On Trying To Figure Out Where the Campaign Stands
by Chris Bowers, Fri Apr 27, 2007 at 12:21:25 PM EDT
Over the last two weeks, I have spent most of my blogging energies on poll analysis in the 2008 Democratic Presidential primary. The goal has been to develop a clear picture of where national Democratic opinion on the nomination currently stands, which has proven surprisingly difficult. Given the eight and a half months between now and the Iowa caucuses, the highly contradictory results produced by different firms, the sample composition arguments that led to the Inflated Clinton Poll Theory, the widely varying degree to which different pollsters push undecideds, and the different candidates and potential candidates included in the questions of different pollsters, this has been a real struggle.
The result of this analysis has been two charts of polling averages--a high-end estimate and a low-end estimate--and extended discussion of the Inflated Clinton Poll Theory both on MyDD and in many other places. I like the charts, and I prefer the high-end estimate to the low-end estimate because it is more distinct and more in line with my general polling philosophy on 2008 national polls. Both actually posit nearly identical margins between Clinton and Obama, 35.7%--28.4% in the high end estimate, and 34.3%--25.9% in the low end estimate, for a very narrow current range of 7.3%--8.4%. Also, I think the Inflated Clinton Poll Theory has generated enough buzz that I expect a test of the theory at some point in May. If the theory is correct, then the campaign is somewhat closer than the 7.3%--8.4% range. If not, then 7.3%--8.4% stands. Either way, trendlines are showing a slow Obama and Edwards rise, accompanied with a slow drain on Clinton and others / unsure. Long-term, the indications are of a three-way race developing, although Obama and Clinton are both noticeably ahead of Edwards.
As for the Inflated Clinton Poll Theory itself, our discussion has led to two important findings. First, the largest source of Clinton's advantage is indeed from those participants in national polls who are not paying close attention to the campaign. We now have data on this both from NBC, which shows Obama ahead among those who are paying close attention, and from Pew, as shown by the pic on the right, which shows which shows both Clinton and Gore performing better relative to Edwards and Obama among those who have not given much thought to the election. As Mystery Pollster notes, how much thought someone has given to the election is "a question generally considered predictive of voter turnout." The same can be said of how much attention someone is paying to the election, which is a slightly different question than "how much thought" someone has given to the election. However, the results are not as clear-cut as the Inflated Clinton Poll Theory assumed:
Given all of this, I think it is time that I move on and start blogging about other subjects again. Through the charts that I will regularly update, we now have very good estimates of where public opinion on the national campaign stands. While the Inflated Clinton Poll Theory is certainly a legitimate enough area of inquiry to deserve a real test, until that test takes place, existing data is inconclusive to the point where it can no longer be assumed the current polling averages are skewed one way or the other. At this point in time, I have to at least tentatively conclude that Hillary Clinton is ahead nationally by around 7-9 points.
I hope you enjoyed this analysis these past two weeks. The only motive I ever had in this process was to try and figure out where the campaign currently stands, and develop a method to monitor its progress over the next nine months. Quite frankly, I think it is sad that accusing people of hidden agendas has become so commonplace and reflexive in America that data-driven discussions of this sort are simply dismissed as "bias" or "wishful thinking" by many. As you may have noticed, I no longer have any tolerance for such accusations, which threaten to suck out both all of the fun and insightful research that these discussions potentially hold. I have done my best to try and figure out where the race stands, and I will continue to do so as long as I am interested in politics.
The result of this analysis has been two charts of polling averages--a high-end estimate and a low-end estimate--and extended discussion of the Inflated Clinton Poll Theory both on MyDD and in many other places. I like the charts, and I prefer the high-end estimate to the low-end estimate because it is more distinct and more in line with my general polling philosophy on 2008 national polls. Both actually posit nearly identical margins between Clinton and Obama, 35.7%--28.4% in the high end estimate, and 34.3%--25.9% in the low end estimate, for a very narrow current range of 7.3%--8.4%. Also, I think the Inflated Clinton Poll Theory has generated enough buzz that I expect a test of the theory at some point in May. If the theory is correct, then the campaign is somewhat closer than the 7.3%--8.4% range. If not, then 7.3%--8.4% stands. Either way, trendlines are showing a slow Obama and Edwards rise, accompanied with a slow drain on Clinton and others / unsure. Long-term, the indications are of a three-way race developing, although Obama and Clinton are both noticeably ahead of Edwards.
As for the Inflated Clinton Poll Theory itself, our discussion has led to two important findings. First, the largest source of Clinton's advantage is indeed from those participants in national polls who are not paying close attention to the campaign. We now have data on this both from NBC, which shows Obama ahead among those who are paying close attention, and from Pew, as shown by the pic on the right, which shows which shows both Clinton and Gore performing better relative to Edwards and Obama among those who have not given much thought to the election. As Mystery Pollster notes, how much thought someone has given to the election is "a question generally considered predictive of voter turnout." The same can be said of how much attention someone is paying to the election, which is a slightly different question than "how much thought" someone has given to the election. However, the results are not as clear-cut as the Inflated Clinton Poll Theory assumed:
So what do these results say about Chris Bowers' theory that national polls are overstating Hillary Clinton's lead? The evidence here is mixed, at best. Obama certainly does better among more attentive voters, although that finding is not particularly surprising given his rapidly growing name recognition in recent months. However, Clinton also does better among the most attentive Democrats. Thus, her margin over Obama among those who pay "a lot" of attention (11 points in the combined March/April data) is actually a few statistically insignificant points higher than her margin among all Democrats (9 points in March, 10 points in April).When this is all combined with the fact that Obama does relatively better among Democratic-leaning independents who largely will not be able to vote in "closed" primaries on February 5th, 2008, than he does among self-identified Democrats who are more likely to also be registered Democrats, the existing evidence no longer provides any clear support to the Inflated Clinton Poll Theory. Simply put, there are conflicting indications as to whether Clinton or Obama would benefit more from a tighter poll sample that focused only on Democrats who are likely to make up the caucus and primary electorate, rather than all Democratic self-identifiers and leaners who are registered to vote.
Now, some cautions about the above. First, those who say they pay a lot of attention to the candidates are more likely to vote than those who do not, but this measure is far from a perfect turnout predictor. Pollsters that use attentiveness to select likely voters usually do so in combination with other measures, such as reports of past voting or future likelihood to vote.
Given all of this, I think it is time that I move on and start blogging about other subjects again. Through the charts that I will regularly update, we now have very good estimates of where public opinion on the national campaign stands. While the Inflated Clinton Poll Theory is certainly a legitimate enough area of inquiry to deserve a real test, until that test takes place, existing data is inconclusive to the point where it can no longer be assumed the current polling averages are skewed one way or the other. At this point in time, I have to at least tentatively conclude that Hillary Clinton is ahead nationally by around 7-9 points.
I hope you enjoyed this analysis these past two weeks. The only motive I ever had in this process was to try and figure out where the campaign currently stands, and develop a method to monitor its progress over the next nine months. Quite frankly, I think it is sad that accusing people of hidden agendas has become so commonplace and reflexive in America that data-driven discussions of this sort are simply dismissed as "bias" or "wishful thinking" by many. As you may have noticed, I no longer have any tolerance for such accusations, which threaten to suck out both all of the fun and insightful research that these discussions potentially hold. I have done my best to try and figure out where the race stands, and I will continue to do so as long as I am interested in politics.
Tags: Barack Obama, Democrats, Hillary Clinton, inflated Clinton poll theory, polls, President 2008, Primary Elections (all tags)
-----------
myDD - skin
-----------
29 Comments