home

OH And TX Polls: Clinton By 16 In Ohio?

By Big Tent Democrat, Section Elections 2008
Posted on Sun Mar 02, 2008 at 09:31:47 AM EST
Tags: (all tags)
Share This: Digg! StumbleUpon del.icio.us reddit reddit

By Big Tent Democrat

Let's start with the most surprising result, a Columbus Dispatch Ohio poll:

A new Dispatch Poll shows that the New York senator would lead in Ohio by as much as 16 percentage points under the usual turnout scenario dominated by loyal, older party voters -- especially if women show up in force, as they have in earlier primary states.

Even if you presume an unusual influx of young and black poll-goers, Clinton still would hold a healthy single-digit lead, the poll indicates.

Clinton by 16? That result would shake up the race. [More...]

But other Ohio polls have it much closer, McClatchey has it 47-43. But turnout models still hold the key as the internals are not very different in the polls.

In Texas, the latest results from McClatchey show a tie. Latinos overwhelmingly prefer Clinton to Obama by 62-30. That proves yet again that who turns out is the big issue.

< Clinton Derangement Syndrome | Bush Moves to Deport Legal Resident Acquitted in Terror Case >

  • The Online Magazine with Liberal coverage of crime-related political and injustice news

  • Contribute To TalkLeft


    • Display: Sort:
    OH And TX Polls: Clinton By 16 In Ohio? | 43 comments (43 topical)

    Readers Must Login or Create an Account to Comment

    polls (5.00 / 5) (#2)
    by Lil on Sun Mar 02, 2008 at 10:01:57 AM EST
    I'm sick of getting my hopes up and yet like an addict I keep coming back for more.

    I expect SUSA to take a last shot (5.00 / 1) (#6)
    by andgarden on Sun Mar 02, 2008 at 10:08:02 AM EST
    for Monday. At least, I hope they do.

    Agreed, plus pointer to pollster's graph/tables (none / 0) (#10)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 10:15:13 AM EST
    Pollster.com has nifty graphs accompanied by tables which show the polling results in reverse chronological order.   Unfortunately, those tables don't show the estimated margin of error in each poll, but at least they're handy for looking at all the results in one place.  Here's Ohio, and Texas .

    Pollster appears to update these every couple of days, perhaps when there are enough new polling results to make it worthwhile.  I would guess that they'll do an update tomorrow, as last-minute results come in.

    Parent

    Re pollster and realclearpolitics' averages (none / 0) (#40)
    by Cream City on Sun Mar 02, 2008 at 01:02:55 PM EST
    and graphs, it's interesting to see differences -- which polls each uses and refuses to use. (Btw, an interesting oddity of realclearpolitics.com is its use sometimes, it seems, of more polls for Dems than for GOP trends, even when the same polls did both -- and especially when it boosts Obama's average.)

    Parent
    There does appear be momentum swinging back (5.00 / 4) (#26)
    by sef on Sun Mar 02, 2008 at 11:33:53 AM EST
    Say what you want about this particular poll & how it was conducted, the mighty mo' appears to be swinging back to Hillary in Ohio & Texas.  IMO she hit a homerun with the red phone ad, she did well on SNL (bootlegs keep popping up on YouTube), and her stump speeches in the last two-three days have been inspiring. Indeed, her performance has been so  good in recent days that it has caused me to say WOW, I'd love for her to be President --- and as most know, I'm a kool-aid swilling Obama guy.

    Had Hillary been doing what she's doing now all along (a combination of tough messaging with Hillary being Hillary instead of Hillary being what her advisers want her to be) this thing would have been over before Iowa.

    I can't wait to see how she does on The Daily Show Monday & whether she can keep up the mo' on the stump.  She's been brilliant.

    thanks, sef (5.00 / 1) (#29)
    by Kathy on Sun Mar 02, 2008 at 11:42:17 AM EST
    for your honest evaluation.  Everyone knows I back Clinton with all my heart (and money and time) and I've been telling myself I must be crazy to think that the momentum is swinging back, but that's really how it feels to me.

