Research Design

Introduction
A researcher’s First challenge is to identify and properly formulate a research problem.
However, to carry out a study successfully, a researcher must also determine a suitable research
design (Asenahabi, 2019). So, prior to delving into a study, the researcher must prepare a plan of
action that outline how the research will proceed.
Before constructing any building, it is essential to know what type of building is required (de
Vaus, 2001, as cited in Bouchrika, 2024): a high-rise, a factory, a school, office complex,
shopping mall, etc., same apply to research. Before starting a study, a researcher must determine
the type of study they are conducting; and that will enable them determine the most suitable
research design.
The research design influence all aspects of a study research; as well as the value of the outcome
and conclusion of the study (Creswell et al., 2018). So, it is critical to ensure that the research
design adopted is appropriate.
Hence, the challenge a researcher face after identifying and formulating a research problem, is to
show how data will be collected, analysed and interpreted. They must determine an adequate
research design that will guide how the research subjects will be brought into the scope of the
study; how they will be situated within the research setting; type of data to be collected and how
they will be collected and analyzed; hence, the results (Creswell et al., 2018).
Definition of Research Design

Research design is the framework that guide the collection and analyses of data required to solve
a research problem. It is the conceptual blueprint that maps the stage-by-stage activities within
which an inquiry is to be conducted. Creswell (2018) defines research design as the overall plan
that provide specific direction and procedures in a research.

The research design help researchers by determining the “best approach” to take, in their efforts
to discover or create new knowledge. In the view of Asenahabi (2019), research design is a plan
adopted by a researcher to guide their data collection and analyses efforts aimed at achieving the
objectives of study in a valid way.
Research design also determine what data is required and what methods are to be used to collect
and analyze the data. Thus, research design helps to prevent frustration by binding the research
together via a structured plan that show how the major parts of a study work together to address a
research problem (Asenahabi, 2019).
Research design also specify strategies for interpreting results to provide adequate basis for
drawing conclusions and making recommendations or showing the implications of a study.
Jongbo (2014) states that an inquiry can be categorized as a research only if it is based on a
precise research problem and implemented based on a valid research design.
Types of Research Design

There are two broad types of research design from which a researcher must decide the most
appropriate for their study:
a) Experimental research design; and
b) Quasi-experimental research design.

Experimental research design: An experimental research is a highly controlled and futuristic

study conducted in a contrived setting wherein, the researcher exercise control over the research
variables and elements. An experimental research design therefore, represents a framework
created to carry out an experimental research.
In an experimental research, one or more variables are manipulated to measure their effect on
other variables. Jongbo (2014) states that experimental research focuses on cause-and-effect
relationship on carefully selected and manipulated subjects.
An experimental research often involve two groups: an experimental group and a control group.
The experimental group is the one that is manipulated while the control group is not
manipulated. Thus, Sjøberg et al. (2007) contend that experimental research is conducted when a
researcher needs to control the research situation with direct, precise and systematic
manipulation of the behaviour of the phenomenon under study. Also, experimental research is
preferred in situations where testing a theory or hypothesis is the focus of a study (Wabwoba &
Ikoha, 2011).
Experimental research is mostly conducted in the basic sciences where it is possible to control or
manipulate the variables and elements of a research. Creswell (2018) identify the following as
different types of experimental research design:
a) Pre-experimental design - where the researcher studies a single group and provides
an intervention during the experiment. In this case, there is no control group to compare
with the experimental group.
b) Semi-experimental design - where the researcher uses control and experimental groups
but does not randomly assign participants to groups.
c) True experimental design - where the researcher uses control and experimental groups
and randomly assign participants to the different groups.
d) Single-subject design - which involve observing the behavior of a single individual or a
small number of individuals over time.

