What is Research? The Research Onion
Session at a Glance
What is Research? The Research Onion; Research vs. Development; BBA & BCA Research Traditions
Paradigm mapping exercise; Initial topic exploration
2 hrs Lecture + 12 hrs Lab/Project
—
Learning Objectives
By the end of this session, you will be able to:
- Define research and distinguish it from everyday inquiry, development, and problem-solving
- Describe all six layers of the Research Onion (Saunders et al.) and explain how each layer shapes a research study
- Identify the key characteristics that make research systematic, logical, empirical, replicable, and objective
- Articulate how research traditions differ between business (BBA) and computing (BCA) disciplines
- Map a given research abstract to its underlying paradigm with justification
Session Planner
Suggested breakdown of the 4-hour contact session. Adjust pacing to suit your cohort.
| Time | Segment | Activity | Mode |
|---|---|---|---|
| 0:00–0:15 | Opening | Welcome, course overview, icebreaker: "What does research mean to you?" | Whole class |
| 0:15–0:35 | Lecture 1 | What is Research? Definition, purpose, characteristics of good research | Lecture |
| 0:35–0:55 | Lecture 2 | The Research Onion: Unpacking all 6 layers with examples | Lecture + Visual |
| 0:55–1:10 | Activity | Research vs. Development vs. Problem-Solving — classify 8 scenarios | Pairs |
| 1:10–1:25 | Lecture 3 | BBA vs. BCA Research Traditions — comparison with real examples | Lecture |
| 1:25–1:40 | Break | — | — |
| 1:40–2:00 | Lab Briefing | Paradigm mapping exercise instructions; topic exploration guidance | Demo |
| 2:00–3:30 | Lab Work | Hands-on paradigm mapping (10 abstracts); individual topic exploration | Individual/Pairs |
| 3:30–3:50 | Discussion | Share paradigm mapping results; discuss disagreements; facilitator synthesis | Whole class |
| 3:50–4:00 | Exit Ticket | Quick self-check + one question about research | Individual |
1. What is Research?
A systematic process of inquiry that involves the collection, analysis, and interpretation of data to answer a question, solve a problem, or contribute to a body of knowledge — conducted in a way that is transparent, rigorous, and open to scrutiny.
1.1 The Purpose of Research
At its core, research exists to create knowledge where none existed before. It is not about confirming what you already believe — it is about discovering what you don't yet know. Research serves five fundamental purposes:
| Purpose | Key Question | BBA Example | BCA Example |
|---|---|---|---|
| Exploratory | "What is happening?" | What factors drive consumer adoption of UPI-based payments among small retailers in Tier-2 Indian cities? | What are the most common categories of bugs reported in open-source Rust projects? |
| Descriptive | "How much? How often?" | What is the average employee attrition rate in Indian IT services firms during 2020–2025? | What is the distribution of programming languages used in AI/ML projects on GitHub? |
| Explanatory | "Why does it happen?" | Why do some Indian startups succeed in raising Series A funding while others fail? | Why do certain neural network architectures outperform others on low-resource NLP tasks? |
| Predictive | "What will happen if...?" | How will a 10% increase in digital marketing spend affect customer acquisition cost? | How will model compression affect inference latency on edge devices? |
| Prescriptive | "What should we do?" | What retention strategy should Indian IT firms adopt to reduce attrition below 15%? | Which caching strategy should be used to minimize response time under high concurrent load? |
1.2 Everyday Inquiry vs. Research
We ask questions every day — "Which phone should I buy?" or "Why is my code running slow?" But everyday inquiry is not research. The table below highlights what distinguishes research from casual questioning:
| Dimension | Everyday Inquiry | Formal Research |
|---|---|---|
| Question | Vague, personal, ad-hoc | Clear, specific, derived from literature gaps |
| Method | Intuition, asking friends, quick web search | Structured, documented, replicable |
| Evidence | Anecdotal, confirmation-biased | Systematic, representative, critically evaluated |
| Generalizability | Personal, context-bound | Aims for broader applicability |
| Scrutiny | No external review | Peer review, supervisor critique, public defence |
| Contribution | Informs personal decision | Adds to the body of knowledge |
What separates research from Googling is methodological rigour — a deliberate, documented, and defensible process. When you write your dissertation, you are not just reporting what you found; you are showing how you found it so that others could, in principle, replicate your path.
