New Iowa, South Carolina Polls
by Chris Bowers, Mon Jun 04, 2007 at 11:30:19 AM EST
Public Policy Polling, an IVR polling firm (automated phone surveys), has released two new early state polls that show different results from other early state polls (I just received the results over email).
Iowa, May 30, 1,238 likely caucus goers, MoE 2.7%. No trendlines
Edwards: 31
Clinton: 17
Obama: 17
Richardson: 10
Biden: 4
Kucinich: 2
Dodd: 1
Gravel: 0
Undecided: 10
Six poll Iowa average: Edwards 27.7%, Clinton 22.8%, Obama 19.8%, Richardson 8.3%
South Carolina, May 31, 531 LVs, MoE 4.2%. No trendlines
Obama: 34
Clinton: 31
Edwards: 15
Biden: 3
Richardson: 3
Kucinich: 1
Dodd: 0
Gravel: 0
Undecided: 12
Four poll South Carolina average: Clinton 30.25%, Obama 26.0%, Edwards 17.5%
Iowa and South Carolina are two of the only states showing conflicting poll leads. In the ten South Carolina polls over the last two months, seven have now shown Clinton ahead, and three have now shown Obama ahead. In the seven Iowa polls in the last two months, five have shown Edwards ahead, and two have shown Clinton ahead. The polls further conflict on the size of the leads they show, and on who is in third place. One South Carolina poll shows Obama in third, and two Iowa polls show Clinton in third.<Br.<Br> In this case, my first reaction is to guess that Obama leads in South Carolina due to the IVR polling style, as IVR polls tend to include a higher percentage of young voters in their sample. Then again, this IVR poll also shows Edwards way ahead in Iowa, and he hasn't been known to sue particularly well among young voters, at least so far and at least nationally. Overall, the fluctuation we see between many early state polls is probably just as connected to a highly volatile early state electorate as it is to different methodologies. As such, I must continue to emphasize the importance of poll averaging as the best current means of gauging the state of the campaign.
http://www.mydd.com/story/2007/6/4/113019/4571