    I really think it started last week with Tina Fey calling it like she sees it.  If not for the writers' strike, there would've been something to "balance" the media bias (or at least shame them into stopping it) and it would've started sooner.

    Parent

    and that CNN debate maybe (5.00 / 1) (#34)
    by diplomatic on Sun Mar 02, 2008 at 11:53:52 AM EST
    she ended it with a standing ovation which is always a good way to get the ball rolling.

    Let's put it this way -- the last week and half could have been much worse for Hillary.

    Hillary has a CHANCE.  That may not have been the case had certain things not gone fallen into place recently.

    I felt she should have capitalized on her comeback after NH and never let Obama catch up to her again, but I don't think anyone expected all the media assistance for Obama.

    Parent

    "usual turnout scenario" is kaput (none / 0) (#1)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 10:00:13 AM EST
    And that's an entirely good thing for Democrats and for democracy in general.  It's what many of us have worked toward for decades: trying to increase the electorate's involvement and commitment, trying to remove barriers to that, trying to ensure open and fair elections, and so on.

    But there's much more to be done: many problems remain, such as gerrymandering, attempts at disenfranchisement via "identity card" schemes, the complete unreliability and insecurity of computerized voting systems,  the plethora of caucus systems and their attendant issues, and so on.

    So when I see things like the Prairie View students in Texas marching en masse to the polls, I'm really encouraged.  It's the payoff for all those protests, all those arrests, all those beatings, all those tear-gassings, that so many of us endured.  I hope TX and OH (and RI and VT) are braced for the avalanche of democracy in action.

    (p.s.: The "Subject" line limit calculation appears to be off: error message says it limits at 50 characters, but it's refusing one that I count as 42.)

    Wasn't Kaput in (5.00 / 1) (#5)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:03:46 AM EST
    MA, NJ and CA.

    Parent
    Ah, but it was (5.00 / 1) (#15)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 10:35:35 AM EST
    Quoting from Matthews, Russert ignored record Democratic turnout in suggesting lack of broad appeal:
    In New Jersey, Democratic primary turnout of 1,104,000 was 69 percent higher than the previous record turnout with 99 percent of precincts reporting. In Massachusetts, the turnout of 1,170,000 was 48 percent higher than the previous record turnout, with 98 percent of precincts reporting.

    In California, Hispanic voters set turnout records:

    Hispanic Democrats nearly doubled their percentage from the 2000 and 2004 primaries. And they matched their percentage among California adults, suggesting the state's fast-growing Hispanic population is finally being reflected at the polls.

    When records are being broken, not just by a few percentage points, but by factors like "double", then it's clearly time to rethink the demographic models.


    Parent

    Wait up (none / 0) (#16)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:40:08 AM EST
    Two points, any increase in tunrout NUMBERS mean the old turnout model are obsolete? That is just plain silly. The demographic breakdwon is they key and the total percentage of turnout.

    As for NJ, this was the first meaningful primary in New Jersey EVER! Of course it set records.

    Your point on California is valid, but hardly helpful to Obama. He did not spur Latino turnout, Clinton did. She destroyed him among Latino voters.

    Parent

    Yes, I'm aware of the difference (none / 0) (#19)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 10:55:20 AM EST
    Yes, an increase in turnout numbers means precisely that.  To do otherwise is presume that turnout demographics scale linearly -- which they don't.  (As the CA numbers particularly exemplify.)

