Correlation vs causation in regard to new answers on qs with accepted answers

Conversation started May 12, 2015 at 23:04.

JNat

May 12, 2015 23:04

@ton.yeung Why not?

We have no way of answering the question

I can give you those numbers and whatnot, but I have no idea, and no way of knowing, if the other questions with accepted answers and no new answers posted after that, haven't had new answers because of the accepted answer

The OP is just asking for a 'best practice'

We can try to use those numbers to guide them

Or tell them what we usually do, and why

or both

but it is ultimately up to them

And we don't have the means to understand why or when new answers are posted after the question has an accepted answer

Or why they aren't

@ton.yeung The question is "Does accepting an answer prevent users from posting new answers?"

The answer is, "we have no way of knowing"

Unless I'm missing something

I miss stuff all the time

@ton.yeung We still have no way of knowing, from what I can tell

Unless you ask every single person that viewed the question "Did you not post a new answer because there was another accepted answer?"

A: Do accepted answers prevent people from giving more answers?

I get your concern — accepting an answer can sometimes drive users with good (or better) answers away from answering the question. I've refrained from accepting answers myself once or twice too (here and here, for instance), but in these instances my main reason for not accepting answers was actu...

> I estimate that up to 20% of the questions with accepted answers have received new answers after having accepted an answer.

If that's where you were going

Taisho

JNat

I didn't want to exaggerate, but maybe a bit over 20% is possible

@ton.yeung Unless I'm misunderstanding what you mean by "follow up answers", I think it does

@ton.yeung Did you read that post?

I just used estimate there because I'm certain that a little above 15% have received new answers

And I estimated that about 5% more are not accounted to in the query, for 'technical reasons'

Relevant link:

Correlation does not imply causation

Correlation does not imply causation is a phrase used in science and statistics to emphasize that a correlation between two variables does not necessarily imply that one causes the other. Many statistical tests calculate correlation between variables. A few go further and calculate the likelihood of a true causal relationship; examples are the Granger causality test and convergent cross mapping. The counter assumption, that correlation proves causation, is considered a questionable cause logical fallacy in that two events occurring together are taken to have a cause-and-effect relationship. This...

I don't care enough to dedicate any more time to this either :P

@ton.yeung I never said that

14 mins ago, by JNat

Unless you ask every single person that viewed the question "Did you not post a new answer because there was another accepted answer?"

@ton.yeung Ok, I just wanted to provide some guidance to the OP, and used the numbers to roughly illustrate what's going on. You have half the work done :P

You, or anyone else who feels compelled to provide a more definitive answer to that user

@ton.yeung From which you can verify correlation, not causation

Which means the question would still not be answered

Unless, again, I'm missing something

Ahah, ok ;)