Chapter 2: Research process

2.1 Introduction
Research is described as a series of systematic steps used to investigate a problem or
opportunity that an organization faces. The general sequence includes:
1. Identifying and defining the problem or opportunity
2. Planning a research design
3. Selecting the research method
4. Choosing the sampling procedure
5. Collecting data
6. Evaluating the data
7. Preparing and presenting the research report
These steps form a broad outline that can be adapted depending on the specific project. The
overall research process is divided into three phases:
 Planning phase: from problem/opportunity identification to selecting the sampling
procedure.
 Execution phase: data collection and data evaluation.
 Report phase: preparing and presenting the research report.
While the core elements are similar, the exact number and order of steps can vary with each
research project.
2.2 Steps in the Research Process
2.2.1 Identifying and Defining the Problem or Opportunity
This is the first and most crucial step. It begins with a question or a need that arises from
curiosity or a practical situation and requires an answer.
Two key actions are involved:
1. Understanding the problem thoroughly
2. Rephrasing it into clear, meaningful research terms
Formulating the problem clarifies what exactly the researcher wants to find out, given the
available time, expertise, and knowledge. It also highlights gaps in the researcher's
understanding (e.g. of statistics or other relevant disciplines).
A major tool here is the review of literature, which means examining existing research
related to the topic. This helps the researcher:
1. Identify which variables are important or unimportant in related studies.
 2. Estimate how much work has already been done and see where further work can
extend it.
3. Organize and systematize existing knowledge, allowing clearer conclusions and
applications.
4. Clarify, redefine, and relate variables in a meaningful context for further investigation.
Literature can be found in journals, books, abstracts, indexes, and periodicals.
For businesses operating in a volatile environment, managers constantly scan the macro
environment to identify problems or opportunities (e.g. poor sales, new market chances).
Problem identification (seeing that “something is wrong” or “something is possible”) must
be followed by problem definition (stating exactly what is to be researched). A vague
problem like “sales are declining” needs to be sharpened.
Often, managers know there is a problem (e.g. new product not selling) but not exactly why.
Here, defining the problem precisely helps the researcher collect the right information and use
resources effectively.
[Link] Exploratory Research
Exploratory research is used in the early stage to understand the topic and clarify the
problem. It helps the researcher arrive at a set of questions that must be answered to solve the
problem or exploit the opportunity.
Characteristics:
 It is informal and initial.
 It involves reviewing existing studies, consulting experts, analyzing the situation, etc.
 It leads to clarity on what information is needed and how the research should proceed.
Two common tools:
 Secondary Data Analysis: using data already collected for other purposes (e.g.
magazines, journals, online articles, company reports). This can help define the
problem and suggest directions.
 Pilot Studies: small, informal studies using actual respondents to gain early insights.
These might involve casually asking for opinions or reactions to a new product or
idea.
Once the problem is clear, the researcher prepares a formal statement of research
objectives.
[Link] Preparing the Statement of Research Objectives
The objectives explain what the research aims to achieve. They:
 Must be precise, specific, brief, and limited in number.
 Serve as guidelines for all subsequent steps.
Objectives typically include:
 The research questions (e.g. “Do customers perceive the new glycerin bath soap as
harmful?”)
 And the hypotheses that will later be tested.
These objectives come from a careful analysis of the purpose of the research.
[Link] Formulation of Objectives
Research is goal-oriented. Objectives give direction to the entire study. Once they are
defined:
 All later activities (data sources, instruments, analysis tools) are designed to meet
those objectives.
For example, to examine how effectively a university examination department functions, the
researcher must clearly state objectives (e.g. quality of question setting, faculty availability,
adequacy of facilities). Based on this, the researcher knows what data to collect and how.
Good objectives are clear, specific, and definite. Vague or broad objectives weaken the entire
research process.
2.2.2 Formulation of Hypothesis
After defining the problem and reviewing the literature, the researcher formulates a
hypothesis – a tentative, testable statement about relationships between variables.
A hypothesis:
 Is a declarative statement (e.g. “Consumers perceive the new glycerin soap as
harmful to skin”).