    (To illustrate that for those who aren't statistics wonks like me, and maybe BTD: suppose 100,000 people vote and 50,000 are women.  Then, four years later, 110,000 people vote -- a 10% increase.  As a first-order assumption in the absence of contradicting evidence, it's reasonable to assume that the 10% increase applied to women as much as it did to men.  And even if that assumption's wrong, it probably won't be terribly wrong because the percentage increase is relatively small.  But if the number jumps to 150,000, a 50% increase, then that assumption becomes much more dubious.  Oh, it still may be completely correct: but it's on thinner ice.  Note for other stat wonks: I'm fully aware that this is not strictly mathematically accurate --  my master's thesis research was in the area of statistical pattern recognition.  I'm just trying to put this in meaningful terms.)

    As to the impact on either campaign, that's not my concern here.  My point is that the turnout increases observed across the country are so large that the models themselves need to be re-examined to find out if they're still accurate.  Some probably will be, some probably won't.

    Parent

    But it also isn't evidence of the opposite (none / 0) (#22)
    by herb the verb on Sun Mar 02, 2008 at 11:28:07 AM EST
    Just because there is a "predicted" increase in voters (also a prediction based on "models" and past data), conversely doesn't "prove" that an older turnout model gets thrown out the window. I'm a statistics junkie too and would submit that data is data and conjecture is not and we should start the discussion with that mutual understanding.

    Past election history in a state is data, statements like "things are different" is not. Election results (and caucus results) in recent or similar primary states are data, conjecture in newspaper articles or blogs about "nothing being the same" are not.

    Parent

    Oh, I quite agree... (none / 0) (#27)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 11:37:51 AM EST
    ...that we should wait for hard data rather than coming to conclusions based on polling.  (Which I believe is what you're advocating, please correct if I've misunderstood.)

    But I'm basing what I'm saying on observed increases in turnout, not predicted increases.  (See citations upthread for a few states; it's not hard to locate similar items covering others.)  So -- again, unless I'm misreading what you're saying, which I assure you is not intentional -- I think we've got hard data.  We may not know much about what it means yet, but I think we already know enough to question the validity of previous models, to ask whether they still apply or whether they don't.

    Parent

    That says nothing about a new turnout model (none / 0) (#25)
    by Big Tent Democrat on Sun Mar 02, 2008 at 11:33:09 AM EST
    You got something better than that?

    Parent
    I wonder (none / 0) (#30)
    by Kathy on Sun Mar 02, 2008 at 11:45:52 AM EST
    on these turn-out models, do they scale them differently for caucus vs primary?  I know this is Ohio, but it speaks to the greater question about polling.  Are women who caucus, for instance, more likely to go Obama than women who vote in primaries?  

    I suppose you'd have to nail it down to a certain "type," which no poll could really evaluate.  I mean, what do you ask--"when was the last time you had a latte?"

    Parent

    Hmm, maybe my point isn't clear (none / 0) (#33)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 11:53:49 AM EST
    Let me try to put it another way: when you are observing something, and you see observed behavior that is only slightly different from previously-observed behavior, it's usually reasonable to presume that your model for previous behavior -- with perhaps a few tweaks -- will adequately explain current behavior.

    But when you see markedly different behavior, it's time to really put the model under a microscope.  Oh, it may still be valid with a few tweaks, but it may have also just been turned into rubbish. And while there's no way to know for sure until you laboriously step through the exercise, generally speaking, the larger the changes, the greater the probability that the model needs revision.

    In electoral terms, we're observing so many large changes that I strongly suspect that previous models aren't going to work.  I'm sure we'll find out in due course if that's correct or not, but that's my working assumption for the moment.  If I'm reading you correctly, you don't share that assumption.  But soon enough, perhaps by the end of this year, we should have a much better idea which one of us is right.  (We're probably both wrong. ;-) )

    Parent

    what do you mean? (none / 0) (#8)
    by Lil on Sun Mar 02, 2008 at 10:13:52 AM EST


    Parent
    What did you not understand? (none / 0) (#9)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:14:40 AM EST


    Parent
    Maybe it's my (none / 0) (#11)
    by Lil on Sun Mar 02, 2008 at 10:17:33 AM EST
    Sunday morning denseness, but I didn't understand kaput except for Ma., etc.

