EPCE 5203 Engineering Research

Methodology
CH-3: Research Design and Data collection
3.1 Meaning of Research Design
• Research design refers to the overall plan or blueprint for conducting a research
study. It outlines the methods and procedures for
• collecting,
• analyzing, and
• interpreting data.
• The design ensures that the research question is answered as accurately and
efficiently as possible. A well-thought-out research design provides structure to
the research and ensures that the findings are valid, reliable, and relevant to the
research objectives.
• Key Elements of Research Design:
1. Purpose of the Research: Whether the study is exploratory, descriptive, explanatory, or
evaluative.
2. Research Question or Hypothesis: The main question or hypothesis the research is
attempting to answer or test.
Cont…
3. Variables: The elements being studied, such as independent variables
(cause) and dependent variables (effect) in an experiment.
4. Population and Sampling: Who or what is being studied (the
population) and how individuals or elements are selected (sampling
techniques).
5. Methods of Data Collection: The approach used to gather data, such as
surveys, interviews, observations, or experiments.
6. Data Analysis: The techniques used to analyze the collected data, such
as statistical analysis for quantitative data or thematic analysis for
qualitative data.
7. Timing of the Study: Whether the research is cross-sectional (data
collected at one point in time) or longitudinal (data collected over time).
3.2 Need for Research Design
• The need for research design is essential because it ensures that a
research project is systematic, organized, and able to produce valid,
reliable, and interpretable results. A solid research design is crucial for
the following reasons:
1. Clear Focus and Direction:
• A research design helps clarify the main research question or hypothesis,
giving the study a clear focus. It outlines the specific steps that will be taken,
ensuring that the research remains aligned with its objectives.
• Without a well-defined design, research efforts can become disorganized,
leading to incomplete or irrelevant data collection.
Cont…
2. Accuracy and Validity of Results:
• A good research design ensures that the methods used to collect and analyze
data are appropriate for answering the research question. This helps produce
valid results that reflect the true nature of what is being studied.
• It minimizes bias and errors, improving the accuracy and credibility of the
research findings.
3. Efficient Use of Resources:
• Research can be time-consuming and expensive, so a well-structured design
ensures efficient use of resources, such as time, effort, and money.
• It helps avoid unnecessary steps and data collection that might be irrelevant
to the research goals, saving resources in the process.
Cnt…
4. Minimizes Confusion and Complexity:
• A research design provides a structured approach to data collection, analysis,
and interpretation, reducing the risk of confusion during the study.
• It ensures that complex research projects remain manageable and follow
logical steps, preventing researchers from becoming overwhelmed by the
complexity of the study.
5. Enhances Reliability and Replicability:
• Research design ensures consistency in how the study is conducted, which
enhances its reliability. If other researchers wish to replicate the study, they
can follow the same design to verify or expand upon the findings.
• Replicability is a key aspect of scientific research, contributing to the
generalization of findings across different settings or populations.
Cont…
6. Ethical Considerations:
• Research design also helps in addressing ethical concerns, ensuring that
participants are treated ethically and that the research complies with ethical
standards.
• By planning ahead, researchers can identify potential ethical issues related to
confidentiality, consent, and harm, and take steps to mitigate them.
7. Helps Handle Unexpected Challenges:
• A solid research design anticipates possible challenges (such as difficulties in
data collection or unexpected results) and includes strategies to deal with
them.
• Having a plan in place helps researchers adapt to unforeseen obstacles
without losing track of their overall research goals.
Cont…
8. Guides Data Analysis:
• The design determines how the collected data will be analyzed and
interpreted, ensuring that the right statistical or qualitative methods are
applied.
• Proper analysis is crucial for drawing valid and meaningful conclusions from
the data.
9. Increases Confidence in Findings:
• A well-constructed design enhances the credibility of the research findings,
both for the researcher and for the broader academic or scientific community.
• It provides a logical and transparent path that others can follow to understand
how the research was conducted and why the conclusions are justified.
Cont…
10. Supports Decision-Making:
• Particularly in applied research, a strong design helps ensure that the
research findings are actionable and can be used to inform decision-making,
whether in business, policy, or other fields.
• It provides evidence-based insights that stakeholders can rely on.

