Nate Silver's flawed model
While many in the media (and Silver himself) openly mock the idea of Republicans' "unskewing polls" (and I am not a fan of unskewedpolls.com by any means), Silver's weighting method is just a more subtle way of doing just that. I outlined yesterday why Ohio is closer than the polls seem to indicate by looking at the full results of the polls as opposed to only the topline head-to-head numbers. Romney is up by well over eight points among independents in an average of current Ohio polls, the overall sample of those same polls is more Democratic than the 2008 electorate was, and Obama's two best recent polls are among the oldest…
In that same post, Silver touts a "SurveyUSA poll showing Mr. Obama with a one-point lead in Florida is really the slightly better result for him." That SurveyUSA poll indeed had Obama up by one point, but had a Democratic party-ID advantage of nine points. In 2008 Democrats had a three-point advantage, and in 2010 the parties were even. So the SurveyUSA poll is good news only if you believe Democrats will not only improve on their 2008 turnout, but triple their turnout advantage over Republicans.
This is the type of analysis that walks a very thin line between forecasting and cheerleading. When you weight a poll based on what you think of the pollster and the results and not based on what is actually inside the poll (party sampling, changes in favorability, job approval, etc), it can make for forecasts that mirror what you hope will happen rather than what's most likely to happen. This is also true of Silver's dismissal of Romney's lead in Gallup this week. While Romney is likely not up by seven points nationally, as the poll predicted, you can't dismiss it while at the same time giving a twelve-day-old Marist/NBC Ohio poll a higher weighting than eight newer polls when Marist has leaned Obama this entire cycle.
> [Original Message]
> From: Kevin Ibeh <k.i.n.ibeh@strath.ac.uk>
> To: usaafricadialogue@googlegroups.com <usaafricadialogue@googlegroups.com>
> Date: 10/27/2012 6:19:52 PM
> Subject: USA Africa Dialogue Series - 'The Rasmussen Difference'
>
> The Rasmussen Difference
> Posted: 10/27/2012 12:42 pm Alan Abramowitz Alben W. Barkley Professor of Political Science, Emory University
> In 2012 as in 2008 and 2010, Rasmussen has been the nation's most prolific polling organization at the state level. Almost every day Rasmussen produces two or three new polls of key battleground states in the presidential election along with their national tracking poll. But in addition to being the nation's most prolific pollster, Rasmussen gets a lot of attention, especially from conservative media outlets and pundits, because its polls consistently produce results more favorable to Republican candidates than the overall averages -- results that frequently don't match the actual election results very well. In 2010, for example, Nate Silver rated Rasmussen as the least accurate and most biased of all polling organizations that conducted a large number of state polls.
> So how is Rasmussen doing this year? Along with its national tracking poll that has typically showed Mitt Romney with a lead that is two to three points larger than the overall average, Rasmussen's polls in the battleground states have also had a consistent Republican lean. I compared Rasmussen's latest results in the nine key battleground states -- Ohio, Florida, Virginia, Colorado, New Hampshire, Iowa, Wisconsin, Nevada and North Carolina -- with the overall averages from the HuffPost Pollster -- an average that includes Rasmussen polls. In every case, Rasmussen's poll showed a better result for Romney than the overall average. The Republican lean of Rasmussen's state polls ranged from one point in Nevada and two points in Florida to four points in Wisconsin and five points in Colorado. Across all nine states, Rasmussen's polls favored Mitt Romney by about three points more than the overall average.
> Read the rest here: http://www.huffingtonpost.com/alan-abramowitz/the-rasmussen-difference_b_2030330.html?utm_hp_ref=@pollster
>
> --
> --
> You received this message because you are subscribed to the "USA-Africa Dialogue Series" moderated by Toyin Falola, University of Texas at Austin.
> For current archives, visit http://groups.google.com/group/USAAfricaDialogue
> For previous archives, visit http://www.utexas.edu/conferences/africa/ads/index.html
> To post to this group, send an email to USAAfricaDialogue@googlegroups.com
> To unsubscribe from this group, send an email to USAAfricaDialogue-
> unsubscribe@googlegroups.com
>
No comments:
Post a Comment