2. The Research Onion
The Research Onion, introduced by Saunders, Lewis, and Thornhill in Research Methods for Business Students, is the most widely used metaphor for understanding research design. Like peeling an onion, you move from the outermost layer (broad philosophical assumptions) to the innermost core (specific techniques for collecting and analysing data). Each layer constrains the choices available in the next.
A six-layer framework that guides researchers from philosophical assumptions (outermost) through research approach, strategy, methodological choice, time horizon, to data collection and analysis techniques (innermost). The key principle: coherence across layers — your philosophy should align with your approach, which should align with your strategy, and so on.
The Six Layers — From Outer to Inner
Research Philosophy
Your assumptions about how knowledge is created. Are you a positivist (objective reality, test hypotheses), interpretivist (subjective meanings, understand context), pragmatist (whatever works), or design scientist (create artefacts)? — We cover these in depth in Week 2.
Approach to Theory Development
Deductive (theory → data → test: start with a hypothesis from theory and test it), Inductive (data → patterns → theory: observe first, build theory from the ground up), or Abductive (surprising fact → best explanation: move back and forth between data and theory).
Methodological Strategy
The overall plan: Experiment, Survey, Case Study, Grounded Theory, Ethnography, Action Research, Design Science Research, or Archival Research. Your strategy is your research's "battle plan."
Methodological Choice
Mono-method (quantitative OR qualitative only), Multi-method (two quantitative methods OR two qualitative methods), or Mixed-methods (both quantitative AND qualitative integrated).
Time Horizon
Cross-sectional (snapshot at one point in time — e.g., a survey conducted once) or Longitudinal (studying change over time — e.g., tracking the same startups over 3 years).
Data Collection & Analysis Techniques
The concrete tools: questionnaires, interviews, focus groups, observations, system logs, benchmark datasets, A/B tests, sensor data. Analysis: statistical tests, thematic coding, performance metrics, content analysis.
The most common mistake in undergraduate dissertations is layer incoherence — for instance, adopting an interpretivist philosophy (seeking subjective meaning) but then using only Likert-scale surveys (a positivist tool) without qualitative depth. Every methodological choice you make in your capstone must be traceable back through the onion layers. "I chose semi-structured interviews because my interpretivist stance requires accessing participants' lived experiences, and a case study strategy allows deep contextual inquiry."
3. Characteristics of Good Research
Not all inquiry qualifies as good research. The academic community has converged on five hallmarks that distinguish rigorous research from sloppy or biased work:
The research follows a planned, structured process — not haphazard or improvised. Every step from problem definition to conclusion is deliberate and documented. Bad research jumps from data to conclusion without showing the path.
Conclusions follow from evidence through valid reasoning. The chain of inference — from premise to evidence to claim — withstands scrutiny. Bad research commits logical leaps: "We interviewed 3 CEOs, therefore all Indian startups are..."
Claims are grounded in observable evidence — data, not opinion. "I believe" is not a research finding. Bad research relies on personal conviction, anecdote, or cherry-picked examples.
Another researcher, following your documented procedure, should arrive at comparable results. This requires transparency about your data, methods, and analysis. Bad research reports findings without revealing how they were produced — the "black box" problem.
The researcher acknowledges and manages their own biases rather than pretending they don't exist. Conclusions are data-driven, not ideology-driven. Bad research starts with the answer and works backwards to find supporting evidence.
