Elon Univ poll-5 southern primary states: Giuliani and Clinton holding their own

[Amenhotep Bakari-Sellers]

Dem primary:

Clinton:     32%
Obama:     14%
Edwards:   8%

Republican primary:

Giuliani:     21%
McCain:     16%
Romney:      3%

North Carolina:

Clinton:      32.4%
Edwards:   13.5%
Obama:      9.5%

South Carolina

Clinton:     16.7%
Obama:     16.7%
Edwards:     5.6%

The poll was conducted Feb. 18-22 by the Elon University Institute for Politics and Public Affairs, which surveyed 719 residents in Florida, Georgia, North Carolina, South Carolina and Virginia.

The margin of error was plus or minus 3.7 percent.

http://www.journalnow.com/servlet/Satellite?pagename=WSJ%2FMGArticle%2FWSJ_BasicArticle&c=MGArticle&cid=1149193381837&path=!localnews&s=1037645509099

[Mr. Morden]

The GOP result is unsurprising.  As I note here:

https://uselectionatlas.org/FORUM/index.php?topic=53960.0

Giuliani leads in the South, but not by as much as he does nationally.

But the NC and SC results for the Democrats are pretty weird.  Much weaker showing for Edwards than the previous polls in those states.  (NC is his home state.  He should be doing better than that.)  Did they only have 719 respondents total for all five states?  If so, the individual state samples are so small as to be worthless, which would explain the weird results.

[Verily]

North Carolina looks weird with Clinton leading, and South Carolina has a LOT of unsures for an earlier primary state. (The MoEs are probably large, though.)

[Mr. Morden]

Regarding the "small sample size" hypothesis, I bet the SC Dem. sample is some small multiple of 18, as 1/18=5.6%, and 3/16=16.7%.  If they really did interview only 719 people in all five states, then I wouldn't be surprised if there were only 36 SC Democrats that they interviewed....or even just 18 SC Democrats, with 1 person saying Edwards, and 3 each for Clinton and Obama.  They can't be so dumb as to report polls with such small sample sizes, can they?

[Filuwaúrdjan]

Uni poll!

[Mr. Morden]

Full poll results given here:

http://www.elon.edu/e-web/elonpoll/022607data.doc

including all the crosstabs for individual states, but no numbers are given for how many respondents in each state.  I'm too lazy to post all the results (someone else can do that), but the short version is: Clinton leads in FL, NC, and VA.  Clinton and Obama tied in GA and SC.  Edwards is in 3rd place everywhere, except NC, where he's in second.  Giuliani leads FL, GA, and NC.  McCain leads SC and VA.  That said, I'm guessing the state samples are so small as to be worthless.

[Reaganfan]

The GOP result is unsurprising.  As I note here:

https://uselectionatlas.org/FORUM/index.php?topic=53960.0

Giuliani leads in the South, but not by as much as he does nationally.

But the NC and SC results for the Democrats are pretty weird.  Much weaker showing for Edwards than the previous polls in those states.  (NC is his home state.  He should be doing better than that.)  Did they only have 719 respondents total for all five states?  If so, the individual state samples are so small as to be worthless, which would explain the weird results.

Edwards was the centrist democrat from the south in 2004. Now he's the anti-war liberal from the south. It doesn't help.

[Eraserhead]

This poll looks like total crap to me.