D. Service Provider Structure Customer Survey Sample Card


The under 25 age group accounts for nearly 50%, which is consistent with the fact that young people are more likely to adopt new services, especially those that involve modern technology. The rest is mainly people aged 44 and under (about 40%).

Table 4.1c. Customer survey sample position structure



Frequency

Percent

Percent


Valid

salaried worker

101

44.3

44.3

Student

104

45.6

45.6

Other

23

10.1

10.1

Total

228

100.0

100.0

Maybe you are interested!

D. Service Provider Structure Customer Survey Sample Card

This structure is also consistent with the fact that the main subjects of debit card services today are students and salary earners (when businesses and socio-economic organizations are gradually shifting to paying salaries through accounts).

Table 4.1d. Structure of sample customer survey card service providers



Frequency

Percent

Percent


Valid

Vietinbank

46

20.2

20.2

ACB

5

2.2

2.2

BIDV

28

12.3

12.3

VCB

29

12.7

12.7

Agribank

77

33.8

33.8

Other

43

18.9

18.9

Total

228

100.0

100.0

It can be seen that this is a suitable structure and reflects the debit card usage structure of customers in Hanoi city area.

Table 4.1e. Time structure of customer survey sample card usage



Frequency

Percent

Percent


Valid

Under 1 year

29

12.7

12.7

1 to 2 years

41

18.0

18.0

2 to 3 years

37

16.2

16.2

3 years or more

121

53.1

53.1

Total

228

100.0

100.0


A large proportion of surveyed customers have used the card for 2 years or more (69.3%).

Thus, after collecting the collected questionnaires and excluding invalid questionnaires, 228 questionnaires remained valid for further analysis. In terms of gender, 40.8% were male, 59.2% were female; the age group under 25 accounted for nearly 50%, from 44 years old and below accounted for about 40%; 45.6% were students and 54.4% were salaried employees; 69.3% of surveyed customers had used the card for 2 years or more. The statistics received showed that the above sample structure was considered appropriate and could represent customers in Hanoi using debit cards of banks. This data allows analysis of the level of response to debit card services from the customer side.

4.1.2. Data processing results

4.1.2.1. Data evaluation

Preliminary statistical description for observed variables Descriptive statistics for TA variable group



N

Minimum

Maximum

Mean

Std. Deviation

Skewness

Kurtosis

TAE

228

1

5

3.88

.815

-.795

1,485

TAR

228

1

5

3.14

.684

-.071

2.020

TA1E

228

1

5

4.00

.771

-.399

.058

TA1R

228

1

5

3.18

.718

.178

1,137

TA2E

228

1

5

3.97

.765

-.420

.223

TA2R

228

1

5

3.23

.730

.043

1,253

TA3E

228

2

5

4.11

.668

-.409

.242

TA3R

228

1

5

3.62

.706

-.296

.571

TA4E

228

1

5

3.90

.821

-.394

-.094

TA4R

228

1

5

3.14

.712

.054

1,318

TA5E

228

1

5

4.08

.779

-.819

1,004

TA5R

228

1

5

3.20

.793

-.081

.572


Descriptive statistics of RL variable group




N


Minimum


Maximum


Mean

Std.

Deviation


Skewness


Kurtosis

RLE

228

2

5

4.21

.810

-.778

-.084

RLR

228

1

5

3.49

.735

-.234

.685

RL1E

228

1

5

4.11

.769

-.504

-.009

RL1R

228

1

5

3.33

.765

.186

.517

RL2E

228

1

5

4.04

.865

-.609

-.114

RL2R

228

1

5

3.21

.834

-.171

.767

RL3E

228

1

5

4.10

.790

-.885

1,348

RL3R

228

1

5

3.43

.758

-.045

.120

RL4E

228

2

5

4.11

.749

-.579

.077

RL4R

228

1

5

3.38

.717

-.076

.273

RL5E

228

1

5

4.03

.785

-.658

.575

RL5R

228

1

5

3.38

.798

-.184

.335


Descriptive statistics of RN variable group




N


Minimum


Maximum


Mean

Std.

