Thursday, May 6, 2021

Weisbach advice

I got a chance to see the page proofs of Michael Weisbach's upcoming book, "The Economists' Craft."  This leapt out at me from the preface: 

A second observation I have made over the years is that, perhaps because of a lack of good advice, many scholars, both doctoral students and faculty members, constantly make the same mistakes. Far too many publicly circulated papers contain incredibly long, mind-numbingly dull literature surveys; introductions that go on and on before they tell the reader what the point of the paper is and why the reader should bother to waste her time on it; data descriptions containing insufficient detail for a third party to replicate the results; tables that are unnecessary, badly labeled, or hard to understand; or overly dry prose written in the passive voice and apparently designed to put the reader to sleep. In addition, many scholars manage their time so badly when giving presentations that they do not get to the main results of their paper until the last five minutes of the talk. Their presentations are often poorly designed, with slides that are incomprehensible or even unreadable owing to their use of fonts so small that participants sitting more than a few rows back cannot read them. Young faculty routinely mismanage their career by not having a coherent research agenda, not getting their papers to journals, or not making connections with people in their field who teach at other universities. Sometimes they do not even bother to show up for seminars in their field at their own university.

 So true. This book's contrary advice will be very useful. 

20 comments:

  1. "Sometimes they [young scholars] do not even bother to show up for seminars in their field at their own university."

    Deepest apologies, but that is stupid behavior. Good such don't get ahead. System is working. :-(

    ReplyDelete
  2. As a young wannabe researcher, I have been repeatedly nudged in the direction of "overly dry prose written in the passive voice and apparently designed to put the reader to sleep" from both faculty and journal reviewers. Personality is verboten, unbecoming of a serious researcher, we are taught.

    When my presentations are well-designed, with slides that are comprehensible and readable, I am guaranteed to hear that my research is too simple or too obvious. Lousy presentations receive no such criticism: what is difficult to understand must be deep and complex and impressively counterintuitive.

    And it's true that I sometimes don't even bother to show up for seminars in my field at my own university. It's because so often the seminars are poorly designed, with slides that are incomprehensible or even unreadable.

    I don't like any of this, but from the perspective of a young wannabe researcher, it's the game I'm forced to play.

    ReplyDelete
    Replies
    1. This resonates with my experience. Any thoughts John?

      Delete
    2. Curious: if you did not attend a presentation, how do you know the slides were bad?

      Delete
    3. This is spot on. As another aspiring researcher I was once giving a talk at a fairly renowned place. An acclaimed professor walked into the room and, before even sitting down, read the title of my presentation out loud commenting: "This is a terrible title! Why are you trying to make it catchy? You need a boring title."

      Delete
    4. The slides are often provided in the email announcing/reminding about the seminar. And even when they're not, Bayes' rule applies: how many bad seminars do you have to sit through before you conclude the next one is also probably going to be bad?

      In any case, it's not terribly difficult or time consuming to do a little bit of research first and determine which speakers are likely to be decent presenters, e.g. by taking a glance at some of their papers.

      Delete
    5. Looks like the causality is wrong: I know the slides are bad, so I won't attend. :-)

      There is the story of somebody not attending Thomas Sargent's seminars when he was visiting Princeton. This shows even mature scholars can fail to attend class because they're smart.

      Delete
    6. This may be an unpopular opinion. For background, I have advised many PhD students and helped many young scholars getting started. As I see it, the problem is not that catchy writing is discouraged. The problem is young scholars with very little writing experience that try to be "engaging", which gets in the way of their content. Writing is hard. First you spend years just learning to get the message across. Then you can worry about style and being engaging.

      John C's early papers were not particularly engaging. Now he is a fantastic writer. It has taken years to develop, but I don't see anybody discouraging his current, very engaging, writing style.l

      Delete
    7. Yeah, I think this is it. Some of the advice sounds like "Young people should write like they're already fully calibrated to the content of the field and the culture of its community, with the confidence of decades of experience." Writing with style is like telling a joke in front of a crowd - better not try unless you know you can pull it off.

      Delete
  3. I emailed David Cox this week and he replied attaching quite a good article he published last year (at age 95!). Among its merits is the start of its *concluding* section:

    "It was a temptation easily resisted to start this article with a historical review. Instead the article concludes with some fragmentary remarks about the very extensive literature and background. Hald (2007) has reviewed developments up to about 1935 with final emphasis on the contributions of R.A. Fisher. Stigler (2016) has written with penetrating insight about more recent developments..."

