3 Purposes

The study of this topic is to systematize the theoretical issues related to online marketing Learn about the current state of online marketing applications of businesses and stores around Huflit From there, understanding this situation will give us a clearer view through analyzing the influence of online marketing tools in online shopping On the other hand, this research also contributes to boosting the online shopping motivation of Huflit students, thereby serving as a basis for businesses and stores around Huflit to improve their sales efficiency The results of this study will provide appropriate recommendations and

10

recommendations for businesses and stores, so that they can more effectively use online marketing tools to attract more customers

3.1.4 Objective

As mentioned, online shopping is popular but still limited Therefore, the objective of the study is to understand and explore online buying behavior as well as the factors affecting the buying behavior of Huflit students From there, propose the most effective direction for the enterprise system

Approach students at HUFLIT to gather all the necessary information for the project and learn and explore the factors related to online marketing

3.1.5 Process

After having the goals and objectives for the topic => give steps to implement

- Step 1: Find out the necessary information to implement the project - In this topic is some basic knowledge about Marketing, Online Marketing, youth trends, current market trends

- Step 2: Take a survey: Collect real information from the subjects you want to research — (in this topic, HUFLIT students) From there, there are factors that can achieve the objectives of the project

- Step 3: Implement the project: After having all the necessary information => start planning to write the project from the collected information => Draw conclusions to achieve the original purpose head

3.1.6 Methods and tools for collecting information

3.1.6.1 Methods for collection information

Information and data are collected through surveys based on pre-designed questionnaires sent online via gmail, on channels such as facebook, zalo, etc

3.1.6.2 Tools for collecting information

11

Information collection tool is a questionnaire used to poll the opinions of the subjects, in which:

- The question type is a structured (closed) question with pre-determined question types and answers and respondents only need to choose, including: two-answer questions, choose one; multiple-answer, single-choice questions; rated on a given scale

- Primary data: based on the survey and survey by questionnaire, this questionnaire will be implemented based on closely following the research topic and will be deployed to the surveyed people through Google Forms It is expected that the survey sample will be from 250-300 samples

3.2 Data processing method

3.2.1 Methods of data analysis

Goertzen (2017) wrote that quantitative research methods are fundamentally concerned with gathering and evaluating structured data that may be represented quantitatively.2 The creation of precise and trustworthy measurements that enable statistical analysis is one of the main objectives Quantitative research is especially good at addressing the "what" or

"how" of a particular issue since it concentrates on data that can be measured Direct,

'

measurable questions frequently use expressions like "what percentage" and "what proportion?" How many, and how much? Quantitative research enables librarians to better understand a population's demographics, gauge the volume of users of a service or product, study patron attitudes and habits, note patterns, or justify anecdotal knowledge The above- mentioned variables can be quantified and supported by measurements such as counts, percentages, proportions, and relationships Quantitative research results reveal patterns and trends in behaviors It's crucial to keep in mind, though, that they don't explain why people feel, think, or act the way they do In other words, while trends in different data sets

or study groups are highlighted by quantitative research, the reasons for observed behaviors are not Qualitative studies, such as focus groups, interviews, or open-ended survey questions, are useful for bridging these information gaps

12

- The method of data collection:

Primary data: based on the survey and survey by questionnaire, this questionnaire will be implemented based on closely following the research topic and will be deployed to the survey respondents through Google Forms It is expected that the survey sample will be from 200-250 samples

- Methods of data analysis:

The simplest and most common method is the descriptive statistical method (in statistical software) to calculate the average number of a certain indicator in the questionnaire or to calculate the frequency of the answers appearing in the questionnaire questionnaires (questionnaire)

Synthesize quantitative data: the quantitative data obtained from the interview questionnaires should be entered into Excel software Depending on the research content and the type of information collected, the research team designs information entry forms

in Excel, coding variables (indicators) to be imported into Excel

=> After synthesizing and processing quantitative data in Excel, this data will be used to conduct analysis in SPSS

3.2.2 Factor analysis

- Reliability coefficient Cronbach Alpha (CA):

Cronbach's alpha coefficient is used to test the internal consistency of the representative variables of each factor Cronbach's alpha coefficient will remove the variables that do not fit Those are variables with Cronbach alpha coefficient less than 0.6

- EFA factor analysis:

