The results in the table above show that all components have Cronbach's alpha values greater than 0.6 and the total correlation value of the observed variables is greater than 0.3. Therefore, no variables are eliminated and can be used for further testing.
2.2.3.2 Cronbac's alpha reliability test for dependent variables
Table 2.8 Measurement of Cronbach's alpha coefficient of dependent variables
Variable
Average scale if variable type | Scale variance if variable type | Variable correlation total | Cronbach's alpha coefficient if variable is excluded | |
Customer attraction: 0.855 | ||||
THKH1 | 11,9933 | 2,544 | 0.710 | 0.810 |
THKH2 | 12,0000 | 2,685 | 0.610 | 0.850 |
THKH3 | 11,9333 | 2,734 | 0.660 | 0.831 |
THKH4 | 12,0333 | 2,422 | 0.813 | 0.766 |
Maybe you are interested!
-
Summary of Cronbach'S Alpha Reliability Coefficients of the Scales -
“Cronbach's Alpha Cost Factor Scale” -
Results of Testing Cronbach's Alpha Coefficient of Independent Variable -
Cronbach's Alpha Reliability Coefficient of Components of the Scale of Factors Affecting Investment Capital Attraction for Tourism in Ba Ria Vung Tau Province -
Cronbach'S Alpha of the Tourist Scenery Factor Scale Table 4.1: Reliability Assessment of the Tourist Scenery Scale
(Source: SPSS 20 data processing results by the author)
The Cronbach's Alpha coefficient scale reached 0.855 and the total correlation of the component variables was greater than 0.3. Therefore, the measurement variables in the customer attraction scale after being assessed for Cronbach's Alpha reliability were all used in the next EFA analysis.
2.2.4 Exploratory factor analysis EFA
Cronbach's alpha reliability test has evaluated the relationship between variables in the same group but has not evaluated the relationship between observed variables. Therefore, it is necessary to conduct exploratory factor analysis (EFA) to extract factors affecting the increase in the number of students at Hong Duc Training and Consulting Center and check the observed variables loaded on multiple factors or the observed variables misclassified by factors.
Exploratory factor analysis is used to shorten and summarize research into concepts. Theoretically, the variables measured by the questions in the interview questionnaire are correlated with each other and therefore they are shortened for easy management. Through factor analysis, we aim to identify and find factors representing the observed variables.
Exploratory factor analysis based on criteria and reliability.
According to Nguyen Khanh Duy (2007), when analyzing exploratory factors, researchers often pay attention to the following criteria:
- KMO coefficient (Kaiser - Meyer - Olkin) ≥ 0.5; significance level of Bartlett test ≤ 0.05
- Factor Loading ≥ 0.5
- Cumulative % Extraction Sum of Squared
loading) ≥ 50%
- Eigenvalue coefficient ≥ 1
- The difference in factor loading coefficients of an observed variable between factors is ≥ 0.3 to ensure discriminant value between factors.
- KMO & Bartlett's test was proposed by Kaiser in 2001 to evaluate the reasonableness of the database, used for factor analysis. The test allows to know whether the database is suitable for factor analysis or not. Kaiser (2001) said that the value of KMO test should be in the range of 0.5 - 1 to be appropriate.
2.2.4.1 EFA factor analysis for independent variables
Table 2.9 Results of KMO and Barlett's test of independent variables KMO and Barlett's Test
KMO Index
0.822 | ||
Bartlett's Test | Chi-square statistics | 2346,443 |
Degrees of freedom (Df) | 253 | |
Significance level (Sig.) | 0.000 |
(Source: SPSS 20 data processing results by the author)
The results of KMO and Bartlett's tests with KMO = 0.822 indicate that factor analysis is appropriate. The Sig. value of Bartlett's test = 0.000 < 0.05 shows that the observed variables are correlated with each other in the population. Thus, the factor analysis data is completely appropriate.