In summary, a well-thought-out research design is critical for ensuring

the success of any research project. It lays the foundation for a
structured, reliable, and ethical study, helping to produce results that
can be trusted and utilized effectively.
3.3 Features of a good design
• A good research design is critical for conducting a successful and
credible study.
• It ensures that the research results are valid, reliable, and applicable.
• Here are the key features of a good research design:
1. Clarity of Purpose:
• The design should clearly define the research objectives and questions. A
well-designed study has a focused purpose, avoiding ambiguity about what
the research aims to achieve.
• The research problem, hypothesis, or questions should be precisely stated to
ensure that all elements of the design align with these goals.
Cont…
2. Rigorous and Systematic:
• A good research design must be logical and methodical, following a structured
approach. It should have clearly defined steps that help achieve the research
goals systematically.
• The design should allow for rigorous testing of hypotheses, ensuring that the
research methods and procedures are scientifically sound.
3. Minimization of Bias:
• The design should aim to reduce or eliminate biases in the research process,
such as selection bias, measurement bias, or response bias. This ensures that
the results are more accurate and represent reality.
• Randomization, blinding, and control groups in experimental designs are
common techniques used to minimize bias.
Cont…
4. Control of Confounding Variables:
• A good research design ensures that confounding variables (factors other than the
independent variable that might influence the dependent variable) are controlled.
This is important for establishing a cause-and-effect relationship.
• Control techniques, such as using control groups, matching, and statistical controls,
help ensure that the research findings are valid.
5. Reproducibility and Reliability:
• The design should enable other researchers to replicate the study under similar
conditions and obtain consistent results. A reliable study consistently produces the
same results when repeated.
• Detailed descriptions of the research methods, sample, and procedures ensure that
others can follow the same design.
Cont…
6. Internal Validity:
• The research design should ensure that the results are valid by accurately reflecting
the relationship between the variables studied. Internal validity refers to the extent
to which the design eliminates alternative explanations for the results.
• In experimental research, internal validity is often strengthened through
randomization and controlling extraneous variables.
7. External Validity (Generalizability):
• A good research design allows the results to be generalizable to a broader population
beyond the sample studied. The sample size and method of selection play a critical
role in determining external validity.
• The research should be applicable to real-world situations or other contexts,
ensuring that the findings have relevance beyond the study environment.
Cont…
8. Appropriateness of Methods:
• The research methods used (qualitative, quantitative, or mixed methods) should be
suitable for answering the research question or hypothesis. Different questions may
require different types of data collection and analysis techniques.
• The design should ensure that the chosen methods are appropriate to the nature of
the data, sample, and overall research goals.
9. Ethical Considerations:
• The design should incorporate ethical principles, such as obtaining informed consent,
ensuring participant confidentiality, and minimizing harm. Research that involves
human or animal subjects should adhere to ethical guidelines to protect participants’
rights and well-being.
• Ethical considerations also involve transparency in the research process, including
disclosing conflicts of interest and ensuring that results are reported honestly.
Cont…
10. Flexibility:
• While research design should be structured, it should also be flexible enough
to accommodate unforeseen challenges or changes that might arise during
the study.
• A good design allows for adaptations or modifications without compromising
the study's overall integrity or objectives.
11. Efficient Use of Resources:
• The design should ensure that the research is conducted efficiently in terms
of time, money, and other resources. A well-planned design helps avoid
unnecessary steps or delays in the research process.
• Effective allocation of resources ensures that the research is feasible and
sustainable within the available budget and timeline.
Cont…
12. Accurate Data Collection and Measurement:
• The methods and instruments used for data collection should be reliable and valid,
ensuring that the data accurately represents the variables being measured. The
design should define clear procedures for measurement to reduce errors.
• For example, standardized surveys, reliable experiments, or validated scales are
common ways to ensure data accuracy.
13. Comprehensive Planning for Data Analysis:
• A good design includes a well-thought-out plan for how the data will be analyzed.
The chosen statistical or analytical methods should be appropriate for the data type
and research question.
• This feature ensures that the results of the study can be interpreted meaningfully
and contribute to answering the research question.
Cont…
14. Realistic and Feasible:
• The research design should be practical, considering the scope of the study
and available resources. It should outline a realistic timeline and workload for
the research team.
• Unrealistic designs that overlook practical constraints (like budget, participant
availability, or time) may fail to deliver reliable results.

