CHAPTER approach the generation and testing of theory from



We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

5.0       Introduction

chapter of the study outlines the research methodology researcher adopted
towards finding answers to the research questions in order to realize the
research objectives. The chapter was structured to consist of type of research,
sampling method, data source, study population, sample size, sampling
technique, data collection instrument (questionnaire), data collection
procedure (respondents), model specification, data analysis, ethical
considerations and conclusion.


5.1       Type of Research

research is generally associated with the positivist/postpositivist paradigm.
It usually involves collecting and converting data into numerical form so that
statistical calculations can be made and conclusions drawn. This study is
quantitative in nature where data obtained from the field were converted into
numerical form with statistical calculations.

and Skoldberg (2011), and Harwell (2011) explicate that quantitative study
comprises standardized variables that offers descriptions of experiences or
views of people that are given numbers by usage of response alternatives that
are determined ahead of time. Christensen et al., (2007) earlier on gave
explanation that the quantitative information is collected through surveys,
interviews and archive material that are frequently obtainable in the form of
tables, diagrams and graphs.

the purpose of this research, the quantitative approach is employed. Regarding
the rationalization, Coll and Chapman (2000) emphasized that research approach ought
to be based on the study objective rather than commitment to a particular
paradigm. Therefore, for the objectives of this study quantitative approach was
used to investigate and analyze the government image influence on the economic
performance, a case of Ghana and the relevance of government image in relation
to marketing strategy.

approach the generation and testing of theory from two ways which are inductive
and deductive (Neuman, 2006). Consistent with Saunders et al. (2009) study,
inductive approach shifts from definite observations to broader generalization
and theories. Gray (2013) stress that inductive research approaches plans are
made for data collection, followed by which data are analyzed to ascertain if
any patterns form suggesting relationships between variables. With these
observations, it may be feasible to create generalizations, relationships and
even theories. Saunders et al. (2009) emphasized that small sample size is apt
for inductive approach. Gray (2013) emphasize that induction is rooted in
multiple cases or observations simply to make certain of a degree of

contrast, deductive research approach shifts towards testing of hypothesis,
then which the principle is substantiated, disproved or modified (Gray, 2013).
Deductive approach hypotheses present an assertion in relation to two or more
concepts that try to explicate the link between them. Concepts themselves are
abstract designs that make up the hypotheses and theories building blocks.
Normally, deductive approach begins from general statement to more specific
one. Occasionally, researchers attempt to observe theories that are more
connected to a topic of concern and subsequently narrow the theories down to
specific research questions or hypotheses that can be tested (Saunders et al.,
2009). To Saunders et al (2009) and also Gray (2013), researchers employ
diverse forms of methods to answer questions or prove hypotheses, in quantitative
ways so as to be able to generalize the findings. Thus, this study adopts the
deductive approach.


5.2       Data

(2009) indicates that there are two main data sources, primary data and
secondary data. Primary data are considered to be originated by the researcher
for a specific purpose of addressing the research problem (Malhotra &
Birks, 2007). It is what the researcher collects based on the study objectives
and questions by means of observations, interviews or questionnaires from the
sample or target population. In contrast, secondary data are gathered for some objective
other than the problem being considered (Malhotra & Birks, 2007). Secondary
data has to do with data that are collected from secondary source for instance
publications, personal records, and survey among others that have previously
been collected by others for another intention. In this study, secondary data
IMF World Economic Outlook Database, 2017 and statistical bulletin extracts of
Ghana while primary data were obtained from the businesses who participated in
the study.


5.3       Study Population

Population is referred
to as any inclusive group of entities such as people, industries, institutions,
businesses or the like that share some common set of characteristics in
agreement with the purpose of the study under investigation and in relation to
which researchers want to be able to draw conclusions and plan to generalize
(Malhotra, Pathak, Nath, Mukherjee & Shanker, 2011). But, the accessible
population of this study consists of businesses who meet the predetermined
eligibility criteria set for the study. The target population used was 11,380 made
up of businesses drawn from various categories. Burns and Grove (2003) define eligibility criteria as “a
list of characteristics that are required for the inclusion in the target


criteria used for the firm’s inclusion in this study are as follows:

The participating business should have existed
and operated at least more 16 years in Ghana and have witness four (4)
successive political administrations, and

The participating business should be in a
formal sector, that is, registered and pay taxes,


Table 1: Exhibition of Study
Population and Sample Size

























(Source: Ghana Social and Living
Standard Reports, 2012)


5.4       Sample

is a technique that presents a range of methods that enable a researcher to trim
down the quantity of data required to be collected by considering only data
from a subgroup rather than the entire likely elements or cases. This technique
saves time and as well provides a superior general accuracy than census as collecting
data from fewer cases means one can collect information that is more detailed
(Saunders et al. 2009). Thus, for this research two hundred twenty two (222) representing
about of 2% of the study population used for the analysis which according to Anderson
(2004) is acceptable in line with students’ thesis regarding population.


