RESEARCH METHODOLOGY ON JOB SATISFACTION
AND EMPLOYEE PERFORMANCE
Research Design
The study used descriptive research design in collecting and analyzing
data. Descriptive study design is a study in which the researcher describes a
picture of a phenomenon under investigation (Kombo & Tromp, 2006). This is
because of its accuracy, cost effectiveness, less repetitiveness and also allowed
proper representation of the three departments in sampling through random
stratified. The method was also appropriate in analyzing quantifiable data.
Descriptive research design involves measuring a variable or a set of variables
as they exist naturally. It is not concerned with the relationship between
variables but rather description of individual variable (Ames et al., 2017). The
method was used in order to obtain information that would help describe job
satisfaction dimensions and employee performance in the telecommunication
sector.
Target Population
A target population is defined as the total number of individuals who share
similar qualities that are being studied in the research (Kothari, 2005).
Psychometricians have defined target population based on the information
gained beforehand regarding the intended use of the study and the relevant
demographics appearing in the study (Duong & von Davier, 2012). In this
study the target population was the staff members of the technical department
at Telkom
Ltd. The technical department is based at Telkom Plaza, Nairobi. The technical
department comprises of 670 staff members.
Table 3. 1: Target population
Employees in the Technical Department Population
1 Supervisors (Managers) 148
2 Technicians 522
Total 670
Source: Author (2022)
Sampling Frame
A sampling frame is essential in outlining the members present in the target
population. In this study the sampling frame captured the different cadres and
number of employees in the technical department at Telkom Kenya Ltd.
Sample and Sampling Technique
In order to collect data sampling is vital as it ensures the data will be collected
from a representative sample of the population. Appropriate sample size for a
research study is 10% of the target population (Cooper & Schindler, 2005).
Moreover, in order to avoid sampling bias, probability methods of sampling
were used. The target population was heterogeneous as there were two
categories of employees in the technical department at Telkom Kenya Ltd.
Subsequently, stratified sampling and simple random sampling were used. The
target population was divided into two strata and then a probability sample
drawn from each group using simple random sampling. In simple random
sampling, there is an equal chance of selection into the sample as the method
forms part of probability sampling technique.
Sample Size Determination
The size of the sample will be determined through scientific methods using
the Nassiuma (2001) formula:
n = NC2
C2 + (N – 1) e2
Where n represents sample size, N the population, C the coefficient of
variation (21% ≤ C ≤ 30%), and e represents the precision level (2% ≤ e ≤
5%). Therefore, using the formula
n = 670 × 0.252
0.252 + (670 – 1) 0.032
n = 63 respondents
Sample size in strata
Employees in the
Technical Department
Population (%) Respondents Per Category
Supervisors (Managers) 148 (22.38%) 14 (22.38%)
Technicians 522 (77.62%) 49 (77.62%)
Totals 670 (100%) 63 (100%)
Source: Author (2018)
Research Instrument
A questionnaire measuring level of job satisfaction and employee
Performance was adopted as the primary data collection instrument. This
research study utilised structured questionnaires in the collection of data from
the sampled respondents at the technical department in Telkom Ltd.
Questionnaires were ideal tools for the collection of primary data in this
research study. The structured questionnaire contained open-ended and closed-
ended questions and this facilitated collection of data in tandem with the study
objectives.
Data Collection Instruments and Techniques
Primary data collection was conducted using questionnaires. The questionnaire
had both open-ended and close-ended questions. Mbwesa, (2006) defines a
questionnaire as pre- formulated written sets of questions to which subjects
record their answers independently. The questionnaire these different parts:
Part one collected the data on the respondent's age, marital status, gender,
education and category of employees. Part two requested respondent's to supply
the information on job satisfaction dimensions and employee on the
performance in the telecommunication sector. The questionnaire as a way of
collecting primary data was chosen because it is easier to administer to a large
population without the presence of the researcher. The researcher also collected
important information from the library books, previous research, internet and
other publications.
Data Collection Procedure
In order to get approval to administer the questionnaire to the respondents,
consent was first sought from the organization. Data collection was conducted
after validation of the research instrument. A letter of introduction obtained
from the university was used in order to get consent from Telkom Ltd to collect
the data. The structured questionnaires were administered to the study
respondents through the personnel depart and were collected after two weeks.
Pilot Test
In order to test whether the questionnaire was sufficient for obtaining data
for the study, it was undertaken through a pilot test. This involved
administering the developed questionnaire to a representative number of the
study respondents. This helped determine its validity and reliability. A pilot
study was used in pre-testing the research instrument. The sample size
appropriate for the pilot test is normally 10% of the total number of
respondents (Baker, 1994). The target respondents for the study were 63
employees and 6 employees were engaged in the pilot study.
Reliability Test
The pilot test findings were tested for reliability in order to determine internal
consistency. This was done using the Cronbach alpha test. Warrens (2015)
assert that a reliable research instrument produces consistent results after being
administered to different samples drawn from the target population. A
questionnaire achieving at least 0.7 in the Cronbach alpha test is deemed
reliable while that producing a value below 0.7 indicates that it needs to be
revised. It is important to determine the reliability as it ensures that the
objectives set for the study are attained.
Validity Test
Validity in qualitative research defines the appropriateness of the tools,
processes and data that is utilised in the study (Leung, 2015). Validity explores
whether the research question is ideal for the desired outcome and if the
methodology adopted is appropriate in answering the research questions.
Content validity was achieved by ensuring that the content of the questionnaire
was in harmony with the study objectives. Construct validity which consist of
divergent and convergent validity was determined by carrying out a pilot study.
Data Processing and Analysis
The filled-in questionnaires were edited for consistency. The data generated
will be quantitative. Therefore, descriptive analysis techniques were used;
consistent with the
