Here are two more articles relevant to my recent post regarding undecideds. (See here and here.) According to the AP/Yahoo poll, almost one in ten voters remain undecided – a slightly higher number than what many of the other surveys are reporting. As the second article indicates, and as I alluded to in my earlier post, the difference is due in part to the question wording that pollsters use when asking respondents for whom they are likely to vote. As David Moore points out in the other article cited above, pollsters routinely push undecideds to make a choice. They also tend not to highlight in their stories the number of respondents who say they might change their mind. Why does this matter? Keep in mind that pollsters ask people who they would vote for if the election was held today – not who they think they will vote for on Nov. 4, the actual election day. In other words, there is room for voters who say they are leaning toward someone to change their mind even this late in the process. Consider that the 2004 exit poll found 9 percent of voters saying they had made up their minds in the three days before the election. In short, looking only at undecideds in any given poll may understate the number of persuadables remaining in the electorate. For example, the most recent ABC poll suggests that there are only 2% undecideds remaining. But in the fine print of the poll, you find an additional 3% who say there is a “good” chance they will change their mind. Among McCain supporters, 4% say they there is a “good” chance they might change their mind, but only 1% of Obama supporters say as much. In the CBS poll, fully 7% of Obama supporters and 9% of McCain supporters say their minds are not made up. In the Fox poll, 17% of Obama and 22% of McCain supporters indicate that they support their candidate only “somewhat.” None of these voters are considered “undecided”, however. But it suggests that there may be more room for movement in support for both candidates than the top-line survey results suggests.
Recent Tweets
- My post on why the presidential race is where political scientists predicted it would be is up at the Economist http://t.co/BY1tofTx 2012-10-10
- In the war on terror, "Obama made it clear he would be as tough, or tougher than Bush." Frontline special. 2012-10-10
- Thanks! MT @DLeonhardt in reply to DLeonhardt 2012-10-10
- More updates...
Pages
Blogroll
- American National Election Studies
- Electoral College map at Taegan Goddard’s Political Wire
- Gallup
- General Social Survey: America’s Research Project
- National Journal Online
- Pollster.com
- Real Clear Politics: Polling
- Roll Call: Congress.com
- Roper Center for Public Opinion Research
- The Diageo Hotline Poll
- The Economist Online
- The Library of Congress: Thomas (legislative info)
- The White House on the Web
- U.S. House of Representatives
- U.S. Senate
Presidential Power Archives
- November 2012
- October 2012
- September 2012
- August 2012
- July 2012
- June 2012
- May 2012
- April 2012
- March 2012
- February 2012
- January 2012
- December 2011
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- May 2010
- April 2010
- March 2010
- February 2010
- January 2010
- December 2009
- November 2009
- October 2009
- September 2009
- August 2009
- July 2009
- June 2009
- May 2009
- April 2009
- March 2009
- February 2009
- January 2009
- December 2008
- November 2008
- October 2008
- September 2008
- August 2008

How do these polls deal with early voting? If the numbers of early voters are as high as reports say, the idea that 10% of voters might change their minds drops significantly – you can’t change your mind once you vote!
Jason – Yes, you are right. The overall number of persuadable voters declines as early voting takes place. In the surveys, pollsters are asking voters if they have voted already, and obviously only asking those who haven’t voted yet how strong their support is. But your point is well taken – 10% of a shrinking pie means less persuadable voters. Note that this is more significant in some states that are showing a large number of early voters (like Nevada). Most of our discussion is in reference to the national tracking polls. At this point, early voting is about 18% of the total turnout in voters for 2004, so assuming equal turnout (and my guess is it will go up this year), the pool of “persuadables” (assuming a 10% rate) is already down 1.8% overall. Don’t know if I’d call that significant – but it is less!
Matt, you have indicated that you thought that Rasmussen and Gallup were the most reliable polls. As of 9 PM Saturday, Rasmussen has the National vote Obama + 5 and Gallup, both Traditional and Expanded at Obama +10. Can you suggest and explanation for this disparity?