3 ms·
Isn't the reasonable definition of reproducible for a poll having overlapping margins of error? A useful poll should do that. The result is qualified, comparis
by maxerickson 10y ago
Isn't the reasonable definition of reproducible for a poll having overlapping margins of error? A useful poll should do that.
The result is qualified, comparisons of results shouldn't ignore that.