Finding a Research Topic Vs Doing the Research

One common question among new and prospective graduate students: “Is identifying a research topic harder than doing the research and then writing papers / your dissertation PhD dissertation?”

The answer to this question very often depends on your source of funding (if any). If you have a research assistant role in grad school, you are mostly likely given a publication-worthy topic(s) on a silver platter. But you may not like the food on that platter as much, at least initially. Kind of like my kid and nearly any vegetable you put in front of him.

But what if you’re shown a grocery store aisle with 1000 different kinds of soup (I know, that’s not many for a normal grocery store, but just play along), and you have to pick just one? Or better yet, you need to come up with a new kind of soup that is sufficiently different from all the other soups, and good enough to deserve a spot on the shelf? Odds are that cooking and eating the soup will be the easy part.

So if you have an external fellowship or a Teaching Assistant position (or are taking out loans to pay for school), then you’re probably going to be faced with this challenging soup task. You have near infinite options for the research you perform. Finding topics that can lead to publication-worthy results that actually matter to your field can actually be the hardest part for many graduate students – including myself. Once I identified good potential soup recipes (specific research objectives), it was far less difficult for me to execute on that idea than it was to come up with the idea itself.

I was fortunate that as part of my external fellowship I was ask (cough required cough) to visit different NASA centers each summer, and during those visits I obtained the majority of the research ideas that became my peer-reviewed publications (and thus dissertation). If you do not have an RA role or an advisor that is a fount of awesome research ideas, definitely pursue internships and/or relationships with folks in industry (and they can even be outside the soup industry if you want). You can build yourself a nice pipeline of soup ideas (and maybe a pipeline of soup too, yum) that you can be confident are 1) desired for real-world applications, and 2) do not currently have any published solutions (assuming your industry contacts have done their homework).

One big challenge when trying to come up with soup ideas as you read the millions of peer-reviewed soup papers: after a recipe idea pops into your head, your first instinct is likely to check if others have pursued and published the same recipe. Campbell probably came up with it 50 years ago. But guess what? All of their recipes from 50 years ago were written in crazy arcane language that is hard to understand. And what about all the other soup manufacturers and boutique soup makers? It can prove extremely difficult to parse all the available literature for proof that no one has already taken your idea. I remember reading many papers when trying to do a search like this, and feeling so discouraged and overwhelmed: “well, maybe this author/paper is doing something like my idea, but I can’t really tell…. his wording and descriptions are so different than I would use…”

So, here’s what I recommend instead: right after a delicious idea pops into your head, build a simple toy example (as simple as you can make it) that demonstrates your solution – which by design should not take long. Afterwards, the problem statement and solution should be much clearer in your mind – all the ingredients and the exact steps to make your soup. Only after you’ve done this, should you search the literature – you’ll probably find it much easier to know if someone has published your recipe. And if you do find someone has published the idea, especially somewhat recently, guess what? You had a publication worthy idea! Someone just beat you to it – which is not the end of the world. Why? Because 1) odds are you can continue to come up with publication worthy recipes, and 2) you can probably derive recipe ideas from this first idea using your toy problem setup, which are much more likely than totally random ideas to find a way to the grocery store shelf.

Good luck in the kitchen! I am really wanting soup right now for some reason….