Deviation


Skewness


Kurtosis

RNE

228

1

5

4.06

.822

-.591

.050

RNR

228

1

5

3.31

.720

.168

.848

RN1E

228

2

5

4.13

.766

-.419

-.656

RN1R

228

1

5

3.33

.834

-.327

.975

RN2E

228

1

5

4.14

.762

-.667

.516

RN2R

228

1

5

3.33

.790

-.003

.429

RN3E

228

1

5

3.95

.838

-.544

.044

RN3R

228

1

5

3.00

.829

.000

.983


Descriptive statistics of AS variable group


N

Minimum

Maximum

Mean

Std. Deviation

Skewness

Kurtosis

ASE

228

2

5

3.95

.801

-.482

-.116

ASR

228

1

5

3.23

.703

.105

.864

AS1E

228

1

5

3.98

.845

-.451

-.256

AS1R

228

1

5

3.20

.717

.208

.835

AS2E

228

2

5

4.05

.778

-.422

-.374

AS2R

228

1

5

3.35

.719

.077

.751

AS3E

228

2

5

4.09

.728

-.357

-.405

AS3R

228

1

5

3.35

.778

.032

.289

AS4E

228

1

5

4.11

.740

-.655

.777

AS4R

228

1

5

3.38

.692

.056

.498


Descriptive statistics of EM variable group


N

Minimum

Maximum

Mean

Std. Deviation

Skewness

Kurtosis

EME

228

1

5

3.87

.830

-.301

-.256

EMR

228

1

5

3.08

.765

.027

1,037

EM1E

228

1

5

3.77

.810

-.280

-.062

EM1R

228

1

5

3.03

.768

.001

1,037

EM2E

228

1

5

3.78

.834

-.367

.148

EM2R

228

1

5

2.97

.815

.069

.724

EM3E

228

1

5

3.75

.792

-.306

.356

EM3R

228

1

5

2.97

.790

.004

.961

EM4E

228

1

5

3.76

.824

-.230

-.208

EM4R

228

1

5

2.95

.738

.225

1,455


The statistical descriptive results for the observed variables show:

- The average customer expectation score in all groups is approximately “high: 4 points” and the difference is small, but the average customer perception score is only approximately “average: 3”.

- All average perception scores are lower than expectations. The average difference is 0.762 points, the highest is 0.95 points (RN3) and the lowest is 0.49 points (TA3). This shows that customers have high expectations (possibly from advertising and marketing programs of banks) but in reality, their perception of the quality of service they receive is still low.


- The dispersion does not vary much and ranges from 0.7 to 0.8.

- The response scores of the observations are distributed relatively symmetrically around the mean (the skewness coefficient is mainly in the range (-1,1). However, the distribution is mostly left-skewed (negative skewness coefficient).

- The response scores of the observations for the variables are relatively concentrated (kurtosis coefficient greater than 1).

The above results show that the survey data can be used to conduct statistical analysis. High symmetry and large kurtosis allow to obtain stable statistical characteristics with little deviation.

Using Cronbach's Alpha coefficient to evaluate the questionnaire, scale and data for each group of variables, we have the following results:

