Discussion for article #229457
Seriously, What is going on with Quinnipiac?
They are either showing a GOP wave of historic proportions that is going to make 2010 look like a mild blip, or they are going to look really bad on the morning of Nov. 5th.
That’s not to say Ernst isn’t leading, Iowa is certainly tight…but other polling has shown Braley taking a small lead, and he also leads by a huge margin in early returns…
They have been showing results that seem to be contradicting most other polls, maybe even their own polling.
I can’t wait for this lunatic to hit the Senate. Another tea party radical is sneaking into office because voters are stupid enough to buy her down-home-hog-castrator-mother bullshit.
Iowa, I love you, but you’re about to go Full Texass.
Never Go Full Texass.
Iowa HAYSEEDS opt for a Michelle Bachmann act-a-like showing just how INTELLIGENT they are…
God help us all…
The TPM headlines on individual polls are ridiculous.
Ernst went from 48 to 49 percent. Braley went from 46 to 45 percent. That’s practically rounding error, not a jump.
The description of this minor change isn’t the real problem, though. The real problem is that the numbers are treated as if they are precise points, when what they really represent is an estimate of the center of a fairly broad range of numbers.
That range itself is determined only partly by the raw polling data. Pollsters know that the respondents to a given poll usually are not a representative sample of the target population, so pollsters adjust the raw data in an attempt to mimic the results they would have had if the respondents had been a representative sample. Nobody knows what a truly representative sample looks like, however, so each pollster has to make its own informed estimates there.
Below is a helpful paragraph from a helpful Fivethirtyeight article. It indicates how wide the confidence interval is for the gap between candidates even when aggregating and tweaking the results of multiple polls by multiple pollsters:
The FiveThirtyEight model, in addition to estimating which candidate is ahead, also evaluates the uncertainty in each forecast based on factors including the volume and consistency of polling in each state and the number of undecided voters. The uncertainty is very high in Alaska. The 90 percent confidence interval on its forecast in Alaska ranges from a Begich win by 6 percentage points to a Sullivan win by 9 points — a 15-point span. By contrast, the 90 percent confidence intervals in Colorado and Iowa, which have more abundant and more consistent polling, span a 9-point range.
http://fivethirtyeight.com/datalab/the-senate-polling-in-alaska-is-making-us-sweat/