Getting Your Head Around Saunders' Framework
Most students pick up Saunders, Lewis, and Thornhill's research methods book because their supervisor told them to. The premise is straightforward enough: pick a philosophical stance, choose a methodology that fits it, and then select methods that align with that methodology. The problem is that step one trips people up constantly. You can read the chapter on research philosophies three times and still be unsure whether your study counts as positivistic or interpretivist. I spent an entire weekend trying to force my mixed-methods project into a purely positivist box before giving up and reading the book again with fresh eyes. The book structures its guidance around what Saunders calls the research onion. You peel back layers starting with philosophy, moving through approach, strategy, choice, time horizon, and finally techniques. Each layer depends on the one before it. If you pick a deductive approach but your philosophy is interpretivist, the whole thing falls apart somewhere between literature review and data collection. I learned that the hard way with a dissertation that had to be substantially revised after my first proposal meeting.
Applying Saunders Et Al 2009 Research Methods to Your Own Study
Start by answering one question honestly: do you believe there is a single objective reality waiting to be measured, or do you think reality is constructed through social interaction and meaning-making? Your answer determines whether you lean toward a positivist or interpretivist position. This is not a trick question. People who work in marketing and consumer behavior often find themselves somewhere in between, and Saunders does acknowledge that pure positions are rare in practice. What matters is that you make an explicit choice and stick with it through the proposal. Once you have your philosophy sorted, the next decision is whether to go deductive or inductive. Deductive means you start with a theory and test it. Inductive means you gather data and build theory from what you find. Saunders devotes several chapters to both, and honestly the examples in the book are where things click. The deductive section walks you through hypothesis formation and testing with concrete business research scenarios. The inductive side covers ground theory and thematic analysis with actual published studies as references. I found myself bookmarking those example sections rather than trying to memorize definitions. The methodology choice comes after that. Saunders lists several standard options: experiment, survey, case study, action research, and ethnography among them. Each has strengths and weaknesses depending on your research question. A case study works beautifully when you need depth over breadth, especially for exploring why something happens rather than how much of something exists. I used a single case study design for a project examining digital transformation in a mid-size retail company. Saunders would have called it an exploratory case study, and it fit perfectly. I collected interview data from twelve employees across three departments and supplemented it with internal documents. The depth I got from that approach would have been impossible with a survey alone.
The Philosophical Layer Is Where People Mess Up
Positivism assumes you can observe phenomena objectively and measure them quantitatively. Interpretivism holds that understanding requires seeing the world through the participants' eyes. Paradigmatic debates between these two have occupied sociology and business research for decades. Saunders presents them fairly and lets you decide. The book's treatment of pragmatism is probably the most useful section for students who feel trapped between the two poles. Pragmatism says pick whatever works for your research question and stop worrying about philosophical purity. I recommend pragmatism if you are doing applied business research and your main goal is producing something useful rather than contributing to foundational theory. Here is something Saunders does not emphasize enough: your philosophy choice affects your sampling strategy. Positivist research typically uses probability sampling to generalize findings. Interpretivist research uses purposive sampling to select information-rich cases. If you tell your advisor you are interpretivist but then use a random sample of five hundred respondents, something is inconsistent. I caught this error in a classmate's proposal and we spent two hours fixing it before the deadline.
Get the Full Details
Timing and Practical Constraints
Saunders discusses time horizons briefly but they matter more than students usually realize. A cross-sectional study gives you a snapshot at one point in time. Longitudinal research tracks change over months or years. Most undergraduate and master's dissertations are cross-sectional simply because the timeline does not allow anything else. That is fine. Do not volunteer for a longitudinal design if you have three months to complete your research. I have seen students take on six-month tracking studies and drown in follow-up logistics. The biggest mistake I see students make is treating the research onion as a checklist rather than a decision tree. They fill in each layer mechanically without checking whether the choices connect logically. Another frequent error is picking a method because it seems impressive rather than because it answers their research question. A survey with three hundred respondents sounds impressive on paper. It means nothing if the survey was poorly designed and the sampling frame was biased. Saunders covers sampling frames, sample sizes, and validity in decent detail. Read those sections carefully. There is also a tendency to conflate approach with methodology. An approach is deductive or inductive. A methodology is case study, survey, or experiment. These are different things. The book makes this distinction but it is easy to blur it under pressure.
What This Framework Does Not Handle Well
Saunders was first published in 2009 and the core framework has held up reasonably well. However, the book does not address online data collection methods in depth. Things like web scraping, social media analytics, and digital ethnography were not prominent in business research methods at the time of publication. If your research involves digital platforms, you will need to supplement Saunders with more recent sources. I used Sanders' framework as my structural guide and then filled gaps with articles from Journal of Business Research and MIS Quarterly covering similar topics. That combination worked for my dissertation on consumer behavior on e-commerce platforms. Another limitation is the book's treatment of action research. Saunders describes it adequately but action research in practice requires a commitment to iterative cycles of planning, acting, observing, and reflecting. Some students treat it as just another methodology label without grasping the ongoing reflective practice it demands. If you choose action research, make sure you actually want to engage with stakeholders in a collaborative change process. Otherwise pick something else.
A Practical Workaround I Found Useful
When I was stuck trying to align my research question with an appropriate methodology, I created a simple matrix. I listed my research questions on the left, potential methodologies down the top, and then scored each combination for fit, feasibility, and data availability. It took about twenty minutes and resolved weeks of confusion. Saunders' book is dense enough that sometimes you need a visual aid to see the connections clearly. I wish I had done this before my first proposal. The matrix approach is not in the book, but it is a practical tool that complements the framework well. If you need the actual Saunders text, the book is widely available through academic publishers and university libraries. Look for the edition published by Prentice Hall. Newer editions exist beyond the 2009 version, so check which one your program recommends. The core framework remains consistent across editions with updated examples and expanded coverage in later versions.
