Methodology and design on performance of turnover tax
Marvin Brown
Liberty University
ACCT 612 - Tax Research and Jurisprudence
March 23, 2022
Research Design
Research design can be described as the plan, structure and strategy of investigation conceived
so as to obtain answers to research questions and control of variance (Kerlinger, 1964). It is a
procedure that provides answers to issues such as techniques to use to gather data, the kind of
sampling strategies and tools used and how time and cost constrain was dealt with (Cooper and
Schindler, 2003).
A descriptive research determines and reports the way things are. Descriptive data is typically
collected through a questionnaire survey, an interview or by observation. On the other hand,
exploration design addresses the need that certain inquiries focus on questions that require
answers in order to understand people, events and situations (Chandran, 2004). The descriptive
research design will be more preferred to other research designs because it reports the status of
study variables.
3.3 Target Population
The study will choose KRA staff in the Domestic Taxes Department Small Taxpayers offices in
Kisumu as the main unit of analysis. Numerous studies have dwelt on taxpayer interviews and
not the staff tasked with implementation of the Turn over Tax collections. The KRA staff will be
selected based on their extensive and direct involvement in efforts to implement TOT and their
in-depth practical understanding of the challenges faced in implementing the TOT policy in
Kenya. The sampling frame will be 235 technical KRA staff in Domestic Taxes Department in
Kisumu region.
Sample and sampling frame
Using the Krejcie and Morgan table, which also shared the same formula with Cochran (1967),
the sample size for 235 staff around Kisumu will be 147, allowing a margin of error of 5% and
with a confidence level of 95%. Questionnaires will be distributed to staff within Domestic
Taxes Department region in Kisumu County.
DataCollectionInstrumentsand procedures
The study will collect primary data with the help of a questionnaire. The questionnaire will have
both open and close-ended questions and employ the Likert scale methodology. In the Likert
scale, the respondents will be expected to either indicate their level of agreement using a five-
point scale namely; Strongly Agree, Agree, Neither Agree or disagree, Disagree and Strongly
Disagree. Secondary data will be obtained from annually turnover tax performance reports of
six-year (2013-2018) revenue authority.
Data CollectionInstruments
The study will rely on both primary and secondary data as they usually reinforce each other
(Stiles & Taylor 2001). The primary data will be collected using questionnaires while the
secondary data will be collected from existing materials within the institutions to be studied.
Secondary, means “next after the first” or “derived” (Hart 2005). He defines secondary data as
data previously assembled for some project other than the one at hand.
Primary data, mainly quantitative, will be collected through questionnaires. Questionnaire as data
collection instrument is preferred as it enables the researcher to collect data from all the
respondents within the limited time frame and responses which the respondent could have felt
shy to give in face-to-face interviews (Kerlinger, 2012). Questionnaire has advantages of low
cost, reduction in biasing error, greater anonymity, considered answers and consultations and
finally accessibility to a wide geographical contact at minimal cost (Nachmias & Nachmias,
2005). The questionnaire method has been selected as it allows the researcher to collect data
systematically and address the research issues in the standardized and economical way. In
addition, the questionnaire has been chosen because of the nature of this study which may
require consultation.
Data Analysis
The data to be generated by the study after fieldwork will be edited, coded then entered into a
computer for processing using the Statistical Package for Social Sciences (SPSS v.21.0). A
master codebook designed to ensure that all the questionnaires are coded uniformly will be used.
Consequently, data will be edited for completeness and consistency before analysis.
Descriptive and inferential statistics will be used to analyse information generated from
respondents. Descriptive statistics refers to, “simple statistical methods, which do not support or
falsify a relationship but help in the description of the data.” Thus, descriptive statistics will
enable the researcher to organize data in an effective and meaningful way. By use of
percentages, frequency distributions, tables, charts, the researcher will categorize the variables.
Y = β0 + β1X1 + β2X2 + β3X3 + β4X4 +ε
Whereby: Y = Turn over tax;
X1 = Tax payer level of education
X2= Business turnover;
X3= Enforcement effort by revenue authority
X4= Tax base expansion
Β0, β1, β2, β3, β4=Regression model coefficients.
ε = Error Term.
