25% - 50% sampling of total is adequate.
-
-
-
for N=200, Slovin’s formula says you suffer .07 margin of error at n=100. For N=450, n=100 still gets you under .09 margin of error
End of conversation
New conversation -
-
-
That the Central Limit Theorem is alive and well.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Consider a poll option with 50% probability (this is the worst case in terms of variance). With n = 100 the standard deviation for the observed proportion is 5 percentage points, so roughly 95% of the time it will be within 10 pp of the final result.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
that your followers are not quite as heterogenous
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
...something something... timezone doesn't matter?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Pretty good heuristic version of the central limit theorem: In any random process, if you measure N counts of something, then you should consider that the true number (i.e., the one that perfectly represents the mean) is within the range N +/- sqrt{N}.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
the number of 9's for your desired confidence interval is satisfied at ~100, unless there are systematic errors or nonstandard distributions (that is my guess)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.