More Bush 2000 voters are defecting than Gore 2000 voters

http://www.gmi-mr.com/en/worldpoll/press_room_wppk_qr.phtml

I posted this poll over in the "The World Likes Kerry" thread...but they asked a question of American voters I haven't seen much in other polls: Are you going to vote differently this time than last?

They asked (in two seperate polls, one in July, one in August)people who voted for Bush (and Gore) in 2000 who they would vote this time.

In July, 10% of Bush 2000 voters said they will vote for Kerry, and 14% were unsure.  Just 4% of Gore voters in July said they will vote for Bush this time, with 10% unsure.

In August, 14% of Bush 2000 voters said they will vote for Kerry, with 12% unsure.  5% of Gore voters in August said they were voting for Bush this time, with 9% unsure.

So, in the August poll, only 74% of Bush 2000 voters are certain to do so again-while 86% of Gore voters are positive about going with Kerry.

I have always believed that there were Bush voters who changed thier minds, but very few who went the other way.  This proves this theory, at least in my mind.

Since the last election was, basically, a tie, this means Bush loses.  There is no way around this, unless Rove really is able to get out more of his base (or he cheats)-and a higher percentage than we are able (I expect higher turn out by both sides this time, at least in their bases-some moderate or fiscal conservatives may be turned off by Bush, so that may hurt him, too).  Plus Nader is weaker this time-a few of his votes will bleed into the Kerry column.

Logically, Kerry is a shoo-in.  Still, we must pretend he actually is down by ten and campaign accordingly, even if he, in fact, is winning.

Tags: (all tags)

Comments

21 Comments

I hope you are right
I too have noticed only defectors away from Bush and not the other way.  On the other hand, the Bush crowd seems more adamantly pro-Bush now.  Last time it seemed they were simply going for someone without the "stain" of Clinton - now it seems they really see the Shrub as some sort of savior.

Bush still loses 54-44-2

by Tito 2004-09-08 04:47PM | 0 recs
You Nailed It
when you said "unless Rove really does manage to get more of his base out."

Supposedly 4 million Evangelicals sat out election 2000, and ever since then Rove has been obsessed with getting this people to show up.

This is why the Gay Marriage amendment was defeated and why several states--including swing states--will have state Constitutional amendments on gay marriage on their ballots this November.

It's also why Bush plays to invitation only audiences. He's not looking for the moderate swing voters. He wants his right wing base to show up; it's the only chance he has to win.

That, I suspect, is also why Bush does better among likely voters than registered voters. A lot of people who plan to vote for Bush in 2004 sat 2000 out. But will they actually register and vote this time? Who can say?

by Cat M 2004-09-08 05:55PM | 0 recs
That's all he's concentrating on...
...Rove has these numbers, too (in fact, I'll bet he has more accurate and more detailed numbers than these).  He's going all out for the base, and not even pretending to care much about the middle.  The problem is, there's just as much GOTV effort on the Dem side-plus a large portion of our deepest base voted for Nader last time.  They won't make the same mistake twice.
by Geotpf 2004-09-08 06:44PM | 0 recs
Oh, and about likely voters...
...if you didn't vote in 2000, some likely voter models assume you are therefore not likely to vote this time (or weigh that heavily in thier overall likely voter calculation).  So your theory there is wrong, IMHO.
by Geotpf 2004-09-08 06:46PM | 0 recs
Re: Oh, and about likely voters...
Not sure I understand what you mean. I'm not talking about calculations as to who will win. I don't think when they report how likely voters responded to polls that they weight them based on whether they think they will actually vote. That would only be necessary if they were trying to project who would win.

In this case, they are only trying to show the popularity of the candidates among likely voters versus among registered voters. I think Bush does better among "likely" voters than Kerry because I think a lot of 2004 Bush voters didn't vote in 2000.

Kerry does better or equal among registered voters in part because some of those who voted for Bush in 2000 aren't going to vote for him in 2004.

Perhaps what we need to see is a poll of how likely "likely" voters are to vote compared to registered voters?

by Cat M 2004-09-08 07:19PM | 0 recs
Re: Oh, and about likely voters...
People who didn't vote in 2000 won't show up as likely voters in any pollster's model.

If Rove found his magic 4 million evangelicals, if they can be convinced to turn out, and if they vote Republican rather than Democrat then including them in polls would actually skew the polls even more toward Bush.

