52/ Aha! OK, this is a valid answer. Thank you!https://twitter.com/AlexNowrasteh/status/993518615180861440 …
-
-
63/ Can you give us the `n`, and tell us how you're defining statistical significance (p = 0.05 ?)? ...or should I dig up the GSS survey data?
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
For the difference between .2 and .23 to be statistically significant, assuming power of .8 and an enrollment ratio of 10:1 (very conservative; ratio of natural-born to naturalized citizens is much higher), you need n > 18,000. And that's not correcting for multiple comparisons.
-
That is, assuming "statistically significant" is defined as p < .05. Default assumption should always be that differences as small as .2 vs .23 are not significant. GSS sample size usually n < 3,000, fwiw.
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.