Exploratory Factor Analysis (EFA) aims to reduce a set of many interdependent observed variables into a smaller set of variables (factors) so that they are more meaningful but still contain most of the content information content of the original set of variables (Hair et al.,

13

1998) Exploratory factor analysis is considered appropriate when the following criteria are satisfied

Important criteria for the largest Factor Loading to be considered:

According to Hair et al (1998), Factor loading is an indicator to ensure the significance level of EFA Factor loading of 0.4 is the acceptable threshold

Kaiser-Meyer-Olkin (KMO) is the index used to consider the appropriateness of EFA: 0.5<KMO < |, then factor analysis is appropriate

Jabnoun & Al-Tamimi (2003) standardizes the difference of the factor loading coefficient

of an observed variable between factors > 0.3 to ensure the discriminant value between factors

Bartlett's test sphericity considers the hypothesis HO: the correlation between the observed variables is zero in the population If this test is statistically significant (Sig < 0.05), the observed variables are correlated with each other in the population and reject the hypothesis

HO

Cumulative % of variance: percentage variation of observed variables (or data) explained

by factors must ensure > 50%

The method of coefficient extraction used is Principal Component Analysis with Varimax rotation to minimize the number of variables with large coefficients at the same factor and the factors are not correlated with each other

Determining the number of factors by the method based on eigenvalue (Determination based on eigenvalue): only keep the factors with eigenvalue greater than | in the analytical model

After analyzing EFA, the research hypotheses are adjusted according to new factors Multiple linear regression analysis method will be applied in assessing the influence of factors on customer satisfaction

- Multivariate linear regression analysis

14

Multivariate regression analysis is a method used to analyze the relationship between one dependent variable and many independent variables

The multivariable linear regression has the form:

Yi= BO + BIXII +B2 X2IT tp Xp1 Tel The purpose of multivariable regression analysis is to predict the extent of the dependent variable (with limited accuracy) when the value of the independent variable is known in advance According to (Hoang Trong and Chu Nguyen Mong Ngoc - 2005) the important parameters in multivariable regression analysis include:

Partial regression coefficient Bk (Beta coefficient): is the coefficient that measures the change in the mean value of Y when Xk changes by one unit, among the remaining independent variables unchanged

Adjusted R square: The coefficient of determination of the rate of variation of the dependent variable explained by the independent variable in the regression model That index is also a parameter that measures the relevance of the regression line according to the R? rule, the closer it is to 1, the more suitable the model is built, the closer to R? the less suitable the model is to the sample data set

The F-test in the analysis of variance is a hypothesis test of the fit of the overall linear model If the hypothesis HO of the F test is rejected, it can be concluded that the multivariable linear regression model fits the data set and can be used

15

CHAPTER 4: ANALYSIS 4.1 Sampling results

Our team uses Google Form to make the questionnaire then we use social networking platforms like Facebook Zalo to send a link to the survey form to students studying at the school After submitting the survey form 236 responses were obtained through the process of eliminating invalid or unsatisfactory survey samples, 200 valid responses were obtained

= Senior

Chart 4 2 Current academic year of students in school

According to statistics, students often use online marketing tools Among them, Senior students account for 62.5%, and Junior students account for 37.5% It can be seen that the ratio is quite high between Junior and Senior students Therefore, we see that Junior and Senior students have more experience and a better understanding of how to use online marketing tools when shopping online

e Frequency of online purchases

17

Chart 4 3 The frequency of online purchases by Huflit students

The statistical analysis results show that online shopping has become a habit for students

to shop Online shopping is more convenient, rather than going to shopping centers and choosing products and spending time and effort choosing the right products 71.5% of school students stated that they made multiple online purchases within a month In addition, 28.5% of students stated that they have made multiple purchases online Therefore, we see that online shopping has become a habit for Huflit students

e Online shopping channels

Chart 4 4 Huflit students' online purchase channels

Data obtained through a survey of students’ opinions The results show that 54 5% of students are accustomed to using e-commerce platforms for online shopping 31.5% of students like to shop online on social media Others like shopping on brand websites What you can see E-commerce platforms are the first choice for students to shop online, because there are often discounts and discount codes Next to it Social media Students also tend

to buy, because there are commodity label advertisements on social media platforms The remaining 14% are purchased on websites, so there are few permanent defects, or only the websites of this brand sell products that students want to buy

