Also this week: the replication crisis vs. significance testing, work vs. thoughts of work, Peanuts vs. Superman (but not that Superman), and more.
Kathryn Paige Harden on what to do with the science of terrible men. I think this is quite good, even if you don’t agree with it (I don’t know enough to fully evaluate it myself). Harden wants to reclaim human behavioral genetics for egalitarian ends.
Deborah Mayo argues that the replication crisis in psychology actually vindicates statistical significance testing.
This is old but I missed it at the time, and it’s perennially relevant: the more hours people actually work, the more they overestimate how much they work. (Or maybe in some cases, the more they lie about how much they work, but I suspect it’s mostly just unconscious overestimation). The linked article makes the interesting broader point that people tend to overestimate the amount of time they spend doing anything they feel they “ought” to be doing. Reinforces this old post of Meghan’s on how you don’t need to work 80 hours a week to succeed in academia–because academics don’t actually work 80 hours/week, even if they think they do. This doesn’t mean you’re wrong to feel overworked! It just means that the lived experience of feeling overworked doesn’t arise because each hour worked automatically translates into one additional unit of “feeling like you’ve worked”. There’s no 1:1 mapping between hours worked and feelings of overwork. I think it’s good to recognize that, because addressing those feelings of overwork probably isn’t as simple as just cutting your work hours.
The Economist’s interactive Covid-19 hospitalization and mortality risk estimator. A very good example of public science. Does a very good job of explaining the underlying data and how the risk estimates were derived from the data. Good reading for a biostats course.
Zeynep Tufekci on mental models, statistical power, and vaccine efficacy. A bit overlong and meandering for my taste, but the core of it would be good reading for intro biostats students.
And here’s Zeynep Tufecki on how the panic over “long covid” is due in large part to sampling bias. Another good example to give to intro biostats students.
Why do some academics fake assumed identities? I mostly avoid linking to discussion of these cases. It’s not as if anyone needs my help finding discussion of them! But in the unlikely event that you want to read more about these cases than you have already, I thought the linked piece was good. This book on literary hoaxes and cultural authenticity is good too. There are both analogies and disanalogies between literary hoaxers, and academics faking assuming identities. See also non-military people who fake military experiences they never had. They skew male, which makes for an intriguing contrast with academics who fake assumed identities.