Quasi-experimental research design: In quasi-experimental research, the research is conducted

in a non-contrived setting, and the researcher do not exert control over the variables and elements
of the research (Abutabenjeh, 2018). In the quasi-experimental research design, a study is
conducted in the natural setting. This type of research is mostly conducted in the social sciences.
Quasi-experimental research design has many variants. Herein, we identify the following, based
on the goals social science research seek to achieve:
a) Exploratory research design;
b) Descriptive research design;
c) Ex post facto research design;
d) Correlational research design;
e) Explanatory research design; and
f) Causal research design.
Exploratory research design: This is a type of research design used when the purpose of a study
is to discover more about a problem or phenomenon. Often, insufficient information exists about
a phenomenon, and in such instances, exploratory studies are conducted to find out more. In
essence, exploratory research design is used to guide studies that seek more or better
understanding of the nature of a problem or phenomenon. Exploratory studies mostly collect data
through:
Interview:
a) Interviews: An interview is a data method that relies on asking questions. Interviews
involve two or more people, one of whom is the interviewer asking the questions. An
interview may be structured, unstructured or semi-structured. Structured interviews have
predetermined questions asked in a predetermined order. Unstructured interviews are
more free-flowing; while semi-structured interviews fall in between.
b) Direct observation: This involves observing and collecting data on a population or
phenomenon without manipulating variables or controlling conditions. It can be
conducted in the natural setting or a controlled setting; and can be done physically or
through the use of technology.
c) Focus group interaction: This involves bringing together a small group of people to
discuss a particular topic, under the moderation of a researcher. The interaction is often
recorded for later analysis.
d) Ethnographic study: This involves conducting detailed observations of a particular
culture or community by immersing in the research context. It is often used to gain deep
understanding of the beliefs, behaviors, and practices of a group.
Descriptive research design: This is a type of research design that guide studies that seek to
describe a problem or phenomenon of interest (Khanday & Khanam, 2019). It is the research
design relied upon when a research problem involves describing the “who, what, where, when
and how” of a phenomenon or members of a population, instead of the why.

In a study where a researcher is interested in providing a comprehensive and accurate picture of

situation or phenomenon, they are acting under a descriptive research design (Bouchrika, 2024).
So, descriptive studies also seek to unravel conditions or relationships that exist; practices that
prevail; believes, attitudes or views held by a population; effects being felt and developing
trends.
Descriptive design provides a researcher with an opportunity to gain insight into the problem
itself. It builds on the groundwork made by exploration, such as providing additional
information, filling in gaps in knowledge, or expanding it. Unique to descriptive research is that
it also aims to collect as much data as possible (Bouchrika, 2024). Example of descriptive
research is market research; wherein, an investor tries to study the current state of a market, and
trends within it.

Descriptive studies mostly collect data using structured questionnaire, rather than observing how
a phenomenon operate in specific situations. Descriptive studies may be:
i) Comparative – when they examine or compare differences that exists between two or
more variables or groups.
ii) Retrospective – if the data analyzed is either collected after the occurrence of a
phenomena; or, if the study focuses on the past.
iii) Prospective – if the data collected and analyzed, focuses on the present.
Descriptive studies collect data through:
a) Survey: A survey is a method of obtaining large amounts of data from a population, in a
relatively short period time (Asenahabi, 2019). A survey is concerned with the present
and attempts to determine the status of a phenomenon being investigated (Singh, 2006).
A survey can be cross-sectional or longitudinal. Cross-sectional means that observations
are done at one or more point in time while longitudinal study means the observations are
carried out at different points in time. Cross sectional research design is descriptive,
exploratory and explanatory while longitudinal survey focus on trend analysis cohort
design and panel design (Jongbo, 2014).
b) Case study (action research): Case study is a design of inquiry in which a researcher
undertakes an in-depth analysis of a process, person, household, organization, group,
industry, culture or nationality to produce immediate and concrete solution to their local
problems (Mugenda & Mugenda, 2003). A case study seeks to provide solutions that
have practical value to the group or firm whose problems the research seek to solve.
Other ways of collecting data in a descriptive study are direct observation, focus group
interaction and ethnographic studies.
A descriptive research design may be adopted when a research intends to:
a) provide better understanding of a population or phenomenon
b) describe relationship between variables
c) describe patterns and trends
d) validate sampling methods and determine the best approach for a study
e) compare data from multiple sources.

