February 27, 2008 4:33

Early Voting in Texas

Posted by Karen Tumulty | Comments (47) | Permalink | Trackbacks (0) | Email This

This analysis, passed along by GWU's John Sides, confirms what we've been hearing anecdotally: Early voting in Texas does not bode well for Hillary Clinton. It is up dramatically in counties with large African-American populations, as well as those with more highly educated and affluent voters--in other words, in areas that would be expected to trend toward Obama. In those heavily hispanic counties that Clinton is counting on, not so much. Sides quotes CUNY's Brian Arbour:

There is no evidence yet that these “new early voters” are really “new voters.” They well could be those who usually vote on election day, who have decided this year to vote early.
Each of these three variables is correlated with each other. The three South Texas (and thus, most likely Clinton friendly) counties—Hidalgo, El Paso, and Nueces—are always at one end of the scale. These are the counties that have the smallest increases. The turnout measure could just be measuring something about political culture that makes voters willing to vote early rather on election day.

We don’t know who or where in each county people are voting early. While these data are consistent with an Obama surge, it could well be that turnout is high in areas in these counties where Clinton will do well. Or that demographic groups that favor Clinton—such as middle income women—are the ones creating the increase.

We have no data yet on who these people are voting for. Thus, I’m not measuring the really interesting question. We’ll still have to wait a week to get an answer.

Still, this can't be making the Clinton campaign any less nervous.

Sphere Related Blogs & Articles

Reader Comments (47)

StewieZ:

Hillary is done. Texas will seal the deal. I just hope that Hillary's supporters will not hand McCain and the republicans the election by refusing to join the rest of us as we work toward national unity and a new tomorrow with President Barak Hussein Obama!

Posted by StewieZ | February 27, 2008 4:46 PM

Elvis Elvisberg

I thought that Obama was winning lately among the less educated and affluent?

Posted by Elvis Elvisberg | February 27, 2008 5:01 PM

stuart_zechman:

(sigh) Here we go again, Karen.

Posted by stuart_zechman | February 27, 2008 5:01 PM

Karen Tumulty:

KT here--

But, SZ, I'm dying to see what you and RKA make of this...:-)

Posted by Karen Tumulty | February 27, 2008 5:06 PM

Paul-no not that one:

Is Texas not a sort of primary/caucus hybrid?
I thought I heard speeches telling supporters that they could vote early but then imporing the voters to show up on election day.
If that is accurate which part is more important?
Sorry if this has been covered already.

Posted by Paul-no not that one | February 27, 2008 5:13 PM

Harry:

No disrespect to a neat post, Karen. But I doubt the Clinton campaign is nervous about this at this point. Despite the brave face, they're not stupid. Their candidate is on the long, slow fade. She's not been able to connect, to mobilize, to energize, and she's been unable to dent her opponent's increasingly shiny armor. It seems that at this point, the interesting story will be with what grace she concedes.

Posted by Harry | February 27, 2008 5:17 PM

jayackroyd

It really looks to me like what has happened, over and over again, is that people trended away from voting for Clinton when it really seemed to be possible that she would lose.

And this is another instance.

Posted by jayackroyd | February 27, 2008 5:23 PM

Paul-no not that one:

Answering my own question for those who are curious and didn't know.
Shorthand- 2/3 of the delegates are from the primary and 1/3 from the caucus held after the polls close that night.

Posted by Paul-no not that one | February 27, 2008 5:29 PM

StewieZ:

jayackroyd - that is human nature, and once you hear what Obama has to say you understand that he cannot, he must not, lose, and you naturally swing to support him. Makes total sense to me!

Posted by StewieZ | February 27, 2008 5:31 PM

Oregon JC:

Sayeth Kos, in simple #s:

"In the top 15 counties with registered Democratic voters, there have been 419,904 early votes cast. Four years ago, the number was 72,688."

Posted by Oregon JC | February 27, 2008 5:38 PM

stuart_zechman:

Karen:

Still pitching that show (LOL, of course)?

But seriously, these numbers are a bunch of nonsense. Brian Arbour is not truly telling us anything about early voters. The problem with his methodology is that he doesn't seem to believe that what he doesn't know matters at all.

Mr. Arbour's entire premise rests on there being a correlation between counties that voted early and their "demographic characteristics", and the assumption that the people who voted early just happen to match those demographics exactly.

Of course, we know no such thing at all. We have no idea whatsoever who those voters actually were. What if the only people who voted early in a county that has "demographic characteristics that favor Barack Obama" were the people who were most likely to vote for Hillary Clinton?

See the problem? By totally discounting what we don't know about the people who actually voted, we're removing a variable that's the lynchpin to this guy's whole proposition.

The real question should be: Of the people who voted early, what were their "demographic characteristics", and do those voters' demographics fit the profile of a likely Obama voter? The demographics of the counties are potentially highly misleading, unless we're talking about counties demographically identical in homogeneity to counties in Norway or Guadalajara or Burundi. The fact is that we don't know who voted.

