9 Essential Steps to Prepare a Foolproof Research Design

Every anthropology dissertation, thesis, or fieldwork proposal starts the same way: with a plan. That plan is your research design, and it only works if you build it step by step instead of jumping straight into fieldwork. Skip a step, and you risk collecting data you can’t analyse, or worse, data that never answers your actual question. Here are the nine steps that turn a vague interest into a research design you can actually defend.

1. Pin down your research problem 🔗

Before you can research anything, you need something worth investigating. This is your research problem, and it usually grows out of the areas and issues you’ve become genuinely interested in during your training. A good research problem isn’t just interesting to you; it has to be focused, rational, and achievable within the time and resources you actually have.

Three questions help you test whether a problem is ready to become a project: What do I want to know? Why do I want to know it? And what might the answers look like? If you can’t answer these clearly, the problem probably needs more narrowing. A well-defined research problem is what stops a project from becoming unfocused or repeating what’s already been said elsewhere.

2. Draft a clear statement of the problem 🔗

Once you have a problem, you need to frame it. The statement of the problem answers a simple question: what exactly will this research investigate? It takes a broad concern and narrows it into something specific enough to actually solve through systematic fieldwork or analysis.

A strong problem statement does several things at once. It should point to a gap in existing knowledge, add something new to what’s already known, leave room for further research down the line, and be genuinely researchable through proper data collection. It should also reflect your own interests and capabilities as a researcher, and it should hold up ethically. A problem statement isn’t just throat-clearing before the “real” research begins; it provides the framework that everything else in your design will be built on, including the significance you’ll later claim for your findings.

3. Review the existing literature 🔗

Before you set foot in the field, you need to know what’s already been written about your topic. A literature review surveys books, journal articles, and other scholarly sources connected to your problem, and it summarises and critically evaluates them in relation to what you’re trying to investigate. This isn’t a one-time task done before fieldwork starts; anthropologists typically return to the literature between field visits too, using it to deepen their understanding as new questions emerge on the ground.

The real value of a literature review is that it shows you where the gaps are. A literature review helps you explain the background of research on a topic and demonstrates why your specific angle matters. Without it, you risk repeating a study someone has already done, or missing a debate your own findings will need to speak to.

4. Set your aims and objectives 🔗

Your aim is the broad outcome you’re working toward. It’s a general statement of what you hope your research will ultimately achieve, and it doesn’t need to be detailed. Your objectives are different: they’re the specific, actionable steps you’ll take to get there and actually answer your research questions.

Objectives work best when they’re focused, connected to each other, realistic given your resources, and precisely worded. A common way to sharpen them is the SMART approach – specific, measurable, achievable, relevant, and time-bound. Being specific about what you plan to achieve, how, where, and by when makes your objectives far easier to evaluate, both for you and for anyone reviewing your proposal later.

5. Decide whether you need a hypothesis 🔗

A hypothesis is a tentative explanation or prediction about your problem, something you accept provisionally until fieldwork either supports or overturns it. In many quantitative social science studies, a hypothesis is essential; it gives you something concrete to test.

Ethnographic anthropology often works differently. Many ethnographers deliberately avoid starting with a fixed hypothesis, preferring to enter the field with an open mind and let theory emerge from what they actually observe. This approach has roots in grounded theory, where researchers build categories and theoretical insights directly from the data they collect, rather than testing a claim decided in advance. Whether you need a hypothesis at all depends on whether your project is exploratory or confirmatory, so decide this early, since it shapes how you write your objectives and choose your methods.

6. Define your unit of analysis 🔗

The unit of analysis is the “who” or “what” your study is actually about – the entity you’re gathering data on and drawing conclusions from. It could be individuals, households, kinship groups, institutions, or even material artefacts, depending on what your research problem demands.

Choosing this carefully matters more than it might seem. If you’re studying shopping behaviour, your unit of analysis is likely the individual; if you’re studying teamwork in an organisation, it’s the group. Get this wrong, and you end up collecting data at one level while trying to draw conclusions at another, which weakens your findings considerably.

How it connects to sampling 🔗

Your unit of analysis is also the unit that shows up in your research questions and hypotheses, which is why it drives your entire data analysis strategy later on. It’s worth distinguishing this from your unit of observation, since a study can collect data from individuals while drawing conclusions about a community or neighbourhood as a whole. Getting this distinction clear early saves a lot of confusion when you sit down to analyse your data.

7. Map out the universe of study 🔗

If the unit of analysis is your “what,” the universe of study is your “how many” and “where.” It’s the entire population your unit belongs to – every person in a country, every household in a district, or every member of a particular ethnic or occupational group, depending on what your problem requires.

You will rarely study your entire universe directly; instead, you select a manageable sample from it. Defining this population clearly from the outset is what makes your sample meaningful, because a sample only tells you something useful if you know exactly which larger group it’s meant to represent. Skip this step and your findings won’t generalise to anyone in particular.

8. Choose your methods, techniques, tools, and analysis plan 🔗

This is where your design gets practical. It helps to keep three terms separate, because textbooks and researchers use them loosely.

Methods, techniques, and tools 🔗

Methods are your broad means of collecting data – observation, case study, genealogy, pedigree analysis, or statistical methods. Techniques are how you actually put a method into action, such as a survey, a census, or an interview. Tools are the physical instruments you use to do that – a questionnaire, an interview guide, or an interview schedule. Anthropological fieldwork often combines several of these, using participant observation alongside structured interviews to capture both lived experience and comparable data across informants.

Analysing what you collect 🔗

Once data is in hand, you’ll analyse it either quantitatively, using statistical tools to identify patterns and relationships in numerical data, or qualitatively, working through narratives, field notes, and interview transcripts to identify themes. Many anthropological studies use both, letting numbers and narratives check and enrich each other.

9. Explain the relevance of your study 🔗

The final step is stepping back and telling your readers why any of this matters. The relevance of the study is your justification: what is to be gained from this research, and who stands to benefit from it? This isn’t a formality tacked onto the end of your proposal. Reviewers, examiners, and funding bodies use this section specifically to judge whether your manuscript deserves their time or money, so it needs to be as carefully reasoned as your problem statement was.

A well-argued relevance section connects back to everything you’ve already built. It should show how your findings will fill the gap you identified in your literature review, and how your specific unit and universe of study make your conclusions meaningful beyond just your own curiosity.

What do you think? 🔗

What do you think? If you were designing a study in your own city or town right now, would you start with a hypothesis, or would you rather let the field guide your theory? And looking at these nine steps, which one do you think most students tend to rush through without realising the cost later?

References

1. https://libguides.usc.edu/writingguide/researchproblem
2. https://library.sacredheart.edu/c.php?g=29803&p=185918
3. https://guides.lib.udel.edu/researchmethods/litreview
4. https://proposaldev.ucdavis.edu/news/grant-writing-tip-think-smart-when-writing-objectives
5. https://methods.sagepub.com/hnbk/edvol/handbook-of-ethnography/chpt/grounded-theory-ethnography
6. https://socialsci.libretexts.org/Bookshelves/Social_Work_and_Human_Services/Social_Science_Research_-_Principles_Methods_and_Practices_(Bhattacherjee)/02:_Thinking_Like_a_Researcher/2.01:_Unit_of_Analysis
7. https://conjointly.com/kb/unit-of-analysis/