Ex post facto research design: This is the research design adopted when the purpose of a study
is to report events or conditions that occurred in the past. Ex post facto studies attempts to
establish facts in order to arrive at a conclusion concerning past events or to predict future
events. It essentially determines and objectively locates, collect, evaluate and synthesizes data
related to past occurrences in order to establish facts or test hypotheses concerning causes,
relationships or trends of those events they may provide better understanding and explain present
events or predict future events. Ex post facto studies rely on data from;
a) Eyewitness accounts (phenomenological research);
b) Documents or records of the past; and
c) Relics (physical remain).
Correlational research design: This is a type of research design that guide investigation into the
extent to which two or more variables have quantitative linear relationship. Correlational studies
investigate if correlations exists between variables, without determining why the correlation exist
(Sassower, 2017, as cited in Bouchrika, 2024). That is, correlational studies basically address the
question of:
a) Is there a correlation between two or more variables or sets of data?
b) What is the direction of the correlation?
c) What is the magnitude of the correlation?
As a non-experimental technique, correlational studies evaluate relationship between variables
using statistical analysis. So, to conduct a correlational study, at least two (a predictor and
criterion) variables must be involved; and statistical techniques will be relied upon to calculate
the direction and extent of relationship between the variables.
Explanatory research design: As name suggests, explanatory research aims to explain a
condition, situation, or phenomenon. It seeks to explain why a condition or situation exists; why
a phenomenon occurs or why one variable relates to another. Explanatory studies are often
conducted to generate ideas to expand a theory. Using this research design, the researchers
explore the limits and boundaries of a subject in order to present results that answer the what,
how, and why of a research problem.
Researchers conduct explanatory research to determine the underlying problem or a new angle to
a problem; which may not always be readily apparent when formulating the research problem.
However, explanatory studies do not seek to provide conclusive answers; they provide avenues
for researchers to plumb the depths of the subject (Bouchrika, 2024). Like descriptive and
correlational studies, explanatory studies also collect data using structured questionnaire.
Causal research design: This type of research design guide studies that examine cause-and-
effect relationship between two or more variables or events. That is, if a change in one variable
or event cause a change in the behaviour of another variable or event. Researchers resort to
causal research when they seek to determine the impact of variable or event on another. Causal
research may be retrospective or prospective.
Causal retrospective studies involves identifying “effects”, and investigating their possible
causes. This approach requires the researcher to start investigation on a problem when the effects
have already occurred, and a need to determine the causes of those effects exists.
Causal prospective studies are those in which a researcher attempts to determine the effects of
given conditions or phenomena or one variable on another. This types of study starts with causes,
and tries to investigate their effects.
Features of a sound research design
Several features demarcate a sound research design. Bouchrika (2024) identify the following:
a) Goals-directedness: A sound research design has one purpose: to collect and analyze data
that address the research problem clearly, accurately and unbiasedly. Achieving this
means successfully specifying what a research seeks to do: to explore, describe, or
evaluate a phenomenon. In any research, the design is basic because everything
eventually emanates from the research design (Vogt et al., 2012). With the right research
design, research results will have less errors; and inferences drawn in the study will be
deemed unbiased.
b) Neutrality: Naturally, formulating a research problem comes with a measure of
assumptions, which is why there is are hypotheses. However, good research design
ensures that results obtained, are as neutral and as objective as possible. A sound research
design allow the researcher to analyze and interpret data without biases.
c) Validity: A sound research design is valid. It indicates the tools and techniques by which
results can be measured. If the design is sound, these tools will be correct and
appropriate, and can help the researcher to measure results.
d) Reliability: A sound research design makes it possible to produce similar results
repeatedly. This means that a sound research design creates an opportunity to establish
standards for collecting and analyzing data and producing identical results.
e) Generalizability: A research design that is sound can be generalized. It can be applied not
only to a part of a study, population or setting; but to all parts with the same measure of
accuracy.
In addition, a sound research design is flexible, appropriate, and efficient. It minimizes bias and
collects and analyses data objectively. Further, a sound design gives the smallest experimental
error, and yields maximal information about different aspects of a research problem.
Research approaches
To achieve the fundamental objectives of a study, a researcher may adopt a quantitative or
qualitative research approach or a combination of both.
Quantitative approach
A quantitative research approach is one that adopts technique and measurements that produce
quantifiable or discrete values (Kothari, 2007). This approach is most suitable when the goal of a
research is to produce actionable insights through statistical analyses.
Objectivity is highly prized in a quantitative research approach. Thus, researchers go to great
lengths to ensure that their results are untarnished by their own presence, behavior, expectations,
biases or presumptions.
Quantitative researchers believe that there is an objective reality which is independent of the
human mind (Rovai et al., 2014); and that for the purpose of a study, this objective reality has to
be broken down into small manageable pieces which should form the research objectives. In
order to approach a study quantitatively, researchers often start with one or more hypotheses on
the relationships between the variables they want to investigate.
Qualitative approach
Qualitative research emphasizes exploring and understanding the meaning a person or group of
people ascribe to a social or human problem (Creswell, 2018). A qualitative research also tries to
determine relationship between variables by recording, analyzing and discovering the
interconnectedness that underpins related subjects (Bouchrika, 2024).
As opposed to the quantitative, the qualitative approach tries to arrive at the conclusion of a
study through induction. Also called “bottom-up" approach, the qualitative approach infer
meaning on the basis of qualitative data collected.
Qualitative research uses open-ended questions and produces non-quantifiable data. Its main
focus is to generate meaning, purpose or reality from opinions and experiences of participants
(Merriam, 2009, as cited in Asenahabi, 2019).
Qualitative research is employed extensively in the social sciences, especially when the purpose
is to observe and uncover the social constructs on which human societies are framed.
Qualitative studies generate large amounts of raw data through phenomenological study, direct
observation, ethnographic study, case study, focus group interaction, etc. In view of the data
collection methods of qualitative research, Blaikie (2009) argue that it is mostly suited to
exploratory studies.

