Types

Now, before you sit down to write a plan, you have to make a few decisions about how you aim to do your research. The single most important thingto understand is that there are different ways to do research and, correspondingly, no one right way to plan it. The first step is to decide what kind of research tradition is yours, and situate your work into it. For School of Design, three models are particularly relevant. The list is not exhaustive, leaving aside such things as the formal sciences.

Traditional hypothetico-deductive research

This is the most typical research format. Basically, your job is to read all relevant existing literature, find problems in it (i.e. identify your claim), develop a hypothesis that you test in your study, and devise a resaerch design that makes it possible for you to test whether your hypothesis is right of wrong. "Hypothesis" is your theoretical answer to the research question; you do not know whether it is true or not befoer you gather data and analyze it.

This tradition comes from the natural sciences, but is the mainstream in most other field of research. Typically, the proof takes a statistical form.

If this is your choice, read methods literature from this tradition, go to library, study literature carefully to see whether it does answer your question, and only then sit down to write a hypothesis (or a series of hypotheses). After this work, you can get to methodology: how to gather data (statistically or through controlled experiments or quasi-experiments), whether you want to construct something, which method to use to analyze data (tip: post-it notes won't do, you need statistics), and what is the crucial test for your hypothesis.

Interpretive research

In particular when studying people, this is an alternative. Under various names, most research in 20thC humanities, the social sciences and philosophy is interpretive in nature. Briefly, you do not try to theorize how people behave and use obscrure theoretical language to write down your hypothesis, but begin with people: their language, their terms, their behaviors, and so forth, and try to explicate data you have gathered.

Don't be fooled. Is this sounds like an easy alternative, it is not. You need to know exactly what you are doing. You need to read youself into some interpretive tradition - like interactionsm - and its methodology, write down a claim to justify your research, plan data gathering carefully andin detail, and describe your analytic plan carefully. I have given several doctoral level classes about qualitative research, and even written books about them. Check them - or anything else - to get the basics right.

Constructive research

Under various guises - most misleading - many designers think that they need to do something "new" outside the traditional sciences. Typically, they want to construct something and integrate it into their research. Actually, there is nothing complex here. This is how most sciences proceed. What, after all, is the difference between constructing an interactive table and a vaccine, and testing it with people? Most work in engineering proceeds exactly like this.

There are all kinds of ways to conduct construcive research. Software engineers and business studies talk about constructive research; social scientists about quasi-experimental research, policy research, or action research; medical researchers about clinical research; artists about practice-based research; and so forth. The question is really what kinds of new knowledge the construction helps to gain: it is new knowledge that justifies time and effort put into construction. Also, what is important to decide is how the construct is integrated into your study: is it like a "treatment" in medical research, a "breaching experiment" like in ethnomethodological sociology; or what. Also, you need to decide how you gather data, and how you analyze it to test your hypothesis, or to create an itnerpretation.

At the SoD, a constructive approach is more than welcomed, unless you make ridiculous claims about novelty, when in fact design researchers have invented very, very little new things.