Of course that is a bunch of awfully big ifs. The Democrats have registered a bunch of people since 2000 including younger voters, latinos, and blacks. True many won't turn out but many will. The Democrats have much more active GOTV programs this year along with more informal GOTV efforts such as people dragging their non-voting friends and relatives to the polls. Finally the assumption that evangelical voters are overwhelmingly Republican are false. True they skew slightly more Republican than the overall population but not overwhelmingly so. Even more interesting is the evangelicals who didn't vote in 2000 are more likely to consider themselves either democrats or economicly liberal.

by ces 2004-09-09 12:24AM | 0 recs
Re: Oh, and about likely voters...
Why wouldn't people who didn't vote in 2000 show up as likely voters??

When I was in college, I spent a little over a year doing polling for the Minneapolis/Star Tribune. A "likely voter" (at least back then) is just someone who says they are likely to vote in the upcoming election. It has nothing to do with whether they voted in 2000. They may or may not be registered.

A registered voter is someone who is currently registered to vote and also has nothing to do with whether they voted in the last election.

There is crossover in the two categories because someone can be registered to vote but not consider themselves a likely voter or they can be a likely voter but someone who is not yet registered.

I should think the "likely" voter number is less reliable because in some states you must register in advance to vote, and if you aren't taking the time to register, will you take the time to vote?

I agree democrats have made some strides on that point too and also I think that a many democrats aren't sitting at home answering telephone polls.

by Cat M 2004-09-09 07:01AM | 0 recs
Re: Oh, and about likely voters...
AFAIK, you MUST be registered to vote in order to vote.  You cannot show up on election day at the polling place and register on the spot.  Registration usually takes time - from a few weeks in some states to more than a month in others.

Registration is necessary so that elections officials can verify that a person lives in the jurisdiction they are voting in - no carpetbagging, even to get the wack-o fundy stealth candidate off your neighbor's school board.  It's also needed to make sure that persons don't vote twice.

Many states require re-registrations as a way to keep the voter rolls from becoming clogged with dead folks or people who moved.  For example, Ohio recently adopted rules that if you have not voted in four years, you must respond via postcard to the voting registrar, or re-register.  I'm concerned that a lof of folks who registered more than four years ago but have not voted since then will suddenly find on November 2 that they needed to re-register, but that the deadline already passed.

by Silent E 2004-09-09 07:56AM | 0 recs
I believe there are some states...
...that allow people to register on election day, but they are not the norm.
by Geotpf 2004-09-09 08:15AM | 0 recs
Re: Oh, and about likely voters...
Actually there are states where you can show up on polling day and register right there.

Minnesota, where I lived, is one of them. Maine is another. Wisconsin (where I also lived) is yet another. I've read that North Dakota doesn't require registration period but I'm not certain whether that is changed.

I don't think any state has closed its registration yet but it must be getting close for some of them.

Ah, just found a website. Six states have same-day voter registration.

http://www.demos-usa.org/page18.cfm

by Cat M 2004-09-09 09:43AM | 0 recs
Re: Oh, and about likely voters...
Interesting - thanks for the info.  It's something I'd definitely like to see implemented.  Forget Motor-Voter - just go with EDR!

(Election Day Registration)

by Silent E 2004-09-09 10:49AM | 0 recs
Here's how Gallup does it...
http://www.emergingdemocraticmajorityweblog.com/donkeyrising/archives/000635.php

Quoting from this site:

So, as a public service, here's how they do it. Let's start with Gallup. According to David Moore of Gallup:

Gallup asks each [RV] respondent seven LV screening questions, and gives each person an LV score of 0 to 7. [Assuming a turnout of 55 percent], the top 55% are classified as likely voters.
In practice that typically means all of the "7"s--given full weight--plus some proportion of those with lower scores (usually the "6"s), who are weighted down so that the size of the likely voter sample matches the projected turnout for the year (apparently 55 percent this year). All other voters are discarded from the sample.

What are the Gallup likely voter questions? Unfortunately, the exact questions and their wording are not released by Gallup along with their polling data, but the questions apparently involve past voting behavior, interest in the election, intention to vote in the election and knowledge of things like the location of the local polling place.

Another article about this:

http://slate.msn.com/id/2104727/%5C

Quoting from that article:

Some pollsters, for example, will ask only whether the voter showed up for the last election and whether they plan on doing so again. Since many people feel slightly guilty about not voting, or want to appear more politically engaged than they really are, such polls tend to dramatically overstate the number of likely voters. Pollsters refer to this approach as a "soft screen."

Other pollsters ask their respondents to qualify their answers, instead of giving simple yes-or-no replies. A typical question in such a poll might be: "Are you very likely to vote, somewhat likely, or not likely?" The answers are then weighted in order to account for the assumption that those answering "somewhat likely" will probably vote in fewer numbers than those responding "very likely." Pollsters may also ask those surveyed to rank their likelihood of voting on a scale of one to 10.