4. Research vs. Development vs. Problem-Solving
One of the most critical distinctions — especially for BCA students building software — is understanding the difference between research (creating generalizable knowledge), development (building a product or system), and problem-solving (fixing a specific issue). These are often confused, with serious consequences for how a capstone project is evaluated.
| Dimension | Research | Development | Problem-Solving |
|---|---|---|---|
| Goal | Create new, generalizable knowledge | Build a working artefact or product | Resolve a specific, bounded issue |
| Question | "What is the nature of X?" or "Does A cause B?" | "How do we build a system that does Y?" | "Why is Z broken and how do we fix it?" |
| Output | Findings, theories, models, principles | Software, hardware, platform, tool | A solution to a particular problem |
| Generalizability | Claims are intended to generalize | The artefact itself is specific | The fix is specific to the situation |
| Evaluation | Validity, reliability, contribution to knowledge | Functionality, performance, usability | Whether the problem is solved |
| Example (BCA) | "How does microservice architecture affect deployment frequency in agile teams?" | Building a CI/CD dashboard for a specific company | Fixing a memory leak in the CI/CD dashboard |
| Example (BBA) | "What is the relationship between ESG disclosure and cost of capital in Indian listed firms?" | Creating an ESG reporting framework for a specific company | Fixing a specific company's high cost of debt |
Your capstone can include development (building a system, designing a framework) or problem-solving (addressing a specific organizational issue), but it must be framed as research. This means: (a) you identify a gap in the literature, (b) your artefact or intervention is positioned as an answer to a research question, and (c) you evaluate it rigorously and discuss what general lessons can be drawn. Building an app without this framing is development, not a capstone.
Quick Check — Classify the Scenario
For each scenario below, decide: Is it Research (R), Development (D), or Problem-Solving (PS)? Click to reveal the answer.
1. A student builds a mobile app for tracking attendance at her college using face recognition. She deploys it and it works.
2. A student investigates whether gamification elements (badges, leaderboards) increase user engagement in enterprise SaaS products. She runs a controlled experiment with 200 users, measures time-on-task and return frequency, and publishes the findings.
3. A BBA student notices that customer churn at her internship company has spiked to 18%. She analyses the customer database, identifies that churn is concentrated among customers who joined in the last 6 months, and recommends a revised onboarding process.
5. BBA vs. BCA Research Traditions
This course serves two cohorts with distinct — but overlapping — research traditions. Understanding the similarities and differences is essential for positioning your own capstone project correctly. Both traditions share the same methodological core (the research onion, validity concerns, ethics), but they apply it to different kinds of problems using different kinds of evidence.
| Dimension | BBA — Business Research | BCA — Computing Research |
|---|---|---|
| Core identity | Research for decision-making | Research for innovation |
| Typical problems | Consumer behaviour, marketing effectiveness, financial performance, HR practices, strategy, organizational culture | Algorithm design, software architecture, ML model performance, cybersecurity, HCI, data systems, networking |
| Dominant paradigms | Positivism (surveys), Interpretivism (case studies), Pragmatism (mixed) | Design Science (artefact creation), Positivism (benchmarks/experiments), Interpretivism (user studies) |
| Primary data sources | Surveys, interviews, financial reports, government databases, company records | System logs, benchmark datasets, code repositories, sensor data, user interaction logs, open datasets |
| Typical sample size | Surveys: n=100–500+; Interviews: 10–30; Cases: 1–5 organizations | Benchmarks: standard datasets; User studies: 10–30 participants; System evaluation: performance-based |
| Analysis tools | SPSS, R, Stata, NVivo, ATLAS.ti, Excel | Python (NumPy, pandas, scikit-learn), R, Jupyter, Git, Docker, LaTeX |
| Dissertation structure | Intro → Lit Review → Methodology → Data Analysis → Discussion → Conclusion | Intro → Lit Review → Methodology/Design → Artefact Description → Evaluation → Discussion → Conclusion |
| Referencing style | APA 7th (preferred), Harvard | IEEE (preferred), ACM |
Design Science Research (DSR) serves as a bridge between BBA and BCA traditions. A BCA student designing a novel algorithm and rigorously evaluating it can frame this as DSR. A BBA student designing a new business model or assessment framework and evaluating its utility can also frame this as DSR. The common thread: creating and evaluating an artefact as a form of knowledge contribution. We cover DSR in depth in Week 12.