Table 2.10 Results after factor rotation
Factor | |||||
1 | 2 | 3 | 4 | 5 | |
NLPV4 | 0.916 | ||||
NLPV2 | 0.904 | ||||
NLPV5 | 0.895 | ||||
NLPV1 | 0.883 | ||||
NLPV3 | 0.817 | ||||
CTDT2 | 0.824 | ||||
CTDT1 | 0.822 | ||||
CTDT3 | 0.814 | ||||
CTDT4 | 0.801 | ||||
CTDT5 | 0.730 | ||||
CSVC3 | 0.839 | ||||
CSVC5 | 0.811 | ||||
CSVC1 | 0.766 | ||||
CSVC4 | 0.746 | ||||
CSVC2 | 0.727 | ||||
GV3 | 0.888 | ||||
GV4 | 0.847 | ||||
GV2 | 0.827 | ||||
GV1 | 0.786 | ||||
HP3 | 0.820 | ||||
HP4 | 0.820 | ||||
HP2 | 0.783 | ||||
HP1 | 0.743 | ||||
( Source: SPSSS 20 data processing results by the author)
Conducting factor analysis for the first time, putting 23 observed variables in 5 independent variables affecting the increase in the number of students of the center into factor analysis according to the Eigenvalue standard greater than 1, 5 factors were created.
Thus, after conducting the EFA, the number of observed variables is still 23, reduced to 5 factors. No observed variable has a factor loading coefficient (Factor Loading) less than 0.5, so the variable is not eliminated, the research topic continues to conduct the next analysis.
The result obtained is total variance extracted = 72.341% ≥ 50%, indicating that these 4 groups of factors explain 72.341% of the variation in the data, the Eigenvalue is greater than 1, indicating that the EFA model is suitable.
First group of factors
Service capacity (NLPV) has Eigenvalue = 6.526 > 1, these are factors related to the capacity and responsiveness of consultants, ...
Including: NLPV4 (Students' academic issues are resolved promptly), NLPV2 (Time from receipt to resolution of issues is quick, saving time), NLPV5 (Students receive timely notifications from the center), NLPV1 (Staff answer your questions), NLPV3 (Staff have an enthusiastic attitude when resolving your problems or questions).
The service capacity factor explains 28.373% of the variation in the survey data. Among the variables on “NLPV”, the observed variable: “Students’ academic problems are resolved promptly” is rated best by many students with a loading factor of 0.916.
Second group of factors
Training program (CTDT) has Eigenvalue = 3.529> 1, this is a group of factors related to training objectives, training program output, training plan, ...
Including: CTDT2 (Training program with clear output standards), CTDT1 (Course curriculum framework is announced in detail to students).
CTDT3 (Training program is renewed to suit each year), CTDT4 (Teaching curriculum is suitable for each course, closely follows, is specific and easy to understand), CTDT5 (Training program is updated regularly)
The training program factor explains 15.346% of the variation in the survey data. Among the variables on “CTDT”, the observed variable: “The training program has clear output standards” is rated best by many students with a factor loading coefficient of 0.824.
Third group of factors
Facilities (CSVC), with Eigenvalue = 2.727> 1, this group of factors are factors related to classrooms, teaching equipment, learning materials, ...
Including: CSVC3 (Teaching materials, learning materials are fully equipped in a timely manner), CSVC5 (The center has spacious and secure parking), CSVC1 (Convenient and easy-to-find customer reception and consultation area), CSVC4 (Classrooms are fully equipped with teaching and learning support facilities), CSVC2 (Comfortable, clean classrooms, suitable number of students)
The physical facilities factor explains 11.857% of the variation in the survey data. Among the variables on “physical facilities”, the observed variable: “Teaching materials and learning materials are fully and promptly provided” has the largest factor loading coefficient of 0.839.
Fourth group of factors
The teaching staff (GV) has an Eigenvalue = 2.199> 1, these are factors related to the knowledge, experience, and teaching of the lecturers to the students.
Including: GV3 (Lecturer always answers students' questions), GV4 (Lecturer has a friendly attitude, always shares experiences and practical knowledge with students), GV2 (Lecturer conveys easy-to-understand and appropriate content), GV1 (Lecturer is an experienced and specialized expert.)