The most thorough polls ask whether respondents know where their polling places are and how they plan to travel there on Election Day. Another tactic is to couch the questions about past voting habits in soothing language, so the respondents don't feel as if an "I didn't vote last election" reply is equivalent to confessing that they're bad Americans. A popular way to phrase the voting-history question, for example, is along the lines of: "In the last election, did something come up that prevented you from voting?"

by Geotpf 2004-09-09 08:12AM | 0 recs
Re: Here's how Gallup does it...
Wow. That really makes me discredit polling even more than I already did. You're saying that for a lot of these polls, likely voters are a subset of registered voters?

That almost makes it irrelevant because there are people who will vote who are not yet registered and certainly this was more true months ago when registration deadlines were far off.

This is what I'm most familiar with:

Other pollsters ask their respondents to qualify their answers, instead of giving simple yes-or-no replies. A typical question in such a poll might be: "Are you very likely to vote, somewhat likely, or not likely?" The answers are then weighted in order to account for the assumption that those answering "somewhat likely" will probably vote in fewer numbers than those responding "very likely." Pollsters may also ask those surveyed to rank their likelihood of voting on a scale of one to 10.

Which has nothing to do with whether someone is registered already and only with whether they believe they will vote in the upcoming election. This makes the most sense to me.

by Cat M 2004-09-09 09:47AM | 0 recs
Re: Here's how Gallup does it...
That almost makes it irrelevant because there are people who will vote who are not yet registered and certainly this was more true months ago when registration deadlines were far off.

But only in the six states you mentioned above where same-day registration is legal.  True, Minnesota, Wisconsin and Maine are swing states, so that may be an important consideration for state-by-state polls there.  Elsewhere, though, the rule holds: if you're not registered before the election (usually more than a few weeks before), you can't vote.

Although, I'm going to have to update my SilentWeb Electoral College Predictions now.

by Silent E 2004-09-09 10:53AM | 0 recs
Re: Here's how Gallup does it...
Well when I was looking into this, I saw that registration deadlines are usually 15 to 30 days before the election. I suppose there may still be people not yet registered who will register before the deadline and are likely voters. I think that's what democrats and republicans are working on so hard right now, registering the unregistered.

I also believe I saw that some states let you register right up to the day before or the week before.

by Cat M 2004-09-09 05:57PM | 0 recs
Re: Here's how Gallup does it...
"You're saying that for a lot of these polls, likely voters are a subset of registered voters?"

As far as I know, that's the case for all of the national polls -- Gallup, Zogby, Harris, Time, Newsweek -- they're all polls of registered voters, with "likely voters" as a subset based on whatever model the particular polling organization uses.  Also, for some of those models, my understanding is that not having voted in the last election always disqualifies someone from being considered a "likely voter."  Since this election is likely to be decided in many of the swing states by first-time voters mobilized in the parties' unprecedented GOTV efforts, that makes these polls suspect, to say the least.
Later,
Alex

by Alex 2004-09-09 10:56AM | 0 recs
adsf
I'd like to know about the 50 some-odd million who didn't show up.
by pdc90dem 2004-09-08 06:27PM | 0 recs
Re: nonvoters
As would I.

It will be interesting to see just how many of those voters turn out this year.

Rove's strategy seems to be to depress turnout as much as possible except among the true believers. He seems to think the Republicans have a larger and more dedicated base in the states Bush needs to win. I'm willing to bet there will be some blatant attempt to scare people away from the polls on election day. Say a terror alert with "specific" information about potential attacks on polling places.

by ces 2004-09-09 12:30AM | 0 recs
Not the first poll to suggest as much
There was a Zogby poll out about a month ago that showed 8% of Bush's 2000 voters going for Kerry this time out, with only 4% of Gore's voters moving in the opposite direction.
by Ben P 2004-09-08 07:05PM | 0 recs
Two things:
  1.  Where are those voters?  If we replay the 2000 election, and Kerry picks up more Bush voters than Bush gets Gore voters, he might not win if none of that margin comes in Florida.  Even if Kerry wins all of Gore's 2000 states, plus NH, he still loses.
  2.  Look at the RV numbers: even if you ignore the LV numbers, Kerry is still getting hammered in the battleground states.  His RV leads are shrinking and in some states they've vanished.  
by Silent E 2004-09-09 05:11AM | 0 recs
Bush's bounce should be completely gone shortly...
...unless the aniversary of 9/11 helps him.  Today's Rasmussen poll has Bush up by only .7%.  Ignoring 9/11, it should be gone by the end of the week.  Even if he does get a bump on 9/11, that bump should be gone by the end of next week, IMHO.

http://www.rasmussenreports.com/Presidential_Tracking_Poll.htm

by Geotpf 2004-09-09 08:19AM | 0 recs

Diaries

Advertise Blogads