4.2.4. Coefficient “Cronbach's Alpha Cost factor scale”
The observed cost variables include 4 variables and the results (Appendix 5) are as follows:
The risk factor group has a "Cronbach's Alpha coefficient" of 0.810, showing a high level of scale reliability, this coefficient is statistically significant.
The correlation coefficient of the total variable "Corrected Item - Total Correlation" of the observed variables measuring the "cost factor" component are all >0.3 (greater than the allowed standard of 0.3). So we will not remove the observed variable of this component.
4.2.5. Coefficient “Cronbach's Alpha Usage decision scale”
Observed variables of factors determining use and spss results (Appendix 5)
The risk factor group has a "Cronbach's Alpha coefficient" of 0.859, showing a high level of scale reliability, this coefficient is statistically significant.
The correlation coefficient of the total variable "Corrected Item - Total Correlation" of the observed variables measuring the "cost factor" component are all >0.3 (greater than the allowed standard of 0.3). So we will not remove the observed variable of this component.
4.3. EFA exploratory factor analysis
After evaluating and testing the reliability of the scale, the "Cronbach's Alpha" coefficient of all variable groups is greater than 0.6, and the "total variable correlation coefficient" is greater than 0.3, meeting the requirements. bridge. The author continues to research and perform EFA analysis on the scales. EFA analysis is to determine which factors influence the decision to use internet banking. Customers may be interested in saving more time, higher work efficiency or being able to make transactions at any time. anytime, anywhere,... factor analysis using the "Principal Component Analich" factor extraction method with "Varimax rotation with Kaiser Normalization", the results are as follows:
Table 4.4: Independent variable KMO results
Factors to evaluate
Table run value | Compare | |
KMO coefficient | 0.868 | 0.5 < α < 1 |
Maybe you are interested!
-
Cronbach'S Alpha of the Tourist Scenery Factor Scale Table 4.1: Reliability Assessment of the Tourist Scenery Scale -
Cronbach's Alpha Reliability Coefficient of Components of the Scale of Factors Affecting Investment Capital Attraction for Tourism in Ba Ria Vung Tau Province -
Summary of Results of Cronbach'S Alpha Coefficient of Impact Factors Scale -
Cronbach'S Alpha Coefficient Test Results of the Ability Scale -
Testing the Reliability of the Scale Using Cronbach'S Alpha Coefficient
Value in testing
Bartlett
0.000 | <0.05 | |
Variance extract | 58.08% | 58.08%>50% |
Eigenvalues | 1.599% | 1.599%>1 |
Source: Results calculated by SPSS data
Table 4.5: Results of “exploratory factor analysis”
Variable
Factor | ||||
1 | 2 | 3 | 4 | |
HI3 | 0.776 | |||
HI4 | 0.754 | |||
HI6 | 0.717 | |||
HI2 | 0.707 | |||
HI5 | 0.695 | |||
HI1 | 0.680 | |||
DSD6 | 0.735 | |||
DSD3 | 0.720 | |||
DSD1 | 0.714 | |||
DSD4 | 0.704 | |||
DSD2 | 0.696 | |||
CP3 | 0.793 | |||
CP4 | 0.788 | |||
CP2 | 0.768 | |||
CP1 | 0.734 | |||
RR1 | 0.795 | |||
RR2
0.709 | ||||
RR4 | 0.708 | |||
RR3 | 0.689 |
Source: Results calculated by SPSS data
The results show that the observed variables have a loading factor greater than 0.5. There are 4 factors representing investment attraction with observed variables.
Factor 1: includes observed variables: HI3, HI4, HI6, HI2, HI5, HI1. Name this factor X 1, Representing the usefulness of Internet Banking service.
Factor 2: includes observed variables: DSD6,DSD3,DSD1,DSD4,DSD2 (removing variable DSD5 loads up 2 factors so remove and run again a second time). Name this factor X 2 , representing the factor of ease of using Internet Banking.
