Philosophy of Research & Types of Research Design
Learning Outcomes
- Understand different philosophical assumptions underlying management research.
- Appreciate how philosophical assumptions influence criteria for judging research quality.
- Develop the ability to recognize and identify latent philosophical assumptions.
- Acknowledge how different research philosophies lead to different methods.
- Consider different approaches in scientific reasoning (deduction vs. induction).
What is Philosophy of Science?
- Definition: Philosophy of science studies the tools scientists use to explain, predict, and understand the natural world.
- In simpler terms, it asks:
- How science works.
- What makes scientific knowledge valid.
- What methods are used to obtain it.
- The Central Goal of Science:
- To promote understanding of the natural world.
- This understanding comes from:
- Explaining natural phenomena.
- Helping scientists make predictions about future events or situations.
Why Do We Need It?
- Knowing the philosophical foundations of research:
- Helps researchers design better studies.
- Clarifies which methods and assumptions are appropriate.
- Ensures the research is meaningful and valid.
- In short: Philosophy of science = thinking about science itself → its logic, methods, and validity.
What is Science? (or What is Not Science)
- Definition: Science = a tool created by humans to help us understand the world.
- It does more than just describe; it explains how and why things happen.
The Role of the Scientist
- Scientists don’t just list every natural event they observe.
- Their job is to:
- Generalize and find explanations that apply broadly to nature.
- Find what principles or rules cause these phenomena.
- Simply listing observations = historian’s job.
- Explaining causes and mechanisms = scientist’s job.
Demarcation Criterion
- Key problem: How to separate science from non-science?
- Solution: Create a criterion (standard) that:
- All true sciences meet ✅
- Non-scientific approaches fail ❌
- This separation is called → Demarcation Criterion
- In short: Science explains why things happen, not just what happened.
- To call something science, there needs to be clear rules (demarcation) to separate it from non-science.
Demarcation Criterion
- The key difference between science and non-science = falsifiability.
- Falsifiable = it must be possible to imagine an observation or experiment that could show the theory is wrong.
- A scientific statement:
- Must be testable.
- Does not have to be actually proven wrong, but it must be possible to prove it wrong.
- Example: "All swans are white."
- Falsifiable: finding one black swan would falsify the theory.
Two Types of Non-Falsifiable Statements
- Tautologies
- Always true by definition.
- Example: "Triangles have three sides."
- Cannot be tested or falsified → not scientific.
- Statements about unobservable phenomena
- Cannot be tested through observation or experiment.
- Examples: "God exists." "God created the world."
- Not falsifiable → not scientific.
- Important: This does not mean they are false or nonsense. They just lie outside the scope of science.
Mini Exam Summary
- What is Philosophy of Science?
- A discipline that studies:
- The tools and methods used to explain, predict, and understand the natural world.
- Central goal of science:
- Promote understanding through explanations of natural phenomena.
- Allow predictions for future phenomena.
- Why important:
- Provides the philosophical foundation for research design.
What is Science (vs. Non-Science)?
- Science = human tool to understand the world.
- Explains both how and why things happen.
- Science ≠ simply listing facts → the goal is to generalize about nature.
- Scientists create theories explaining the causes behind phenomena.
- Simply cataloging events = historian’s job, not scientist’s.
Demarcation Criterion
- Key concept: Falsifiability (Karl Popper).
- A statement is scientific if:
- It can be tested.
- It is possible to imagine an observation that would prove it wrong.
- Falsifiable ≠ already proven false — only testable.
- Purpose: separate science from non-science.
Two Types of Non-Falsifiable Statements
- Tautologies
- True by definition.
- Example: "Triangles have three sides."
- Not testable → not scientific.
- Statements about Unobservable Phenomena
- Cannot be observed or tested.
- Examples: "God exists." "God created the world."
- Not falsifiable → not scientific.
- Note: Being non-scientific ≠ being false.
- Science = Falsifiable, Testable, Explains HOW and WHY
The Hallmarks of Scientific Research
- These are the key characteristics of what makes research good or scientific.
- Term
- Purposiveness
- Meaning: Clear goal or objective for the research.
- Rigor
- Meaning: Logical, careful, and thorough process; conclusions based on solid reasoning.
- Testability
- Meaning: Hypotheses must be testable through empirical methods.
- Replicability
- Meaning: Other researchers should be able to repeat the study and get similar results.
- Precision and Confidence
- Meaning: Results should be accurate and findings reported with certainty levels.
- Objectivity
- Meaning: Results should not be influenced by personal biases.
- Generalizability
- Meaning: Findings should apply to other contexts or populations beyond the sample studied.
- Parsimony
- Meaning: Explanations should be simple and not unnecessarily complicated. (Use as few assumptions as possible.)
- Short trick for exam memory: P.R.T.R.P.O.G.P. (Purposiveness, Rigor, Testability, Replicability, Precision, Objectivity, Generalizability, Parsimony)
- You can also remember this as: "Proper Research Tries Really Precisely On Giving Parsimony."
