|
Some polling data the media seems to have missed
Imagine that.(posted by a pro-Clark blogger)
Rassmussen conducted a national poll of voters from both parties (as well as none and other, presumably) to assess what percentage view favorably or unfavorably the various potential 2008 candidates. Of the ten Democrats on whom they collected data (Biden, Clark, Clinton, Edwards, Gore, Kerry, Kucinich, Obama, Richardson, Vilsack), only THREE were viewed favorably by a significantly larger number of voters, statistically speaking, than viewed them unfavorably. In other words, only THREE had more people like 'em than not.
Care to guess who those three are?
Well, I won't keep you in suspenders. But I will withhold some detail until I can apply a little analysis. So in alphabetical order, the only three with more to the good than to the bad, in numbers large enough to matter, are Clark, Edwards and Obama.
Now for the details.
As with any statistical data (and no doubt at the root of the old admonition against figures who lie and liars who figure), there are a number of ways look at the facts, each way answering different questions, based on different assumptions. Here are the ones that struck me as most significant and why. Your milage may vary.
Most favorable. Simply put, the three potential candidates whom more people favor than don't, currently enjoy the most people who like them in this order:
1. Barack Obama 51%
2. John Edwards 47%
3. Wes Clark 39%
These numbers don't account for how or even whether voters' opinions are informed, only that they are willing to tell the pollster they like a guy (or gal) for any reason at all. At the end of the (primary) day, this is the only figure that matters. This soon before, not so much.
Least unfavorable. A wholly different concept, and possibly more important if one believes people are more inclined to vote aginst one candidate than to vote for a different one:
1. Wes Clark 28%
2. Barack Obama 32%
3. John Edwards.42%
Again, this doesn't account for how or whether opinions are informed. And since all voters were polled, it must by definition include everyone from any party (or no party) who will not admit to viewing any Democrat favorably, whether they know them or not. For this latter reason, it doesn't have much affect on primary/caucus results (although if Democrats voted with a little more savvy, it probably would).
Best liked. A synthesis of "most favorable" and "least unfavorable," the net percentage points between the two values is probably the best snapshot of who is the most popular and/or likeable at the time of the poll.
1. Barack Obama 19 pts
2. Wes Clark 11 pts
3. John Edwards 6 pts
Don't believe me? Look at Hillary Clinton's numbers at the link below.
Most favorable extrapolated. In this concept, one assumes that current opinion reflect a "true" value, such that those who currently admit to no opinion will eventually form opinions in proportions similar to those who have already made up their minds:
1. Obama 60%
2. Edwards 52%
3. Clark 52%
In many cases it's reasonable to believe that public opinion is likely to remain relatively static, especially when a candidate relatively well known for what he or she actually is. A relatively blank slate has much more room for variance.
Growth Potential. The opposite of extrapolating current data, growth potential is a simple measurement of the percentage of voters who admit to no opinion. It is not, however, as a simple concept in what it means to a candidate's future prospects. Growth potential assumes the candidate will be actively courting the "no opinion" voters; the more of them there are, the more opportunity he or she has to impress them favorably. It futher assumes that as a generally likeable guy or gal, the candidate stands a reasonably good chance of making those favorable new impressions. In this respect, the three are ranked:
1. Wes Clark 33%
2. Barack Obama 17%
3. John Edwards.12%
Growth potential in extremely important this early in the race. Theoretically, a guy who's rated 40% favorable and 25% unfavorable (expressed as 35%, or the total of the two subtracted from 100) has the potential to reach an 75/25% split (granted not likely, but theoretically possible); whereas one who is already 50% favorable and 40% unfavorable can at best aspire to reach 60/40%. Each is almost certain to fall short of their optimum, but the one with greater potential is likely to finish higher. For this reason, over a year before the first caucus, a large potential for growth can only be a good thing for any candidate so long as he or she maintains significantly greater favorability than not.
It's worth noting that poll respondents tend not to want to admit they have no opinion, even when they don't. That is why standard presidential polls this early show very small percentages for "undecided" when it should be fairly obvious that most people have not made up their minds. It's also why those same polls are a more effect barometer of name recognition than anything else.
Unfortunately, it's also worth noting that since candidates typically do not reach their full growth potential by primary/caucus day, name reconition often remains as large of a factor in the results that count as it does in presidential polling.
If for those who are curious as to how the other seven Democrats stacked up, a more a complete chart is buried in the Rassmussen archives. In a nut shell, there are five (Biden, Clinton, Gore, Richardson and Vilsack) for whom there is no statically significant diffterence between the numbers who favor them over the numbers who don't. They're a statistical wash. Of those five, two (Clinton and Gore) have almost no growth potential at all.
Last, and in this case least I'm afraid, there are two (Kerry and Kucinich) of whom significantly more people seem to think they suck than don't.
Posted by Jai on January 2, 2007 8:28 PM | Permalink
|
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
