Comment 49722

By Borrelli (registered) | Posted October 18, 2010 at 11:55:40

Dear Reality Check:

Only residents who said they were likely to vote were polled. That is why the undecided numbers are important. This means there is a good chunk of sample intending to vote but who are parked on the sidelines waiting, and there's no reason to think those votes will be evenly distributed.

No candidate has license to break out the champagne with this poll. If Undecided was a candidate, maybe they could. The margin of error is such that Bratina and Eisenberger are essentially tied (+/- 3.1), and Dianni is not far behind.

And if you use the true margins for each percentage, Larry's not doing well among decided voters. Fun fact: as a rule polling companies tend to report the highest confidence interval as the global "margin of error". However, each percentage statistic has an error term associated with it, so if you wanted the unvarnished numbers:

Bratina 23.4% +/- 2.62 Eisenberger 20.8% +/- 2.51 Di Ianni 18.3% +/- 2.39

Comment edited by Borrelli on 2010-10-18 10:55:59

Permalink | Context

Events Calendar

There are no upcoming events right now.
Why not post one?

Recent Articles

Article Archives

Blog Archives

Site Tools