Pollster.com

Articles and Analysis

 

Winthrop/ETV: NC, SC, VA (9/28-10/19)

Topics: PHome

Winthrop/ETV
9/28-10/19/08
Mode: Live Telephone Interviews
(source)

North Carolina 744 LV, 3.6%
Obama 45, McCain 44

South Carolina 617 LV, 3.9%
McCain 55, Obama 35

Virginia 665 LV, 3.8%
Obama 45, McCain 44

 

Comments
thorfinn:

This is from a month ago? It's not very useful.

____________________

mysticlaker:

HAHA. three weeks in the field...

____________________

Mike A.:

ha I was a little worried than I saw this is over 3 weeks. weak :)

____________________

Dan:

what a stupid poll.
it moved VA to light-blue though. poo.

____________________

bpd1069:

What a useless poll

____________________

Schadenfreude:

I'm not buying those SC numbers. That would put Obama's share of the white vote at something pretty close to 10%.

African-Americans are about 30% of the population, Hispanics about 4%. Given trends in other states, that should give Obama 28-30% right there, leaving him only 5-7% out of 65% of the population.

____________________

NYCREALAMERICAN:

The color on pollster or RCP doesn't mean poopy. The trend has been all BO, and all polls conducted since 10/17 have BO leading by an avg of 4 in NC and 7 in VA.

But if your worried - donate make phone calls, go door to door.

____________________

Northeastern Republican:

hm that is kinda strange. why would they poll for so many weeks?

still, it is suprisng that the entire period they polled was when obama was surging to +10 in the polls and it still only shows a 1 point obama advantage. wish they'd provide weekly break downs so we could see what weeks were the ones that were more favorable to obama or more favorable to mccain. if it was this last week that was more pro mccain then that could spell virginia trouble for obama bc there was another poll out this week calling it a tight race in virginia - although there was also i think 2 others showing it to be an obama blow out. id be suprised if mccain pulls off a virginia will but looks like it COULD be turning into less of an obama blow out. COULD. we need to see some more recent polls i think.

____________________

Hope Reborn:

soon to be Boomshack's poll of the year... that tard does after all live in the past, in the days when Palin wasn't Failin and McNopoly still had a chance...

____________________

Viperlord:

Discount this poll for God's sake, it's ancient history. Like McCain.

____________________

Batony:

Actually this is pretty good news for McCain...b/c the past few weeks haven't been good at all for him.

All of a sudden I have noticed tons of McCain /Palin signs everywhere in Northern Va. Before now it was Obama/Biden all over the place.

____________________

Schadenfreude:

They used this to downgrade VA to a lean? Somebody is failing to pay attention.

____________________

Viperlord:

THIS POLL IS WAY OLD PEOPLES. IGNORE IT.

Can you people read it now?

____________________

BrookLynda:

It took three weeks to talk to 665 LV's in VA? That's like 220 a week. WTF??

____________________

boomspin:

THIS POLL SHOWS OBAMA'S COLLAPSE!

____________________

ZanzibarBuckBuckMcFate:

One old (and obscure) poll seems like a pretty slender excuse to justify downgrading Virgina's status to "lean".

____________________

RaleighNC:

Uh-oh

____________________

RaleighNC:

You can't have it both ways, ZANZIBAR. If one polls tips a state blue, it should work in reverse, right? The great thing about pollster.com (despite some of the liberal minion commenters) is that they use every poll out there and let the chips fall where they may. Good for them.

____________________

PHGrl:

Who are these guys? never heard of them.. such a small sample over such a long period of time.. what did they do? strap some poor graduate students to a call station for 3 weeks?

____________________

ZanzibarBuckBuckMcFate:

RaleighNC:
You can't have it both ways, ZANZIBAR. If one polls tips a state blue, it should work in reverse, right? The great thing about pollster.com (despite some of the liberal minion commenters) is that they use every poll out there and let the chips fall where they may. Good for them.

