Chapter Three
METHODOLOGY
3.0 Introduction
The methodology approach and the processes required to conduct the study with regard to
determining how leadership style affects employee performance in retail and trading Small and
Medium Enterprises (SMEs) in Mogadishu, Somalia is presented in this chapter. The chapter
discusses the research design, research population, sample size, sampling techniques, research
instrument, the validation and reliability of the instrument, data collection methods, data analysis
procedures, ethical considerations and the limitations of the study. In this chapter, procedures and
methods of data collection and data analysis for the purposes of accomplishing the objectives of the
study are explained.
3.1 Research Design
Research design is the general outline or blue print of a researcher's approach toward addressing
research questions and are goals of the study. Research design is a framework for collecting,
measuring and analyzing data systematically (Creswell, 2014).
The type of research aims of this research is Quantitative Study. The quantitative approach was
appropriate as it allows for numerical information to be gathered which can be analyzed statistically
to find the relationship between the variables. The study centred on the key focus of transformations
leadership, transactional leadership, or Laissez faire leadership on employee performance.
Descriptive survey research design was used in the study. This descriptive survey design was
chosen because it enables the researcher to gather information from a sample of respondents to
describe the various conditions, opinions, attitudes and relationships that exist at the current time.
Shows how it is used in research in business and management, as it is a good research design in
collecting data from many respondents in a short time period.
Primary data was collected from the employees who are working in these selected retail and trading
SMEs in the city of Mogadishu Somalia. The snowball sampling technique was employed to gather
information from the respondents with the use of a questionnaire with questions that focused on
perception of employees and leadership styles. Given the objective of the survey, the cost effective,
time efficient and ability to produce "standardized" data for statistical analysis, the survey method
was deemed appropriate.
3.2 Research Population
The entire group of individuals who exhibits the characteristics under investigation and who can
yield data for the research is the research population. A population is defined as (Creswell 2014)
refers as "all individuals with common characteristics of interest to the researcher. This study was
conducted in selected retail and trading SMEs of target population Mogadishu, Somalia of total 135
employees. They were chosen because these workers are involved directly as part of the
organizational activities and work on a variety of leadership styles practiced by managers and
supervisors. They have thus a favorable standing to contribute with information about the impact of
leadership styles on employee's performance.
Employees of various departments and operational units were part of the population in the selected
SMEs. They engage in a range of sales, customer service, finance, administrative, procurement,
inventory related and other business related tasks. It was crucial to involve staff members from a
range of functions as this helped to capture a range of views around leadership practice and staff
performance. Retail and trading SMEs were selected because they play a major role in creating jobs
and contributing significantly to the economic development and growth of Mogadishu. The right
setting in which to explore leadership practices is being employed in these enterprises—where
employees routinely are in contact with organizational leaders, and where they are directly impacted
by organizational leadership decisions and management style.
3.2.1 Sample Size
A sample is a subset of the members of a population that is chosen to study for a research.
Researchers can collect data from all the individuals in the population if they have time and
resources, but since it is often impractical to collect these data this approach is not possible and the
researcher has to sample a portion of that population to assemble a representative population.
In this study the sample size was calculated by using Yamane (1967) formula that is a simplified
formula to determine the sample size.
N
N=
1+ N (e)2
Where:
n = Sample size
N = Population size
e = Margin of error (0.05)
Substituting the values:
135
N=
1+135(0.05)2
135
N=
1+0.3375
135
N=
1 .3375
N= 100.93
N=101
Therefore, the sample size for this study was 101 respondents.
The selected sample size was considered adequate because it represents a substantial proportion of
the target population and provides reliable data for statistical analysis.
3.2.2 Sampling Procedure
Sampling is the process of selecting respondents from the target population and the process is called
sampling procedure. The respondents were selected using simple random sampling technique from
the population. Simple random sampling was used in this study because it provides equal
probability of an employee taking part in the study. With this technique, there will be a reduced
level of selection-bias and a greater opportunity to obtain a true sample of your population. The
names of the employees working in the identified retail/ trading SMEs were taken from an
employee register and respondents were selected randomly.
The simple random sampling was adequate since the community in question had a relatively
homogenous population such as all the employees of the organization with similar work
environments and organizational characteristics. Moreover, the finds and flowcharting method
improved the fairness and objectivity of the selection procedure.
3.3 Research Instrument
Research instrument is the instrument used to gather data from the respondents. A structured
questionnaire was used as the instrument for collecting the data in this study. The questionnaire
comprised five-point Likert Scale which were closed ended questions. The questionnaire had been
separated into segments based on the research variables- Transformational, Transactional, Laissez
faire or absentee leadership and Employee performance.