So interesting post, Karen, but this looks like rank speculation again.

I welcome any (ahem) suggestions as to how my reasoning might be in error...

Posted by stuart_zechman | February 27, 2008 5:39 PM

stuart_zechman:

Oh, and Mr. Arbour isn't taking into account gender as a demographic characteristic of those counties either, btw...

Posted by stuart_zechman | February 27, 2008 5:42 PM

Oregon JC:

I daresay that Stuart is perfectly "right" but as a judge once told me as I lied to him that I wasn't drag-racing, that I was merely in a hurry to get to a party (my best friend tried to take on the '74 Le Mans in his Dodge custom van, mind you), if it looks like a duck and walks like a duck it's probably a duck. Now, argue with that logic.

I rest my case.

Posted by Oregon JC | February 27, 2008 5:55 PM

KathyR:

Do we know for sure that early votes will be counted? In some places absentee ballots aren't counted if it's deemed they wouldn't make a difference. I suppose this is because the envelopes would have to be opened, etc.

I know that both campaigns have urged their supporters to vote early, so I presume they know they'll be counted - but I wonder if any of you know the history in Texas re: early votes being counted regardless of day-of-voting outcome.

Posted by KathyR | February 27, 2008 6:07 PM

Martin Gale

"Exit polls" a week before the real vote....

Didn't the networks stop publishing exit poll data while a vote was in progress for fear of influencing the outcome? I guess print's standards aren't as high as TV's these days. Considering these aren't even exit polls, it's probably a pretty good guess.

Posted by Martin Gale | February 27, 2008 6:15 PM

QUESTION HILLARY

http://news.yahoo.com/s/nm/20080227/pl_nm/usa_politics_lewis_dc_2;_ylt=AvfVUo5At0hfZevXQwFYYThh24cA

More good news for the Clixon lieberry.

Posted by QUESTION HILLARY | February 27, 2008 7:27 PM

Aaron:

The R^2 are horrible, and there's too much correlation between the variables. That means that there is probably not a lot of predictive power in this model. (I'd have to look at the underlying data to be sure, but that would tend to be the case.)

I think that third caveat means that Brian Arbour is aware of Stuart_Zechman's first major concern. As for gender, he must have the data. Perhaps he's been convinced by the post-Potomac primary punditry that women are evenly split between the two candidates. As he says, "These numbers don’t prove anything yet." I agree with that.

Posted by Aaron | February 27, 2008 7:44 PM

Malcolm:

KathyR,
Do they really not count absentee ballots in primaries that are not winner-take-all? Do you know of examples of this is any Democratic primary this year?

Stu,
Brian Arbour seems to be bending over backward to emphasize that the obvious inference that everyone is drawing is on very unsteady ground. But I'd like to add that the one piece of demographic information that we do have to go on at this point, namely, the county of the voters, does correlate well with different levels of support for different candidates. It's as if we knew that large numbers of blacks or older women were voting early - we would extrapolate from that that Obama or Clinton, resp., is likely to be doing well because of their high levels of support from these groups. Or to put it another way, on what basis could we conclude that the demographics of early voters aren't similar in composition on average to that of the counties as a whole?

And what makes you think that Burundi, with its split between Hutus and Tutsis, is so homogeneous?

Posted by Malcolm | February 27, 2008 9:00 PM

four legs good:

Told ya!!

:-)

Secretary of state said this evening that voter totals, so far, are over 600.000.

We have 2 days to go. Oh my oh my. I do believe the firewall is on fire.

Posted by four legs good | February 27, 2008 9:38 PM

stuart_zechman:

Malcolm:

Brian Arbour seems to be bending over backward to emphasize that the obvious inference that everyone is drawing is on very unsteady ground.

This is true.

...what basis could we conclude that the demographics of early voters aren't similar in composition on average to that of the counties as a whole?

None. There is no basis for that conclusion, simply because one can't prove a negative. Of course, that's a logical fallacy, and gets us nowhere near actually knowing anything about anything. What you seem to be doing is justifying speculation, which you can do, obviously, with a premise like "well, we don't know for certain that X is not the case".

And what makes you think that Burundi, with its split between Hutus and Tutsis, is so homogeneous?

AHA!!

I knew that somebody would come up with that!
I was also waiting for a "one of these is not like the others" with the inclusion of Guadalahara in the list.

...you got my point though, right?

Posted by stuart_zechman | February 27, 2008 10:11 PM

StewieZ:

I can taste the Obama victory and the rise of the Obamanation!!!! I love it love it love it!

Obama 08!

Si se puede!

Posted by StewieZ | February 27, 2008 10:39 PM

stuart_zechman:

T4T!!

