Doing Good Better: Effective Altruism and a Radical New Way to Make a Difference

by William MacAskill

I suspect the apparent effectiveness of Scared Straight can be explained by a phenomenon called regression to the mean. If you play a truly excellent round of golf one day, you’ll probably play worse the next time you play because that excellent round was statistically unlikely and you should expect to see a more typical performance the next time. Similarly, people who are undergoing a bout of particularly severe depression will on average be happier three months later, because they are likely closer to their average level of happiness. And, similarly, if you select a group of juveniles to go through a reform programme because they’ve committed an unusually high number of misdemeanours in a given time period, they’re likely to exhibit something closer to typical behaviour in the following months.