No one's stomach gets upset in a Waymo. Self-driving cars drive well, so they're going to make people more comfortable when they arrive places.

Remember: There's no (trial-based) evidence that parachutes save lives when people fall from planes.

This is a mockery of scientific reasoning. It says there's no evidence for tons of things that have great evidence, just because it's not trial-based. Plenty of people turn their pale skin brown or black with melanotan. Saying there's no evidence should be disqualifying.

Everyone working in a STEM field should read this - ‘writing is thinking’

Here's something heartening: Individuals who experienced more economic growth as kids tend to see the world as less zero-sum. That is, they're less likely to see someone's gains as another's loss.

It's been funny to watch the Coldplay CEO scandal develop, but the punishment - the sheer scale of being ritually humiliated by the entire internet - doesn't fit the crime.

Young adults drink way less than they used to. Look at the non-drinking result from this recent poll. More than half of them didn't drink in the last month!

Current dietary guidelines say for women to consume no more than one alcoholic drink per day, and for men to only double that. Those guidelines might be scrapped soon, and replaced with "drink in moderation". That likely means going above current guidelines.

Two weeks ago, I posted about this @CellCellPress paper. As it drew more attention than I'm used to, I wrote a longer piece with my thoughts. Beyond the points made before, I also address some of the feedback on the original tweet👇

One of the most harmful things econometricians can teach students is the plug-in characterization of 2SLS -- without first showing how it is simply an IV estimator with a particular choice of instruments. The plug-in approach often leads to bad habits down the road.