CHAPTER THREE

METHODOLOGY

3.0 Introduction
According to Polit and Beck (2004) Research methodology refers to ways of obtaining,
systematizing and analysing data. Creswell (2003) portrays methodology as a coherent group of
methods that harmonize one another and that have the capability to fit to deliver data and
findings that will reflect the research question and suits the researcher’s purpose. Bowling
(2002) explains that methodology is the complete structure of the research study; the size and
sample methods, the practices and techniques utilized to collect data and the process to analyse
data.

By choosing the research methodology, the researchers can formulate the path to be used in
conducting the study and reporting the findings. The methodology helps in the search of
literature, development of research questions and the creation of the most suitable study design.
It also assists in the interpretation of the results and the publication of the findings in journals.
References. Jha, N. (2008).

This chapter presents the research design, study population, sample size and selection, data
collection methods, data collection instruments, validity and reliability, procedure of data
collection such as interview questionnaire survey and documentary review were employed to
gather data for the project , data management and analysis and measurement of variables, data
structures that were employed in research study to determine the requirements, system analysis
and design as well as system development, testing and evaluation was done through individual
module testing and unit testing of the components of the system.

3.1Research design
The study used a qualitative survey research design to provide a more comprehensive picture
of each aspect of the study. A research design is a detailed outline of how an investigation
took place (Creswell, 2012). The study also used a survey research method, which was a
nonexperimental, descriptive research method. Surveys were useful when the researchers
wanted to collect data on phenomena that could not be directly observed. Other advantages of
the survey were; it was economical, and allowed a collection of a reasonable amount of data
from the study population, and also facilitated standardization of the data collected
(Bhattacherjee, 2011) and a disadvantage of the survey method was its artificiality.

3.2 Study Population

The target population contained members of a group that the researchers were interested in
studying (Shahrokh & Dougherty, 2014) and the results of the study were generalized to this
population, because they all had significant traits in common. To identify the target
population, the researchers specified the criteria that determined which individuals were to be
included and which individuals were not to be included. They considered a portion of the
population to which they had reasonable access; a subset of the target population. For
example, respondents (both residents and EC staff) were from Busia District where the
researchers’ studied from. Therefore, the researchers identified 270 individuals from whom
the sample size was got as shown in table 3.1.

3.2.1 Sample Size

A sample is the actual group from which data is collected (Shahrokh & Dougherty, 2014) and
it is a subset of the population and should be a typical representation of the larger group. At
least 30% of the total population is representative (Shahrokh & Dougherty, 2014). Thus, the
study was conducted on a convenience sample of eighty (81) respondents, who included both
78 residents and 4.5 staff (that is, EC officials at Busia) formed the sample of the study. The
target population was selected because it was central and pivotal to the topic under study and
because of that, the researchers were in position to obtain the relevant information to the
study. The sample size was adequate enough to generate unbiased results.

Table 3.1: sample size

Respondents Population Percentage (%) Sample Size Sampling technique
Staff (EC Officials) 15 30/100*15 4.5 Purposive
Residents 260 30/100*260 78 Simple random
Total population 275 82.5
Source: Primary data (2022)
3.2.2 Sampling Techniques
Sampling was concerned with the selection of a subset of individuals from within a statistical
population to estimate characteristics of the whole population (Shahrokh & Dougherty,
2014). For this study, purposive sampling technique was used so as to gather the necessary
information from the staff. This enabled the researchers to select the respondents basing on
their ability to provide relevant and dependable information. This sampling design was used
because it enabled the researchers to obtain real facts on the study under investigation since
only respondents who with information on the topic under investigation were purposely
sampled.

Simple random sampling technique was also used to collect information from residents
(Sekaran & Bougie, 2013). This technique had high generalizability of findings; hence it was
suitable for a large study population. The researchers allocated a number to every district
respondent of the accessible population, place the numbers in a container then picked
numbers at random. The subjects corresponding to the numbers picked were included in the
sample.

3.3 Sources of Data

The researchers used both primary and secondary data. Primary data about the existing paper-
based voting system was obtained from the district documents, interviews with members of
staff and questionnaires with residents. Secondary data was used from online journals, text
books, articles and other literature available about e-voting systems.

3.4 Data collection Techniques

To collect data from Busia, the researchers relied on primary and secondary data collection
methods (Sekaran, 2003). The methods that were employed included documentary review,
interviews and questionnaire surveys.

3.4.1 Interview Method

The common way of collecting qualitative data was by conducting interviews because it
enabled the researchers to acquire loads of insightful primary data in a very short time. It was
also possible in this technique to ask follow-up questions if the answers from the respondents
would not be enough (Bryman, 2002). The interview method refers to the method of data
collection through verbal interaction, face-to-face and by telephone (Sekaran, 2000; Amin,
2005).

Primary data was collected using the interview guide (appendix B) to guide the researchers in
conducting face-to-face interviews. This guide consisted of a list of questions which the
researchers used in collecting data (Jacobsen et al, 2002). These questions helped the study to
stay focused and thus enabled the researchers to collect necessary information.

3.4.2 Questionnaire Survey Method

The researchers also used a questionnaire survey method to collect primary data from the
respondents. It was used because respondents could easily read and write the answers. The
questionnaire was a written set of questions that was used to obtain information about the
study resources from the selected respondents. A questionnaire (appendix A) was used
because it helped to investigate motives and feelings in Likert scaling (Creswell, 2014). The
questionnaire had closed-ended questions divided into sections that represent the different
variables of the study. The open-ended questions provided a chance for the respondents to
provide detailed information regarding the study.