Table 4.2. Cronbach's Alpha coefficient of variable groups


Group

Cronbach's

Alpha

Variable

Cronbach's Alpha

when leaving out variables

TAE

.873

TA1E

.849



TA2E

.843



TA3E

.850



TA4E

.837



TA5E

.852

TAR

.822

TA1R

.782



TA2R

.766



TA3R

.804



TA4R

.771



TA5R

.809

RLE

.917

RL1E

.900



RL2E

.897



RL3E

.895



RL4E

.897



RL5E

.902

RLR

.843

RL1R

.803


Group

Cronbach's

Alpha

Variable

Cronbach's Alpha

when leaving out variables



RL2R

.819



RL3R

.819



RL4R

.802



RL5R

.816

RNE

.918

RN1E

.866



RN2E

.868



RN3E

.912

RNR

.877

RN1R

.824



RN2R

.769



RN3R

.882

ASE

.915

AS1E

.880



AS2E

.887



AS3E

.891



AS4E

.901

ASR

.837

AS1R

.785



AS2R

.798



AS3R

.790



AS4R

.801

EME

.929

EM1E

.897



EM2E

.898



EM3E

.907



EM4E

.929

EMR

.877

EM1R

.847



EM2R

.834



EM3R

.824



EM4R

.862


All groups of questions with Likert scales have Cronbach's Alpha coefficients from 8.22 to 9.29. Except for the RN3R variable of the RNR group, in the groups, there is no variable with a Cronbach's Alpha if Item Deleted greater than the group's general Cronbach's Alpha (Appendix 3). However, because the RNR group has only 3 variables and when removing this variable, the Cronbach's Alpha coefficient does not increase significantly, so there is no need to remove the RN3R variable. The results of the reliability analysis of the groups of variables including the group's composite variable give similar results. Thus, according to Ngo Van Thu (2015), it can be concluded that the questionnaire, scale and data are reliable and can be used for analysis.

4.1.2.2. Group aggregate variables

Usually previous studies do not examine these variables but calculate them by the arithmetic mean of the variables in the group.

For example, Parasuraman and colleagues use the following calculation:

TA- Average gap score for tangible items = (TA1+TA2+TA3+TA4+TA5)/5 RL- Average gap score for reliability items = (RL1+RL2+RL3+RL4+RL5)/5 RN -Average gap score for response items = (RN1+RN2+RN3/3

AS - Average gap score for assurance items = (AS1+AS2+AS3+AS4)/4

EM -Average gap score for empathy items = (EM1+EM2+EM3+EM4)/4

From there, the overall service quality is calculated by averaging the scores of the above 5 components. OSQ- Overall service quality = (TA+RL+RN+AS+EM)/5

To evaluate and adjust these variables, we can use different tools such as regression analysis, using variance structure model, correlation analysis. Here, the thesis uses correlation analysis between group aggregate variables and component variables. The results show that it is possible to use aggregate variables for general evaluation because the variables in the groups are all positively correlated, statistically significant (level 1%) with the group aggregate variable (Appendix 2). For the variable evaluating the general quality of service response, the thesis chooses the factor analysis model using the principal component method to synthesize and determine the weight as the importance of the groups of variables. From there, the impact of each component variable in all groups can be calculated.

It can be seen that the group aggregate variables on desire are more closely correlated with the variables within the group than the variables on the perception of service quality that customers receive. However, with this close correlation, the group aggregate variables can be used directly for analysis.


The results of calculating the internal weights of the variable groups are as follows:

Table 4.3. Weights of component variables of variable groups



Group detail variables


Total

Group aggregate variables

1

2

3

4

5

TAE

0.25

0.20

0.18

0.18

0.19

1.00

TAR

0.22

0.23

0.19

0.20

0.16

1.00

RLE

0.22

0.22

0.20

0.18

0.18

1.00

RLR

0.23

0.22

0.15

0.19

0.21

1.00

RNE

0.33

0.31

0.35



1.00

RNR

0.35

0.33

0.32



1.00

ASE

0.28

0.24

0.25

0.23


1.00

ASR

0.28

0.26

0.23

0.23


1.00

EME

0.26

0.26

0.24

0.24


1.00

EMR

0.26

0.28

0.24

0.22


1.00

4.1.2.3. Create analysis variables

- Creating GAP variables: The gap variables (R –E) are created by calculating the difference between the desired and perceived levels. These variables are denoted according to the original variables with the addition of the letter “G” at the beginning of the variable name. For example: GTA=TAR-TAE.

- Create Rate Statisf variables: The score ratio variables (R/E) are created by calculating the difference between the desired and perceived levels. These variables are denoted according to the original variables with the addition of the letter “R” at the beginning of the variable name. For example: RTA=TAR/TAE.

4.1.2.4. Database information used for analysis

The basic data includes 109 variables (excluding the vote variable), including:

- 5 identifier variables: Customer grouping, service provider.

- 52 primary variables: Desired scores and perceived scores obtained.

- 52 secondary variables: distance and ratio of desired score and perceived score obtained (Appendix 4).


