logoalt Hacker News

jrflowersyesterday at 11:51 PM1 replyview on HN

Imagine posting “sorry that the facts bother you” and then linking to

- A study with a sample a size < 50

- A study that says that medication improves outcomes over CBT

- A study that says that evidence for CBT improving ADHD symptoms comes from studies with such small sample sizes that the conclusions could be the result of bias

The only way someone could conclude “CBT has the same outcome as medication” from the studies you linked to would be to not read them. The first two don’t really say that and the third one literally refutes that position.


Replies

duskdozertoday at 3:39 AM

>The only way someone could conclude “CBT has the same outcome as medication” from the studies you linked to would be to not read them.

Fortunately for them, that's often the case. I've seen at least a couple internet arguments with LLM-generated "sources" that didn't actually exist.