Model Sum of Squares df Mean Square F Sig.
1 Regression 46.213 10 4.621 76.592 .000b
Residual 23.652 392 .060
Total 69.865 402
a. Dependent Variable: Decision
b. Predictors: (Constant), Marketing, Program, Fee, Recommendation, Location, Facility, Connection, Brand, Quality, Teacher
Source: Compiled by the author Analysis of variance (ANOVA) is a statistical method for analyzing the total scale of variation of the dependent variable (the sum of which the total scale of variation is defined as the sum of squared deviations from the number (average of it) into parts and each is attributed to the variation of a particular explanatory
variable or a group of explanatory variables. The rest cannot be attributed to any variable called unexplained variation or residual. This method is used to test hypothesis H0 to determine whether the samples were drawn from the same population. The test results tell us whether the samples obtained are correlated or not.
With ANOVA test, H0: "Average is equal"
If Sig <= 0.05: reject H0, it is eligible to confirm there is a difference between the groups for the dependent variable. In contrast, if Sig> 0.05: accepting H0, it is not eligible to confirm the difference between groups for the dependent variable. In this case, Sig = 0.00 <0.05, so the regression model makes sense.
Table 4-49 Model Summaryb
Model R R Square
Adjusted R Square
Std. Error of the Estimate
Durbin- Watson
1 .813a .661 .653 .24564 1.954
a. Predictors: (Constant), Marketing, Program, Fee, Recommendation, Location, Facility, Connection, Brand, Quality, Teacher
b. Dependent Variable: Decision
Source: Compiled by the author R squared adjusted is 0.653 = 65.3%. Thus, the independent variables in the regression affect 65.3% of the change of the dependent variables.
Table 4-50 Coefficientsa
Model
Unstandardize d Coefficients
Standardized
Coefficients t Sig.
Collinearity Statistics
B
Std.
Error Beta
Toleranc
e VIF
1 (Constant) -.15 4
.155 -.993 .32
1
Program .227 .019 .365 11.80
4
.00 0
.901 1.109
Fee .048 .020 .077 2.438 .01
5
.866 1.154
Location .165 .019 .268 8.548 .00
0
.876 1.142
Brand .027 .019 .044 1.376 .16
9
.851 1.176
Facility .088 .025 .111 3.491 .00
1
.861 1.161
Connection .010 .022 .015 .463 .64
4
.820 1.220
Recommendatio n
.059 .019 .095 3.145 .00
2
.950 1.052
Teacher .161 .027 .199 6.003 .00
0
.784 1.275
Quality .069 .021 .106 3.277 .00
1
.818 1.223
Marketing .154 .019 .275 8.290 .00
0
.784 1.275
a. Dependent Variable: Decision
Source: Compiled by the author
Testing the phenomenon of multi-collinearity:
o VIF coefficients for independent variables are less than 10, so no collinearity occurs.
Testing variance independence
o According to Nguyen Dinh Tho (2011), the Durbin-Watson value is used to test the variance independence, this value usually fluctuates in the range of 1-3, or adjacent 2, this value is accepted. According to the analysis results, the Durbin-Watson value is 1.954, so the variance independence is not violated.
Regression results:
o The regression results show that the Connection, Brand variables do not make sense in the model because the test sig t is greater than 0.05.
The remaining variables all have an effect on the dependent variable since the t-test tig of each independent variable is less than 0.05.
o According to the regression equation on the weights of the factors affecting the decision on choosing an English center of students in Ho Chi Minh City are arranged from the strongest one to the weakest one as follows:
Training program: 0.227
Location: 0.165
Teacher: 0.161
Marketing: 0.154
Facilities: 0.088
Training quality: 0.069
Recommendation: 0.059
Tuition fee: 0.048
To determine the importance of each variable to the dependent variable in the comparative relationship between the independent variables, we use the standardized regression coefficient (Beta). Training Program is the most important factor because there is a standardized Beta coefficient of 0.365; Marketing was the
second most important due to the standardized Beta coefficient of 0.275; The third most important Location due to the standardized Beta coefficient is 0.268. We have no factor which has negative standardized Beta.
The standardized regression equation:
o Decision = - 0.154 + 0.227 * Program + 0.165 * Location + 0.161 * Teacher + 0.154 * Marketing + 0.088 * Facilities + 0.069 * Quality + 0.059 * Recommendation + 0.048 * Fee
The non-standard regression equation shows the change of the dependent variable when an independent variable changes and the remaining independent variables remain the same.
Standardized regression equation:
o Decision = 0.365 * Program + 0.268 * Location + 0.199 * Teacher + 0.275 * Marketing + 0.111 * Facilities + 0.069 * Quality + 0.095 * Recommendation + 0.106 * Fee
In addition, all beta coefficients of the independent variables in the regression equation were positive, showing a positive correlation between these variables and the dependent variables. Therefore, the hypotheses the author proposed initially are accepted.
Table 4-51 Conclusion of the Hypothesis examination
Hypotheses Result
Hypothesis H1 Accepted
Hypothesis H2 Accepted
Hypothesis H3 Accepted
Hypothesis H4 Accepted
Hypothesis H5 Accepted
Hypothesis H6 Accepted
Hypothesis H7 Accepted
Hypothesis H8 Accepted
Hypothesis H9 Not Accepted
Hypothesis H10 Not Accepted
Source: Compiled by the author
Summary of chapter 4:
This chapter presents all research results.
First, the data has been cleaned before processing and results in statistical inference. The description of the research subjects is done on demographic variables (universities, grades, genders).
The determination of reliability and value of the scale using Cronbach's Alpha coefficient and EFA factor analysis has confirmed 36 observed in 10 factors.
The regression using Enter method gives the results that remove 2 elements and determine the intensity of the factors that influence the decision on choosing an English center of students in Ho Chi Minh City as well as.
The next chapter will discuss the research results, conclusions and give some implications to help centers develop to better meet the needs of learners to increase the competitiveness of the foreign language center.