Originally published on March 16th, 2014 at generalcirculation.blogspot.com
Great conversations can be had over tea. My roommate and I will often find ourselves gravitating toward the kitchen in the late evening, orbiting the tea kettle until it whistles and enjoying the brief distance from our work.
During a recent nighttime tea break, my roommate (who is a masters student in public policy) admitted to a fascination with PhD students--what do we spend our time doing, and how do we know what to do on any given day?! My off-the-cuff answer surprised me: "well," I said, "we spend most of our time trying to figure out what questions to ask."
The scientific method--as laid out by that most reliable of sources, Wikipedia--starts with defining a question (e.g. Why did the chicken cross the road?).
Once you have a question, you: -do some information gathering on your question (What was on either side of the road? What might be some primary motivators for a chicken? What are some existing hypotheses for why the chicken crossed the road?) -come up with a hypothesis (The chicken crossed the road to get to the other side) -design and perform a test for your hypothesis (I welcome suggestions for how to test this hypothesis) -analyze the results of your test -draw some conclusions about whether or not your hypothesis was right -rinse and repeat from the hypothesis formation point until you have a correct hypothesis -and, finally, publish your results.
We all learned the scientific method in elementary school, right? But what I've only recently realized is that step one, defining the question, is actually the hardest part.
Defining your question is only the first step in the multiple step scientific process, the end result of which (publishing your results) is what ultimately gets you hired in an academic job or gets you tenure (though there's an interestingongoingdebate on whether this is the best mode of evaluation). With the time pressure of trying to finish a PhD or beat the tenure clock, it can be so tempting to rev through that first step so that you can get moving on the rest of the process.
Defining a good question is hard. It requires reading lots of journal articles--the result of other people successfully getting all the way through the scientific process--so that you know what questions have already been asked. (And, let me be honest, trying to read scientific journal articles sometimes makes my brain feel like swiss cheese.) It requires going to scientific seminars to figure out where the interesting questions in your field may lie (and not spending them making to do lists instead of listening intently). Defining a question requires much more mental energy and focus for me than sitting down to write a piece of code. But without a fully baked research question, you can end up wasting a lot of time.
My seemingly glib response to my roommate has reasserted itself several times this week. On Monday, I met with some collaborators to figure out how to set up some model runs we want to do. We'd been going back and forth and confusing ourselves about how to structure things for about two weeks. In the end, it all came down to the fact that we hadn't thought thoroughly enough about what question we were trying to ask. On Tuesday, I had a meeting to pitch an idea for a policy paper--a pitch that I don't think ended up being very successful, partly because I didn't properly articulate the question I thought it would answer. On Thursday, I met with a professor about a possible final project idea for a class. That meeting went very well, largely because the whole project was inspired by a singular question that I wanted to figure out the answer to.
I'm a naturally impatient person; delayed gratification is not my strong suit. But the virtues of a well-defined, interesting research question may be enough to overcome even my flighty disposition. I resolve to spend more time figuring out what questions to ask. Perhaps if that were a more universal resolution, we could have been spared ever being asked the question: why did the chicken cross the road?