Tuesday, September 19, 2006

Survey USA Poll: Granholm even, Stabenow ahead (Released 9.18.2006)

New numbers out, courtesy of Survey USA (conducted between September 15 and September 17):


Dick DeVos (R) - 47%
Granholm (D) - 47%
Undecided - 2%
Margin of error - 3.7%

With high name recognition for both candidates only 2% of voters are undecided, showing that this is a very heated and polarized race.


Bouchard (R) - 41%
Stabenow (D) - 54%
Undecided - 2%
Margin of error - 3.6%

Unless Bouchard can raise his visibility quickly, it will become increasingly unlikely that the national GOP will intervene in this race.


Cox (R) - 49%
Williams (D) - 41%
Undecided - 4%
Margin of error - 3.7%

Pretty self-explanatory. I can't imagine most voters know who Williams is yet.


Yes - 33%
No - 32%
Not Sure - 35%
Margin of error - 5.8%

Only 40% of the poll's participants claimed to know what the MCRI is. Over a third of voters who knew what the MCRI is don't know how they will vote on this yet, and there is a high margin of error. Nobody knows how they are going to vote on this yet. Its the same story as always: its tough to poll ballot proposals, no idea what is going to happen with this.

Overall, I trust the methodology of this poll a lot more than EPIC-MRA last week. In my opinion, IVR is preferable over traditional phone polling. In addition, they appear to try harder to get a truly random, representative sample. I'm also glad that they ask the "critical" questions right off the bat, without any leading questions. I would still like to know the number of people that they considered a likely voter for each demographic (gender, race, age, geography, and so on), so that we can determine for ourselves if it seems representative of the population.

You can view the crosstabs here. Not a lot very interesting or unexpected, though.

It is important to note that it is not enough to just try and turn out a target region or demographic. However, this information can help us to identify who and where it is generally worth spending more resources trying to turn out.

In terms of the Governor's race, Republicans support DeVos more than Democrats support Granholm. We have voters aged 50 and up, whereas DeVos is is more popular with individuals under 50. Black voters, Hispanic voters, and other minorities seem to favor Granholm. Although individuals in "Generation Y" tend to support Granholm, people in college or with some college experience favor DeVos by a wide margin (sucks for me). We have strong support in Wayne County and Urban areas. DeVos is ahead nearly everywhere else, with one notable exception: Interestingly, Granholm is within the margin of error with DeVos in his hometown of Grand Rapids.


At 7:42 PM, Anonymous Anonymous said...

As a college student (albeit college in wisconsin, I am a michigan native and live in Saginaw when I am home) I still support Jennifer Granholm. After having worked on the DeVos campaign, I have discovered that I do not want to support him.


Post a Comment

<< Home