CEO's from a large number of respondents could be subjected to research in an efficient and
economical way, that is why the questionnaire was chosen. It also generates standard answers which
are easily codeable for analysis with statistical packages. In addition, questionnaires have the effect
of keeping the respondents anonymous which will improve the chances of getting their honest and
truthful answer.
3.3.1 Validity and Reliability of the Instrument
Validity: how well a research tool assesses what it was designed to assess. Validity is the ability to
ensure that the conclusions that are drawn from the collected data are accurate and suitable (Hair et
al. 2022). The questionnaire was checked by the research supervisor and the academic experts, to
make it content valid, where relevance, clarity and appropriateness of the questionnaire items were
checked. Prior to the final distribution of the questionnaire, they suggestions and recommendations
were taken on board.
Reliability – it is the stability or consistency of the research instrument in measuring the variables in
the research. The reliability test was performed by the values of Cronbach's Alpha and Composite
Reliability with SmartPLS software. (Hair et al. 2022) suggest that a score of 0.70 or above is good
reliability. All the reliability results met the recommended criteria, showing that the instrument had
good reliability and reliability to collect data and analyze the data in this study.
3.4 Data Gathering Procedures
Data collection means the procedure which involves the collection of data from the respondent with
the intent of accomplishing the research goals. The researcher first secured the approval of the
relevant authorities and SMEs management of the selected SMEs. Questionnaires were designed
and sent to the respondents who were selected after obtaining approvals. The respondents were
informed about the purpose of the study beforehand.
The questionnaires were directly administered to the employees who are working in the selected
retail & trading SMEs. Questionnaires were ample in length for the respondents to complete in
time. Upon completion of the questionnaires, they were retrieved, examined for completeness and
transformed for data entry and analysis.
After collection, the data were coded and analyzed using software Microsoft Excel and SmartPLS.
3.5 Data Analysis
Data analysis is the process of organizing, summarizing and interpreting collected data by
answering the research questions and the research objectives. Data collected were subsequently
coded, pre-processed in MS Excel and then analyzed using SmartPLS 4 software. The data on
respondent characteristics and the respondents' answers were summarized using descriptive
statistics.
Indicator loadings, reliability analysis, convergent validity and discriminant validity were used to
assess the measurement model. Both Cronbach's Alpha and Composite Reliability were used to
evaluate reliability, and Average Variance Extracted (AVE) was used to evaluate convergent
validity. Fornell-Larcker Criterion and the Heterotrait-Monotrait Ratio (HTMT) were used for the
evaluation of the discriminant validity. Path coefficient, coefficient of determination (R²), effect size
(F²), predictive relevance (Q²) and model fit indices were used to analyze the structural model. The
analyses were performed to find out the impact of transformational leadership, transactional
leadership and laissez-faire leadership on performance of the employees.
3.6 Ethical Considerations
Ethical issues will be considered if research is to be carried out ethically and professionally.
Participants were asked to participate in the study and the purpose and objectives of the research
were explained prior to administering the questionnaire. The respondents were made aware of the
fact that the information to be supplied would solely be for academic purposes. The confidentiality
and anonymity of respondents was ensured throughout the study. Participants were not asked to
provide any particular information like names or contact information. All data collected was treated
confidentially and only seen by researcher and academic supervisor.
Moreover, the participants were free to either not take part in the study or drop out of it without any
drawbacks at any moment.
3.7 Limitations of the Study
Although we sought to achieve the high quality and accurate research process, there were some
limitations of the study. The study only measured retail and trading SMEs from a selected group in
the city of Mogadishu in Somalia. Hence, the results might not be representative of the other SMEs
in the other sector and regions of Somalia.
Second the study used a self-report method of data collection based on a questionnaire.
Consequently, answers may have been skewed because of some respondents' giving biased answers
or their not having understood the questions on the questionnaires. Thirdly, constraints on the
amount of time and money available narrowed the focus of the study and the number of
organizations used in the research.
Lastly, only transactional leadership, transformational leadership and laissez-faire leadership were
studied. This study was conducted on other factors determining employee performance such as
organizational culture, employee motivational factors, compensation, job satisfaction and training
programmes were not taken into consideration. The study has the limitation that it did not examine
how these factors affect retail and trading SMEs in other city areas of Somalia such as Bardera, Q
Ismailiy, Afgoyin, and Hargheis in Somaliland, but it still contained valuable findings on the effect
of leadership styles on the performance of the employees of retail and trading SMEs in the city of
Mogadishu in Somalia.