Chapter I Introduction
Chapter 3 Methodology
The aim of this study is going to overview of literature about service quality and that impact on the banking service. Study the impact of service quality on customer satisfaction and customer loyalty. Testing the different between demographic factors in customer loyalty
To achieve the research objectives, this study is going to answer following questions:
RQ1: Do the service quality impact on the customer satisfaction that used EIB service?
RQ2: Do the customer satisfaction impact on customer‘s loyalty in EIB service?
RQ3: Is the any difference from customer demographic (Gender, Age, Education, Income) in customer’s loyalty?
3.2 Sample designed
The target population is customer who is currently using or used retailing service in Eximbank service. This study is going to conduct the survey with 250 customer who are using the Eximbank service in Hanoi. The survey is conducted on May and June, 2015.
3.3 Questionnaire designed
This study uses SERVQUAL model of Parasuraman (1988) to develop the questionnaire.
There are 22 items distributing in 5 factors. The customer satisfaction variable these study uses scale of Bachelet (1995). The customers loyalty is based on the Chen and Tsai (200&) scale . all the item was modified to fix the context of Vietnam banking service sector. The table 3.1 shows the variables and its items.
Table 3.1 Variables and its items
Variable Items Code
Reliability
EximBank guarantee security issue REL1
EximBank provide timely service as committed REL 2 The deposit products and services of EximBank
were performed accurately REL 3
EximBank Provide complete information, accurate
and timely REL 4
Bank guarantee security issue REL 5
Responsiveness
Diverse products and services to meet customer
needs RES 1
There are many customer care program RES 2 Professional staff to handle rapid accurate RES 3 Staff actively advise suitable products for
customers RES 4
Assurance
Staff ready to serve and assist customers ASS 1 Resolve complaints quickly and rational ASS 2 Guide to procedures for clients fully
understandable ASS 3
The staff is professional and qualifications ASS4
Empathy
Enthusiastic, considerate and fun staffs EMP1 Employees interested in the individual
requirements of the customer EMP 2
Always listen to customers EMP 3
Always polite, friendly customer EMP 4
Tangible
The location of transaction is convenience and in
accordance with customer needs TAN1
Spacious transaction, using modern technology TAN 2 Nice layout of the counter, identifiable for
Customers reasonable TAN 3
The staff dresses polite TAN 4
Paper forms to be used in transactions were
designed easily TAN 5
Customer satisfaction
VietinBank offers more benefits than costs me
money CUS1
VietinBank is the best among competing banks CUS 2 VietinBank is much better than what I expected CUS 3
Customer Loyalty
Ready to become long-term customers of banks CL1 Ready to introduce family and friends CL2 If there are minor issues and comments will
continue to use the services of the bank CL3
3.4 Analysis
3.4.1 Reliability analysis
Cronbach's coefficient alpha only trust that the measure has linked together or not; but did not observe any necessary variables and observed variables left to be retained. Meanwhile, the calculation of the correlation coefficient between variable-general will help sort out the variables do not observe significant contribution to the description of the concept to be measured.
The criteria used to evaluate the reliability scale:
- Type the observed variables have a correlation coefficient small turn-total (less than 0.3); criteria when selecting the scale greater reliability 0.6 Alpha (Alpha larger the internal consistency reliability at higher) (Nunally & Burnstein 1994).
- The level of the Alpha value: greater than 0.8 is a good measurement scales; from 0.7 to 0.8 is used; 0.6 or more can use the case study is a new concept or new in the context of research (Nunally, 1978; Peterson, & Deal, 1998).
- The observed variables are correlated variables-total small (less than 0.4) are considered junk variable will be removed and the scale is accepted as Alpha reliability coefficient satisfactory (greater than 0.7 ).
Based on this information, researchers conducted assessment scales based on the following criteria:
- Type the observed variables with correlation coefficients less than 0.4 turn-general (these variables do not contribute much to the description of the concept to be measured, and many previous studies have used this criterion) .
- Choose a reliable scale greater than 0.6 Alpha (the concepts in this study is relatively new to the subject of research when respondents).
3.4.2 Regression analysis
Regression analysis was to find dependencies of a variable, called the dependent variable in one or more other variables, called the independent variable aims to estimate or predict the expected value of the dependent variable when anticipated value of the independent variable.
Some other names of the dependent variable and the following independent variables, Dependent variable: variable is explained, the forecast variables, variable regression, variable response, endogenous variables. Independent variables: explanatory variables, predictor variables, repressor, variable or variable control agents, exogenous.
Although the regression analysis based on the idea of the dependence of one economic variable on other economic variables but technique itself regression analysis does not imply causality. Good examples of the confusion between these concepts conduct regression of burglary in a city with a number of police officers of the city. Y is called the burglary in a
year and X is the number of police officers. When we Regression Y on X, if we find a positive relationship of Y and X variables have statistically significant, the regression analysis to conclude: "Increase the number of police officers will increase the number burglary ".
Clearly, this analysis identified mistakes in a causal relationship. Increased police numbers is due to the strengthening of the police force in the context of increased burglary. So we have to correct the regression of police according Y. burglary or X according to regression analysis before we have correctly identified a causal relationship.
3.4.3 T - Test, ANOVA
The contents of this section in order to test the differences between qualitative variables with quantitative variables, for example the difference in satisfaction with banking services A customer objects (such as gender, age, income ...) or not. To achieve this we conducted analysis of variance ANOVA and independent-sample T - test. The difference was statistically significant with a 95% confidence level (or levels of meaning Sig. <0.05).
There are two procedures for the analysis of variance: ANOVA factor ANOVA and multiple factors. Custom research conducted analysis of variance one element or two elements, particularly for undergraduate thesis or Master of Science in net toward applied research, sample surveys are not too complicated, this type of study user testing variance factor will be done because we just test the qualitative variables to classify observations into different groups. Inspection includes inspection ANOVA between variance homogeneity of the group has a significant factor sig. > 0.05.
The two selected variables (e.g. gender only two shown is Status of Men and Women), therefore we will use the test of Independent-sample T - test (test the hypothesis of two overall average possible) to find differences with quantitative variables.