4.2 Reliability test

The factors are preliminarily evaluated through Cronbach's Alpha reliability coefficient to eliminate unsuitable variables first variables with a correlation coefficient of total variables (Corrected Item-Total Correlation) less than 0.4 will be excluded and the criteria for choosing the scale when having alpha reliability of 0.6 or higher (Nunnally & BernStein.1994) Results of performing Cronbach Alpha on SPSS software We can see the results of the reliability analysis of the following factors:

e Scale of online marketing tool aspects

Cronbach's Alpha analysis results of online marketing tool aspects are presented in Table 4.1 with specific results as follows: (See details in Appendix 3.1)

- Website aspect has Cronbach's Alpha (0.693); observed variables WEB1, WEB2, WEB3, WEB4, WEBS have total correlation coefficients over 0.4

- Social Media aspect has relatively high Cronbach's Alpha (0.784); Close variables in this component have high total variable correlation coefficients are greater than 0.5 So, this is

a good scale

- The aspect of E-commerce Exchange has Cronbach's Alpha of 0.754; The observed variables in the component have the total correlation coefficient greater than 0.4 The achieved indicators are in accordance with the criteria, so this is a good scale

19

- The Search Engine aspect has a Cronbach's Alpha of 0.629; observed variables SE1, SE2, SE3 all have correlation coefficients of all variables greater than 0.4; except for variables SE4, SES has coefficients respectively 0.232; 0.183 is much lower than other variables in the population, but when these two variables are removed from the scale, the Cronbach's Alpha coefficient of the scale does not increase significantly, so we still keep these two observed variables

- The Email Marketing aspect has a Cronbach's Alpha of 0.794, which is a relatively high index; All observed variables have total correlation coefficient exceeding 0.4 This shows that these observed variables all perform their measurement function well

Table 4 1 Cronbach's Alpha coefficient of the scale of online marketing tools (N =

200) Scale Corrected Cronbach's Scale Mean if

Code Variance if | Item-Total Alpha if Item

e Risk Acceptance Scale

The Cronbach alpha coefficient of the Risk Acceptance Scale reaches 0.931; The correlation coefficient between variables in this scale ranges from 0.671 to 0.92, which is very high; This indicates that the observed variables have high reliability At the same time, the alpha coefficient of the scale is greater than 0.8, so the conclusion of the scale can be well drawn (see Table 4.2) (see Appendix 3.2 for details)

Table 4 2 Cronbach's alpha coefficient of the scale of Risk Acceptance (N = 200)

4.3 Exploratory factor analysis (EFA)

After testing the reliability of scale, exploratory factor analysis (EFA) conducted principal component analysis using Varimax rotation Firstly, we used KMO and Barrett's tests to evaluate the correlation of EFA analysis In EFA analysis, the Kaiser Meyer Olkin coefficient is an indicator that considers the applicability of factor analysis KMO values must be between 0.5 and 1 to be appropriate

If these values are less than 0.5, factor analysis may not be suitable for the data

e Scales for online marketing tools

After using Cronbach alpha analysis to check reliability, the scale for online marketing tools was measured using 31 observation variables from the 5 components of the scale Factor analysis is used to evaluate the degree of component convergence of observed variables

In the HO hypothesis proposed in this analysis, there is no correlation between the 31 observed variables The testing of KMO and Bartlett in factor analysis showed that the hypothesis was rejected (sing=0.000); The KMO coefficient is 0.592 (>0.5) This result indicates that the observed variables are generally correlated, and factor analysis (EFA) is appropriate (Table 4.3)

Table 4 3 KMO and Bartlett Test - Scale of aspects of online marketing tools

23

KMO and Bartlett's Test

Table 4 4 Factor analysis results - Scale of online marketing tool aspects

Independent Code Factor Loading N

WEB2 0.523 Website WEB3 0.619 5

WEB5 0.600

SM2 0.503 SM3 0.558 SM4 0.686 7

SM6 0.730 SM? 0.725

BCE! 0.638 ECE2 `

24

0.693 SEI 0.487 SE2 0.586

SE4 0.548 SES 0.521 EMI 0.545 Email Marketing EM2 0716

EM3 0.638 EM4 0.690

e Risk Acceptance Scale

The Risk Tolerance Concept Scale consists of five observation variables: R1, R2, R3, R4, and R5 In the HO hypothesis proposed in this analysis, there is no overall correlation

