Home / Consumption

Analysis of Factors Affecting Consumer Attitudes


Table 4.18: Summary of Cronbach Alpha coefficients of the scales


STT

Ingredient name

Number of observed variables

Cronbach Alpha

1

Information

3

0.884

2

Entertainment

2

0.804

3

Trust

3

0.913

4

The annoyance

3

0.936

5

Personalization

2

0.723

6

Promotional gifts

3

0.703

7

Permission

2

0.815

8

Consumer attitudes

4

0.717

9

Consumer experience

3

0.759


Total

25


Maybe you are interested!

Analysis of Factors Affecting Consumer Attitudes

4.5.2 Exploratory Factor Analysis (EFA)


The main objective of this section is to determine which of the factors hypothesized in the proposed research model have an impact on consumer attitudes and behavior towards Mobile Marketing activities.

4.5.2.1 Analysis of factors influencing consumer attitudes


In the proposed research model mentioned in chapter 2, there are seven factors (corresponding to 21 observed variables) that are assumed to have an influence on consumer attitudes. However, as proposed in the section on testing the reliability coefficient of the scale, the author removed 3 observed variables, INFO3, ENT1, PER3, due to the total correlation of variables < 0.4. After removing the above 3 variables, the official scale has 18 observed variables left. Applying the method of rotating the factors, the results show that all 18 variables are significant.


significance (value > 0.5) and extracted into 5 factor groups with total extracted variance = 78.165% meeting the requirement of greater than 50%. (Table 4.19)

Table 4.19: Matrix of factors affecting consumer attitudes after rotation (Rotated Component Matrix a )



Component

1

2

3

4

5

Advertising messages provide the information I need

.843





The information of the advertising message is up-to-date.

.822





Advertising messages give me a lot of marketing information.

.795





school






The promotional messages I receive often have content and

.648





writing style appropriate for my age






The promotional messages I received were relevant to my needs.

.633





my personal






I am attracted to text message winning programs.


.641




I find it fun to read the promotional messages I receive.


.594




Okay






The promotional messages I receive often come with gifts.


.575




Promotions (ringtones, wallpapers, games, codes)






reward…)






I am willing to receive promotional messages if I am accepted.


.569




more gifts






The gift sent with the promotional message is of very low value.


.558




I find the SMS advertising program reliable.



.765



higher than advertising programs on the media






Other (TV, radio, newspaper...)






I trust brands that are advertised through social media.



.750



message






I think text message advertising is very reliable.



.681



I only allow the units I have registered to send messages





.682

ads send me messages






I do not allow all units to send advertising messages





.559

for me






Sending advertising messages during working and off hours




.716


It bothers me a lot.






I get annoyed when I receive promotional messages.




.698


Advertising messages are just annoying.




.692


Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.

a. Rotation converged in 6 iterations.


The analysis results show that, unlike the research model proposed in chapter two, the factor analysis procedure only extracted 5 factors that ensure significance from the original 7 factors. The newly identified factors include both their inherent observed variables and the observed variables of other factors. Therefore, the term factor group will be used to refer to these newly extracted factors. The five groups of extracted factors after rotating the factors will be renamed as: (1) Perception of information value, (2) Perception of entertainment value, (3) Perception of trustworthiness, (4) Perception of annoyance and (5) Permission.

In addition, the measurement results show that the “Personalization” scale in the initially proposed theoretical model has been removed and combined with the “Information” scale to become a new scale called “Perception of information value”. This shows that although the theory of Mobile Marketing separates these two contents, in Vietnam, consumers identify the “Personalization” of the Mobile Marketing program with the “Information” it brings. The “Promotional gifts” scale is also combined with the “Entertainment” scale to create the “Perception of entertainment value” scale because respondents consider promotional gifts to also serve entertainment purposes. This is also the difference between Vietnamese consumers and foreign consumers: “Entertainment” is considered a spiritual perceived value, while promotional gifts are considered material values. Other scales such as “Annoyingness”, “Trust” and “Permission” still fit the original theoretical model.


Table 4.20: Cronbach Alpha coefficient of newly extracted components


STT

Extractable component name

Number of variables

Cronbach alpha

1

Perception of information value

5

0.830

2

Perception of entertainment value

5

0.784

3

Trust

3

0.913

4

The annoyance

3

0.936

5

Permission

2

0.815


Total

18


4.5.2.2 Analysis of factors influencing consumer behavior


Based on the results of Cronbach Alpha coefficient testing of the scales, to improve the reliability of the scales, the author removed the two observed variables ATT5 and EX4 in the section of factors affecting consumer behavior.

Similar to the above section, to perform factor analysis, the author conducts the factor rotation procedure. After 3 rotations and removing variables with results < 0.5, the results obtained only 5 significant observed variables from 7 variables included in the initial analysis corresponding to 2 extracted components with a total extracted variance = 84.002% meeting the requirement of greater than 50%. (Table 4.21)

Table 4.21: Matrix of factors influencing consumer behavior after rotation (Rotated Component Matrix a )



Component

1

2

I have used the information of advertising messages to choose products,

.778


service for me



I have a lot of experience with text message advertising.

