Effect Of E-banking On Bank Profitability
[A STUDY OF GUARANTY TRUST BANK PLC ENUGU]

RESEARCH METHODOLOGY

3.1     Introduction

The term methodology is used to describe the activities involved in collecting the required information for this research work. This chapter describes how the study was carried out by showing the methods and procedures used for the research and collection of data for the study. It includes the description of the research design, sources of data, instrument for data collection and data analysis and techniques. 

3.2     Research Design

The research design employed by the researcher is ex post-facto research which aims at determining or establishing or measuring the relationship between one variable and another or the impact of one variable on another (Onwumere, 2009).

3.3     Nature and Sources of Data

The nature of data for the analysis of this study is secondary and data for this study was accessed from Guarantee Trust Bank Annual Report from 2014 to 2017.

3.4     Model Specification

A regression model has been employed, the essence of regression is to use a mathematical equation to express the nature of the relationship existing between variables and ultimately to use this equation to predict the value of one variable given a specific value of the other variable (Ugbam, 2001). This research work uses a three-model regression to capture the interaction between: internet banking and ROA; internet banking and ROE; and ATM transactions and ROE. The basic aim of the regression model in this study is to investigate empirically the extent to which the predictor variable explains the variation in dependent. The model will be estimated using the coefficients of the independent variable and its level of significance. This test provides an empirical platform in drawing generalization for this study. The variable to be predicted is called the dependent variable while the variable whose value will be used in the prediction is called the independent variable (Ugbam, 2001).

In analyzing data, the simple regression model will be employed which is:

Y = bo+ b1X + µ.

Where:

Y = the variable we are trying to predict

b0 = the intercept

b1 = the slope

X = the variable we are using to predict Y

µ = the error term

The intercept (b0) is the value of the dependent variable when the independent variable is equal to zero while the slope of the regression line (b1) represents the rate of change in Y as X changes. Because Y is dependent on X, the slope describes the predicted values of Y given X.

The above model can thus be applied in this study as:

ROA = b0 + b1Web + µ.……………………………………………….…Eqn. (I)

Where

ROA – Return on Assets {Dependent Variable}

Calculated by Net IncomeTotal Assets x 100

Web – Total value of Web Transactions (Independent Variable)

ROE = b0 + b1Web + µ...…………………………………………....... Eqn. (II)

Where

ROE – Return on Equity {Dependent Variable}

Calculated by Net IncomeShareholders'Equity x 100

Web – Total value of Web Transactions (Independent Variable)

ROE = b0 + b1ATM + µ...………………………………………….... Eqn. (III)

Where

ROE – Return on Equity {Dependent Variable}

Calculated by Net IncomeShareholders'Equity x 100

ATM – Total value of ATM transactions of GT Bank (Independent Variable)

3.5     Techniques of Data Analysis

Techniques of data analysis employed by the researcher are the ordinary least square (OLS) method and Granger Causality Test with the aid of Statistical Package for Social Sciences (SPSS) Version 25. The researcher chose OLS because it minimizes the squares of the residuals. The formulas for obtaining the estimates of the beta coefficients, standard errors, etc. are all based on this principle. The aim of using this method is to minimize the error in our prediction of the dependent variable, and by minimizing the residuals, error will be minimized. By using the "squares" the researcher is precluding the problem of signs thereby giving positive and negative prediction errors the same importance.