CHAPTER THREE
RESEARCH DESIGN AND METHODOLOGY
3.1. Research Approach
This study was employed quantitative research approach through structured questionnaires with
closed Likert type statement; it is a means for testing objective theories by examining the
relationship among variables. These variables, in turn, will be measured, typically on
instruments, so that numbered data will be analyzed using statistical procedures.
3.2. Research Design
The study adopted empirical investigation with descriptive and explanatory research design,
since the major focus of the research war to investigate factors affecting consumer choice of
supermarket. Explanatory studies and descriptive designs were used to allow for the gathering of
information, summarize, present and interpret it for the purpose of clarification (Creswell,
(2003). Descriptive research design was used in order to describe the population, situation, or
phenomenon that is being studied. On the other hand, explanatory research was deployed to
explain, understand and predict the cause-and-effect relationship between the variables.
Therefore, in order to accurately describe the effect & relationships between the independent
variables and the dependent variable descriptive and explanatory research studies were employed
which helps to use both qualitative and quantitative data analysis.
3.3. Sample Size and Sampling Techniques
3.3.1. Target Population
The target population is the population to which a researcher wants to generalize the result of the
study. The population of interest in this study consists of customers visiting supermarket stores
in Addis Ababa city (customers visit supermarket stores in Addis Ababa daily for the purpose of
shopping). It may not be possible to establish the actual number of customers shop at the stores
on specific day. Therefore the researcher is forced to determine sample based on infinite number.
3.3.2. Sampling Size Determination
Since, exact number of target population is difficult to get, the researcher used to take sample
based on infinite number of population. As Alreck & Settle (2005) noted that the choice of
sample size is made after considering statistical precision, practical issues and availability of
resources. According to Malhotra & Peterson (2006) there is no a single and precise way to
determine the size of sample; hence there are a number of inadequacy for deciding on sample
size. The researcher used Cochran (1963) formula to determine the sample size for the proposed
study based on a 95% desired confidence level and a 5% desired level of precision. Accordingly,
the following formula gives the size of the sample in case of infinite population when we are to
estimate the proportion in the universe.
Z ² PQ
n=
(e)²
The value for Z is found in statistical tables which contain the area under the normal curve (the
abscissa of the normal curve that cuts off an area α at the tails (1 - α equals the desired
confidence level, i.e. 95%).
1.96 ²∗0.85∗0.15
n= =195.92 ≈ 196
(0.05)²
From the total target population 196 were considered as sample size of the study.
3.3.3. Sampling Techniques
For the purpose of study, the researcher were used non probability sampling particularly
convenience sampling technique to select the supermarkets in Addis Ababa city, and also in
order to reduce biased data and time taken to collect data for the study the researcher was used a
simple random sampling technique to select customer because respondents are relatively similar
and are selected as they happen to be in the right place at the right time (Creswell, 2003).
3.4. Type and Source of Data
There are two types of data, namely primary and secondary data. A researcher might use either
both or one of the types of data depends on the research type and data collected by researcher
(Saunders et.al, 2007). For this research purpose primary data will be collected through
standardize questionnaire. Primary data are originated by a researcher for the specific purpose of
addressing the problem at hand (Malhotra and Birks, 2006). The sources for primary data in the
case of the study were from customers of supermarkets in Addis Ababa.
3.5. Data Gathering Tools
As the study was used quantitative method questionnaire were designed as a data gathering
instrument for the quantitative study because quantitative method has numerical data or
information that can be quantified to enable research questions to be answered. In the study
structured questionnaires were used in which there are definite, concrete and predetermined
questions. These questions will present exactly the same wording and the same order to all
respondents, which are simple to administer and analyze the data ((Cameron, 2008) and
(Lancaster G., 2005). The forms of the question are closed ended types. To gather primary data
Likert scale will be used. The respondents asked to rate the extent to which they feel motivated
by rating 1 to 5 (grading scale 1 is strongly disagree, 2 disagree, 3 Neutral 4 agree, 5 strongly
agree). The Likert scale will be employed to examine the relationship between variables. The
questionnaire will also include some questions about educational back ground of respondents,
occupation of the respondents, and age of the respondents.