5.5       Sampling

non-probability sampling technique was used for this study since the dominant
aspect of non-probability sampling is that it provides researchers greater
freedom and flexibility in selecting the individual population units than
probability sampling (Tongco, 2007). Proportional purposive sampling techniques
were considered appropriately for this study. The researcher adopted this
approach to ensure all business types are represented in the population structure
of Ghana registered businesses. In this case, allocated percentages were
assigned to all types of business based on their numbers. For the purposive sampling
technique, the researcher had to ensure that only responding businesses with
the requisite knowledge and information were included in the study.


5.6       Data
Collection Instrument

instruments employed for data collection in this study was questionnaire and statistical
bulletin extracts. Data collected with questionnaire was designed using
close-ended questions. The questionnaire utilized 5 point likert-scale
measurements where 5 assumes strongly agree, 4-agree, 3-neither agree nor
disagree, 2-disagree and 1-strongly disagree. The justification of using
questionnaire as data collection tool was to be able cover large number of the
respondents within earliest possible time at the minimum costs. Questionnaire
is also noted for its objectivity and respondent’s privacy in administration
which is all essential for this study.

performance indicators were all obtained from the IMF World Economic Outlook
Database and Ghana statistical bulletins while the government image data were
obtained from the respondents.


5.7       Data
Collection Procedure    

to the designing of the questionnaire, there was a thorough review of the
literature as well as the validity and reliability tests carried out to check
its authenticity. Care was taken in designing the questionnaire to ensured
objectivity and consistency on the part of the researcher, and also ensured a
high level of standardization of the research process and observance of ethical
rules. The businesses were stratified based on type of business. With the help
of social network, that is, Twitter and LinkedIn about 70% of the questionnaire
were administered while the remaining 30% in the form of hard copy were also completed
by some businesses in the remote part of the country where internet networks
almost non-existing. Each respondent business completing the hard copy was
given a limited time to do so.

the economic performance (GDP/capita), the researcher extracted data from the
statistical bulletin generated by the Ghana Statistical Service (GSS) and the IMF
World Economic Outlook Database. Data obtained from these institutions were
compared to confirm their validation.


5.8       Model

discover the long run relationship between the variables, multiple regression
analysis was used. In this study, five (5) variables were used namely GDP/capita,
nation-branding, trust, media-relation, and consonance. The economic
performance data used have the span of sixteen (16) years from 2000 to 2016.
After the selection of the variables, function of the study can be described in
the following way:



(Gross Domestic Product per capita) = Economic Performance

= Trust

= Nation-branding

= Media-relation

= Consonance


specifying the function in linear form with an addition of error term (, the multiple
regression model can be written in the following way:


 = the model’s constant

,,, and  are the regression coefficients of respective
explanatory variables, and

 = the error term


5.9       Data

the level of best practice, the thematic categorization of the data was done in
order to mirror the research purposes (Namey, Guest, Thairu & Johnson,
2008). This research used quantitative approaches in its analysis where
frequencies, means and standard deviations use for interpretations. With
regards to quantitative analysis, the researcher applied two methods: First, diagnostics
test to identify multicollinearity of the predictors. Second, multiple
regression analysis was used to estimate the causal relationship between economic
performance (GDP/capita) and predictor variables. In the regression analysis,
influence of each predictor on GDP/capita was determined.


5.10     Ethical Considerations

recognizing the ethics of research work, Economic and Social Research Council
(ESRC) Framework for Research Ethic (FRE), 2010 was duly followed. In this
framework, the researcher obtained all the data through legal means and also
ensures that appropriate arrangements were put into effect to maintain the
integrity, sanctity and security of the acquired data. Furthermore researcher
maintained the high ethical standard of research framework by not using any
aspect of the research results as a form of caustic tool to demonize any
sampled firm per its performance. As a final point, all materials and data
sourced for study were duly referenced in accordance to the required
specifications demanded by the University.


5.11     Conclusion

this chapter, it described the research methodology used for the research. It
began with introduction of the chapter and also discussed the research design.
It followed with the research type where quantitative which is generally
positivist was used. Additionally, other sections which the chapter captured
include source of data, study population, sample size, sample technique, model
specifications, data collection procedure, data analysis and ethical
considerations. The methodology this study employed is compatible with other
analogous studies. The rationale is to guarantee that results would be valid
and reliable, and could be used as bases for comparable studies.