 Suggests a relationship between variables based on prior research, theory, or
observation.
 Should be formulated before data collection, not after.
Hypotheses clarify the research question and help ensure that the research design covers all
relevant aspects. For example:
 Research question: “Why are refrigerator sales going up in winter?”
 Hypothesis: “Sales are rising due to off-season discounts.”
This makes the problem clearer and helps the researcher include, for example, questions about
discounts in the questionnaire.
Sometimes a hypothesis adds extra dimensions to a problem (e.g., adding lifestyle and
geography to the study of tea vs coffee consumption), making the study richer. However:
 Not every research question needs a hypothesis.
 Very vague or trivial hypotheses (e.g. “Sales will increase” vs “Sales will not
increase”) add little value.
  Researchers should avoid formulating hypotheses that have already been conclusively
proven elsewhere.
Before moving on, researchers must:
1. Evaluate the value of the information being sought via cost–benefit analysis. If the
cost of getting the information is higher than the benefit, they should stop.
2. Ensure that the required information doesn’t already exist; otherwise, doing new
research is unnecessary.
2.2.3 Identifying, Manipulating, and Controlling Variables
In research, especially experimental, you deal with variables—characteristics that can change
and be measured, controlled, or manipulated. Key types:
 Dependent Variable: The outcome that is predicted or explained; it changes when the
independent variable changes.
 Independent Variable: The factor that the researcher manipulates or selects to see its
effect on the dependent variable.
 Extraneous (Relevant) Variable: Other variables that may unintentionally influence
the dependent variable. The researcher tries to control them because their effect is not
of primary interest.
Proper identification and control of variables is essential to conducting valid experiments.
2.2.4 Formulation of a Research Design
Once the problem is defined and hypotheses are set, the researcher develops a research
design—a detailed blueprint of how the study will be carried out.
A research design:
 Specifies how the researcher will test relationships between independent and
dependent variables.
 Is based on earlier objectives and hypotheses.
 Includes details such as where to get information, the time period, budget,
measurement techniques, and the sampling process.
The choice of research design depends on:
 The objective of the study
 The importance of the decision
 The cost of the research
 The availability of data
Experimental design, in particular, aims to give the researcher maximum control over
independent and extraneous variables to ensure that changes in the dependent variable are
really due to the manipulation of the independent variable.
In general, a good research design:
 Explains how answers will be found.
 Sets out logistics, measurement procedures, sampling strategy, analytical framework,
and timeframe.
 Must be valid, workable, and manageable.
A poor design can lead to misleading results, so empirical studies are judged largely by the
quality of their design.
2.2.5 Constructing Device for Observation and Measurement
After choosing the research design, the researcher selects or constructs tools to collect data
and measure variables. Common instruments:
 Questionnaires
 Interview schedules
If suitable tools do not already exist, the researcher must develop them. These tools mostly
rely on asking people questions (rather than direct observation) to obtain needed information.
Good instruments ensure accurate and reliable data collection.
2.2.6 Selecting the Research Method
The researcher then decides on the basic method for the study. Main methods include:
 Surveys
 Experiments
 Secondary data studies
 Observation techniques
The choice depends on the research objectives, costs, data availability, and the urgency and
importance of decisions.
[Link] Surveys
A survey collects information from a sample of respondents using a questionnaire. It is
typically used to gather primary data (data collected for the first time for the specific study).
Surveys can be conducted:
 Face-to-face
 By telephone/mobile
 By email or online
Each mode has its own advantages and disadvantages (cost, speed, depth of interaction), and
researchers choose based on their needs.
[Link] Experiments
Experiments study cause-and-effect relationships by manipulating one variable and
observing changes in another.
Examples:
 Changing incentives (independent variable) and measuring sales performance of
salespeople (dependent variable).
 Test marketing a new product in selected areas to see how it performs.
Experiments provide stronger evidence about what actually causes changes in outcomes.
[Link] Secondary Data Studies
Here, the researcher analyzes existing data related to the research topic. For example:
 Using records on market setup, distribution networks, and past sales to forecast
demand for a new soap.
