Some people were shocked today by a poll released by Democratic polling firm Public Policy Polling that showed Republican incumbent Senator Richard Burr ahead of Democratic Secretary of State Elaine Marshall by only one point 43%-42%. That result made good headlines, but it has as much basis in political reality as Ronald Reagan coming back from the dead to run for President in 2012.
Let’s look inside the numbers and methodology a minute and show just where this poll goes off course and how it’s numbers are irrelevant.
First, it is a poll of ALL registered voters. There is no screen on the poll to determine if the respondent is even likely to vote in the 2010 election. This method was fine in 2008 when turnout was off the charts and it appeared everyone was going to vote, but in 2010, polling all registered voters is not an accurate election prediction model. Nobody thinks turnout is going to be anywhere near 2008 levels, why does PPP poll as if it is going to be? By not including a likely voter screen, this biases the poll heavily toward the Democratic side as Civitas polling (and just about every other polling firm) has shown Democratic voter enthusiasm is severely depressed while Republican enthusiasm is sky high.
Second, the percentage of African-Americans in the poll is 22%. Again, this was the turnout percentages of African-Americans in 2008, not anything close to what it is expected in 2010. Does PPP really think all those first-time Obama voters are going to come out and vote again in 2010? Well, that’s exactly what they are hypothesizing in their poll today. (African-American turnout in 2004 was a little under 19%, 2006 was around 17%).
Since African-Americans vote 95-97% Democratic, this poll artificially inflates the overall percentage Marshall receives by roughly 3-5 points on this statistic alone.
PPP’s poll today should actually be seen as fabulous news for Richard Burr, because even under the political environment in 2008 that PPP tests, Burr is winning. But 2010 is shaping up to be nothing like 2008 — the voter mood is completely different, Obama isn’t on the ticket and Democratic voters, especially young and minority, are not enthusiastic about voting.
Therefore, while this poll may have been relevant two years ago, it’s methodology does not fit the current political landscape.