Factors affecting the intention to use an automated car

Exploratory factor analysis EFA with Independent variables.The number of factors analysis with 5 independent variables includes Performance Expectancy, Effort Expectancy, Social Influenc

MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY UNIVERSITY OF

TECHNOLOGY AND EDUCATION

FACTORS AFFECTING THE INTENTION TO USE AN AUTOMATED

Subject: Data Analysis Class Code: DANA220606E_22_2_01FIE

Student Performance: Group 6 Lecturer: Ph.D TRUONG THI HOA

Ho Chi Minh City, March 14, 2023

LIST OF MEMBERS PARTICIPATING IN THE ASSIGNMENT SEMESTER II, SCHOOL YEAR: 2022-2023

GROUP 6

No Name of Members

ID Contribution Contents ContributionRate 1 Phạm Hữu Anh 21124313 - Run data

3 Bùi Nhật Thành 21124030 - Run data

- Exploratory factor analysis

Research topic: Factors affecting intention to use an automatic car in Thu Duc City- Ho Chi Minh City Note Leader: Phạm Hữu Anh I TESTING RELIABILITY OF MEASUREMENT SCALES (CRONBACH’S ALPHA) 1 1.1 Performance Expectancy Scale 1

1.2 Effort Expectancy Scale 3

1.3 Social Influence Scale 4

1.4 Facilitating Conditions 5

1.5 Hedonic Motivation Scale 6

1.6 Behavioural Intention Scale 7

II EXPLORATORY FACTOR ANALYSIS EFA 8

2.1 Exploratory factor analysis EFA with Independent variables 8

2.2 Exploratory factor analysis EFA with dependent variable 18

III TESTING RESEARCH MODEL AND HYPOTHESIS 21

I TESTING RELIABILITY OF MEASUREMENT SCALES (CRONBACH’S ALPHA) 1.1 Performance Expectancy Scale

I would use the time during which a conditionally automated car is driving for other activities

I expect that a conditionally automated car would be useful in meeting my daily mobility needs

Using a conditionally automated car would help me reach my destination more safely

Using a conditionally automated car would help me reach my destination more comfortably

I assume that a conditionally automated car would be useful in my daily life

Look at the table’s result, keeping these five items (the Corrected item – Total correlation ≥ 0.3) 1.2 Effort Expectancy Scale

Reliability Statistics

1.3 Social Influence Scale I assume that people whose

opinions I value would prefer that I use a conditionally automated car

I expect that people who influence my behavior think that I should use a conditionally automated car

I expect that people who are important to me think that I should use a conditionally automated car

I could acquire the necessary

knowledge to use a conditionally automated car

I would expect the use of a conditionally automated car to be compatible with other digital devices I use

I would expect to have the necessary knowledge to use a conditionally automated car

I would be able to get help from others when I have difficulties using a conditionally automated car

Look at the table’s result, keeping these four items (the Corrected item – Total correlation ≥ 0.3) 1.5 Hedonic Motivation Scale

1.6 Behavioral Intention Scale

I intend to use a conditionally

I plan to buy a conditionally

automated car once it is available 11.54 6.096 .746 .832

Assuming that I had access to a conditionally automated car, I predict that I would use it

I would use a conditionally automated car during my everyday trips

Look at the table’s result, keeping these four items (the Corrected item – Total correlation ≥ 0.3)

Conclusion: After analyzing Cronbach’s Alpha of each scale, we find that no variable, is eliminated and the scales have no changes after reliability testing.

II EXPLORATORY FACTOR ANALYSIS EFA

2.1 Exploratory factor analysis EFA with Independent variables.

The number of factors analysis with 5 independent variables includes Performance Expectancy, Effort Expectancy, Social Influence, Facilitating Conditions, and Hedonic Motivation The process has 5 parts:

Part 1: Analysis of the Performance Expectancy Variable KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy .832

Bartlett's Test of Sphericity

Approx Chi-Square 520.278

The factor analysis method has the standard in the KMO that must be greater than 0.5 and the P-value (Sig.) of the significance level must be less than 0.05

The result table shows:

Kaiser-Meyer-Olkin Measure of Smapling Adequcy (KMO) value = 0.832 ( >0.5).

-> Acceptable The data used for factor analysis are sufficiently intercorrelated to justify the factors.

