Latest Issue Polls
Latest Presidential Tracking Polls 2008
|
Looking For A Specific Archive?Primaries 2012: Primary Polls Election 2008: Articles | Presidential Contest Polls | Congressional House Polls | Congessional Senate Polls | Governor Polls Primaries 2008: Candidates | Blogs | Democratic Primary Exit Polls | Primary Polls Primaries 2004: Primary Polls Who gets your vote in 2012?
Scientific Polls Suggests that Undecided Republicans May Flock to ... Ron Paul and Mike Huckabee10/01/07 I could not believe it when I took a look at the Newsweek poll (source) and saw some pretty lame analysis of it. This is how we analyzed it.
The first thing we noticed was that the Undecided percentage among likely voters was much less than of the total. That makes a lot of sense. Many things could be in play. Likely voters may feel embarrassed to say that they are so they declare a preference. Or that they are just more in tune with the political candidates and have a set desire on who to vote for. Those that don't follow the political campaigns would be undecided. The next thing we noticed was that Romney, Thompson, Giuliani, and McCain were each roughly stuck about the same among likely voters and all voters. You cannot hold anything against those candidates because they are the best known candidates in the race. The two candidates who moved considerably when looking only at likely voters were Mike Huckabee and Ron Paul but at the expense of who? Well, at the expense of "Undecided". Unsure voters dropped by about 30% while Huckabee doubled in his support while and Ron Paul more than doubled. With rounding, Ron Paul may have even tripled (i.e. 1.8% to 5.4%). So let's use some dummy math to show our point. Imagine you have 100 voters, 60 of which are likely and they are asked their preferences. Well then Mike Huckabee would have 6 that said YES for him in the original pool of 100 voters considering he had 6% overall. But if you look at the small subsample of 60 voters, for him to get 12% that would mean he has to get 7.2 votes. Remember the original 6 voters? Even if they were 100% likely to vote, Huckabee would still need to derive 1.2 more voters from somewhere. So there is a considerable jump in support when adding in a small number of undecided voters. The same could be said of Ron Paul but dealing with percentages as low as 2% and 5% would not be as convincing but the relative weights of what we showed above with Mike Huckabee would be just as impressive as Ron Paul. So in conclusion, in order for Huckabee and Ron Paul to gain as they did among likely voters; the following two things would have to have happened.
So if the Undecided population dropped from 21 to 15 and we observed the gains that Huckabee and Paul did, imagine what would happen if that 15 dropped to 0% during the Iowa caucus when you must vote on one candidate? Then Huckabee could be as high as 20% and Ron Paul as high as 10%. As we get closer to the election, that very well may happen. The other thing to note is how many voters are supporting the top 4 of Romney, McCain, Giuliani, and Thompson but knowing very little about Ron Paul and/or Mike Huckabee? And are they likely to switch after Paul's/Huckabee's name recognition grow in the state? Questions like these should be asked. Is it no wonder that we claimed these two candidates the real winners of the Iowa straw poll. (source). A little bit of analysis can make you look like a genius or a fool. Of course the pundits and analysts will claim this analysis is all a bunch of sh** because the MoE of the likely voters was 9% so they will tell us that we are being irresponsible for doing any analysis on it. I have the emails to prove it. But we don't claim this to be prophecy but rather a possibility. I mean, we feel that it is much better to report what we observe and provide both sides and let you make the decision as to whether you will accept it or not. Come on, on the flip side, I think it is irresponsible to be given data but be unwilling to do any hypothesizing from it. In addition to the large MoE, how they determined who is likely could be questionable. You nor I really know that until election day. But just because there are uncertainties in polls does not mean we have to completely ignore its results. Besides ABC, CBS, and Fox would never promote nor analyze the polls from another network. How lame. So in conclusion, I give you a quote by Herbert Asher, a Political Science professor at Ohio State University that I have published before but do so again because it is so true. "Often, however, the media do not actually interpret poll results, but instead present simple, straightforward descriptive statements about them... Measures of association, correlation analysis, and multivariate statistical analyses are virtually nonexistent" - Asher, Ohio State University
| Traffic During 2008 Election 2008 Predicted Electoral Math 2008 Presidential Candidates
2008 Battleground States
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