In summary, a good research design is clear, systematic, ethical, and

adaptable. It reduces bias, controls confounding variables, and ensures
that the research methods align with the research questions, allowing
for valid, reliable, and generalizable findings.
3.4 Different Research Designs
• There are several different types of research designs, each suited to specific
kinds of research questions and objectives. These designs can broadly be
categorized into quantitative, qualitative, and mixed-methods approaches,
depending on the nature of the research.
• Here’s an overview of the major types of research designs:
1. Quantitative Research Designs
• Quantitative designs focus on numerical data, measurable variables, and statistical analysis.
They are often used to test hypotheses, measure variables, and examine relationships
between them.
a. Descriptive Research Design
• Purpose: Describes characteristics or functions of a population or phenomenon being
studied.
• Example: Surveys, case studies, or observational studies that provide details about a group
without manipulating variables.
• Uses: Market research, demographic studies, public health assessments.
Cont…
b. Correlational Research Design
• Purpose: Examines the relationship between two or more variables to determine whether
they are correlated (i.e., whether they move together).
• Example: A study investigating the correlation between hours of study and academic
performance.
• Uses: Studies where variables can’t be manipulated (e.g., social science, behavioral research).
c. Experimental Research Design
• Purpose: Tests causal relationships by manipulating one variable (independent variable) and
observing its effect on another (dependent variable).
• Example: A clinical trial testing the effectiveness of a new drug by comparing it to a placebo.
• Features: Randomization, control groups, and manipulation of variables are common
features.
• Uses: Psychology, biology, medical studies, and other sciences aiming to establish cause and
effect.
Cont…
d. Quasi-Experimental Research Design
• Purpose: Similar to experimental design but lacks random assignment to control and experimental
groups. It examines cause-and-effect relationships when randomization is not possible.
• Example: A study comparing the effects of different teaching methods on student performance in
different schools (where schools are pre-assigned to a group).
• Uses: Education, social sciences, and other fields where ethical or practical constraints prevent
randomization.
e. Longitudinal Research Design
• Purpose: Studies the same subjects or phenomena over an extended period to observe changes or
developments over time.
• Example: A long-term study tracking health outcomes in individuals who quit smoking.
• Uses: Medical research, developmental studies, and behavioral sciences where changes over time are
significant.
f. Cross-Sectional Research Design
• Purpose: Collects data at one point in time from a sample that represents a larger population. It
provides a "snapshot" of variables at a specific moment.
• Example: A survey measuring voter preferences before an election.
• Uses: Public opinion polls, epidemiological studies, and market research.
Cont..
2. Qualitative Research Designs
• Qualitative designs focus on understanding social phenomena through non-
numerical data, such as interviews, observations, and texts. These designs are
exploratory and focus on depth and context.
a. Ethnography
• Purpose: Involves in-depth study of people or cultures through direct
observation, participation, and immersion in the environment being studied.
• Example: A researcher living in a remote village to study social customs and
daily life.
• Uses: Anthropology, sociology, and cultural studies.
Cont…
b. Case Study Research Design
• Purpose: Provides an in-depth exploration of a single case or a small number of cases
in a real-world context.
• Example: Analyzing the business strategies of a successful company.
• Uses: Business research, psychology, law, education, and other fields where intensive
study of one or a few cases is insightful.
c. Phenomenological Research Design
• Purpose: Focuses on understanding lived experiences from the perspective of
individuals experiencing a particular phenomenon.
• Example: Studying the experience of grief among people who have lost a loved one.
• Uses: Psychology, nursing, and social sciences that focus on subjective human
experiences.
Cont…
d. Grounded Theory
• Purpose: Aims to generate theory grounded in data from participants.
Researchers collect data and develop a theory based on observed patterns.
• Example: Developing a theory on how people adapt to workplace stress by
conducting interviews.
• Uses: Sociology, organizational studies, and psychology.
e. Narrative Research Design
• Purpose: Collects and analyzes stories or personal accounts to understand
how individuals interpret their life experiences.
• Example: Gathering life histories from war veterans to understand the impact
of war on personal identity.
• Uses: History, psychology, education, and literature.
