^@Aella_Girl @literalbanana seems like your kind of thing
-
-
-
very clear illustration of a tricky point!
End of conversation
New conversation -
-
-
What could have been fun is to ask the same people the same two questions but separated by dozens of other questions.
-
They do that in a bunch of personality test
End of conversation
New conversation -
-
-
I ran an economics experiment where we got radically different results depending on how we delivered the instructions. It wasn't the paper we wanted to write, but it still got published so
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What this shows is that almost nobody knows how much we spend on defence, how much defence stuff costs, or how much we'd need to spend to achieve x objective. But I used to play 5-a-side with a defence contract lawyer and he can confirm we taxpayers get rotten value for money.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
As you add resolution the results have the opportunity to become more granular, just like digital media is hi-def or lo-res depending on its resolution. Lo-res can be manipulated by bad actors but it's also an outcome of bad design (when hi-def results are desired).
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Not enough data - give full percentile breakdown
-
The percentages are given for each response in the linked article.
End of conversation
New conversation -
-
-
This is well known in the polling world. Question design 101: response options of quant questions need to be exhaustive and mutually exclusive.
-
Of course, this knowledge can also be used to obtain data that pushes desired narrative. Example from a recent poll:pic.twitter.com/j7KOOPhkkd
End of conversation
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.