Factor 3: includes observed variables: CP3, CP4, CP2, CP1. Name this factor X 3 , representing the costs of using Internet Banking.
Factor 4: includes factors: RR1,RR2,RR4,RR3. Name this element X 4 representing the steps to perform Internet Banking transactions.
The values of the four factors X 1 , X 2 , X 3 , X 4 are automatically calculated by SPSS 20.0 software using the regression method of component observed variables.
Thus, the scale of reliability and tests of EFA exploratory analysis, recognize 4 factors representing the usefulness and ease of use of internet banking services, risks and costs when using internet banking. , through transaction operations. The number of observed variables for the factors is 19 variables.
4.3.1. “EFA factor analysis” for the dependent variable Table 4.6: KMO results table of the dependent variable
Factors to evaluate
Table run value | Compare | |
KMO coefficient | 0.300 | 0.3 ≤ α ≤1 |
Sig value in check
Definition of Bartlett
0.000 | <0.05 | |
Variance extract | 77,984 | 77.984%>50% |
Eigenvalues | 2,340 | 2,340%>1 |
Source: Results calculated by SPSS data
Table 4.7: Results of "EFA exploratory factor analysis of decision scale"
Variable
Factor | |
1 | |
QD3 | 0.888 |
QD1 | 0.886 |
QD2 | 0.876 |
Source: SPSS data calculation results The results of "exploratory analysis" for the decision scale have a value of sig=0.000 < 0.05 (the variables in the population have a relationship with each other), and the system KMO number = 0.300. Scale measuring customer decisions regarding usefulness, ease of use, risk limitation, and cost extracted into 1 factor from 3 observed variables, with loading factor
The factors of the three variables are quite high (all greater than 0.8).
4.4. Test the model and research hypotheses
4.4.1. Correlation analysis
Before conducting multiple linear regression analysis, we must consider the correlation between variables
Table 4.8. Results of correlation analysis
Correlate
Y ( decided | F 1 ( The useful) | F 2 (the easy to use | F 3 (cost) | F 4 (risk) |
determined) | use) | |||||
Y ( decision) | Pearson Correlation | 1 | 0.692 ** | 0.681 ** | 0.462 ** | 0.453 ** |
Sig. (2- tailed) | 0.000 | 0.000 | 0.000 | 0.000 | ||
F 1 ( The useful) | Pearson Correlation | .692 ** | 1 | .317 ** | .320 ** | .396 ** |
Sig. (2- tailed) | .000 | .000 | .000 | .000 | ||
F 2 (ease of use) | Pearson Correlation | .681 ** | .317 ** | 1 | .323 ** | .304 ** |
Sig. (2- tailed) | .000 | .000 | .000 | .000 | ||
F 3 (cost) | Pearson Correlation | .462 ** | .320 ** | .323 ** | 1 | .304 ** |
Sig. (2- tailed) | .000 | .000 | .000 | .000 | ||
F 4 (risk) | Pearson Correlation | .453 ** | .396 ** | .304 ** | .304 ** | 1 |
Sig. (2- tailed) | .000 | .000 | .000 | .000 | ||
**. Correlation is significant at the 0.01 level (2-tailed). | ||||||
Source: SPSS data calculation results Looking at table 4.8, we see that the independent and dependent variables are correlated with each other. In which the independent factor "usefulness" (0.692, p < 0.05) correlates most strongly with the dependent factor "decision" and "Decision" correlates weakest with the independent factor " risk” (0.453, p < 0.05).
The above analysis results show that the values are symmetrical across the diagonal and the
The value located on the main diagonal reaches the value 1 (satisfying the condition -1 ≤ r ≤+1) (Hoang Trong and Chu Nguyen Mong Ngoc, 2008). This matrix shows that there is a correlation between the variables "decision" (dependent variable) with the independent variables ease of use, usefulness, risk, analysis does not find a correlation between the variable " cost” and “decision” (Sig value 0.692>0.05). However, an r value indicating there is no linear relationship does not necessarily mean that the two variables are truly unrelated. In addition, there is no correlation between the independent variables, so it can be concluded that there is no multicollinearity phenomenon between the independent variables.