Posted by stuart_zechman | February 27, 2008 11:03 PM

four legs good:

Kathy R, early voting isn't absentee voting. That's something different.

We actually have a two week period here where you can go to several central county wide polling places and cast a ballot. On a regular voting machine.

If you miss the early voting period, you can still go cast your ballot on the regular election day at your precinct. And then of course for the primary, on the dem side, you can go caucus that night. You must vote in the primary (dem) to participate in the caucus.

And yes, they do count the votes.

Posted by four legs good | February 28, 2008 12:19 AM

Malcolm:

Stu,
I get your point, but just because it's speculation doesn't mean it's groundless speculation.

And why the T4T shout out? He didn't post on this blog. Or are you trying to encourage him to contribute?

four legs good,
Thanks for clearing that up.

Posted by Malcolm | February 28, 2008 2:13 AM

stuart_zechman:

Malcolm:
Sorry, "StewieZ" is "stuart__zechman" who is "Time4Tolerance".

I was warning the troll.

Posted by stuart_zechman | February 28, 2008 9:36 AM

docbiohazard:

Stu,

I would say your concerns were fully addressed and acknowledged in Dr. Arbour's caveats, so I fail to see your concern... unless you denounce (or reject?) statistical analysis of campaigns and elections as a whole field of study?

DocBH

Posted by docbiohazard | February 28, 2008 10:00 AM

StewieZ:

Sadly, stuart, you appear to be mistaken on this. But I'll forgive you in the spirit of Democratic unity behind President Obama.

Posted by StewieZ | February 28, 2008 11:02 AM

stuart_zechman:

DocBH:

I can perhaps concur with my concerns being addressed, but:

unless you denounce (or reject?) statistical analysis of campaigns and elections as a whole field of study

...that's a bit of a strawman, don't you think?

For example:
If a geologist comes out with a theory, say that there was never such a thing as the super-continent Gondwana, and I object to that theory on the basis that the evidence to reach that conclusion is incomplete, or that required data is missing, that doesn't indicate at all that I believe that the entire scientific method is bankrupt, or that I denounce and reject geology.

I do think, though, that we should get more in the habit of repeating the phrase "denounce and reject", because it has so many uses, and is really funny. I really did LOL when I read that in your post. Good one.

Posted by stuart_zechman | February 28, 2008 11:49 AM

docbiohazard:

...that's a bit of a strawman, don't you think?

Yep. Thats what I get for posting before I'd had my morning coffee. Oh well, I'll denounce and reject that portion of my earlier post.

DocBH

Posted by docbiohazard | February 28, 2008 2:53 PM

stuart_zechman:

DocBH:

If I post prior to coffee intake, the volume of commentary that I wish could be rescinded far surpasses any portion of your post that you may have wanted to edit, so I'll take this opportunity to denounce and reject any pre-coffee assertions that I may have made in error at this time.

I love this new "denounce and reject" thing, BTW.

Posted by stuart_zechman | February 28, 2008 3:59 PM

stuart_zechman:

...when I said "prior to coffee", I meant just then.

That last post was...bad.

Posted by stuart_zechman | February 28, 2008 4:45 PM

Malcolm:

Stu,
To summarize, what we know is that early voting in counties that are likely to be pro-Obama is up. While this is not STRONG evidence that Obama will do better in Texas than expected, it is still evidence of that.

Posted by Malcolm | February 28, 2008 8:18 PM

StewieZ:

President Obama is going to own Texas!!! His time is now. The question is will the Hillary supporters do the right thing and come join us and will Hillary do the right thing and drop out gracefully?

Posted by StewieZ | February 28, 2008 11:14 PM

goldstonesoft:

M4V Converter free download center- ALL M4V conversion tools which help you feel free to convert M4V to MP3, AVI, WMV, MPEG, FLV, 3GP, WMA. And useful Guide on How to Deal with M4V files.
m4v converter
convert m4v
Ultra Quicktime M4V converter currently is the best M4V converter which can convert Quicktime movie M4V to AVI, MPEG, MP4, MPG, WMV, ASF and Vob. As we know Quicktime movie usually has the video formats of M4V, MOV, QT, MP4, and M4V, With this M4V Converter, you can feel free to Convert them all with fast speed and high output quality.
m4v to avi
m4v to wmv
With the fast and powerful QuickTime video decoder inside, Ultra QuickTime M4V Converter supports almost all MOV, QT, MP4, M4V files, even QuickTime Player has not been installed. Integrated High-speed MPEG-2 encoder which let you convert M4V to DVD-Video files(VIDEO_TS, AUDIO_TS) and VCD/SVCD image(*.bin,*.cue), so you can burn VCD/SVCD/DVD disc easily from QuickTime files by using third-party buring tools. It is a software program for converting video formats at fast speeds and high quality. Very user-friendly interface and Quality Profiles.
m4v to mpeg
m4v to mp4