Yesterday evening on the front page of the Columbus Dispatch were the results of their latest Ohio polling, thru Sept 29th. The results were 51/O 42/R.
After I puked a little in my mouth, I began to read the breakdown, methods etc.
The poll was registered voters and was mailed out across the state. 705 of the respondennts were D, 563 were R, 362 were I or other.
When they discussed potential errors, they described ‘Nonrespondent Bias’, as a possible source. What that means is the poll assumes that those who did not respond would have responded in the same way as those who did.
THE RESPONSE OF THIS POLL WAS 15%....that means 85% of the people who received this poll in the mail considered it junk mail. Now who do you think is more likely to toss aside political junk mail, especially from a liberal media outlet? Whose out there working and living and doesn’t have time to lay around the house and respond to polls?
This whole NONRESPONDENT BIAS is very intriguing to me. We don’t really hear much about it. I really think it could be one of the most overlooked errors in polling. JMHO
Chris Wallace probably laughed at the concept too, but this should put a chill in his britches if nothing else does.
If the public won't tell you what they think, then who will, and why? Are current polls better than prayer? Worse than prayer? No different than prayer? You be the judge.