The faculty factor accounted for 9.563% of the variation in the survey data. Among the variables on “Lecturers”, the observation variable “Lecturers always answer students’ questions” had the largest loading factor of 0.888.
Fifth group of factors
Tuition policy (HP) has Eigenvalue = 1.657> 1, these are the factors
related to tuition, scholarships, promotions.
Including: HP3 (Discount policy, incentives for students who continuously study at the center), HP4 (Scholarship awards for students with high exam results), HP2 (Reasonable pricing method), HP1 (Price suitable for each course)
Tuition policy factor accounts for 7.203% of the variation in survey data. Among the variables on “HP”, the observed variable “Discount and preferential policies for students studying continuously at the center” has the largest loading factor of 0.820.
2.2.4.2 EFA factor analysis for dependent variables
Table 2.11 KMO value of observed variables KMO and Bartlett's Test
KMO Index
0.662 | ||
Bartlett's Test | Chi-square statistics | 325,319 |
Degrees of freedom (Df) | 6 | |
Significance level (Sig.) | 0.000 |
(Source: SPSS20 data processing results by the author)
The test results give us KMO coefficient = 0.662 and the Barlett's Test results also show that Sig. = 0.000 has rejected the hypothesis that the variables are not correlated with each other, so factor analysis is appropriate.
Table 2.12 Results of customer attraction analysis
Observation variable
Load factor | |
Overall, how do you see the ability to attract customers at the Center? Hong Duc Training and Consulting is good (THKH4) | 0.912 |
Hong Duc Training and Consulting Center is a reliable place to study (TKH1) | 0.849 |
Are you willing to recommend your course to your acquaintances when they have a need? (THKH3) | 0.809 |
Hong Duc Training and Consulting Center always does the right thing. what has been committed. (THK2) | 0.771 |
Eigenvalues = 2.8 | |
Extracted variance: 70,004 | |
(Source: SPSS20 data processing results by the author)
The study obtained the results Eigenvalues = 2.8 > 1 and total variance extracted = 70.004% showing that the conditions of factor analysis are suitable.
Comment:
The EFA exploratory factor analysis process identified 5 factors affecting the ability to attract customers at Hong Duc Training and Consulting Center, which are "Training program" , "Lecturer team" , "Service capacity", "Tuition policy", "Facilities" . Thus, after studying the EFA exploratory factor analysis, it showed that there was no change compared to the beginning and no observed variables were removed from the model during the reliability testing and exploratory factor analysis.
2.2.5 Regression analysis
2.2.5.1 Testing the correlation between independent variables and dependent variables Table 2.13 Pearson correlation analysis
THKH | CTDT | GV | NLPV | HP | Facilities | |
Pearson correlation | 1 | 0.474 xx | 0.405 xx | 0.376 xx | 0.510 xx | 0.492 xx |
Sig.(2-tailed) | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | |
N | 150 | 150 | 150 | 150 | 150 | 150 |
(Source: SPSS 20 processing results by the author)
Based on the above analysis results, we see:
- The Sig.(2-tailed) values of the new factors are all smaller with a significance level of α = 0.05, showing that there is a linear relationship between the independent variable and the dependent variable. There is the strongest correlation between THKH and HP with a coefficient r of 0.510.
- Pearson correlation coefficient is also quite high (5 factors are greater than 0) NLPV, CTDT, CSVC, GV, HP so we can explain that the independent variables can explain the dependent variable "attracting customers".
2.2.5.2 Building a regression model
After conducting exploratory factor analysis (EFA) to discover new factors affecting the dependent variable "Customer attraction ability" , the study conducted linear regression to determine the level of influence of new factors on students' satisfaction with training quality.
The regression model is built with the dependent variable being “ customer attraction ability” – THKH and the independent variables extracted from the EFA factor including 5 variables: "Training program" - CTDT , "Lecturer team" - GV "Service capacity" - NLPV , "Tuition policy" - HP , "Facilities" - CSVC with corresponding coefficients β 1, β 2, β 3, β 4, β 5.
THKH= α + β 1 CTDT + β 2 GV + β 3 NLPV + β 4 HP + β 5 CSVC + ei