Think Deeper — Cross Questions
Discuss in pairs or small groups before sharing with the class. These questions bridge the lecture content with your own capstone thinking.
Think about a topic you might want to research for your capstone. Working from the outside in, sketch your initial choices for each layer of the Research Onion. Where do you feel most uncertain? Why?
A BCA student says: "I'm building an AI-powered crop disease detection app for farmers. That's my research." Is this statement adequate as a research framing? What questions would you ask to help this student reframe it as research rather than development?
Can a study be systematic and empirical but still be bad research? Think of an example. What characteristic is missing, and why does it matter?
Business research and computing research share the same methodological core, but they have different epistemic cultures — different norms about what counts as convincing evidence. Based on your discipline, what kind of evidence would convince you that a research finding is trustworthy? Compare with someone from the other cohort.
Knowledge Check — Interactive Quiz
Test your understanding of the core concepts. Select an answer for each question.
Q1. A researcher collects survey data from 500 consumers to test whether price sensitivity predicts brand switching behaviour. Which layer of the Research Onion does the survey belong to?
Q2. Which of the following is the BEST example of research (as opposed to development or problem-solving)?
Q3. A study claims: "We interviewed 25 startup founders, and 80% said they prefer remote work. Therefore, remote work is better for startups." Which characteristic of good research is MOST clearly violated?
Q4. Which statement about the Research Onion is TRUE?
Q5. A BCA student wants to study "the factors that influence open-source contributors' sustained participation in a project." Which research tradition does this problem MOST align with?
Lab Activity — Paradigm Mapping & Topic Exploration
Part A: Paradigm Mapping Exercise (60 min)
Below are 10 research abstracts — 5 from business research, 5 from computing research. For each abstract, identify:
- The research philosophy (positivist, interpretivist, pragmatist, or design science)
- At least two other layers of the Research Onion visible in the abstract
- One sentence justifying your classification
"This study examines the relationship between corporate social responsibility (CSR) disclosure and firm profitability among NIFTY 500 companies. Using panel data from 2019–2024, we test the hypothesis that higher CSR disclosure scores are positively associated with ROA and Tobin's Q, controlling for firm size, leverage, and industry."
"We present a novel cache eviction algorithm, Adaptive-LRU, designed for read-heavy key-value stores. We implement the algorithm in Redis and evaluate it against standard LRU, LFU, and ARC using the YCSB benchmark across three workload patterns. Results show Adaptive-LRU reduces miss rate by 12–18% under skewed workloads."
"Through in-depth semi-structured interviews with 20 first-generation women entrepreneurs in Tier-2 Indian cities, this study explores the lived experience of navigating gendered expectations while building ventures. Using interpretative phenomenological analysis (IPA), we identify four superordinate themes..."
"We conduct an ethnographic study of three agile software development teams over six months, observing daily stand-ups, sprint planning, and retrospectives. Through thematic analysis of field notes and 30 interviews, we describe how team rituals develop, evolve, and sometimes break down under deadline pressure."
"This study employs a convergent mixed-methods design to understand fintech adoption among MSMEs. We administered a survey to 400 MSME owners (quantitative) and conducted 15 follow-up interviews (qualitative). Quantitative results identify trust and perceived ease of use as primary drivers; qualitative findings reveal how trust is constructed through peer networks."
"We designed and developed SecureVote — a blockchain-based e-voting system that provides end-to-end verifiability while maintaining voter privacy. We evaluate the system through security analysis (threat modelling), performance testing (throughput and latency under simulated load), and a usability study with 40 participants using SUS and think-aloud protocol."
"Using an action research methodology, we collaborated with the HR department of a large Indian manufacturing firm over 12 months to design, implement, and refine a competency-based performance management system. Each cycle of diagnose → plan → act → evaluate yielded refinements."