4.4.2. Regression analysis
The research model and hypotheses are tested using regression methods. The "regression analysis" method aims to determine the important role of each factor in customer evaluation when using Internet Banking services and the service evaluation components ("Useful", "Easy to use". Use”, “Cost”, “Risk”). Tests are applied through the adjusted R 2 determination number and the F test
4.4.2.1 Check the suitability of the overall model
The results of regression analysis show that the regression model has coefficients R 2 and adjusted R 2 in table 4.9. We see that the adjusted R 2 is smaller, using it to evaluate the suitability of the model will be safer because it does not inflate the fit level of the model (Hoang Trong & Chu Nguyen Mong Ngoc, 2008).
The meaning of adjusted R 2 : for the variable the independent explanatory variables are included
percentage (%) of the variation of the dependent variable.
In the study, we have adjusted R2 = 0.741, which means that 4 independent variables: DSD, HI, RR and CP can explain 74.1% of the change in Decision (Decision to use Internet Banking); The remaining 25.9% is due to the influence of variables outside the model that the project has not found and due to random errors.
Table 4.9: Model summary table
Model Summary
Model | R | R 2 | R 2 brand adjust | Standard error estimate |
1 | 0.861 a | 0.741 | 0.737 | 0.37580 |
Source: SPSS analysis results
The ANOVA analysis table of the regression model shows that the regression model has a test of F= 200.370, Sig.= 0.000 < 0.05, showing the overall suitability of the regression model.
Table 4.10: ANOVA analysis table
Model
Sum of squares | Degrees of freedom Df | Square medium | F-statistic value | Significance level Sig | |
Regression | 113,188 | 4 | 28,297 | 200,370 | 0.000 b |
Remaining | 39,543 | 280 | 0.141 | ||
Total | 152,731 | 284 |
Source: analysis results from SPSS
4.4.2.2. Testing research hypotheses
According to table 4.11, testing the research hypotheses shows that the four factors ease of use, usefulness, risk, and cost have an impact on the decision to use internet banking services.
Table 4.11. Summary table of regression results
Model
Unstandardized coefficients | Coefficient standardization | T | Significance level | Collinearity statistics | ||
B | Deviation standard | Beta | Degree accept | Launch coefficient grand | ||
Std Error | of variables | variance VIF | |||||
Constant | -.198 | .131 | -1,504 | .134 | |||
Existence useful | .402 | .030 | .468 | 13.55 1 | .000 | .774 | 1,292 |
Easy to use use | .388 | .028 | .463 | 13.80 4 | .000 | .823 | 1,216 |
Risk | .077 | .031 | .085 | 2,478 | .014 | .787 | 1,271 |
Expense | .107 | .026 | .137 | 4,071 | .000 | .821 | 1,218 |
Source: SPSS analysis results
Results from the linear regression model show that there are 4 independent factors with positive standardized Beta coefficients, which have an impact on individual customers' decision to use internet banking services.
The results of the regression model show that usefulness is the component with the highest standardized regression coefficient (coefficient β = 0.468; sig.=0.000), meaning that the component has the greatest impact on the decision. use internet banking services and this is a positive impact, so hypothesis H2: usefulness has a great influence on the decision is accepted. The usefulness of internet banking services creates more customer interest in the decision to use, helping customers do more work, helping customers save transaction time and waiting for their turn to transact. at the counter, helping customers make transactions anywhere...
The second strongest factor is “ease of use” with a standardized beta coefficient
=0.463; sig.=0.000, quality is the ease of use of the product that attracts customers with simple IB operations that are generally easy to use and convenient, so hypothesis H1 is accepted.
The third strongest impact factor on cost also has an impact on the decision with standardized beta coefficient = 0.137; Sig.= 0.000 and hypothesis H4 is accepted