    Parent
    Did you look at the comment (none / 0) (#12)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:20:24 AM EST
    I responded to? That might clear it up for you.

    Parent
    ok (none / 0) (#13)
    by Lil on Sun Mar 02, 2008 at 10:22:08 AM EST
    I see the subject of original post. Sorry.

    Parent
    No problem (none / 0) (#14)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:22:58 AM EST


    Parent
    Maybe not kaput (none / 0) (#32)
    by muffie on Sun Mar 02, 2008 at 11:49:44 AM EST
    but certainly it shifted from 2004.  For instance, MA had +4% female voters and -7% for voters 65 and older.  In CA, these numbers were closer to the 2004 turnout model (+1% and -3% respectively).

    Anyway, at least it's nice that the article acknowledges what a difference the turnout model makes.  


    Parent

    Certainly good news for Hillary (none / 0) (#3)
    by fladem on Sun Mar 02, 2008 at 10:02:10 AM EST
    though the Ohio Poll is a MAIL poll, not a telephone survey.

    The Columbus Dispatch poll's methodology (4.50 / 2) (#4)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:03:20 AM EST
    has been proven sound in the past.

    I especially like the upfront admittance that the turnout model is the thing.

    Parent

    I should add (5.00 / 1) (#7)
    by Big Tent Democrat on Sun Mar 02, 2008 at 10:08:06 AM EST
    that the mail in model does not change the selection of the representative sample selection. It is no different that the selection made on who to call.

    I believe the argument for mail in is that the response rates are higher. To wit, those who favor it say its methodology say it is better at solving the biggest problem polling now has, very high non responses.

    Parent

    Very different (none / 0) (#28)
    by fladem on Sun Mar 02, 2008 at 11:39:40 AM EST
    dates for this poll as well.  It runs from Feb 21 to yesterday.  

    In the past the Ohio poll has been decent for general election results, in part because its results are captured over a long time period, and thus less likely to capture temporary swings.  This model may not be as good at capturing volatility in a primary.

    In 2004 the last Ohio Poll had Bush up 1 - which was pretty accurate.  BTW- here is something I never had answered by the Ohio was stolen crowd:
    In the last 10 polls taken before Ohio in 2004, Kerry led in only 1 - a USA Today poll.  The 10 poll average comes out to Bush +2.1 - almost exactly what the final results where.  
    http://www.realclearpolitics.com/Presidential_04/oh_polls.html

    Parent

    Dispatch Says Ohioans Think Obama Will Win In Fall (none / 0) (#17)
    by AdrianLesher on Sun Mar 02, 2008 at 10:43:10 AM EST
    This was to me the interesting part of the Dispatch article:

    But there's one undeniable fact among Democrats: They overwhelmingly think that Obama will win the November election, regardless of whom they support in the primary. Even nearly a third of Clinton supporters in the poll expect Obama to be the next president.
    Since the media has been running with (5.00 / 1) (#23)
    by RalphB on Sun Mar 02, 2008 at 11:29:21 AM EST
    the "inevitable" Obama for a few weeks now, I would expect people to think he'll be the nominee.  That will attract voters who want to vote for the winner, but I don't know how many of those are voting this cycle?


    Parent
    Slanted quotes (none / 0) (#18)
    by 1jane on Sun Mar 02, 2008 at 10:52:25 AM EST
    After reading the article BT refers to i must cry foul or hold up the red card. Here is the actual quote.

    "A new Dispatch Poll shows that the New York Senator WOULD lead Ohio by as much as 16 percentage points, UNDER THE USUAL scenario dominated by loyal, older party voters - especially if women show up in droves - THERE IS NO USUAL SCENARIO."

    Further down in the same article:

    "There is an undeniable fact among Democrats. They overwhelmingly think that Obama will win the November election, regardless of whom they support in the primary."

    Reads differently doesn't it?

    Check the primary sources on this site because what you read may have been altered to fit the bias of the post.

