business research methods

Business decisions get made on flimsy evidence every day — a hunch, a survey with three respondents, a case study of one company. This book gives you the toolkit to tell solid research from shaky research, and to produce the solid kind yourself. It suits MBA students, managers commissioning studies, and anyone who has to defend a finding in front of sceptical colleagues.

1. 1Chapter 2

   Two engines of research: testing ideas or building them

   Picture a researcher at a whiteboard with two very different plans of attack. One starts at the top: a general theory about how organisations behave, which gets boiled down into a specific, testable hypothesis. Data is gathered, results are compared against the hypothesis, and the theory is confirmed, modified, or rejected. The other path runs the opposite way: the researcher starts in the field, collects observations about a concrete situation, and works upward toward a general theory that the evidence suggests. Bell, Bryman and Harley present these two routes — deduction and induction — as the classic contrast in how business research gets designed.

   The idea

   Deduction and induction are not rival camps so much as two directions of travel between theory and evidence. Deduction risks confirming what you already believed if you rig the test; induction risks mistaking a local pattern for a universal law. Most real projects mix both, moving down and up the ladder repeatedly. Knowing which direction you are travelling keeps you honest about what your evidence can actually support.

   Try this: Write down your research question, then label it: are you testing a theory you already hold, or trying to build one from what you observe? If you cannot say, you are not ready to design the study.

2. 2Chapter 2

   Positivism and interpretivism: two ways of knowing a company

   In the philosophy chapter, the authors stage the field's oldest argument. On one side stands positivism, treating the social world as something that can be observed and measured from the outside, the way a natural scientist measures a chemical reaction. On the other side stands interpretivism, insisting that organisations are made of people who attach meanings to what they do — so understanding requires grasping those meanings, not just counting them. Bell, Bryman and Harley lay out this epistemological divide as the foundation beneath every choice a business researcher makes about methods.

   The idea

   Your philosophy of knowledge is not academic decoration; it silently decides what counts as a good answer. A positivist study can tell you how often something happens but struggles with why it matters to the people involved. An interpretivist study can explain meaning but may struggle to say how widespread it is. Naming your position before you collect data prevents the commonest embarrassment: claiming your method proves something it was never built to show.

   Try this: Before your next project, write one sentence: 'I will treat the phenomenon as measurable behaviour' or 'as meaningful action.' Notice how it changes what you plan to collect.

3. 3Chapter 8

   Qualitative interviews: getting past the official version

   The qualitative interviewing chapter opens a door most outsiders never see through. Bell, Bryman and Harley describe the semi-structured interview as a conversation guided by a list of questions and topics, but flexible enough to follow up on whatever the interviewee reveals. The interviewer can reorder questions, probe unexpected answers, and let the person speak in their own terms. This flexibility, they explain, is precisely what lets researchers reach the meanings and interpretations behind people's answers — the stuff a rigid questionnaire never captures.

   The idea

   The power of the semi-structured format lies in its controlled looseness: enough structure to compare interviews, enough freedom to chase what you did not expect. The catch is skill. A poor interviewer gets rehearsed, official-sounding answers; a good one creates the conditions for people to explain what actually drives them. Preparation — knowing the topic well enough to improvise follow-ups — is what separates the two.

   Try this: Draft five open questions for an interview, then practise rephrasing each one three ways so you can adapt mid-conversation instead of reading a script.

4. 4Chapter 9

   Focus groups: listening to people think out loud together

   In the focus group chapter, the authors describe a method built on interaction rather than one-to-one questioning. A moderator brings a small group together, poses a topic, and lets the participants respond to each other — questioning, agreeing, disagreeing, refining their views as they go. Bell, Bryman and Harley present this group dynamic as the method's defining feature: it reveals how people form and justify opinions in company, not just what they think in isolation. The scene is a table of colleagues building an answer none of them would have produced alone.

   The idea

   Because participants react to one another, focus groups expose the social reasoning behind attitudes — the compromises, the peer pressure, the arguments people find persuasive. That makes them powerful for exploring how consumers or employees talk about a topic, and risky if you treat the output as a statistically representative sample. The moderator's skill in managing dominant voices and drawing out quiet ones decides whether you get a conversation or a monologue.

   Try this: Run a 20-minute discussion with three colleagues on one question, and note who changed their mind and why — that shift is the data.

5. 5Chapter 7

   Sampling: who you ask decides what you can claim

   The sampling chapter confronts a deceptively simple problem: you cannot study everyone, so who do you study? Bell, Bryman and Harley lay out probability sampling, where every member of the population has a known chance of selection, against non-probability sampling, where selection relies on the researcher's judgement or convenience. The choice, they show, is not technical trivia — it determines whether findings can be generalised beyond the people actually studied. A sample chosen for access tells a different kind of truth than one chosen at random.

   The idea

   Sampling is where research honesty is won or lost. Probability samples let you generalise with quantified confidence; non-probability samples are often the only feasible route into hard-to-reach business settings, but they demand that you claim less. The mistake to avoid is dressing a convenience sample in the language of statistical certainty. Match the strength of your conclusions to the strength of your sampling logic.