"One poll" didn't tip the state in question dark blue, it was a series of them, and all far more current than this one. I don't really care one way or another what color the state is as such, I just think it's unwarranted to change its status merely on this basis.

____________________

ZanzibarBuckBuckMcFate:

And in any event, going by the "one poll" rule, Georgia should now be yellow at least, if not light blue, since a poll was just released showing Obama with a small (very small) lead.

____________________

RaleighNC:

ZANZIBAR...you must be new here. One poll can tip a state that doesn't have a lot of polls; however, one poll for a state with a lot of polls is averaged in a lost in the sheer volume of polls. If you want to do this yourself, just add up the numbers for Obama and McCain in any state then divide by the number of polls you have. You should get something close to what pollster.com has. I think they may weight their polls, but all that is too technical for me! Anyways, the trend in VA is slightly to McCain and the latest batch of polls show that it's closer than it was, but Obama still ahead. Simple.

____________________

Catherine Christman:

Folks,
I represent the "ETV" part of the Winthrop/ETV Poll. We are the public television network that is statewide in SC. I am posting the rationale for our length in the field. I think you will find it educational and thought-provoking. I think it brings up many points you may not have ever considered about polls that are turned around very quickly. I think you will find the logic worth the read.
Catherine Christman/VP of Communications/ETV

Questions Regarding Length of Time in Field

While this is a long time for a political survey to be in the field, it does two things: 1) Increases the ‚Äúrepresentativeness‚ÄĚ of the sample without weighting and 2) Smoothes the erratic jumps and dips seen when overnight polls are aggregated. First, most polls done in a few days call during the daytime and must use a complicated weighting scheme for the data. For example, retirees and stay-at-home moms are more likely to be reached during the day and this over-sampling must be accounted for in the final results. The Social & Behavioral Research Lab (SBRL) at Winthrop University calls only on weekday evenings and during the day and evenings on weekends. This allows respondents to be randomly sampled in their proper proportions without additional weighting. While the trade off between speed of data gathering and un-weighted accuracy of the sample is unacceptable to some, un-weighted sample quality holds a supreme position for us since this data will be used for academic research after the conclusion of the election. Second, a longer time in the field has a "smoothing effect" on the data. Trends are more fluid compared to the choppy ups and downs seen in many quick snapshot polls. However, spending too long in the field can mask significant shifts. The consumer of this data should use these results to paint a fuller picture of trends seen from the multitude of quick snapshot polls; she or he should not assume these results supplant any other methodologically sound data.

____________________

Schadenfreude:

@Catherine Christman

I'm not sure that methodology is terribly useful in a political campaign, which is all about short-term changes in opinion. Great for market researh, maybe. Determining whether or not Virginia will vote for McCain or Obama on Nov. 4th, not so much.

I would also like to see your demographic breakdown (especially for South Carolina), which I could not find on your site.

____________________

falcon79:

@ catherine christman:
thank you for the information.
i can definitely see your rationale.

____________________

Mike In Maryland:

Catherine Christman,

Balderdash.

When you call only weekday evenings, how many people do you exclude - people working evenings? People out shopping, on a date, whatever?

What are the demographic breakdowns? Did you over- or under-sample any ethnic group? Age group?

Do you include or exclude cell phones?

Do you call back for missed or unanswered calls?

There were a lot of things that happened during the time you were in the field:
- Three Presidential debates;
- One Vice-Presidential debate;
- On the final day you were in the field, there was a major endorsement;
- 'Joe the Plumber' was created, then exploited;
- Major changes in campaign strategies were announced;
- Etc.

Political polling over a several week period can prove to be effective during the early part of the campaign, or months prior to an election, when events are not constantly changing. Think summer prior to the primaries. Polling over a several week period ending within three weeks of the general election (using the excuse it '[s]moothes the erratic jumps and dips') can hide major trends occurring while the poll is in the field. A shift of just .25 point a day over a 20 day period translates into a 5 point shift.

There seem to be serious problems in your methodology, and serious problems with your excuse for the length of time for conducting the poll.

And again, where are the demographics?

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR