It's like ppl just register that I'm asking through twitter, know that my twitter account doesn't have a good sampling of the population, and then assume this means that selection bias is a good theory for any sort of twitter poll finding. No! This isn't how it works!
-
Show this thread
-
And I'm NOT saying that my childhood abuse/bdsm poll is conclusive, there's a lot of possible explanations for it and I'd want to replicate this in a better, more careful and detailed survey to check on other causes. But yelling 'selection bias' just makes me wanna punch you
2 replies 0 retweets 42 likesShow this thread -
the way it's delivered to me is often a bad, lazy critique that doesnt understand how bias works or how it impacts things we check for. If you tell me *how* it might be selection bias then im happy to engage: what is your theory for the selection pressures on my respondents?
2 replies 0 retweets 52 likesShow this thread -
If you're like 'hey aella, i think this might be mostly just due to selection bias because you have a high following in this one BDSM recovery from childhood abuse forum that you didn't know about", I'd be like whoah, that's a great point I had no idea
9 replies 1 retweet 75 likesShow this thread -
Replying to @Aella_Girl
There's another possibility -- when you ask two questions together, you may get a disproportionate interest in answering in the way that ties them together by people who believe the two things are connected and want to tell the world through your survey.
2 replies 0 retweets 6 likes -
Replying to @TryTryAgain17 @Aella_Girl
If you ask for A|B, A|Not B, Not A|B, Not A|Not B, if for some reason a lot of people who are A|B think it's *true and important that A caused B for them*, they may be more motivated than the other three groups to answer, so you may get a disproportionate number of their answers.
1 reply 0 retweets 1 like -
Replying to @TryTryAgain17 @Aella_Girl
Exactly, it's clear the hypothesis being tested in the question. It's very likely that people who match the hypothesis will be motivated to answer.
1 reply 0 retweets 2 likes -
Replying to @DanielCoopreal @TryTryAgain17
Aella Retweeted Aella
Aella added,
1 reply 0 retweets 0 likes -
Replying to @Aella_Girl @DanielCoopreal
I get that that's your general approach, but looking at this particular survey we can't see that there is such a check at work in your response.
1 reply 0 retweets 0 likes -
Replying to @TryTryAgain17 @DanielCoopreal
the difference in % answers when i do the spot checks in other polls is usually much smaller than 15%, so I assume it's unlikely that this one (out of all the other loaded ones I've done) is subject to this 'identity effect' or whatever. But if i replicated with a study id
1 reply 0 retweets 0 likes
explicitly design it to try to prevent that
-
-
Replying to @Aella_Girl @DanielCoopreal
If it were me, I'd have a hard time feeling any confidence in the results. You may be onto something, but this is just so easy to manipulate. You feel this is loaded and other surveys have been loaded, but maybe this is extra loaded? Also, you have 83K followers. About 2K ...
1 reply 0 retweets 0 likes -
... voted on this. If 4% of the total fewer voted Yes|Yes, the ratio would look the same for Yes|* as No|*. Could there be 80 people who want to make this point?
1 reply 0 retweets 0 likes - Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.