...using a national poll, they either have an oversample (to get useful numbers) or they should stay far away from making the comparison.
-
-
Replying to @zeynep
TIL I learned people will even resist the basic fact that polls have a margin of error that gets (much) larger as the sample gets smaller.
2 replies 5 retweets 29 likes -
Replying to @zeynep
You seem to be conflating
@sarahkendzior tweeting a flawed study with the study itself.1 reply 1 retweet 6 likes -
Replying to @gilliatt @sarahkendzior
The study isn't flawed. Tiny subgroup *comparisons* as Sarah did is flawed. The study is fine.
3 replies 0 retweets 22 likes -
Replying to @zeynep
The point is,
@sarahkendzior didn’t do the subgroup comparisons. She posted an image with a table from the report, including %.2 replies 0 retweets 0 likes -
Replying to @gilliatt @sarahkendzior
No, she did the subgroup comparisons in her tweet. The study is fine.
3 replies 0 retweets 7 likes -
Replying to @zeynep @sarahkendzior
And BTW, I’m not shouting about this. This whole incident seems like a perfect example of the hazards of Twitter.
1 reply 0 retweets 1 like -
Replying to @gilliatt @sarahkendzior
Reporting percentages is okay; what you cannot do statistically is compare them. Honestly—this isn't controversial.
1 reply 0 retweets 1 like -
This is actually a common error, by the way. Trump Twitter went viral with the "mirror" of this error last week.
1 reply 0 retweets 3 likes -
I think one can say, "oh, okay, those numbers were within margin of error of zero; correct tweet; done."
1 reply 0 retweets 2 likes
One can remain interested in the topic; just there is no data on this on that chart.
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.