Forming a hyper-precise numerical summary during a research crisis can improve an article’s chance of achieving its publication goals.

Speaking of measurement and numeracy . . . Kevin Lewis pointed me to this published article with the following abstract that starts out just fine but kinda spirals out of control:

Forming a military coalition during an international crisis can improve a state’s chances of achieving its political goals. We argue that the involvement of a coalition, however, can have unintended adverse effects on crisis outcomes by complicating the bargaining process and extending the duration of crises. This argument suggests that crises involving coalitions should be significantly longer than crises without coalitions. However, other factors that affect crisis duration are also likely to influence coalition formation. Therefore, taking into account the endogeneity of the presence of a coalition is essential to testing our hypothesis. To deal with this inferential challenge, we develop a new statistical model that is an extension of instrumental variable estimation in survival analysis. Our analysis of 255 post–World War II interstate crises demonstrates that, even after accounting for the endogeneity of coalition formation, military coalitions tend to extend the duration of crises by approximately 284 days.

Approximately 284, huh? What’s the precise number, 283.734908243098230498?

Somehow I’m reminded of my favorite sentence from any quantitative research ever:

Participants reported being hungrier when they walked into the café (mean = 7.38, SD = 2.20) than when they walked out [mean = 1.53, SD = 2.70, F(1, 75) = 107.68, P < 0.001].

P.S. I have not read the paper on military coalitions and it might well be wonderful and important research. If you’re interested in the topic, go read it! Make your own call on its quality and on the relevance of this research to the real world. I just think that this “approximately 284 days” thing is hilarious, and I say this as someone who’s approximately 174.3 centimeters tall.

P.P.S. Lewis responds:

Cool. But what is the ideal phrasing? I assume you wouldn’t say “coalitions SIGNIFICANTLY extend crisis duration.” Is it ok to say “extend crisis duration by almost a year on average”?

My reply: To start with, I’d express the result in the past tense as these are past data! “Crises with military coalitions lasted 300 days more on average.” Something like that.

9 thoughts on “Forming a hyper-precise numerical summary during a research crisis can improve an article’s chance of achieving its publication goals.

  1. “approximately 284 days” — I can also imagine something like “approximately 280 days” hiding hilarity. For example, my university newspaper once had the following line: “Approximately 20 people were there” [so far so good, but then it continued] “19 men and 1 woman”

  2. Off topic, but relevant to this blog because it is, apparently, another example of a scientist wanting to avoid review of his work. (There was also recent discussion that Mann’s climate change opinions were more reliable than the IPCC’s.)

    Michael Mann’s lawsuit against Tim Ball was dismissed with prejudice because Mann refused to turn over his code by the court-ordered deadline. Mann sued Ball nine years ago for disparaging his climate change work (“the hockey stick”).

    Here is a restrained description of the opinion: https://www.powerlineblog.com/archives/2019/08/michael-mann-refuses-to-produce-data-loses-case.php

    Here is a much more nasty and detailed discussion: https://wattsupwiththat.com/2019/08/22/breaking-dr-tim-ball-wins-michaelemann-lawsuit-mann-has-to-pay/

    Ask yourself. Does Mann sound like a responsible scientist, or does he sound like some other scientists discussed on this blog? Should we be basing enormous decisions on Mr. Mann’s opinions?

  3. What excuse does he give for not sharing his code? Also, if the court ordered him to share it but the failed to, isn’t that contempt of court or something?

    Btw: My guess is that this climate change frenzy is hitting a fevered pitch because they want to implement some policies to take credit for when the next grand solar minimum occurs. I’ve been keeping track of the sunspot observations[1] and it increasingly looks like that could be in the next decade. The upcoming Cycle 25 is now predicted to be the weakest solar cycle in 200 years.[2]

    [1] http://www.sidc.be/silso/datafiles
    [2] https://www.nasa.gov/feature/ames/solar-activity-forecast-for-next-decade-favorable-for-exploration

  4. Isn’t it obvious? 284 days plus or minus 12 hours. Because in international affairs the coalition-forming and the military crisis will tend to be in different time zones.

  5. Something to consider: if you promote the practice of rounding to the nearest digit that uncertainty will allow, which I think is what you are asking for, there will be two unintended consequences: 1) people who dont share that belief will give authors who do round, if it works out favorably for the authors, a really hard time and unfairly accuse them of padding their stats, and 2) in practice many authors will only do the rounding if it makes their number look better, not worse, and then there will be yet another filter to think about when estimating the edlin factor.

    Basically, this is one of those practices where if you were the dictator of stats and could make everyone do it, it would be a good idea, but since you aren’t it will just make things worse.

    • Rick:

      Could be. My problem with a statement such as “tend to extend the duration of crises by approximately 284 days” is one of the problems that Orwell was talking about in his famous essay, Politics and the English Language, which is that the statement is written in a pseudo-authoritative style.

      – What is meant by “tend to”? What’s really happening is that they are extrapolating from historical averages. So why not say that directly?

      – “Approximately 284” is another phrase that inappropriately implies authority, as if the authors have the exact number in their pocket. If in the abstract you want to say 284, or 290, or 300, just say it! The uncertainty interval will be in the full article anyway.

      I have no reason to think these authors are doing anything wrong on purpose; I’m sure they’re just trying to be precise and accurate. The trouble is that they’re not being precise and accurate; they’re doing one thing and saying another. Bad habits, as social scientists are always being encouraged to make big sweeping statements rather than to humbly report what they found.

      That said, you could be right in what you say in your comment.

      • ““Approximately 284” is another phrase that inappropriately implies authority, as if the authors have the exact number in their pocket.”

        I don’t know. I don’t see how “approximately 284” implies authority, an exact number in the authors’ pocket. To me it sounds like the authors are emphasizing that there is uncertainty to the number, and the 284 is their central estimate.

        And I don’t see why rounding to 300 is better. That’s just an artifact of our use of the decimal system, which is, in turn, an artifact of hand anatomy. If we worked in base 4, 284 would be a round number.

        • Clyde:

          It’s about communication. If the article was written to be read by computers, and the numbers in the article were to immediately go into some software that computed averages and differences and so forth, then, sure, they could report 283.734908243098230498 and that should be fine. But the article is written to be read by humans. Report 283.734908243098230498 and the implication is that this last digit conveys some meaning.

Leave a Reply to Anoneuoid Cancel reply

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