4.2. Statistical analysis of perceived and desired gap in card service quality

4.2.1. Statistical description of indicators

The general assessment variables of all groups show that the desire is higher than the satisfaction level of customers. Most of the median values ​​as well as the mean values ​​are within 1 level. The desire is always high (level 4) and the satisfaction is only average (level 3). The small sample standard deviations and smaller than the statistics from the pilot survey show that customers have relatively uniform desires and perceptions of actual quality. The asymmetry coefficients of the first two indicators have negative values, indicating that many people have lower expectations and perceptions than the average, while for the last three indicators, the perceptions are skewed to the right (positive values). The kurtosis coefficients of the perception scores are positive and larger than this coefficient, corresponding to the desire score. This shows that customers may have low centrality but perceive service quality relatively more centrally.



Mean


Median


Mode


Stdv


Skewness


Kurtosis

TAE

3.88

4.00

4

.815

-.795

1,485

TAR

3.14

3.00

3

.684

-.071

2.020

RLE

4.21

4.00

5

.810

-.778

-.084

RLR

3.49

3.49

3a

.735

-.234

.685

RNE

4.06

4.00

4

.822

-.591

.050

RNR

3.31

3.00

3

.720

.168

.848

ASE

3.95

4.00

4

.801

-.482

-.116

ASR

3.23

3.00

3

.703

.105

.864

EME

3.87

4.00

4

.830

-.301

-.256

EMR

3.08

3.00

3

.765

.027

1,037

Table 4.4. Statistical description of primary variables by group Statistics


Table 4.5. Detailed statistical description of primary variables



Mean

Median

Mode

Stdv

Skewness

Kurtosis

TA1E

4.00

4.00

4

.771

-.399

.058

TA1R

3.18

3.00

3

.718

.178

1,137

TA2E

3.97

4.00

4

.765

-.420

.223

TA2R

3.23

3.00

3

.730

.043

1,253

TA3E

4.11

4.00

4

.668

-.409

.242

TA3R

3.62

4.00

4

.706

-.296

.571

TA4E

3.90

4.00

4

.821

-.394

-.094

TA4R

3.14

3.00

3

.712

.054

1,318

TA5E

4.08

4.00

4

.779

-.819

1,004

TA5R

3.20

3.00

3

.793

-.081

.572

RL1E

4.11

4.00

4

.769

-.504

-.009

RL1R

3.33

3.00

3

.765

.186

.517

RL2E

4.04

4.00

4

.865

-.609

-.114

RL2R

3.21

3.00

3

.834

-.171

.767

RL3E

4.10

4.00

4

.790

-.885

1,348

RL3R

3.43

3.43

3

.758

-.045

.120

RL4E

4.11

4.00

4

.749

-.579

.077

RL4R

3.38

3.00

3

.717

-.076

.273

RL5E

4.03

4.00

4

.785

-.658

.575

RL5R

3.38

3.19

3

.798

-.184

.335

RN1E

4.13

4.00

4

.766

-.419

-.656

RN1R

3.33

3.00

3

.834

-.327

.975

RN2E

4.14

4.00

4

.762

-.667

.516

RN2R

3.33

3.00

3

.790

-.003

.429

RN3E

3.95

4.00

4

.838

-.544

.044

RN3R

3.00

3.00

3

.829

.000

.983

AS1E

3.98

4.00

4

.845

-.451

-.256

AS1R

3.20

3.00

3

.717

.208

.835

AS2E

4.05

4.00

4

.778

-.422

-.374

AS2R

3.35

3.00

3

.719

.077

.751

AS3E

4.09

4.00

4

.728

-.357

-.405

AS3R

3.35

3.00

3

.778

.032

.289

AS4E

4.11

4.00

4

.740

-.655

.777

AS4R

3.38

3.00

3

.692

.056

.498

EM1E

3.77

4.00

4

.810

-.280

-.062

EM1R

3.03

3.00

3

.768

.001

1,037

EM2E

3.78

4.00

4

.834

-.367

.148

EM2R

2.97

3.00

3

.815

.069

.724

EM3E

3.75

4.00

4

.792

-.306

.356

EM3R

2.97

3.00

3

.790

.004

.961

EM4E

3.76

4.00

4

.824

-.230

-.208

EM4R

2.95

3.00

3

.738

.225

1,455


4.2.2. Analysis of variance of variables by grouping variables

4.2.2.1. Analysis of the influence of gender on component scores

Table 4.6. Results of variance analysis of variables according to customer gender


Tests of Between-Subjects Effects

Source Dependent

Variable

Type III Sum

of Squares

df

Mean

Square

F

Sig.