   Try this: For your next survey, write one sentence naming your population and one naming your sampling method — then check whether your planned conclusions outrun that method.

6. 6Chapter 10

   Structured interviews and questionnaires: asking the same question the same way

   In the survey chapter, the authors turn to the machinery of standardised questioning. Bell, Bryman and Harley describe structured interviews in which each respondent is asked exactly the same questions, in the same order, with the same wording and the same answer options. The interviewer's role is to deliver the script without bending it. This uniformity, they explain, is what makes responses comparable across hundreds of people — the foundation for turning individual answers into patterns you can count.

   The idea

   Standardisation is both the strength and the ceiling of the structured interview. Because nothing varies, differences in answers reflect the respondents, not the interviewers — that is what makes comparison legitimate. The cost is depth: no room to probe a surprising answer or rephrase a question someone misread. The method buys breadth at the price of nuance, and the researcher must decide which the question needs.

   Try this: Take three questions from a survey you plan to run and check: could two different interviewers read them and get the same meaning? Rewrite any that fail.

7. 7Chapter 12

   Case studies: going deep where it counts

   The case study chapter makes the argument for depth over breadth. Bell, Bryman and Harley describe the case study as an in-depth examination of a single case — an organisation, a location, a person, or an event — studied in its real-life context. The researcher can draw on multiple sources of evidence to build a rich picture of that one case. Rather than sampling many units thinly, the case study concentrates attention where the phenomenon lives, accepting the trade-off that what is gained in depth is not automatically generalisable in breadth.

   The idea

   A single case, examined thoroughly in context, can reveal mechanisms that a broad shallow survey never sees — how a change actually unfolded, who resisted it, what it depended on. The trade-off is generalisation: one case tells you what is possible, not how common it is. The craft lies in choosing a case that is revealing rather than merely convenient, and in being explicit about what your conclusions can and cannot claim beyond it.

   Try this: Pick one recent change in your organisation and write a one-page account of it from three different sources — meeting notes, an email, and one conversation — then note where the accounts disagree.

8. 8Chapter 5

   Ethics: the permission that never expires

   In the ethics chapter, Bell, Bryman and Harley set out the principles that govern how research participants must be treated. Informed consent means participants understand what the research involves and agree to take part; confidentiality means their identities are protected; anonymity means even the researcher cannot link responses to names. These principles, the authors explain, apply across the whole research process, from first contact to publication. The scene is a researcher weighing every stage of a project against the people who made it possible.

   The idea

   Ethical principles are not a form to sign at the start; they are commitments that shape decisions at every stage, including how findings are written up. Anonymity and confidentiality differ — one severs the link entirely, the other promises to protect it — and confusing them can betray participants. In business settings, where jobs and reputations ride on what people say, the stakes of getting this wrong are unusually high.

   Try this: Before your next interview, write down exactly what you will tell participants about who will see their words — and check whether that promise survives your write-up plan.

9. 9Chapter 6

   Mixed methods: combining numbers and narratives

   In the research design chapter, the authors present mixed methods research as the deliberate combination of qualitative and quantitative approaches within a single study. Bell, Bryman and Harley describe designs where both kinds of data are collected and integrated, so that each compensates for the other's blind spots. The scene is a researcher with two toolkits, choosing which to open first and how the findings from one will shape the other. The method's promise is a fuller account than either tradition delivers alone.

   The idea

   Mixing methods is not about hedging; it is about answering questions that have both a 'how much' and a 'why' inside them. A survey can show the size of a problem while interviews explain its texture — but only if the two are genuinely integrated, not stapled together at the end. The design decision that matters is sequencing and connection: what each strand must deliver, and when, for the other to use it.

   Try this: Write your research question twice — once as a measurable claim, once as an open 'how' or 'why' — and design one data source for each.

10. 10Chapter 14

    Writing up: research that nobody reads is research that never happened

    The final chapter turns to the moment every project ends at: the write-up. Bell, Bryman and Harley describe the structure of a research report — the introduction, the literature review, the methods section, the findings, the discussion, and the conclusion — as the framework within which a study's story gets told. The scene is a researcher with a folder of data facing the blank page, deciding what goes where and in what order. The authors present this structure as the discipline that turns months of work into something a reader can follow, check, and use.

    The idea

    The standard report structure is not bureaucracy; it is a contract with the reader. The methods section lets others judge whether your findings deserve trust, the findings section separates what you found from what you think it means, and the discussion earns the right to interpret. Writers who blur these boundaries — smuggling conclusions into the findings, or hiding methodological weaknesses — invite the scepticism that sinks otherwise good work.

    Try this: Outline your next report under the six headings — introduction, literature review, methods, findings, discussion, conclusion — before writing a word, and put every piece of data in exactly one section.

MBA and business students who need to design, conduct, or evaluate research, and managers who commission studies and must judge whether findings hold up. Also valuable for anyone writing a dissertation who wants a clear map of the whole process.

FAQ