Cont…
3. Mixed-Methods Research Design
• Mixed-methods designs combine both quantitative and qualitative
approaches to provide a comprehensive understanding of the research
problem.
a. Convergent Parallel Design
• Purpose: Collects both quantitative and qualitative data simultaneously,
analyzes them separately, and then merges the results to provide a
comprehensive interpretation.
• Example: Conducting surveys (quantitative) and interviews (qualitative) to
study customer satisfaction.
• Uses: When researchers want both numerical data (what happens) and
contextual data (why it happens).
Cont…
b. Explanatory Sequential Design
• Purpose: Begins with quantitative data collection and analysis, followed by
qualitative data collection to further explain or clarify the initial results.
• Example: A study first surveys a population (quantitative), then interviews a subset
to understand why specific survey patterns emerged.
• Uses: Ideal for research that needs detailed explanation or interpretation of
numerical findings.
c. Exploratory Sequential Design
• Purpose: Starts with qualitative data collection to explore a phenomenon, followed
by quantitative data collection to test or measure the initial findings.
• Example: Conducting focus groups to identify issues and then designing a survey
based on the qualitative findings to quantify the issues.
• Uses: When little is known about a topic and the researcher wants to explore first,
then quantify findings.
Cont…
4. Other Specialized Designs
a. Action Research
• Purpose: Involves a cycle of planning, acting, observing, and reflecting to solve a
problem or improve practice, often in collaboration with the subjects being studied.
• Example: Teachers using action research to improve classroom practices.
• Uses: Education, social work, community development.
b. Ex Post Facto Research Design
• Purpose: Studies events or phenomena retrospectively, where the cause and effect
are already established. Researchers can’t manipulate variables but look back at the
data to understand relationships.
• Example: Examining the relationship between exposure to an environmental toxin
and the development of a health condition.
• Uses: Public health, epidemiology, and policy research
 Cont…
Summary of Major Research Designs
Research Design Main Focus Data Type
Descriptive Describe characteristics Quantitative
Examine relationships between
Correlational Quantitative
variables
Experimental Test cause-and-effect relationships Quantitative
Quasi-Experimental Test effects without full control Quantitative
Longitudinal Study changes over time Quantitative
Cross-Sectional Snapshot at a single point in time Quantitative
Cultural understanding through
Ethnography Qualitative
immersion
Case Study In-depth study of one/few cases Qualitative
Grounded Theory Develop theory from data Qualitative
Combine qualitative and
Mixed Methods Mixed
quantitative data
Each of these designs is suited to different types of research questions and contexts. The choice of design
depends on the nature of the research problem, the objectives, and the type of data needed to answer the
research questions.
3.5 Collection of Primary Data and Secondary
Data
• When conducting research, data can be collected from two main sources:
primary data and secondary data.
• Both types of data have distinct characteristics and methods of collection.
1. Primary Data:
• Primary data refers to the original data collected firsthand by the researcher
specifically for the purpose of the study. This data has not been previously published
or used by other researchers. Primary data collection allows researchers to address
their specific research questions and objectives in a direct and personalized way.
2. Methods of Collecting Primary Data
There are various methods for collecting primary data, depending on
the research design and nature of the study. Some common methods
include:
Cont…
a. Surveys and Questionnaires
• Description: Involves asking questions of participants in written form (online, paper, or
electronic) to gather their responses.
• Types: Can include close-ended questions (e.g., multiple choice) for quantitative data or
open-ended questions for qualitative insights.
• Examples: Market research surveys, employee satisfaction questionnaires.
b. Interviews
• Description: Involves direct, one-on-one conversations between the researcher and the
participant, either structured (pre-determined questions) or unstructured (open discussion).
• Types:
• Structured interviews: Pre-set questions, focused on quantitative data collection.
• Semi-structured or unstructured interviews: Open-ended questions, focused on gathering
qualitative data.
• Examples: Interviews for gathering customer feedback, job interviews, in-depth interviews in
social research.
Cont…
c. Observations
• Description: Involves systematically watching or recording behavior, events, or conditions in
their natural setting without direct interaction with participants.
• Types:
• Participant observation: Researcher becomes part of the group being studied.
• Non-participant observation: Researcher observes from a distance without interacting.
• Examples: Observing classroom behavior in educational research, watching consumer
behavior in a store.
d. Experiments
• Description: Involves manipulating one or more variables under controlled conditions to
observe their effect on a dependent variable.