Corrected Model

TAE

1,164

1

1,164

1,757

.186

TAR

.645

1

.645

1,383

.241

RLE

.472

1

.472

.719

.397

RLR

.001

1

.001

.001

.975

RNE

.001

1

.001

.001

.977

RNR

.068

1

.068

.132

.717

ASE

.076

1

.076

.118

.731

ASR

.723

1

.723

1,469

.227

EME

.094

1

.094

.136

.712

EMR

.106

1

.106

.180

.672


With a significance level of 5%, it can be concluded that the component scores are not affected by customer gender (Sig > 0.05). With detailed variables in the components, there are only 2 variables, TA3E, which is "The physical facilities in service activities are very attractive" with a significance level of 0.038 < 0.05, meaning that male customers have different expectations about the physical facilities of the service than female customers (Appendix 5).

4.2.2.2. Analysis of variance of variables by age group

Table 4.7. Results of variance analysis of variables by customer age group


Tests of Between-Subjects Effects


Type III Sum of Squares

Df

Mean Square

F

Sig.

EM2E

9,644 dollars

4

2,411

3,624

.007

EM1E

8,696 as

4

2,174

3,453

.009

EM3R

8.102 axes

4

2,026

3,378

.010

TAE

8.511 a

4

2,128

3.333

.011

TA3R

5,339 hours

4

1,335

2,761

.029

EME

7.065 aq

4

1,766

2,638

.035

TA4R

5,044 j

4

1,261

2,558

.040

AS4R

4,697 apps

4

1,174

2,520

.042

EM3E

5.962 watts

4

1,491

2,433

.048


Only 9 variables in the table above have differences in scores between age groups (Sig

<0.05) (Appendix 6).

4.2.2.3. Variance analysis of variables according to customer position

Table 4.8. Results of variance analysis of variables according to customer status



Source

Type III Sum of Squares


df

Mean Square


F


Sig.

Corrected

TAE

10,086 a

2

5,043

8,060

.000

Model

TAR

4.129 b

2

2,065

4,557

.011


TA1E

7,477 c

2

3,738

6,596

.002


TA1R

4,600 d

2

2,300

4,600

.011


TA2E

8.057 e

2

4,028

7,267

.001


TA2R

4.281 f

2

2.141

4.129

.017


TA3E

4,061 g

2

2,031

4,695

.010


TA3R

1,618 hours

2

.809

1,632

.198


TA4E

5.067 i

2

2,533

3,853

.023


TA4R

5,587 j

2

2,794

5,747

.004


TA5E

6,333k

2

3.166

5,424

.005


TA5R

2,572 l

2

1,286

2,061

.130


RLE

4,325 m

2

2,162

3.365

.036


RLR

1,120 n

2

.560

1,038

.356


RL1E

5.157 degrees

2

2,578

4,498

.012


RL1R

3,092p

2

1,546

2,677

.071


RL2E

6,033 q

2

3.016

4,147

.017


RL2R

1,879 r

2

.940

1,354

.260


RL3E

3.686 s

2

1,843

3.002

.052


RL3R

.320 t

2

.160

.276

.759


RL4E

3,361 u

2

1,681

3,050

.049


RL4R

2.208v

2

1.104

2,166

.117


RL5E

8,141w

2

4,070

6,955

.001


RL5R

1.298 x

2

.649

1,018

.363


RNE

3,584 y

2

1,792

2,693

.070


RNR

2,620 z

2

1,310

2,565

.079

Comment


Agree Privacy Policy *