P-value (Sig.= 0.000 < 0.05) -> Acceptable Measured variables are sufficiently intercorrelated to justify the factors.

Total Variance Explained

-> Acceptable The data used for factor analysis are sufficiently intercorrelated to justify the factors.

P-value (Sig.= 0.000 < 0.05) -> Acceptable Measured variables are sufficiently intercorrelated to justify the factors Total Variance Explained

Component Initial Eigenvalues Extraction Sums of Squared Loadings

Extraction Method: Principal Component Analysis.

This variable should be kept because it has based on the eigenvalue greater than 1 Explaining the total variance equals 71.886% and acceptable (>50%) in EFA.

Component Matrix

Component

1 I assume that people whose opinions I value would prefer that I use a

I expect that people who influence my behavior think that I should use a

I expect that people who are important to me think that I should use a

I would recommend a conditionally automated car to others 827 Extraction Method: Principal Component Analysis.

a 1 component extracted.

4 factors are kept because their overview of principal component analysis is higher than 0.5 Part 4: Analysis EFA of the “Facilitating conditions” variable

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling

The factor analysis method has the standard in the KMO that must be greater than 0.5 and the P-value (Sig.) of the significance level must be less than 0.05

The result table shows:

Kaiser-Meyer-Olkin Measure of Smapling Adequcy (KMO) value = 0.770 ( >0.5).

-> Acceptable The data used for factor analysis are sufficiently intercorrelated to justify the factors.

P-value (Sig.= 0.000 < 0.05) -> Acceptable Measured variables are sufficiently intercorrelated to justify the factors Total Variance Explained

Component Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative %

Extraction Method: Principal Component Analysis.

This variable should be kept because it has based on the eigenvalue greater than 1 Explaining the total variance equals 70.021% and acceptable (>50%) in EFA.

Component Matrix

Component

1 I could acquire the necessary knowledge to use a

I would expect the use of a conditionally automated car to

be compatible with other digital devices I use .827 I would expect to have the necessary knowledge to use a

I would be able to get help from others when I have

difficulties using a conditionally automated car .813 Extraction Method: Principal Component Analysis.

a 1 component tracked.

4 factors are kept because their overview of principal component analysis is higher than 0.5 Part 5: Analysis EFA of the “Hedonic Motivation” variable

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling

The factor analysis method has the standard in the KMO that must be greater than 0.5 and the P-value (Sig.) of the significance level must be less than 0.05

The result table shows:

Kaiser-Meyer-Olkin Measure of Smapling Adequcy (KMO) value = 0.720 > 0.5 -> Acceptable The data for factor analysis are sufficiently intercorrelated to justify the factors.

P-value (Sig.= 0.000 < 0.05) -> Acceptable Measured variables are sufficiently intercorrelated to justify the factors Total Variance Explained

Component Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative %

Extraction Method: Principal Component Analysis.

This variable should be kept because it has based on the eigenvalue greater than 1 Explaining the total variance equals 75.195% and acceptable (>50%) in EFA.

Component Matrix

Component

Using a conditionally automated car would be fun 883 Using a conditionally automated car would be entertaining 848 Using a conditionally automated car would be enjoyable 870 Extraction Method: Principal Component Analysis.

All of the explainings of the total variance have equals higher than 50% and are acceptable All of the factors in each variable have the principal component analysis, which is higher than 0.5

There are no dropped variables

2.2 Exploratory factor analysis EFA with Dependent variable

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy .813

Bartlett's Test of Sphericity

Approx Chi-Square 442.471

Factor analysis for dependent variable is Behavior Intention.

The factor analysis method has the standard in the KMO that must be greater than 0.5 and the P-value (Sig.) of the significance level must be less than 0.5

The result’s table shows:

Kaiser-Meyer-Olkin Measure of Smapling Adequcy (KMO) value = 0.813 ( >0.5).

-> Acceptable The dada used for factor analysis are sufficiently intercorrelated to justify the factors P-value (Sig.= 0.000 < 0.05) -> Acceptable Measured variables are sufficiently intercorrelated to justify the factors.

Total Variance Explained

Component Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative %

Extraction Method: Principal Component Analysis.

A factor is based on eigenvalues greater than 1 The total of variance extracted is 72.539% (>50%), explaining the data in the 4 observed variables participating in EFA.

