Further evidence that creativity and innovation are stimulated by college sports: Evidence from a big regression

Kevin Lewis sent along this paper from the Creativity Research Journal:

Further Evidence that Creativity and Innovation are Inhibited by Conservative Thinking: Analyses of the 2016 Presidential Election

The investigation replicated and extended previous research showing a negative relationship between conservatism and creative accomplishment. Conservatism was estimated, as in previous research, from voting patterns. The voting data used here were from the 2016 US Presidential election. The number of patents granted per county in the United States was used as estimate of creative and innovative accomplishment. Using a 2-level multilevel approach, in which state-level influences are taken into consideration, various control variables were tested, including socioeconomic status (SES), education, income, and diversity. The results confirmed a negative relationship between conservatism and the number of patents granted. Therefore, in counties and states with high conservatism, fewer patents were granted, even after controlling for SES and population. Patents were positively related to racial diversity and education. Practical implications include the benefits of liberal thinking outside of the political arena. Liberal thinking is very likely associated with flexibility, tolerance, and openness, and according to the present results, creative accomplishment. Limitations of the research and future directions are discussed.

I’d really like to think this a parody but it just might well be serious. I wonder what Susan T. Fiske would think of it. On one hand, it’s ridiculous. On the other hand, it’s a peer-reviewed publication with p less than 0.05 so it’s got to be true. It’s a tough call.

Meanwhile, I have an idea that, outside of certain big cities, the number of patents in a county is associated with the presence of college sports teams. I conjecture that the presence of college sports stimulates the sort of creative thinking. Somebody get a p-value on that, ok?

P.S. Some would say it’s uncool to mock a paper in an obscure journal. And, indeed, I’ve tried to show some courtesy by not mentioning the authors’ names, as I’m sure they’re doing their best (or perhaps doing an awesome parody; it’s hard for me to tell as the article is beyond a paywall). But I don’t buy the argument that we should be gentle on bad research just cos it’s in an obscure publication. I’ve published in obscure journals myself, and I still hope these articles I’m writing can make some positive contribution. If you really don’t want your work criticized in public, you shouldn’t publish it at all. Or, to put it another way, if you’re ok with citations and positive press, you should be able to handle criticism.

22 thoughts on “Further evidence that creativity and innovation are stimulated by college sports: Evidence from a big regression

  1. “Kevin Lewis sent along this paper from the Creativity Research Journal:”

    The link doesn’t work for me. I looked up the journal, and it seems like the paper is listed there but is not available anymore (?).

    http://www.tandfonline.com/toc/hcrj20/current

    What i found interesting though, is that it seems that the 1st author of the paper is actually one of the 2 editors of the journal itself. If this is correct, i find that strange.

    http://www.tandfonline.com/action/journalInformation?show=editorialBoard&journalCode=hcrj20

      • Hmm, perhaps it’s just me then. I tried to access other articles listed on the journal page i linked to above and am able to access those, just not the paper in question. I get an error message, either through clicking on your link or clicking on the paper listed on the journal page:

        “Error

        There appears to be a problem with the page you are looking for. Either the content has moved, or the URL is incorrect. Please check the URL, or try again.”

        Thank you for checking though!

  2. Another example of Terry’s Law. Manifestly silly academic papers are publishable if they produce a correct soundbite.

    Soundbite 1: “conservatives are stupid”
    Soundbite 2: “racial diversity is good”

    The authors wanted a publication, so they whipped up a silly paper with correct conclusions. An academic journal gets to sneer at conservatives as uncreative, and, as an extra lagniappe, laud racial diversity. Political activists get to sneer condescendingly by citing to an academic paper no one will read.

    • +1

      Though I’ll bet a lot of people read it via the link in the inevitable Daily Kos / Alternet / Free Thought Project / etc. article that gets shared all over Facebook and Twitter…

      • Exactly.

        Daily Kos et al. will report it as a “scientific finding” and no one who gets their information from those sources will read the actual paper, or even be capable of reading the paper.

        That’s why there is such a demand for these results-oriented papers.

  3. This paper (or the abstract, at least) is patently silly. For instance, it doesn’t take into account the process through which people consider, apply for, and receive patents. It helps to have other inventors around (who can guide you through the process); I’d think this would be likelier in urban areas, where there are more inventors clustered together and more companies involved with invention. It does not seem that the authors considered proximity of other inventors–or of institutional support–at all.

  4. I think it’s a shame that many researchers ignore how many good sources of public data there are to answer social science questions, but then I see something like this. Sometimes the price we pay for free public data is that the data do not allow bulletproof conclusions, at least not to the extent we might have if we could design the perfect study. But sometimes the data do not permit any conclusions and we just shouldn’t use them.

  5. Have not read the full paper, but it doesn’t seem like they are making any fatal flaw here. If anything, they should be using more control variables (doable) and making sure not to overstate their results or claiming causality (again, maybe they do). But it doesn’t seem like a terrible contribution to make, especially given it is a somewhat obscure journal. We already have some prior knowledge that conservatism is negatively correlated with creativity—at least in Western circumstances—so it doesn’t seem to be too out there of a claim to make, even if it likely does not control for enough variables.

    • Mark:

      A big trouble here is that it is people and organizations that do patents, yet the analysis is performed on the county level, and there are tons of differences between counties. Also, the paper says, “Liberal thinking is very likely associated with flexibility, tolerance, and openness,” but, as a commenter above pointed out, patents are associated with restriction and closure: the whole idea of a patent is that you get to make money off an invention, and nobody else gets to use it without your permission. This is the opposite of flexibility and openness! So I think the theory here is pretty hopeless in that it could explain just about any data pattern or its opposite.

      But you’re not the only person out there who thinks that there could be merit in this paper, which to me seems transparently hopeless. To start with, there are the three authors of the paper plus the journal editor who accepted it. I think we’re going through a change in science, where people are gradually realizing the problems with analyses like that in the paper under discussion, or power pose, or himmicanes, etc.

    • One way to see the silliness of the paper is to consider the other conclusions the authors could have probably reached if they had wanted to. The engine driving these seems to be that college towns produce a lot of patents, so you could probably get similar results for other variables associated with college towns.

      For instance:

      College football teams (as pointed out by Andrew)
      Sales of Decembrist albums (popular with college students)
      Income inequality (very high in many college areas such as Berkeley and Madison)
      Segregation (geographical separation of town and gown)
      Low upward mobility for poor people (a Raj Chetty finding iirc)

      But, of course, the authors didn’t want to find those findings, so those findings were not found.

  6. Their study sets out to prove: “that Creativity and Innovation are Inhibited by Conservative Thinking”

    But then they seem to conclude that: “Liberal thinking is very likely associated with flexibility, tolerance, and openness, and according to the present results, creative accomplishment.”

    I have not read the paper, but does less conservative thinking imply more liberal thinking? Can you draw conclusions on liberal thinking by testing conservative thinking? Is my thinking off?

Leave a Reply

Your email address will not be published. Required fields are marked *