    David Cox, "Statistical Significance," Annual Review of Statistics and Its Application. 2020. 7:1–10. https://doi.org/10.1146/annurev-statistics-031219-041051

    ReplyDelete
  4. For some reason, most reviewers regard dull, dry, complex, long sentences as the "academic, scientific style". If you try to put anything witty or attempt to engage the reader you are dismissed as non-serious, charlatan of some sort. My papers were rejected many times for this, and now my advice to the young is: BE DULL. Whenever there is a complex, latin, greek word - use it instead of normal English. Sadly, but it gets through much easier that way.

    ReplyDelete
  5. People write long introductions and unnecessary interviews to 1) hedge against the distinct possibility that their work will be rejected for not citing their reviewers' work and 2) demonstrating to editors that you understand your "contribution" to the literature. An all too common refrain from editors is that the paper simply doesn't make a sufficient case for how it "contributes" to the literature. It is hard to make that case unless you show how your paper adds value relative to the 943 other papers in the literature.

    The passive voice has an important place in academic writing that is supposed to convey a disinterested, external assessment of facts and data. The modern drive to extirpate from academic writing is misplaced. Not every piece of writing can read like a comic strip or John Grisham novel.

    The lack of replicability of results in economics is an enormous problem that would be easy to resolve. Any paper that is to be published in a refereed journal should provide the specific code used to generate the results as well as the data when it is publicly available. If the data are not public the code should be provided as well as mock data that can be used to execute the code. So many, many papers have been published with big mistakes because journals do not demand that data and code be made available to reviewers and the public. Some journals are moving in this direction but it should be a universal standard for publication.

    ReplyDelete
    Replies
    1. If you can't explain your contribution without the long into and unnecessary lit reviews, then you don't really understand your contribution.

      Delete
    2. In econ/finance, long literature reviews (2-4 pages) is the standard. It is not about not knowing what your contribution is. Fields like AP or CF have many papers that address similar issues, and it is nice for the reader to know how the paper fits in this (large) literature. Also, you have to option to skip the lit review and make your own judgement, so long lit reviews should never be an issue. To me, some of the comments here address personal tastes, and not what the academic community, as a whole, wants. Lasse Pedersen (http://docs.lhpedersen.com/How_to_Succeed_in_Academia.pdf) gives a sobering view on how to be an academic.

      Delete
  6. John, Thanks so much for your help with the book. Readers of his blog should know that John put in an incredible amount of time and effort into helping me make the book better!

    Mike Weisbach

    ReplyDelete
    Replies
    1. No favors, you wrote a great book. The comments rolling in are interesting, on pressures people feel to do otherwise. I hope the politics of economics are not so changed that they are not right -- that simple, clear, honest, politically neutral writing research and presentation, and good citizenship, still are the key to eventually greater influence.

      Delete
  7. As a PhD student, I concur that this is good advice. Most presentations are boring, long winded and do not get into the meat of the papers we're supposed to be discussing.

    You have about 1 or 2 minutes to convince people they should listen and the ability of people to actively listen is a window of 40 to 50 minutes, so if you only get to results after 45 or 50 minutes, no one will remember what you said. More to the point, the whole "fun" of economics is the interplay of theory and data... so why the hell wait so long? In a typical seminar, you'll also always get astoundingly stupid discussions about assumptions 5 minutes into it. They have no clue where this is going, but they want to dispute the assumptions. Here's a better way to do it: suspend disbelief, see how far you run -- then come back to the assumptions.

    The comment about papers is also true. One of my own current papers contains way too much stuff for my own taste -- and we got slammed asking for more. It's as if people think being critical means asking for what someone didn't do. It's definitely easy to play that game because it's always true that you have to draw somewhat arbitrary lines somewhere and stuff gets left out, so you might attribute some of that tendency to the peer review process.

    But I would disagree with many people here that we must comply. What I learned trying to publish those heavy empirical studies on macroeconomic forecasting is that we just loose people in the weeds trying to do too much in too many dimensions in just one paper. If you look at some of the more successful economists in academia, they rarely do something insanely complicated that is presented in a clunky paper that never ends.

    ReplyDelete
  8. It appears to be a self-help book for juniors, but I'm sure seniors will read it too. Which will make it a great discussion book, helping people to get on the same page. We can't have young people thinking they need to do crap in order to succeed.

    ReplyDelete
  9. Much of what he is complaining about is the result of the incentive system in place. Who hasn't greatly expanded their lit section due to a petty referee? Young economist are responding to the incentive structure in place and it's senior economists who put or keep that structure there.

    ReplyDelete
  10. We'd love to have a guest blog from you! :)

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.