More polling shenanigans
It's panic time, and they're really pulling out all the stops.
I used to get paid to do public opinion polling, and there is absolutely no good reason for the finished product to show the consistent direction and magnitudes of sample bias these polls show. There are some bad reasons, though.
Let me (1) word the questions, (2) determine the sequence of the questions, and (3) draw the sample, and I'll write up the general conclusions of the poll right then; the actual polling will be a formality. The poll and its conclusions will be nothing but junk, but no one will know that from the splash headline -- which is all most people read.
Skewed Sample Data Used In PPP Wisconsin "Do Over" Poll
The pollster for DailyKos might be biased? No!
It's all too funny.
Byron
I used to get paid to do public opinion polling, and there is absolutely no good reason for the finished product to show the consistent direction and magnitudes of sample bias these polls show. There are some bad reasons, though.
Let me (1) word the questions, (2) determine the sequence of the questions, and (3) draw the sample, and I'll write up the general conclusions of the poll right then; the actual polling will be a formality. The poll and its conclusions will be nothing but junk, but no one will know that from the splash headline -- which is all most people read.
Skewed Sample Data Used In PPP Wisconsin "Do Over" Poll
Public Policy Polling (PPP) is a long-time Democratic pollster, which last year became the official pollster for DailyKos.
The pollster for DailyKos might be biased? No!
It's all too funny.
Byron
posted by Byron | Tuesday, March 01, 2011
0 Comments:
Post a Comment
<< Home