Paradigms in Social Research
- Level
- Ontology
- What it asks: What is reality?
- Short explanation: The nature of reality and social beings. Are facts objective or socially constructed?
- Epistemology
- What it asks: How can we know reality?
- How we gain knowledge about reality. What counts as valid knowledge?
- Methodology
- What it asks: How to answer research questions?
- The strategies or plans for research (qualitative, quantitative, mixed).
- Methods and Techniques
- What it asks: Which tools to collect data?
- The specific instruments (surveys, interviews, experiments, observations).
- Ontology = the nature of reality and existence.
- In research, we ask:
- "What kind of reality am I studying? Is there one reality or many?"
Ontology Type
- What is Truth?
- What about Facts?
- Realism
- What is Truth?: There is one single truth.
- What about Facts?: Facts exist and can be discovered.
- Internal Realism
- What is Truth?: Truth exists, but it is not fully accessible (hidden or complicated).
- What about Facts?: Facts exist but cannot always be directly observed.
- Relativism
- What is Truth?: There are many "truths" (depends on people).
- What about Facts?: Facts depend on observer’s viewpoint.
- Nominalism
- What is Truth?: There is no universal truth.
- What about Facts?: Facts are fully constructed by humans.
- Situations
- Realism
- "Is climate change real?" Yes — one scientific truth exists.
- Relativism
- "Is climate change real?" Depends — different groups may interpret data differently.
Epistemology
- Epistemology = how we know what we know.
- Two big approaches:
- Positivism
- Reality exists objectively (independent of people).
- We observe it using scientific, objective measurements.
- Our knowledge comes from observation, measurement, experiments.
- Social Constructionism
- Reality is constructed by people.
- Knowledge comes from people’s interactions, discussions, and shared meanings.
- It’s about understanding how people create reality.
Epistemology: Positivism vs Social Constructionism (Comparison)
- Question
- "What is marriage?"
- Positivist answer: Legal and biological facts.
- Social Constructionist answer: Social institution created and interpreted by cultures.
- Aspect
- The observer
- Positivism: Must stay independent.
- Social Constructionism: Is part of what is studied.
- Human interests
- Positivism: Should be irrelevant.
- Social Constructionism: Are central to understanding reality.
- Explanations
- Positivism: Must find causality (X causes Y).
- Social Constructionism: Focus on understanding situations as they are.
- Research progresses
- Positivism: Form hypothesis → test it.
- Social Constructionism: Collect data → develop ideas.
- Concepts
- Positivism: Must be clearly defined to measure.
- Social Constructionism: Should reflect different stakeholder views.
- Units of analysis
- Positivism: Simplify into measurable parts.
- Social Constructionism: Study full complexity of situations.
- Generalization
- Positivism: Based on statistics.
- Social Constructionism: Based on theory-building.
- Sampling
- Positivism: Large random samples.
- Social Constructionism: Small, carefully chosen cases.
- Objective
- Positivism: Objective
- Social Constructionism: Subjective
- Scope:
- Positivism: Quantitative
- Social Constructionism: Qualitative
- Relations of researcher with subject:
- Positivism: Independent
- Social Constructionism: Interactive
- Explanation:
- Positivism: Cause-Effect
- Social Constructionism: Understanding meaning
Positivist Research Design
- Positivist methods incorporate the assumption that there are true answers, and the job of the researcher is to start with a hypothesis and seek data to confirm/disconfirm it (to develop theory)
- Rigor and replicability, reliability and generalizability
- Methodologies: experimental/ quasi- experimental methods and surveys ( Quantitative approaches)
The Positivistic Approach (Process Flow)
- Feature
- Assumption
- Explanation: There is one true answer that can be discovered.
- Role of researcher
- Explanation: Start with theory → collect data to confirm/disconfirm it.
- Goal
- Explanation: Develop theory based on testing.
- Focus on:
- Explanation: Rigor, replicability, reliability, generalizability.
- Methodologies:
- Explanation: Experiments, quasi-experiments, surveys (= quantitative methods).
- Main Idea: Reality exists objectively → one truth.
- Researcher tries to find this truth.
- Research starts with a hypothesis (already existing theory).
- The researcher collects data to test whether this hypothesis is true or false.
- Key characteristics:
- This is called: Deductive Approach
- Deductive = start with theory → test with data.
Constructionist Research Design
- Feature Explanation
- Assumption: Reality is socially constructed.
- Role of researcher: Explore and understand different interpretations.
- Goal: Illuminate how reality is created by people.
- Methodologies: Action research, archival research, ethnography, narrative methods (= qualitative methods).
- Main Idea: Reality is not fixed → different people may have different interpretations.
- The researcher's role is to explore multiple "truths".
- Focus on understanding people’s perspectives and social meanings.