.752


I am satisfied with the information provided by the promotional message


.691

I find using text messages for advertising a good idea.


.633

I don't care about promotional messages


.605

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.

a. Rotation converged in 3 iterations.


Based on the meaning of the observed variables, we can determine that the two newly extracted factors are identical to the two proposed factors, the experience factor and the attitude factor. Therefore, the author decided to keep the names of both factors above. The experience factor includes 2 observed variables and the attitude factor includes 3 observed variables. The procedure for testing the Cronbach Alpha reliability coefficient of the two newly extracted components shows that both newly identified factors have high reliability and ensure significance in the analysis. (Table 4.22)

Table 4.22: Cronbach Alpha coefficient of newly extracted factors


STT

Extractable component name

Number of variables

Cronbach alpha

1

Experience

2

0.841

2

Attitude

3

0.826


Total

5


4.5.2.3 The research model and hypothesis are adjusted according to the survey reality.


H1

H2

Attitude

H6

H3

H7

H4

H5

Perception of information value

Perception of entertainment value

The annoyance

Behavior

Trust

Experience

Permission

In chapter 2, the author proposed a research model, but after conducting factor analysis and verification procedures based on actual survey results, the research model was adjusted to be more suitable. (Figure 4.6).


Figure 4.6: Adjusted research model


In addition, after adjusting the theoretical model, the new theoretical model has differences compared to the original theoretical model. Therefore, the study will add new research hypothesis groups for the newly developed factor groups: "Perception of information value" and "Perception of entertainment value". The new research hypothesis group will include the hypothesis group

- H1: Perception of information value affects the attitude of consumers in Hanoi's inner city towards Mobile Marketing activities.

- H2: Perception of entertainment value affects the attitude of consumers in Hanoi's inner city towards Mobile Marketing activities.

- H3: Trust affects the attitude of consumers in Hanoi inner city towards Mobile Marketing activities.

- H4: Distraction affects the attitude of consumers in Hanoi inner city towards Mobile Marketing activities.

- H5: Permission affects the attitude of consumers in Hanoi inner city towards Mobile Marketing activities.

- H6: Consumers' attitudes in Hanoi's inner city area affect their behavior towards Mobile Marketing activities.

- H7: The experience of consumers in Hanoi's inner city area affects their behavior towards Mobile Marketing activities.

4.5.3 Regression analysis


In the adjusted research model (Figure 4.6), the independent variable attitude is affected by 5 dependent variables, so to estimate the research model of the influence of factors on the attitudes of Vietnamese consumers


For Mobile Marketing, the thesis uses multiple linear regression equation. The equation has the form:


ATT = β0 + b1INFO + b2ENT + b3CRE + b4ANN + b5PER


In there:

ATT: Attitude

INFO: Perception of information value

CRE: Trust

ENT: Perception of entertainment value

PER: Permission

ANN: The Annoying

Meanwhile, consumer behavior is a dependent variable of two independent variables: attitude and experience. Therefore, to estimate the research model of the influence of factors on Vietnamese consumer behavior towards Mobile Marketing, the study also uses multiple linear regression equations. The regression equation has the form:

CB = β1 + b6ATT + b7EX

In there:


CB: Behavior

ATT: Attitude

EX: Experience

In addition, as mentioned in the data processing techniques section in the research methodology chapter, the author uses multiple regression method to test the research model because the multiple regression method allows building a correlation model with many factors that affect the dependent variable.

It can be said that the multiple regression model closely reflects the overall model and can clearly assess the importance of the concepts that need to be studied that have their own correlation with the dependent variable, Hoang Trong (2008) [11]. To


To evaluate the suitability of the linear model, we use the coefficient R, R2 (with 0 < R2≤ 1), adjusted R2 and standard error. The regression equation is estimated based on data collected through the results of a survey of 835 sample elements.

4.5.3.1 Regression analysis of factors influencing consumer attitudesfor Mobile Marketing activities

Table 4.23 and Table 4.24 below show the results of regression model analysis with the dependent variable being “consumer attitude” and 5 independent variables being (1) Perception of information value, (2) Perception of entertainment value, (3) Perception of trustworthiness, (4) Perception of annoyance and (5) Permission.

Table 4.23: Statistical results of factors affecting consumer attitudes towards Mobile Marketing

Model Summary



Model


R


R Square


Adjusted R Square

Std. Error of the Estimate

1

.820a​

.672

.665

.503

a. Predictors: (Constant), Perceived information value, Perceived entertainment value, Trust, Annoyance, Permission

ANOVA b


Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

365,603

5

24,374

96,359

.000 a

Residual


Total

178,580


544,183

48


53

.253



a. Predictors: (Constant), Perceived information value, Perceived entertainment value, Trust, Annoyance, Permission

b. Dependent Variable: Consumer Attitude

Comment


Agree Privacy Policy *