3.6. Data Gathering Procedure
Before collecting the data, the instruments of the data were prepared carefully and permission of
the selected firm will be taken. Then the objectives of the study will be explained to subjects.
Based on their willingness to participate in filling questionnaire the questionnaires will be
distributed.
3.7. Data Analysis Tools
After the data collection, both descriptive and inferential statistical techniques were employed
with the aid of Statistical Package for Social Sciences (SPSS) version 22 to analyze the data. In
order to describe the characteristics of the study variables mean, standard deviation, frequencies,
and percentages are intended to be used to analyze and present the collected data in the study.
And also the collected data were analyzed by Regression analysis and correlation analysis
because it helps to investigate effect and relationship between the independent variable of
Consumer Choice Factors and dependent variable Choice of Supermarket.
3.7.1. Multiple Regressions Assumptions
In addition, before making the regression analysis some regression assumptions were tested in
order to check the data accuracy. In order to get the reliable and dependable result of the
analysis, all the assumptions of the multiple regressions should be fulfilled before making the
regression analysis interpretation. Therefore, before going to answer the research questions the
researcher was tested pre regression assumptions such as Multicollinearity, Normality,
Homoscedasticity and Heteroscedasticity assumptions.
3.7.2. Model and Estimation Techniques
In order to found the cause and effect relationship between dependent and independent variables,
the study was used multiple regression model to measure the level of significant relationship
between the dependent and independent variables. Therefore, based on the conceptual model of
the study expressed on Figure 2.1, mathematically the relationship between Consumer Choice
Factors and Choice of Supermarket is expressed in the multiple regression equation as:
Y = X0 +X1 (IL) + X2 (IPR) + X3(ICS) + e
Where: Y= Consumer Choice of Supermarket.
IL = Location Factor
IPR = Product Range Factors
ICS = Customer Service Factors
E = Level of Precision
X0= the constant parameter.
X1= Coefficient of Location Factor
X2= Coefficient of Product Range Factors
X3= Coefficient of Customer Service Factors
In accordance with the above mathematical model the constructed research hypothesis were
tested by considering significance level of each constant parameter in multiple regression
analysis.
3.8. Validity and Reliability Tests
3.8.1. Test of Validity
Validity is the most critical criterion and indicates the degree to which an instrument measures
what it is supposed to measure. Malhotra (2010) mentioned about three types of validity in his
study: content validity, predictive validity, and construct validity. This study addressed content
validity through the review of literature and adapting instruments used in previous research.
3.8.2. Test of Reliability
After the test of validity of the research instrument the next step is to cross check the constancy
and reliability of the instruments in order to do so the questioners has been reproduced and
distributed to a sample of 12 respondents for the pilot test. It is used to test the degree to which
instruments items are homogeneous and reflect the same underlying construct(s). As multiple
items in all constructs were used, the internal consistency/reliabilities of factors of Consumer
Choice of Supermarket were assessed with Cronbachs Alpha and the reliability values for all
constructs are confirmed as greater than 0.7, which are considered ideal (Pallant 2005).
Therefore, it can be concluded that the data collection instruments were acceptable as reliable.
The following table shows the summary of reliabilities of all constructs.
Table 3.1: Reliability of constructs
Construct Variables Cronbach’s Alpha Number of
Result items
Influence of Location (IL) 0.776 16
Consumer Choice Influence of Product Range (IPR) 0.718 10
Factors Influence of Customer Service (ICS) 0.855 20
Total 2.35
Source: Research Data, 2024
3.9. Ethical Considerations
A formal letter was written from Maryland international College, Department of Business
Administration to concerned authorities of the supermarkets in Addis Ababa city. The data
collection was started after getting consent from the parties mentioned above. In addition to this,
name of the customers (selected for the sample) was not included to maintain confidentiality.
