Tuesday, June 11, 2013

It is time to quit!


There is an interesting Freakanomics podcast about "The upside of quitting."
It is about the tension between "sunk costs" and "opportunity costs."
It points out how we often refuse to quit something [a job, project, relationship, investment] because we have invested so much time, energy, or money it [sunk cost]. But, we overlook the fact that not quitting has a significant [opportunity] cost because we could instead invest our time, energy, or money in something that will yield better returns in the future.

What does this have to do with science and academia?

Steve Levitt claims that a key to his success in economics research is that he is good quitter.
He fails fast. He has many ideas, but quickly decides which ones are not going to work out, and drops them fast. He thinks his competitors hang on longer, keeping on hoping something will come of their cherished idea that looked so promising. This wastes precious time that good have been spent on the good ideas.

What are some other areas where this is relevant to science and academia?

-specific research projects
-specific collaborators who don't pull their weight
-graduate students you are supervising and not making progress 
-applying for specific grant programs
-applying for faculty jobs
-teaching courses with consistently low enrolments

Success in science does involve a lot of perseverance, failure, and frustration.
However, never giving up on some of the above can be a mistake.
Furthermore, it is important to make a rational calculated decision, rather than just letting things drag on forever or living in denial.

Are there things you wish you had quit earlier?

1 comment:

  1. With efforts that span many long years like Andrew Wile's proof of Fermat's last theorem or more recently the 500-page, 10-year-in-the-making 'supposed' proof of the ABC conjecture by Shinichi Mochizuki (http://en.wikipedia.org/wiki/Shinichi_Mochizuki), it seems to me that an individual researcher's projects can stretch a lot without losing scientific value if among other factors they have precisely defined goals ...

    ReplyDelete