Read on Twitter
How to generate ideas@ubcVSE students have started to ask me how to generate research ideas, now that they are moving to 2nd or 3rd year. I don't know whether I answer this question well, but here is what I tell them: [1/N]
Suppose you are broadly interested in topic X (e.g. policing, religion, entrepreneurship). Read econ papers on X -- especially those well-published and highly cited. Do this not for ideas, but to know what has already been done. Your goal now is to generate *new* ideas. [2/N]
Now read work on topic X by academics outside of, but close-ish to, econ (e.g. psych, sociology, anthropology). I got ideas this way by reading "The Person and the Situation" and Haidt's books. Read the academic papers referenced in those books. [3/N]
Recognise your comparative advantage. Grew up in (small non-US) country Y? Read the national newspaper. You'll identify policy experiments in Y quicker than others. Spent your childhood in scouting? Look into work on how scouting moulds leadership traits. [4/N]
You have your first idea. Write it down in a notebook or Excel sheet. Now try to kill it, and quickly. Pitch it to friends and to profs. Listen carefully to their feedback. If the idea isn't salvageable, shelve it, and go back to generating ideas. [5/N]
This process is really hard, but I think might be easier if you know going into that (a) most of your ideas will be bad (especially initially) and (b) most of the good ones will not work out. This is natural, and given this, it really helps to work through ideas quickly. [6/N]
Now you have an idea that survived the first round of feedback. Try to kill it again. Suppose it is a natural experiment and there is a key balance check -- without that, the project is DOA... [7/N]
...do the balance check first, and carefully! Forget about everything else that sounds more fun -- reading that massive book on the policy, web-scraping the outcome data, etc. You can waste less time this way. [8/N]
How do you stick to this process over the summer? It depends. Many students may be burned out from first/second year. Your job is to recover -- go on holiday, work on whatever will rekindle your interest in economics. [9/N]
Other students want to dive in. Your job is to bring some discipline to a process without immediate rewards. You could try committing to 2 hours per day of ideas-focused work. You could also organise weekly 1st year or 2nd year idea-sharing with classmate feedback. [10/N]
Idea generation is probably the central bottleneck in PhD programs and yet we mostly don't try to teach it. But I'm new to this game -- and only gradually learning whether idea generation is even "teachable". Given that, I would love to hear what others think! [N/N]
Addendum: what I have said here is not new, and I just remembered that @PaulFNiehaus has written eloquently about idea generation (and related issues) here: https://medium.com/@paul.niehaus/doing-research-18cb310529e0. Read his piece too -- you won't regret it! [N+1/N]