• Types:
• Laboratory experiments: Highly controlled, conducted in a lab setting.
• Field experiments: Conducted in a natural setting but with control over certain variables.
• Examples: Clinical trials, psychology experiments.
Cont…
e. Focus Groups
• Description: Involves gathering a small group of people (usually 6-12) to
discuss a specific topic, facilitated by a moderator. It helps collect group
opinions, ideas, and perceptions.
• Examples: Focus groups in product development, policy evaluation.
f. Social Media and Online Platforms
• Description: Collecting data through online channels, including social media
posts, user comments, and online feedback tools.
• Examples: Analyzing user comments on a product or gathering data from
Twitter or Facebook about public opinion on a trending topic.
Cont…
2. Secondary Data
• Secondary data refers to data that has already been collected and published
by others. It was originally gathered for purposes other than the current
research but can be reused and analyzed for new purposes. This data is
usually readily available and can come from a variety of sources.
• Sources of Secondary Data
• The sources of secondary data are wide-ranging and can come from the following:
a. Published Sources
• Government Reports and Publications: These include census data, health
statistics, or national economic reports.
• Examples: U.S. Census Bureau reports, World Bank data, United Nations
publications.
Cont…
b. Books and Journals
• Academic Books and Journals: These contain research findings, theories, and
data published by scholars, often used for literature reviews.
• Examples: Articles in scholarly journals like Nature, Harvard Business Review,
or research papers published in online databases.
c. Institutional and Organizational Reports
• Description: These include reports generated by businesses, research
institutes, and organizations such as annual reports, market research findings,
and internal studies.
• Examples: Industry reports from companies like Gartner or McKinsey, or
internal organizational data like sales reports.
Cont…
d. Databases and Repositories
• Description: Access to structured data that has been compiled and stored in
databases for use by researchers.
• Examples: Economic databases like Eurostat or global health data from WHO,
or research data stored in repositories like Google Scholar, JSTOR.
e. Websites and Social Media
• Description: Information from websites, blogs, and social media platforms
can serve as secondary data for analyzing trends, opinions, and behaviors.
• Examples: Analyzing product reviews from Amazon, tracking trends on
Twitter, or gathering website traffic data.
Cont…
f. Media Sources
• Description: Newspapers, magazines, television broadcasts, and other media
outlets provide secondary data that reflect current events, public opinions,
and trends.
• Examples: Articles from The New York Times, televised interviews on news
channels, or editorial articles from magazines.
g. Historical Records and Archives
• Description: Historical records such as letters, photographs, official
documents, and census records can serve as valuable secondary data sources.
• Examples: Archival documents from national archives, historical census data,
or ancient manuscripts stored in libraries.
 Cont…
Key Differences Between Primary and Secondary Data
Aspect Primary Data Secondary Data
Collected firsthand by the Collected by others, available for
Source
researcher reuse
Collected specifically for the Originally collected for a different
Purpose
research problem purpose
Often more expensive and time-
Cost Usually less expensive or free
consuming
Takes more time to collect and Can be accessed and analyzed
Time Requirement
analyze quickly
Researcher has full control over Researcher has no control over the
Control
data quality quality or methods used
Surveys, interviews, observations, Books, government reports,
Examples
experiments journals, websites, databases
Cont…
Conclusion
• The choice between primary and secondary data depends on the research
objectives, available resources, and time constraints. While primary data is
more specific and tailored to the research problem, it can be time-
consuming and costly. On the other hand, secondary data is cost-effective
and accessible but may lack relevance or precision. Many research projects
use a combination of both to strengthen their findings and conclusions.

Assignment
Enumerate the advantage and Disadvantages of Primary and Secondary
Data??
3.6 Methods of collecting primary data
• Primary data refers to the information collected directly from first-
hand sources specifically for the purpose of the research.
• There are several methods to collect primary data, depending on the
research design and the nature of the information sought.
• Below are the most common methods of collecting primary data:
1. Surveys and Questionnaires
• Surveys and questionnaires are one of the most widely used methods for
collecting primary data, especially in quantitative research.
• They are used to gather responses from participants regarding their opinions,
behaviors, or characteristics.
Cont…
a. Structured Questionnaires
• Description: These consist of pre-formulated, close-ended questions with limited
response options (e.g., multiple choice, yes/no).
