Home / Credit Card

Scale of Factors Affecting Credit Card Service Quality at Saigon Commercial Joint Stock Bank



Spending

Under 2 million

23

9.3

From 2 to under 5 million

83

33.5

From 5 to under 8 million

96

38.7

From 8 million and up

46

18.5

Maybe you are interested!

Scale of Factors Affecting Credit Card Service Quality at Saigon Commercial Joint Stock Bank

Source: SPSS processing results

4.2.2. Analysis of Cronbach's alpha reliability coefficient

Cronbach's alpha reliability coefficient is a statistical test coefficient of the degree of tightness and correlation between observed variables in the scale. The purpose of this step is to test whether the observed variables have the same explanation for a concept that needs to be measured. This method allows to eliminate inappropriate variables, limiting garbage variables in the research process.

Accordingly, observed variables with a Corrected Item Total Correction coefficient greater than 0.3 and a Cronbach's alpha reliability coefficient greater than 0.6 are considered acceptable and suitable for analysis in the next steps (Nunnally, 1978; Peterson, 1994; Slater, 1995). Normally, scales with a Cronbach's alpha coefficient of 0.7 to 0.8 are usable. However, many researchers believe that if the Cronbach's alpha coefficient is greater than 0.95, it will show that many variables in the scale overlap, meaning that there are variables measuring the same content. Accordingly, the results of the analysis of the Cronbach's alpha reliability coefficient of the research model are as follows:

Table 4.2: Cronbach's alpha reliability coefficient of factors


Observation variable

Scale mean if variable excluded

Scale variance if variable is excluded

Total variable correlation

Cronbach's alpha if variable is excluded

Facility scale: Alpha = 0.817

VC1

14.7298

7,866

.716

.752

VC2

14.8185

7,809

.635

.773


VC3

14.8266

7,836

.671

.763

VC4

14.7379

8,000

.546

.801

VC5

14.8710

8,186

.497

.816

Reliability scale: Alpha = 0.873

TC1

13.8669

11,735

.785

.826

TC2

14.0766

12,484

.619

.865

TC3

13.9234

12,168

.703

.845

TC4

13.9194

10,989

.809

.818

TC5

14.1331

12,027

.603

.872

Responsiveness scale: Alpha = 0.826

DU1

15.0081

10,955

.549

.822

DU2

14.9032

12,250

.578

.803

DU3

14.8065

11,371

.743

.759

DU4

14.7460

12,312

.575

.804

DU5

14.7621

11,316

.705

.767

Service Competency Scale: Alpha = 0.726

PV1

6.9879

1,866

.567

.615

PV2

7.0081

1,854

.555

.631

PV3

6.9315

2,145

.526

.666

Empathy scale: Alpha = 0.767

DC1

10.9234

3,593

.591

.700

DC2

10.7379

4.105

.494

.750

DC3

10.8750

3,843

.681

.657

DC4

10.9032

4,007

.521

.736

Credit card service quality scale: Alpha = 0.730

CL1

7.3548

1,623

.631

.555


CL2

7.3266

1,832

.496

.706

CL3

7.1734

1,415

.550

.660

Source: SPSS processing results

According to table 4.2 we have the following comments:

• Facility scale: Cronbach's alpha coefficient = 0.817 greater than 0.6; total item correlation coefficient of all measured variables meets the standard of greater than 0.3. Conclusion: Facility scale meets the requirements and variables VC1, VC2, VC3, VC4, VC5 continue to be analyzed in the next steps.

• Reliability scale: Cronbach's alpha coefficient = 0.873, greater than 0.6 and the total correlation coefficient of the measured variables all meet the standard of greater than 0.3. Conclusion: The reliability scale meets the requirements and the variables TC1, TC2, TC3, TC4, TC5 continue to be analyzed in the next steps.

• Responsiveness scale: Cronbach's alpha coefficient = 0.826, greater than 0.6 and the total correlation coefficient of the measured variables all meet the standard of greater than 0.3. Conclusion: The responsiveness scale meets the requirements and the variables DU1, DU2, DU3, DU4, DU5 continue to be analyzed in the next steps.

• Service capacity scale: Cronbach's alpha coefficient = 0.726, greater than 0.6 and the total correlation coefficient of the measured variables all meet the standard of greater than 0.3. Conclusion: The service capacity scale meets the requirements and the variables PV1, PV2, PV3 continue to be analyzed in the next steps.

• Empathy scale: Cronbach's alpha coefficient = 0.767, greater than 0.6 and the total correlation coefficient of the measured variables all meet the standard of greater than 0.3. Conclusion: The empathy scale meets the requirements and the variables DC1, DC2, DC3, DC4 continue to be included in the analysis in the next steps.

• Credit card service quality scale: Cronbach's alpha coefficient = 0.730, greater than 0.6 and the total item correlation coefficient of the measured variables all meet the standard of greater than 0.3. Conclusion: The credit card service quality scale meets the requirements and the variables CL1, CL2, CL3 continue to be analyzed in the next steps.


