Hacker Read top | best | new | newcomments | leaders | about | bookmarklet login

I don't think A/B testing is suitable for this. You could never collect enough data to make any statistically significant observations.


sort by: page size:

I mostly agree with you, but I'm not sure if there is data to prove. I would be interested in running such an A/B test myself.

Though I understand the idea, I'm not too familiar with how A/B tests are performed in practice. I would've assumed you'd need a lot of data to get statistically significant results. Is the volume of data that you need to achieve that statistical significance well-understood and do you actually have the traffic to collect that amount of data? (I guess I imagined that it would require huge amounts of data.)

Can't you do A/B testing on it?

Sounds like you should be able to A/B test this.

A/B testing?

A/B testing?

you could look at a/b testing. it might be interesting to learn how to set that up... (not sure how you'd use it to measure this particular case though)

agreed. was thinking the same thing but you beat me to it. anybody who has ever evaluated an A/B test would look at this and immediately question the significance/validity of the result. if only somebody would actually back up their claim with ALL the data for once (i.e. variance and specific sample sizes)

Is there any A/B testing that backs this notion up?

You should try doing some A/B testing

Hmm, thank you - that's really very interesting indeed. I'd not thought of doing A/B tests like that. Thank you very much!

I'd be interested to see some stats from some a/b testing on this.

A/B tests are only useful when you can gather statistically significant amounts of data from them. For a lot of small websites, or infrequently used features on larger websites, that is not the case.

> On top of that, doing anything economical with the analytics is very rare.

If this was true, then A/B testing would be useless, while it's quite the opposite.


This would be an easy A/B test. If you do it, I'd love to hear what you found.

A/B testing probably wouldn't have caught this.

No, this is A/B testing. Why do you think it is A/B testing?

Wouldn't bad A/B testing be a 50/50 solution. Is making decisions on data without statistical significance any better than throwing darts blind?

You just invented A/B testing. Congratulations.
next

Legal | privacy