• Purpose: Collects quantitative data.
• Examples: Customer satisfaction surveys, employee feedback forms.
• Advantages: Easy to analyze, scalable for large samples.
• Disadvantages: Limited depth of information.
b. Unstructured Questionnaires
• Description: These include open-ended questions that allow respondents to provide
more detailed answers in their own words.
• Purpose: Collects qualitative data.
• Examples: Feedback forms asking for detailed suggestions.
• Advantages: Allows for more detailed responses and insights.
• Disadvantages: Harder to analyze due to variability in responses
Cont…
c. Online Surveys
• Description: Digital forms that can be distributed through email or online
platforms, where participants submit responses electronically.
• Advantages: Low cost, wide reach, automated data collection.
• Disadvantages: May exclude participants without internet access.
d. Telephone Surveys
• Description: Data is collected through phone interviews, often structured or
semi-structured.
• Advantages: Can reach geographically dispersed participants, quick
turnaround.
• Disadvantages: Can be expensive and subject to non-response bias
Cont…
2. Interviews
Interviews involve direct verbal interaction between the researcher
and the respondent, either face-to-face, over the phone, or through
video conferencing. Interviews are especially useful for collecting in-
depth qualitative data.
a. Structured Interviews
• Description: The researcher asks a set of pre-determined questions in a fixed order,
similar to a survey but conducted verbally.
• Purpose: Collects quantitative data.
• Examples: Formal job interviews, product feedback interviews with fixed questions.
• Advantages: Easy to analyze, consistent data collection.
• Disadvantages: Limited flexibility for probing deeper into responses.
Cont…
b. Semi-Structured Interviews
• Description: The researcher follows a pre-set list of topics but has the flexibility to
probe deeper or explore new topics based on the interviewee’s responses.
• Purpose: Collects qualitative data.
• Examples: Interviews with experts, open-ended product evaluations.
• Advantages: More depth than structured interviews, flexibility to explore new topics.
• Disadvantages: Time-consuming, harder to standardize across multiple respondents.
c. Unstructured Interviews
• Description: The interview is more conversational, with no pre-determined
questions. The researcher may guide the discussion based on the interviewee's
responses.
• Purpose: Collects qualitative data.
• Examples: Life history interviews, ethnographic interviews.
• Advantages: Provides rich, detailed data and deeper understanding.
• Disadvantages: Difficult to analyze and compare responses across multiple
participants.
Cont…
3. Observations
Observation involves the researcher recording behavior, events, or
conditions as they occur in real-time. It is particularly useful for studying
naturally occurring phenomena.
a. Participant Observation
• Description: The researcher becomes actively involved in the activities of the group
being studied while observing their behavior.
• Purpose: Collects qualitative data through immersion.
• Examples: An anthropologist living within a tribal community to observe their
culture.
• Advantages: Provides deep, contextual insights.
• Disadvantages: The researcher’s presence may influence behavior (observer effect),
and it can be time-consuming.
Cont…
b. Non-Participant Observation
• Description: The researcher observes subjects without directly interacting with them. The researcher remains
detached and does not intervene in the environment.
• Purpose: Collects objective behavioral data.
• Examples: Observing how customers interact with products in a retail store.
• Advantages: Reduces bias and allows for natural observation.
• Disadvantages: The researcher has limited control over events, and it can be difficult to interpret behaviors.
c. Controlled Observation
• Description: The researcher observes participants in a controlled environment, often in a laboratory or
structured setting.
• Purpose: Collects data under pre-determined conditions to ensure control over variables.
• Examples: A psychologist observing children’s reactions to stimuli in a lab setting.
• Advantages: Allows for precise control over variables.
• Disadvantages: Behavior may be artificial or influenced by the controlled setting.
d. Naturalistic Observation
• Description: The researcher observes participants in their natural environment without any manipulation or
control.
• Purpose: Collects data on naturally occurring behaviors.
• Examples: Studying animal behavior in the wild or watching how people interact in public spaces.
• Advantages: Provides insights into real-world behavior.
• Disadvantages: Limited control over external variables and difficult to replicate.