- No strict hypothesis at the start → researcher is open to what data reveals.
- Key characteristics:
- This is called: Inductive Approach
- Inductive = start with data → build theory from findings.
Deductive vs Inductive Reasoning
- Deductive Reasoning
- You start with theory.
- Based on the theory, you create hypotheses (predictions).
- Theory → Hypothesis → Data → Test
- Theory: "Stress causes lower productivity."
- Hypothesis: "Employees under stress will have lower performance scores."
- Collect data: Analyze if results support the hypothesis.
- This is the typical process in positivist research.
- Theory comes first.
- Inductive Reasoning
- You start with observations/data.
- You analyze data and look for patterns.
- From the patterns, you generate a theory.
- Data → Pattern → Theory
- You interview 50 managers about leadership styles.
- You find that most talk about empathy.
- From this, you develop a theory that "Empathetic leadership improves team satisfaction."
- You collect data to check if these predictions are confirmed.
- This is used in constructivist research.
- Observations come first.
Why is Pure Induction Problematic?
- Because if you start only with data, you may never falsify anything.
- You only build theories without challenging them.
- Induction helps build theories, but later we still need deduction to test them.
The Abductive Approach
- What is abduction?
- Abduction happens when you observe something surprising or unexpected.
- Instead of starting from theory or data, you try to make the best possible explanation for that unexpected observation.
- Abduction helps to generate new theories when existing theories cannot explain what you see.
- You observe: "A company suddenly performs extremely well during a financial crisis."
- This is unexpected (surprising observation). You don’t have a ready theory.
- You generate an explanation: "Maybe they invested in crisis-resistant sectors."
Comparison Table
- Reasoning Type
- Deduction
- Process: Theory → Hypotheses → Test with data
- Typical Use: Testing known theories (Positivist)
- Induction
- Process: Data → Patterns → Build theory
- Typical Use: Discovering patterns (Constructivist)
- Abduction
- Process: Surprising data → Best possible explanation
- Typical Use: Exploring new unexpected phenomena
- Apply theory
- Deduction: Test hypotheses
- Induction: Find patterns
- Abduction: Explain surprises
- Deduction: Confirm existing knowledge
- Induction: Generate new theory
- Abduction: Propose possible explanation
Pragmatic Approach to Research
- Key idea: Pragmatism is practical: It does not care too much about choosing between positivism or constructivism.
- Feature
- No strict position
- Explanation: Does not follow one strict philosophy.
- Objective + subjective
- Explanation: Combines both objective (positivist) and subjective (constructivist) knowledge.
- Theory Practice
- Explanation: The purpose of theory is to help solve real-world problems.
- Used by managers
- Explanation: Helps managers make decisions based on useful knowledge, not on strict philosophical rules.
- Main principles:
- Mixed Methods
- What is it?
- Mixed methods = combining both quantitative (numbers, statistics) and qualitative (interviews, meanings) approaches in one study.
- It combines positivist and constructionist elements.
- Why use mixed methods?
- Real-life problems are often complex → using only one method might not be enough.
Design Considerations in Mixed Methods
- Design Feature
- Sequencing
- Alternatives: Which comes first?
- Dominance
- Alternatives: Which method is stronger?
- Sequencing
- Qualitative first
- Quantitative first
- Both at the same time
- Dominance
- Predominantly qualitative
- Predominantly quantitative
- Balanced
- Arguments for and against mixed methods
- Mixed methods research has the potential to throw new perspectives on research questions, to increase the credibility of results, to demonstrate generalizability, and to provide deeper insights.
- But there are also plenty of reasons for being cautious about their wholesale adoption.
Types of Research Questions: Quantitative vs. Qualitative
- Quantitative Research Questions ("To what extent?")
- These questions ask:
- How much does one variable influence another?
- To what extent, how much, how strong, relationship between variables
- They assume:
- Variables can be measured.
- The relationship is known or assumed from theory (deductive logic).
- You usually have:
- Independent variable (cause).
- Dependent variable (effect).
- Connected to positivism + deductive approach.
- Examples:
- To what extent does community-based shared value influence the success of social enterprises?
- To what extent reciprocal altruism influences the success of social enterprises?
- Qualitative Research Questions ("How?" and "Why?")
- These questions ask:
- Focuses on: Meanings, processes, perceptions.
- Exploring new or complex phenomena.
- No predefined variables → theory is built from data (inductive logic).
- Connected to social constructionism + inductive approach.
- Examples:
- How do consumers perceive ‘Made in China’ products?
- How does innovation take place in the software sector in emerging markets?
- How do online consumers experience privacy issues during online transactions?
- How do SMEs in food and drinks sector experience the use of digital warehouse for exporting in China?
Super Summary Table
- Question Type
- Quantitative
- Typical Words: To what extent / How much
- Linked Approach: Deductive / Positivist
- Qualitative
- Typical Words: How / Why
- Linked Approach: Inductive / Constructivist