Also this week: measuring journal reputations, BEST PAPER TITLE EVER, and more.
Programming note: Jeremy is very busy today, so comment moderation may be slow.
Over at EcoEvoEvoEco, Andrew Hendry is running a blogpost forum on concrete, practical proposals to improve data integrity in published papers, particularly in the context of collaborative work. Follow the link to see the contributions to the forum so far, and for instructions on how to propose your own contribution. As Andrew emphasizes, this forum is inspired by, but not specifically directed at addressing, the Jonathan Pruitt situation. I like this, because the Jonathan Pruitt situation is very unusual, and we don’t yet know all the details. So I think it can be an occasion to reflect on current scientific practices. But I don’t think that, in and of itself, it should have much effect on the outcome of those reflections. If there are systemic changes to our current scientific practices that would be good to make on balance, well, they’re good independent of the Jonathan Pruitt situation. Conversely, if there are systemic changes to our current scientific practices that would be bad to make on balance, well, they’re still bad independent of the Jonathan Pruitt situation. Just my two cents.
Quanta magazine covers doubts about whether GWAS can really uncover the signal of past selection in human genetic variation.
Pop songs are getting gloomier, by many different measures. It’s a long-term trend. Might have to add this to my list of handy examples for teaching basic statistical ideas.
Using polling data to determine sociology journal reputations. Curious what you’d find if you did the same thing in ecology.
Apparently, scraping public websites is now legal in the US. Question to someone who knows more about this than me: does this apply to JSTOR or scientific publisher websites? Has the way been cleared for anyone to do text-mining of the scientific literature, without first needing publisher permission?
Here’s a story of how SoftBank’s venture capital fund is “blitzscaling against itself”. It isn’t pitched as an illustration of how individual-level selection can undermine group-level selection. But it’s an excellent illustration nonetheless.
BEST PAPER TITLE EVER. Leave nominations for second place in the comments. 🙂