Cont…
4. Experiments
• Experiments involve the manipulation of one or more variables to test their
effect on other variables, allowing researchers to establish cause-and-effect
relationships.
a. Laboratory Experiments
• Description: Conducted in a controlled environment where the researcher
manipulates variables and observes their effects.
• Purpose: Collects data to test hypotheses and establish causality.
• Examples: Testing the impact of different advertising strategies on consumer
behavior.
• Advantages: High level of control over variables, replicable results.
• Disadvantages: May lack external validity because the artificial setting might not
reflect real-world conditions.
Cont….
b. Field Experiments
• Description: Conducted in a natural setting (outside the lab) but still involves manipulation of
variables.
• Purpose: Tests hypotheses in real-world conditions.
• Examples: Testing the effect of different store layouts on customer purchases.
• Advantages: Combines the control of an experiment with real-world relevance.
• Disadvantages: Less control over external factors, may be harder to replicate.
Others include
5. Focus Groups
6. Social Media and Online Data Collection
7. Delphi Method
Conclusion
• The method chosen to collect primary data depends on the research objectives, available
resources, and the nature of the information needed. Quantitative methods like surveys
and experiments are suitable for collecting measurable data, while qualitative methods
like interviews and observations are more appropriate for gaining deeper insights into
human behavior and experiences. Some studies may even combine multiple methods to
achieve a more comprehensive understanding of the research question.
3.7 Selection of appropriate method for data
collection
Choosing the right method for collecting data is critical to ensuring that the data gathered is
reliable, valid, and appropriate for the research objectives. The selection of the data collection
method depends on various factors including the research design, the type of data needed
(qualitative or quantitative), the available resources, and the nature of the research question.
Key Factors in Selecting an Appropriate Data Collection Method
1. Research Objectives
2. Nature of Data Required
3. Resources Available (Time, Budget, Expertise)
4. Population and Sample Size
5. Type of Respondents
6. Availability of Data
7. Desired Accuracy and Precision
8. Ethical Considerations
Cont…
Choosing the Appropriate Data Collection Method
• Here’s a brief guide on selecting appropriate methods based on different
research conditions:
1. Surveys
• Best For: Large, geographically dispersed populations; collecting quantitative data on
opinions, behaviors, or characteristics.
• Use When: You need standardized data that can be easily compared across a large number of
respondents.
• Example: Conducting a national customer satisfaction survey for a product.
2. Interviews
• Best For: Exploring in-depth insights from individuals; understanding complex processes,
behaviors, or motivations.
• Use When: You need detailed, qualitative data and are working with a smaller, focused group
of people.
• Example: Interviewing experts in a specific field to gather insights for policy development.
Cont….
3. Focus Groups
• Best For: Group discussions that explore multiple perspectives on a topic, product, or
issue.
• Use When: You want to generate a variety of opinions and ideas or explore group
dynamics.
• Example: Conducting focus groups to understand user experience with a new app.
4. Observations
• Best For: Capturing natural behaviors in real-time; studying processes in natural
environments.
• Use When: You need to study behavior as it naturally occurs without interference.
• Example: Observing customers in a retail store to study buying patterns.
5. Experiments
• Best For: Testing hypotheses, identifying cause-and-effect relationships, or
comparing different interventions.
• Use When: You need precise, controlled data to establish causality.
• Example: Testing different marketing strategies in a controlled setting to determine
which one drives higher sales.
Cont…
6. Secondary Data Analysis
• Best For: Studies where existing data can provide sufficient answers to the research question.
• Use When: Primary data collection is too costly or unnecessary, and reliable secondary
sources are available.
• Example: Analyzing census data to understand demographic shifts in a region.
7. Delphi Method
• Best For: Reaching a consensus among experts on a complex issue.
• Use When: You need expert opinions over multiple rounds, especially on forecasting or
decision-making problems.
• Example: Using the Delphi method to predict technological advancements in the next
decade.
8. Case Studies
• Best For: In-depth exploration of a single instance or a small number of cases in detail.
• Use When: You want to explore complex issues in specific, real-world contexts.
• Example: Studying the success factors of a startup company to understand business growth
patterns.
Cont…
Conclusion
• Selecting the appropriate method for data collection is crucial for the
success of a research project.
• The choice depends on the research objectives, type of data required,
resources available, and the target population.
• By carefully considering these factors, researchers can ensure that the
chosen method is aligned with their needs and will yield accurate, relevant
data.