The study issued out 75 questionnaires, which were successfully completed by the
respondents (residents), hence the response rate generated was 100%. Five (05) other
respondents (EC officials) were interviewed.

3.4.3 Documentary Review Method

Documentary review method refers to the analysis of documents that contain information
about the phenomenon under study. Payne and Payne (2004) described the documentary
review method as the technique used to categorize, investigate, interpret and identify the
limitations of other methods like findings in previous studies. Data collection involved
review of documents to gather secondary data that was used in the study. Documents that
were reviewed included textbooks, journals and articles.

The documentary review checklist was used to collect secondary data. It constituted the list of
items of information that were obtained from documents, records and other materials. In
order to secure measurable data, the items that were included in the schedule were limited to
those that could be uniformly secured from a large number of case histories or other records.
3.5 Data Analysis
The data was analyzed and interpreted using excel spreadsheet, which was easier to use. The
patterns of the results were used to highlight the key features of the designed framework.
Excel was easy to use and could enable the researchers to generate results of analysis inform
of bar graphs and statistical tables.

To analyse qualitative data, the researchers organized statements and responses to generate
useful conclusions and interpretations based on the research objectives (Sekranan, 2003).
Qualitative data analyses for the study objectives involved analysis of themes of interview
data. Interview response were reviewed, sorted and classified into related themes. Once the
themes were established, data was evaluated and analyzed to determine consistency,
credibility and usefulness of the information to support the qualitative data requirements for
the study.

To analyse quantitative data, values were entered in excel corresponding to different variables
on the questionnaire. Results of analysis were presented inform of bar graphs and statistical
tables.

3.6 System Analysis and Design

The district e-Voting system was designed based on the functional and non-functional
requirements which described the parameters and the data that were incorporated into the
system. The researchers used DFDs (context diagram and level one data flow diagram) and
entity relationship diagram (ERD) to model the conceptual design of the system. According
to Dillman (2012), DFDs and ERDs were used because they are easier to understand by
technical and nontechnical audiences; provided a high-level system overview, complete with
boundaries and connections to other systems and provided a detailed representation of the
system components.

3.7 System Design Approach

The researchers used an agile software development model and followed the systems
development life cycle (SDLC), which was composed of a number of clearly defined and
distinct work phases, to plan for, design, build, test, and deliver the district e-Voting system.
Jacobson et al. (2008) assert that agile development model is a type of Incremental model.
Therefore, unlike other software development models, such as the waterfall model, the
researchers used agile model because very limited planning was required to get started with
the project. Agile assumes that the end user needs are ever changing in a dynamic business
and IT world (Jacobson et al, 2008).

3.8 System Implementation

The district e-Voting system was implemented using the following technologies;
i. Hyper Text Markup Language (HTML) for front end. JavaScript was used to develop
the modules of the district e-Voting system since it is an object-oriented
programming language. Its ability to display information in the browsers also enabled
users to use the system.
ii. Cascading Style Sheet (CSS) was used in formatting the user interfaces of the district
eVoting system.
iii. PHP Hyper-Text Pre-Processor (PHP) that is a server-side scripting language used in
connecting the front end to the back end of the district e-Voting system.
iv. Bootstrap was used for responsiveness of the district e-Voting system.
v. AngularJs was used to validate the district e-Voting system using regular expressions.
vi. My Structured Query Language (MySQL) for the backend [Database].

3.9 System Testing and Validation

The district e-Voting system was tested by the researchers, using sample data, on the basis of
error rates to verify that it met design requirements. Thereafter, the district e-Voting system
was taken to the users for validation to ensure that it operates to their satisfaction.

Testing was done before and after the district e-Voting system was put to use. It was done in
phases.

i) Unit Testing: Unit testing was carried out on individual modules of the district e-
Voting system to ensure that they were fully functional as separate units.
ii) Integration Testing: The researchers carried out integration testing after the different
modules were integrated to make a complete system. Integration testing was carried out to
ensure that the modules are compatible and that they could be integrated to form a complete
working district e-Voting system.
iii) User Acceptance Testing [UAT]: The researchers further carried out user acceptance
tests to check for any errors that could arise.

3.10 Ethical Consideration

The researchers sought authorization from Busia before collecting data indicating that the
study was purely for academic purposes and the researchers also respected the respondents
and assure them that information given would have no effect on them and their businesses. In
addition, the researchers promised the respondents that information given would be treated as
provided with maximum confidentiality.

3.11 Limitations of the study

The first limitation was locating respondents who were willing to participate in the study.
Distrust by respondents and time to carry out the study posed as a challenge. Another
challenge was language barrier and hence respondents could not interpret the wording on the
questionnaires properly.

The researchers had to seek assistance from research assistants to help them with the location
and distribution of questionnaires to the respondents. These problems were solved by sparing
time by the researchers with the respondents to explain the purpose of the research and the
interpretation of the items on the questionnaires.

3.12 Conclusion
In conclusion, the researchers used the qualitative and quantitative research methods,
applying both secondary and primary data collection approaches in order to interact as much
as the researchers could with the empirical and theoretical levels. Whenever the first-hand
data or secondary data was collected, the researchers kept the criteria of validity and
reliability in their minds. Therefore, to meet these demands, all respondents in Busia were
informed about the content of the analysis before the actual interviews.