Causal inference as a blind spot of data scientists

Re:Causal inference as a blind spot of data scientists by jiggawatts on Sunday October 15, 2023 @05:31PM (#37893566)

The “gold standard” has failure modes that seem to be ignored.

E.g.: making UI elements jump around unpredictably after a page load may increase the number of ad clicks simply because users can’t reliably click on what they actually wanted.

I see A/B testing turning into a religion where it can’t be argued with. “The number went up! It must be good!”

[ Parent | on HN ]

Re:Causal inference as a blind spot of data scientists by bertil on Sunday October 15, 2023 @05:49PM (#37893717)

That’s generally because the metrics you are looking at do not represent what users care about. It’s different than the testing methodology, often overlooked, and a lot more important.

I’ve argued that A/B testing training should focus on that skill a lot more than Welch’s theory, but I had to record my own classes for that to happen.

[ Parent | on HN ]

Re:Causal inference as a blind spot of data scientists by travisjungroth on Sunday October 15, 2023 @07:43PM (#37894455)

But those metrics are hard to move, so you target secondary metrics.

The problem with that strategy becomes obvious when you spell out the consequences: measurably improving the product is hard, so you measure something else and hope you get product improvements.

[ Parent | on HN ]