• Often, a combination of methods may be the best approach to capture
both qualitative and quantitative insights, providing a comprehensive view
of the research problem.
3.8 Testing of Hypothesis
• Hypothesis testing is a fundamental aspect of statistical analysis in research,
helping to determine whether there is enough evidence to reject a null
hypothesis (H₀) in favor of an alternative hypothesis (H₁). This process allows
researchers to make inferences about populations based on sample data.
1. Understanding Hypotheses
a. Null Hypothesis (H₀)
• The null hypothesis represents a statement of no effect, no difference, or no relationship. It is
the hypothesis that researchers aim to test against.
• Example: H₀: There is no difference in the average test scores of students who study with a
tutor compared to those who study alone.
b. Alternative Hypothesis (H₁ or Hₐ)
• The alternative hypothesis is what the researcher aims to support, indicating the presence of
an effect, a difference, or a relationship.
• Example: H₁: Students who study with a tutor have higher average test scores than those who
study alone.
Cont…
2. Steps in Hypothesis Testing
Step 1: Formulate Hypotheses
• Clearly state both the null and alternative hypotheses based on the research
question.
Step 2: Select a Significance Level (α)
• The significance level is the probability threshold for rejecting the null
hypothesis, commonly set at 0.05 (5%). This represents a 5% risk of
concluding that a difference exists when there is no actual difference.
Step 3: Choose the Appropriate Test
• Select a statistical test based on the type of data and the research design.
Common tests include:
• t-test: For comparing means between two groups.
• ANOVA: For comparing means among three or more groups.
• Chi-square test: For categorical data.
• Correlation and regression analyses: For examining relationships between variables.
Cont…
Step 4: Collect Data
• Gather the data needed for analysis using the selected data collection
methods.
Step 5: Calculate the Test Statistic
• Use the appropriate statistical test to calculate the test statistic (e.g., t-value,
F-value, z-value).
Step 6: Determine the p-value
• The p-value indicates the probability of observing the test results under the
null hypothesis. It helps in deciding whether to reject or fail to reject the null
hypothesis.
Cont…
Step 7: Make a Decision
• Compare the p-value to the significance level (α):
• If p-value ≤ α: Reject the null hypothesis (evidence suggests an effect or difference
exists).
• If p-value > α: Fail to reject the null hypothesis (insufficient evidence to support the
alternative hypothesis).
Step 8: Interpret the Results
• Discuss the implications of the findings in relation to the research question
and hypotheses. Consider the practical significance of the results in addition
to statistical significance.
Cont…
3. Types of Hypothesis Tests
a. One-tailed Tests
• A one-tailed test is used when the alternative hypothesis specifies a direction
(greater than or less than).
• Example: H₁: The mean score of students with tutors is greater than that of
those without (H₁: μ₁ > μ₂).
b. Two-tailed Tests
• A two-tailed test is used when the alternative hypothesis does not specify a
direction, only that there is a difference.
• Example: H₁: The mean score of students with tutors is different from that of
those without (H₁: μ₁ ≠ μ₂).
Cont…
4. Assumptions of Hypothesis Testing
• Each statistical test comes with its own set of assumptions that must be met for the
results to be valid:
• Normality: Many tests assume that the data follows a normal distribution.
• Independence: The observations should be independent of one another.
• Homogeneity of Variance: In tests comparing groups, variances among the groups
should be approximately equal.
5. Common Pitfalls in Hypothesis Testing
• Misinterpretation of p-values: A common misconception is that a p-value indicates
the probability that the null hypothesis is true. It merely indicates how consistent the
data is with the null hypothesis.
• P-hacking: This refers to manipulating the data or analysis until nonsignificant results
become significant. This practice can lead to misleading conclusions.
• Ignoring Effect Size: Statistical significance does not imply practical significance.
Researchers should also consider effect size to assess the magnitude of differences
or relationships.
Cont…
Conclusion
• Hypothesis testing is a systematic method for evaluating claims about
population parameters based on sample data.
• By formulating null and alternative hypotheses, selecting appropriate
tests, and interpreting results, researchers can draw conclusions and
make informed decisions about their research questions.
• Understanding the process and its underlying assumptions is essential
for conducting robust research and ensuring valid conclusions.