7 Concrete Strategies for Jumpstarting Research Ideas

By Jessica Jeffers

In my third year, I started looking in earnest for some good research ideas. But I didn’t know where to start. People kept saying things like “It will come” which was, to put it mildly, not very helpful. Here are seven concrete strategies to help you jumpstart the idea process. Disclaimer: this is based on my experience in my field (finance) – I definitely recommend asking older students in your field for tips. If they are giving you the good old “Oh, just keep thinking,” try showing them this list as a starting point, and see where they agree/disagree.

1.     Start with the basics: keep an idea journal in which to write down ideas or questions as they come up (seminars, talking to a friend, browsing facebook). Choose something that you can access easily from different places/ at different times, since ideas can pop up unexpectedly. I like Microsoft OneNote which I keep on Dropbox. Other ideas: Evernote, a note-taker on your phone, a small moleskin. It helps to have different tabs to keep track of various topics as your ideas grow. 

2.     Pick a couple of top journals in your field. Read through abstracts of a) most cited articles and b) forthcoming articles and jot down the research questions. Then go back through and see if you identify broad patterns in the questions and highlight questions you find interesting. Keep these in mind as you read the news and talk to friends. Proving a top article wrong is also a possible path to an idea.

3.     Spend a half hour to an hour every morning reading some news source(s) relevant to your field. For example, I read the New York Times’ DealBook, but you can also try blogs or RSS feeds of articles related to your research interests. Bookmark articles that are thought-provoking in a "Research Ideas" folder. Clip particularly interesting paragraphs or figures into your idea journal. 

4.     Set up coffee with 1-2 friends in your department to discuss ideas on a weekly basis. Before you meet up, pick a couple of ideas in your journal and try to write out an abstract for them. Why do people care? How is this a contribution? What is the coolest finding you could get?

5.     Become a data hound (or model hound or lab experiment hound as applicable). By that I mean become an expert at what is out there. This can give you ideas directly, or later when an idea comes up you will know what is available for making that idea happen. For data, go through WRDS, data.gov, and the library website (you'd be surprised what they have!). Ask older students. Apply for data often and early, especially if you’re in a field like health where data takes a long time to come through. Google anything you think or hope might exist. Keep notes and bookmark online resources in a special folder. 

6.     Think about your ideal experiment or setting and find out if it exists! For example, if your ideal setting is a lottery in school assignments or a change in medical billing practices, start by googling “lottery in school assignment” or “laws on medical billing practices.” You may not directly find what you are looking for, but in the process still learn interesting facts or necessary jargon. Policies, laws, and experiments are particularly good candidates for this. You can try looking for specific states, but also don’t be afraid to look outside of the U.S.

7.     If you come up with an idea you think is interesting and aren't sure if it's feasible, give yourself 1-2 weeks to work on a mini-proposal for the topic. Compile potential data sources, analyses, and background information and write up a few page summary. Then go talk to a professor about it! Even if this doesn't lead to your dissertation topic, it can be useful to quickly determine how worthwhile a topic is, and in doing your background research you might happen upon another interesting question. Doing this “up front” work now will also save you a lot of time later if you discover that many people have already written about this topic, or it really isn't as promising as you once realized.