Friday links: experiments vs. observations vs. p-hacking, Andrew Gelman vs. Nate Silver, new Expression of Concern for Jonathan Pruitt, and more

Also this week: stuff we’re not linking to.

From Jeremy:

Yet another Expression of Concern for a paper co-authored by Jonathan Pruitt, for which he collected the data. This time it’s DiRienzo et al. 2012 Animal Behaviour. Read the admirably detailed notice, which first author Nick DiRienzo screenshotted and tweeted. This one is for anomalous patterns of repeated digits in purportedly stopwatch-recorded decimals. The anomalous data weren’t used in the paper, but other data from the same experiments were, and so the existence of the anomalies raises concerns that those other data aren’t reliable either. Pruitt couldn’t explain the anomalies, and apparently did not agree to the Expression of Concern. Kudos to Nick and co-author Ann Hedrick for doing the right thing here. This is especially rough on Nick, because this was his first independent project in grad school. (Aside: If you are new to this whole saga, see here and here for some background. And make sure you’re sitting down.)

In economics, randomized controlled trials are less commonly p-hacked than are studies using observational methods to infer causality (differences-in-differences, instrumental variables). Somebody should look at how the frequency of p-hacking in ecology varies with the study design. I for one would be very interested in the outcome. (ht @noahpinion)

Andrew Gelman vs. Nate Silver on uncertainty in US Presidential election forecasting. Raises broader issues about uncertainty in any forecast.

This week in Stuff I Am Not Linking To: apparently a prominent scientist/activist on Twitter spent years…uh…inventing and pretending to be another, non-existent, prof on Twitter? I know little about this and DEFINITELY have nothing to say about it, and I plan to keep it that way. So I’ll just recommend this book I’ve been reading. It’s interesting, and seems like it might be relevant to the stuff that I am not linking to.

From Meghan:

There’s a new preprint on safe fieldwork strategies for at risk individuals that might be of interest to many folks.

3 thoughts on “Friday links: experiments vs. observations vs. p-hacking, Andrew Gelman vs. Nate Silver, new Expression of Concern for Jonathan Pruitt, and more

  1. also this week in Stuff I Refuse To Link To: Remember that Harvard Law School prof who was scammed by a young woman with whom he was having an affair and who may have been a serial scam artist? And how New York Magazine wrote a story about it that included quotes from an interview with the prof? The prof is now *suing New York Magazine* for…(wait for it)…*breach of contract* and…(wait for it some more)…*sexual harassment*.

    Yes, really.

    • Also there was some mass food fight recently about whether worms are good, or something? And some mass food fight about whether 2+2=4? It has been a banner week for Online Stuff I Am Glad To Be Entirely Ignorant Of.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.