Qualitative research according to Rovai et al. (2014), is inductive in nature and is based on
several underlying assumptions:
a) reality is a social construct;
b) variables are difficult to measure, complex and interwoven;
c) there is a primacy of subject matter; and
d) data collected will consist of an insider’s viewpoint.

Mixed-method
The quantitative and qualitative approaches ought not to be considered as strict, divergent
dichotomies or categories. Rather, they should be seen as representing distinctive ends on a
continuum (Creswell, 2018). In a mixed-method approach, the researcher uses any or both
quantitative and qualitative approaches as they deem fit for their study.
Burke-Johnson et al. (2007) view the mixed-method as a practical approach to research wherein,
the researcher combines elements of qualitative and quantitative research approaches for the
broad purposes of breadth and depth of understanding and corroboration. In a mixed-method
research, qualitative data collection tools bring in data without predetermined responses while
quantitative research brings in closed-ended data (Creswell, 2018); these are then triangulated in
order to get to reality as close as possible.
The mixed-method approach care little about which approach is better. Instead, it uses the
method that appears to best suit the task at hand. As a result, pragmatists use whatever tools,
reasoning and techniques that are appropriate to the situation without worrying over what
approach they are using. Central to this idea is that they recognize that each approach has its
strengths and weaknesses, but can also complement each other.
Types of mixed method research
Asenahabi (2019) identify three basic types of mixed method research:
a) convergent parallel mixed method;
b) explanatory sequential mixed method; and
c) exploratory sequential mixed method.
Convergent parallel mixed methods: In a convergent parallel mixed methods, the researcher
converges or merges quantitative and qualitative data in order to provide a comprehensive
analysis of the research problem. Creswell (2018) explain that convergent parallel mixed method
studies collect both quantitative and qualitative data, analyzes them separately, and then
compares the results to see if the findings confirm or disconfirm each other.
Explanatory sequential mixed methods: This is a mixed method approach in which the
researcher first conducts quantitative research, analyzes the results and then builds on the results
to explain them in more details with qualitative research. According to Creswell (2018)
explanatory sequential mixed method involves a two-phase project in which the researcher
collects quantitative data in the first phase, analyzes the results, and then uses the results to plan
(or build into) the second, qualitative phase.
Exploratory sequential mixed methods: In the exploratory sequential approach, the researcher
begins with a qualitative research to explore the views of participants. The data are then
analyzed, and the information used to build into a quantitative research (Creswell, 2018). The
qualitative phase is used to:
a) build an instrument that best fits the variables and sample under study;
b) identify appropriate instruments to use in the follow-up quantitative phase, or
c) specify variables that need to go into a follow-up quantitative study.