4.2.3. Exploratory factor analysis EFA

4.2.3.1. Scale of factors affecting credit card service quality at Saigon Commercial Joint Stock Bank

Exploratory Factor Analysis - EFA is a technique used to reduce a set of interdependent observed variables into a smaller set of variables that still contains most of the content of the original set of variables (Hair & ctg (1998,111), Multivariate Data Analysis, Prentice-Hall International). This method helps evaluate the two factors of convergent validity and discriminant validity of the scale. Accordingly, the conditions in EFA factor analysis must satisfy the following factors:

- Factor loading:

Is the coefficient used to ensure the practical significance of factor analysis. In which, we evaluate 2 values ​​of the scale: convergent value and discriminant value. Accordingly, the appropriate Factor Loading coefficient must be greater than or equal to 0.5. According to Hair & ctg (1998,111), a Factor Loading coefficient greater than 0.3 is considered the minimum level, a Factor Loading coefficient greater than 0.4 is considered important and a Factor Loading coefficient greater than or equal to 0.5 is considered to have practical significance.

- KMO coefficient (Kaiser Meyer Olkin):

As a coefficient used to examine the suitability of EFA factor analysis with research data, KMO must have a value between 0.5 and 1 for the analysis to be considered appropriate (Hoang Trong & Chu Nguyen Mong Ngoc, Analyzing research data with SPSS, Volume 2, page 31, 2008, Hong Duc Publishing House).

- Bartlett quantity:

Is a statistical quantity used to examine the hypothesis of correlation between observed variables. With the Ho hypothesis, the correlation between observed variables in the population is 0. If the Sig coefficient is less than or equal to 0.05, we reject the Ho hypothesis and the test is statistically significant and we can use the EFA analysis results for the following studies (Hoang Trong & Chu Nguyen Mong Ngoc, Analyzing research data with SPSS, Volume 2, page 30, 2008, Hong Duc Publishing House).


- Total variance extracted:

It is a coefficient indicating how much of the variation in the data each factor explains. A total variance extracted greater than 50% is considered satisfactory.

- Eigenvalue convergence:

The coefficient represents the amount of variation explained by each factor, and must be greater than or equal to 1.

Accordingly, the scale of factors affecting the quality of credit card services at SCB consists of 5 factors with 22 observed variables. After the scale was tested by Cronbach's alpha coefficient, it showed that no variables were eliminated. Therefore, the observed variables that achieved reliability were included in the EFA factor analysis. The method of factor analysis is performed as follows:

- 1st time: A set of 22 observed variables after meeting the reliability test criteria were included in factor analysis:

Table 4.3: Results of factor analysis of factors affecting credit card service quality at SCB, first time

STT

Parameter

Value

Satisfy the conditions

1

KMO

0.839

≥ 0.5

2

Sig. of Bartlett's Test

0.000

≤ 0.05

3

Eigenvalues

1,518

> 1

4

Total extracted variance

64.009%

≥ 50%

Source: SPSS processing results Using Varimax rotation matrix method shows that 22 observed variables are grouped

into 4 groups as follows


Table 4.4: Rotated matrix of factors affecting credit card service quality at SCB, first time




Factor



Variable






1

2

3

4

5

TC4

.900


TC1

.894

TC3

.714

TC5

.708

TC2

.658

DU3


.792


DU5


.792

DU1


.716

DU4


.674

DU2


.651

VC1



.847


VC3


.760

VC4


.745

VC2


.695

VC5


.450

.482


DC3




.840


DC1




.724

DC4




.689

DC2




.689

PV3

.791


PV1

.782

PV2

.741

Source: SPSS processing results

According to tables 4.3 and 4.4 we have the following observations:

KMO coefficient reached 0.839, so factor analysis is suitable for research data.

Bartlett's test with significance level Sig = 0.000, less than 5%, shows that this test is statistically significant and the observed variables are correlated with each other.

The factor analysis results also show that the total variance extracted is 64.009%, greater than 50%. This means that 1 factor explains 64.009% of the variation in the data.

The Eigenvalue is 1.518, greater than 1. Therefore, the factor analysis result is appropriate.

The observed variables VC1, VC2, VC3 and VC4 have factor loading coefficients that meet the requirement of greater than 0.5. Variable VC5 has a coefficient of 0.482, less than 0.5, not meeting the requirement. Therefore, the second factor analysis was performed by removing this variable.

- 2nd time: A set of 21 observed variables after the first EFA factor analysis and removing variable VC5 were included in the second analysis. The results are as follows:

Table 4.5: Results of factor analysis of factors affecting credit card service quality at SCB, second time

STT

Parameter

Value

Satisfy the condition

1

KMO

0.827

≥ 0.5

2

Sig. of Bartlett's Test

0.000

≤ 0.05

3

Eigenvalues

1,517

> 1

4

Total extracted variance

64.958%

≥ 50%

Source: SPSS processing results


Through the Varimax rotation matrix method, 21 observed variables are grouped into 4 groups and named specifically as follows:

Table 4.6: Rotated matrix of factors affecting credit card service quality at SCB, second time


Variable



Factor


Element

1

2

3

4

5

TC4

.901


Trust Symbol TC

TC1

.895

TC3

.718

TC5

.709

TC2

.661

DU5


.798


DU Symbol Response

DU3


.790

DU1


.709

DU4


.687

DU2


.662

VC1



.856


Facilities Symbol VC

VC4



.770

VC3



.750

VC2



.684

DC3




.845


Empathy DC Symbol

DC1




.726

DC4




.689

DC2




.687

Comment


Agree Privacy Policy *