"This paper investigates whether code generated by LLM-based coding assistants (GitHub Copilot, CodeWhisperer) contains more security vulnerabilities than human-written code. We designed a controlled experiment where 60 developers completed identical programming tasks, half with and half without AI assistance. Each submission was analysed using static analysis tools and manual code review."
"Through a multiple case study design examining four Indian unicorn startups (one from fintech, one from edtech, two from e-commerce), we investigate how organizational culture evolves during hypergrowth phases. Data sources include 40 interviews, internal documents, and publicly available founder communications."
"We analyse the energy consumption of five popular JavaScript frontend frameworks (React, Vue, Angular, Svelte, Solid) when rendering identical web application workloads. Using a controlled measurement setup, we collect power data across 10,000 page render cycles. Statistical analysis (ANOVA with post-hoc Tukey HSD) reveals significant differences."
Part B: Initial Topic Exploration (Lab hours)
Use the remaining lab time to begin exploring potential capstone topics. For each topic idea, answer:
- What is the broad domain? (e.g., consumer behaviour, cybersecurity, fintech, ML model optimization)
- What specific problem or question interests you? (one sentence)
- Why does this matter? (who would care about the answer?)
- Where would you start looking for existing literature? (name at least two databases)
- Which research philosophy seems most appropriate? (preliminary guess)
Prepare a one-page summary. You will refine this into a formal problem statement in Week 3.
Exit Ticket
Complete before leaving. Submit to your facilitator.
- In your own words (2–3 sentences): What is research, and how is it different from everyday problem-solving?
- Name all six layers of the Research Onion from outermost to innermost.
- Rate your confidence in distinguishing BBA and BCA research traditions: Not confident / Somewhat confident / Confident / Very confident
- One question you still have about research methodology:
- One topic you're considering for your capstone (even if very tentative):
Key Takeaways — Week 1
Research is not Googling, not opinion, not building something. It is a systematic, logical, empirical, replicable, and objective process of creating knowledge.
Every methodological choice must align through all six layers. Philosophy constrains approach, which constrains strategy, which constrains technique. Incoherence is the #1 methodological error in dissertations.
Building an app is development unless it is framed as answering a research question, grounded in literature, and rigorously evaluated, with generalizable findings.
BBA and BCA research share the same methodological foundation (onion, validity, ethics) but differ in problems, evidence types, tools, and referencing styles. DSR bridges both worlds.
Facilitator Notes
Preparation Checklist
- Print or share the 10 paradigm mapping abstracts (Part A of lab). Prepare answer key with classifications.
- Ensure students have access to at least two academic databases (Google Scholar + institutional access to Scopus/IEEE/ProQuest).
- Prepare a live demonstration of the Research Onion using a sample topic — e.g., "What factors influence customer satisfaction with food delivery apps?" — peeling each layer.
- For BCA-strong cohorts: spend extra time on Research vs. Development distinction with computing-specific examples.
- For mixed cohorts: use the Cross Questions (CQ4) to spark a productive BBA-BCA dialogue about evidence standards.
Common Student Difficulties
- Confusing "research" with "reading about a topic": Emphasize the systematic, empirical nature of research. Reading 10 articles and summarizing them is a literature review, not research.
- The Research Onion feels abstract: Use a concrete worked example through all six layers. Better: have students apply each layer to their own tentative topic.
- BCA students resisting non-coding work: Acknowledge the impulse. Then reframe: the methodology is what distinguishes a capstone from a personal project. Employers value the ability to think rigorously.
- BBA students unsure about quantitative methods: Reassure them that Unit 3 provides extensive hands-on training. For now, focus on conceptual understanding.
Pacing Tips
- If time is tight: shorten the "Characteristics of Good Research" section (it's intuitive for most students) to leave more time for the onion and the paradigm mapping lab.
- If the cohort is engaged by the BBA vs. BCA comparison: let the discussion run — this early cross-disciplinary dialogue sets the tone for the entire course.
- The exit ticket is essential — it surfaces misconceptions before Week 2's deeper dive into paradigms.