Component Matrix

Component 1

I intend to use a conditionally automated car in the future 838 I plan to buy a conditionally automated car once it is available 863 Assuming that I had access to a conditionally automated car, I predict that I

I would use a conditionally automated car during my everyday trips 874 Extraction Method: Principal Component Analysis.

Explaining of the total variance have the equals higher than 50% and acceptable Factors in each variable have the principal component analysis, which is higher than 0.5

III TESTING RESEARCH MODEL AND HYPOTHESIS Correlations

PE Pearson Correlation 1 733** 697** 719** 710** 667**

** Correlation is significant at the 0.01 level (2-tailed).

PE (Performance Expectancy) – EE (Effect Expectancy) H0: rab = 0 (there is no linear relationship between the two variables) H1: rab ≠ 0 (there is a linear relationship between the two variables) Sig (0.000) < 0.05 → Reject H0, Accept H1

Pearson correlation = 0.733

-> PE (Performance Expectancy) and EE (Effort Expectancy) have a positive relationship SI (Social Influence) and FC (Facilitating Conditions)

H0: rab = 0 (there is no linear relationship between the two variables) H1: rab ≠ 0 (there is a linear relationship between the two variables) Sig (0.000) < 0.05 → Reject H0, Accept H1

Pearson correlation = 0.788

-> PE (Performance Expectancy) and EE (Effort Expectancy) have a positive relationship HI (Hedonic Motivation) and BI (Behavior Intention)

H0: rab = 0 (there is no linear relationship between the two variables) H1: rab ≠ 0 (there is a linear relationship between the two variables) Sig (0.000) < 0.05 → Reject H0, Accept H1

1 HM, EE, PE, SI, FCb Enter

R2 = 0,705 (70.5%): The model is able to explain 70.5% of the observed data This linear regression model fits the sample's data set.

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.

CHECK THE VIOLATION OF ASSUMPTIONS ASSUMPTION 1: CHECK SCATTERPLOT

Looking at the scatter plot, we see a uniform dispersion Thus, the assumption of constant variance of the regression model is not violated.

ASSUMPTION 2: DURBIN-WATSON VALUE

Durbin-Watson = 1.870 -> Weak positive autocorrelation: Acceptable

Through the above test results, the assumptions of the linear regression function are not violated and the built regression model is consistent with the overall.

ASSUMPTION 3: NO OR LITTLE MULTICOLLINEARITY: THE INDEPENDENT VARIABLES ARE NOT HIGHLY CORRELATED WITH EACH OTHER [COV(XI, XJ) =0]

Coefficientsa Model Unstandardized Coefficients Standardized

t Sig Collinearity Statistics

a Dependent Variable: BI

The variance inflation factor (VIF) reached the maximum value of 4.222 (less than 10), showing that these independent variables have a weak relationship with each other, so there is no multicollinearity phenomenon outside Therefore, the relationship between the independent variables does not significantly affect the explanatory results of the regression model.

The unnormalized regression equation has the form:

Y = 2.612E-016 + 0.103*PE + 0.022*EE – 0.049*SI + 0.298*FC + 0.518*HM The normalized regression has the form:

Y = 0.103*PE + 0.022*EE – 0.049*SI + 0.298*FC + 0.518*HM

ASSUMPTION 4: RESIDUALS ARE NORMALLY DISTRIBUTED

Observation of the histogram of the normalized residuals shows that the normal distribution of the residuals is approximately standard Mean= -1.57E-16 (mean is close to 0) and standard deviation Std Dev = 0.989 (standard

deviation) close to 1) Therefore, it can be concluded that the hypothesis of the normal distribution of the residuals is not violated.

The P-P plot also shows that the points of the residuals are scattered not far but randomly around the diagonal (the expectation line), so the assumption of the normal distribution of the residuals is satisfied.

ASSUMPTION 5: VIOLATION OF ASSUMPTIONS

• Looking at the scatter plot, we see that there is a uniform dispersion Thus, the assumption of constant variance of the regression model is not violated.

• In addition, the Durbin - Watson test (d) gives the result d=1.870 (1<d<3), so it can be concluded that the residuals are independent of each other or there is no correlation between the residuals.

• Through the above test results, the assumptions of the linear regression function are not violated and the built regression model is consistent with the overall.

Assumption 5: SATISFIED