How does early voting affect exit polls on election day?

[jpgray]

Just curious if anyone has an idea. Are the pollsters conducting such during early voting and then combining data on the day?

[Schulzz]

1. Great question. I would also like to know this.

It will be important how to deal with the exit polls. If they don't include early voting at all, Obama should outperform most exit polls.

[ecstatic]

4. Maybe my brain is fried right now, but I thought it would be just the

opposite. So much early voting is going on, especially here in Georgia. A lot of Obama voters are being urged to vote early to avoid long lines and not knowing their registration status. So that makes me think that on election day, the exit polls might indicate stronger support for McShame since Obama supporters would have already turned out in large numbers already for absentee/early voting.

[glowing]

2. They are keeping track.. This is a big deal election.. They keep track

of the amt of Dems and Pugs showing up to vote..

[skavoovi]

3. Exit Polling and vote manipulation

My concern is that if Democratic turnout is high in early voting, then Nov 4th exit polls won't be a true reflection of voting. This could result in lower results in exit polling which will help support the idea of a 'Bradley effect" being responsible for an Obama defeat when votes are manipulated....

[bigwillq]

5. Hello Nutmegger!

Welcome to DU! :hi:

[ddeclue]

6. Historically it skews it Republican...

Democrats tend to outvote Republicans in EV.

on the other hand Republicans tend to outvote D's in ABV.

If we keep up in ABV (which is starting to be true in FL races in the last 2 years) then EV will get more D votes in than R votes BEFORE election day resulting in a smaller # of D's voting ON election so if the pollsters are relying on a simple exit question on election day and not factoring in the D's and R's that EV'd and ABV'd then the numbers will generally skew to the R's.

I ran the ABV program for Orange County, FL in 2004 for Kerry and have helped many candidates with their ABV chase programs and have tracked the EV's and ABV's (although not this time) for Orange for the last 4 years and what I've said generally holds true in FL.

Doug D.
ABV Director for Kerry Campaign,
Orlando/Orange County FL, 2004

[Jackinbox]

8. Thanks for the info

I thought EV and ABV were pooled together. Isn't EV just filling out an ABV in person?

I will be keeping a watchful eye on the FL returns on Election Day. In 2004 when I saw the returns coming in from Pinellas, Hillsborough, Orange and Volusia counties, I knew we were in trouble based on the results from 2000. If Pinellas, Hillsborough and Volusia start coming in good and Orange is at least close, we'll be fine.

[ddeclue]

11. The SOE's distinguish the two in their daily reports.

[BzaDem]

7. They do phone polls in advance to determine who voted early.

This was mentioned on CNN and MSNBC on multiple primary election nights. The polling is pretty accurate.

[ShadowLiberal]

9. The Politico did an article on this a day or two ago

Basically the article was about fear among pollsters that so much early voting is going to mess up the exit polls. The reasons for it were the following.

-Obama supporters are far more enthusiastic to vote, and more willing to tell pollsters who they voted for, hence it's easier to get them to participate in exit polls (and over sample them).

-Groups that generally support McCain (like the elderly, etc) are less likely to participate in exit polling

-In most of the primary battles between Hillary and Obama (with Obama's supporters more enthusiastic and motivated then Hillary's) the exit polls overstated Obama's support in most states (I think the article said 26 out of 28 states). The article also said that it wasn't until later in 2004, after early exit poll results were already leaked, that pollsters realized they were oversampling women voters, who were more likely to support Kerry, and more willing to say who they voted for.

According to the article they were going to carefully look over exit poll results on election day before releasing them, with a bunch of people looking at both who answered their exit polls, and who didn't, to see if they're possibly oversampling one group or under sampling another (since exit pollsters can say the gender, roughly where the person's age was, and how educated they looked, etc).

[meow2u3]

10. This is the real "Bradley effect"

It's the massive repuke early voting--not race--that lost the election for Bradley.

I think it's going to be the Obama effect; early voting reportedly is favoring Obama, except in those areas that are either stealthily purging voters without notifying them, intimidating voters from casting their votes for Obama, or where trick voting machines are switching their votes for Obama to McLame.

[Ichingcarpenter]

12. affect vs effect?

If you are talking about a result, then use the word "effect."

It is appropriate to use the word "effect" if one of these words is used immediately before the word: into, no, take, the, any, an, or and.

Example: The prescribed medication had no effect on the patient's symptoms.
Example: In analyzing a situation, it is important to take the concepts of cause and effect into consideration.


Affect can also be used as a verb. Use it when trying to describe influencing someone or something rather than causing it.


I still have no clue... those two words have 'affected' me for a long time
Now, should have I said 'effected' me?