Secondary data studies often use mathematical models to project trends and are cheaper and
faster than collecting new data.
[Link] Observation Techniques
Observation involves watching respondents’ behavior without interfering.
Examples:
 Observing how shoppers move and choose products in a supermarket.
 Counting vehicles passing a junction.
Its advantage is that it does not depend on what respondents say. However, it cannot reveal
internal factors like attitudes or motivations.
[Link] Analyzing Research Designs
No single research design is perfect for every problem. The best way is often to “work
backward” from the desired end result and select a design that can provide that result most
effectively and economically.
Example:
Studying tobacco use in public places might be better done by observation than by surveys, to
save costs and avoid relying on self-reports.
Once the method is chosen, the researcher proceeds to design sampling procedures.
[Link] Selecting the Sampling Procedure
Sampling means choosing a small subset of the population to represent the whole. It’s a
critical step because decisions will be based on this sample.
Key points:
 The sample must reflect the characteristics of the population.
 Poor sampling leads to misleading results.
  The population must be clearly defined first, or research results will not help
decision-making.
 Sample size matters: larger samples tend to be more precise but are also more
expensive. A well-chosen probability sample is usually sufficient.
Types of sampling:
 Probability sampling: every element has a known, equal chance of selection; helps
ensure representativeness.
 Non-probability sampling: representativeness is not guaranteed; selection may be
based on convenience or judgment.
All steps up to sampling belong to the planning phase. After sampling, the researcher moves
to data collection in the execution phase.
2.2.7 Data Collection
With the sample prepared, the researcher actually gathers data. The choice of data collection
method depends on the research technique:
 Surveys
 Observations
 Experiments
 Case studies
…etc.
The researcher must also decide how to administer tools (individually, in groups, by mail, by
phone, online). In this phase, issues like staff recruitment, training, response rates, cost, and
data quality must be carefully managed.
Data collection errors can occur when:
 Respondents misunderstand questions.
 Interviewers record answers wrongly.
To reduce such errors, data collection usually includes:
 Pretesting: trying out the data collection plan on a small sub-sample to check for
problems, refine questions, and plan tabulation. If pretest results are unsatisfactory, the
research design may have to be adjusted.
 Main study: full data collection after fixing issues found in pretest.
2.2.8 Evaluation of the Data
After data are collected, they must be processed so they can be used for decision-making. This
involves:
 Editing: checking the data for completeness, consistency, and obvious errors (e.g.
answers in wrong columns).
  Coding: assigning symbols or numerical codes to responses so they can be entered
into a computer and analyzed.
Modern technology allows data to be entered directly into computers during interviews (e.g.
computer-assisted interviewing), reducing human error and speeding up processing.
2.2.9 Data Analysis and Interpretation
Analysis is the process of applying quantitative and qualitative techniques to the edited and
coded data to understand what they say about the research problem and hypotheses.
This can involve:
 Simple tools (frequency tables)
 More complex techniques (multivariate analysis)
 Univariate, bivariate, or multivariate analyses, depending on how many variables are
considered at once.
Statistical analysis often aims to reject the null hypothesis and support the alternative
hypothesis, using either parametric or non-parametric tests depending on data type.
The ultimate goal is to make sense of the data: to see what light they throw on the problem
and to draw conclusions accordingly.
2.2.10 Drawing Conclusion
Based on the analysis, the researcher draws conclusions and generalizes them to the
population.
In this step:
 Hypotheses are either accepted or rejected.
 Findings are related back to the theory or earlier research from which the hypotheses
arose.
 Sometimes, new findings suggest that existing theories need modification or
extension.
2.2.11 Preparing and Presenting the Research Report and Publication
This is the final major phase. The research report can be oral or written, but it should:
 Clearly describe what was done (methods),
 What was found (results), and
 What was concluded (interpretation and recommendations).
A good report usually includes:
 Objectives of the study
 Summary of research design and methods
 Major findings
 Limitations of the study
 Recommendations or implications