25

between the five observed variables The testing of KMO and Bartlett in factor analysis showed that this hypothesis was rejected (sig=0.000), and the KMO coefficient was 0.857 (>0.5) This result indicates that the observed variables are generally correlated, and factor analysis (EFA) is appropriate (Table 4.5)

Table 4 5 KMO and Bartlett Test- Risk Acceptance Scale KMO and Bartlett's Test

Approx Chi-Square 987.554

df 10 Sig .000

Through EPA analysis using principal component extraction method and Varimax rotation method, an element 3995 was extracted from eigenvalue with a variance of 64.217% (>50%) In addition, the factor loading coefficients of the observed variables are quite high (0.771-0.958), so all variables can be accepted in the scale (Table 4.6) (see Appendix 4.2

for details)

Table 4 6 Factor Analysis Results - Risk Acceptace Scale

Dependent Code Factor Loading N

RI 0.958 R2 0.923 Risk Acceptance R3 0.923 5

R4 0.883 R5 0.771

26

Eigenvalues 3.995

Percentage of Variance Explained (% phuong sai trich) 64.217

4.4 Correlation analysis: The relationship between variables

The correlation matrix in table 3.8 presents the Pearson correlation coefficients (r) between the research variables The coefficient is considered significant if the p-value is less than

or equal to 0.05 In Appendix 5.2, we see all VIFs < 2, so there is no multicollinearity phenomenon - the phenomenon between independent variables is closely correlated, thus increasing the standard deviation of the coefficients regression and reduce the statistical value of the test of significance In addition, the analysis results also show that there is a correlation between the independent variables - aspects of online marketing tools and the dependent variable - risk acceptance First, Website has a strong positive correlation with risk tolerance (r=0.428;p<0.01); next is Search Engine (r=0.179; p<0.01); and moderately positive correlations, specifically including the following relationships: E-Commerce Exchange (t=0.145;p<0.01), Email Marketing (r=0.126; p<0.05); email (r= 0.101; p<0.01) The above results show that the respondents perceive that the positive values of online marketing tools will reduce risk acceptance

Table 4 7 The correlation between online marketing tools and risk acceptance Correlations

4.4 Repression analysis

The purpose of analyzing this section is to establish a model to determine the relationship between customers’ risk tolerance and factors, and to confirm the importance of each factor that affects risk tolerance In other words, Regression analysis will demonstrate the correctness of the conceptual model in the current research environment and find a unified model to explain the relationship between factors affecting risk tolerance Provide suggestions to improve the effective use of online marketing tools for small and medium- sized manufacturing enterprises This analysis was conducted using multivariate regression techniques

General regression model after EFA analysis: R=f(WEB,SMECESEEM)

Through multiple linear regression, consider which factors truly affect direct risk tolerance:

28

R=B0+B1WEB+B2SM+B3ECE+B4SE+B5EM

Among them, B0, B1, B2, B3, B4, and B5 are non standardized regression factors Evaluate the applicability of multiple linear regression models: Multiple linear regression analysis (enter), which included all variables (in terms of online marketing tools), showed an appropriate regression model for validating the theoretical model (sig, F=0.000) and explained 64.2% of the dependency difference - Risk acceptance (Adjusted R? = 0.642) (Table 4.8)

Table 4 8 Model determination coefficients Model Summary

R Adjusted R Square | Std Error of the

Variance Proportions Eigenvalu | Condition

The regression coefficients of 5 aspects of online marketing tools are listed in Table 4.8, specifically Website (B=0.542:p<0.05); Social Media (B=0.515; p<0.05); E-Commerce Exchange (B=0.112;p<0.05); Search Engine (B=0.32; p<0.05); Email Marketing (B=0.053; p<0.05) positively affects risk acceptance (details in Appendix 5.2.2)

From the above results, the equation for risk tolerance for online marketing tools is written

as follows:

R= 8.251 +0.542 WEB + 0.515 SM + 0.32 SE + 0.112 ECE + 0.053 EM

Thus, there are 5 factors affecting customer's risk acceptance including Website, Social Media, E-Commerce Exchange, Search Engine, Email Marketing In which, Website is the biggest influence on customer's risk acceptance

Based on the results of multiple linear regression and analysis, 5 aspects of online marketing tools that affect guest risk acceptance are described by the following model:

31