Y
Hacker News
new
|
ask
|
show
|
jobs
by
Fomite
3652 days ago
There are a number of study designs that don't require a null-intervention control group. Some of those would have been interesting, and might actually have yielded something closer to the true effect.