    That first quote (none / 0) (#20)
    by lilburro on Sun Mar 02, 2008 at 11:06:13 AM EST
    is not in the article.

    And doesn't the second quote simply get its support because Obama is the frontrunner?

    Parent

    You are suspended for the day (none / 0) (#21)
    by Big Tent Democrat on Sun Mar 02, 2008 at 11:09:42 AM EST
    for lying and insulting me.

    Do not post anymore today.

    Parent

    doesn't look like a power trip to me (none / 0) (#46)
    by tree on Sun Mar 02, 2008 at 05:00:27 PM EST
    The poster put up a bald-face lie and called it a "quote" from the Dispatch, and then had the gall to accuse BTD of misquoting. I'm happy to be at a site that doesn't make you spend all your time refuting garbage, because they delete the worst cases themselves.

    Parent
    I feel Obama has no chance against McCain (none / 0) (#35)
    by diplomatic on Sun Mar 02, 2008 at 11:58:30 AM EST
    Putting together the whole picture of what that general election matchup would be like:

    Barack and Michelle vs McCain and Cindy.

    The media darling vs the GODLY media darling.  
    The war veteran/POW/person who has actually visited Iraq vs the guy with no military experience/less than one term in the senate/never convenes meetings of his subcomittee/never been to Iraq.

    It's frightening.  People need to wake up and just visualize it.  Then visualize how the right wing machine will try to tear Obama apart.  And here's the most troubling part: we have no idea if he's ready to handle that and fight back.  He has not been tested.  He has gotten an easy ready so far.  He seems ashamed of even being called a liberal.  "Supposed liberal" is what he said recently.

    God help us all.

    Way off topic ... (none / 0) (#36)
    by plf1953 on Sun Mar 02, 2008 at 12:03:32 PM EST
    And if you make such categorical statements, please provide evidence or keep them to your self.

    That said, have you got any evidence (and links to it) to support your assertions?

    Please respond to the comment ... (none / 0) (#48)
    by cymro on Sun Mar 02, 2008 at 06:09:53 PM EST
    ... you are addressing. If you comment on the original post, as you did, then you cannot criticize it for being "off topic." It is the topic.

    Parent
    How to tell a patriot (none / 0) (#38)
    by RiderOnTheStorm on Sun Mar 02, 2008 at 12:10:30 PM EST
    It's not whether they sing the national anthem, or say the pledge of allegiance, or wear flag lapel pins, or any of those superficial things.  It's not whether they say the right things or the expedient things or the trendy things or anything like that.  It's not whether they align themselves with other supposed patriots or not.  It's not whether they're veterans or not.  It's not whether they're popular or unpopular.  It's not...a very long list of things.

    It's whether they will uphold and defend the Constitution, and thus by extension, the rule of law.

    So when I'm trying to assess any candidate's patriotism, that's what I look for.  Will they do it, even when it costs them -- even when it costs them dearly, even when it costs us dearly?

    I'm willing to die to uphold and defend the Constitution.  I expect no less of my President.

    You don't mention (none / 0) (#39)
    by along on Sun Mar 02, 2008 at 01:00:25 PM EST
    that the poll excludes independent voters.
    The poll only goes to registered Democrats and Republicans for a primary election.

    The article itself acknowledges the importance of this part of the electorate:

    So Tuesday's election likely comes down to how many independent voters decide to grab a Democratic ballot and vote for Illinois Sen. Barack Obama. Secretary of State Jennifer Brunner's prediction of a record 52 percent turnout means that well more than a quarter of Ohio's 5 million-plus nonpartisan voters will vote.

    According to Mark Blumenthal, all the other pollsters--except for the UofC Ohio Poll--sample "lean Democratic" or "independent" likely voters. That part of the electorate could make up from 17% to 26% of the vote total on Tuesday.

    It would seem that is why the numbers for this poll are so different from other recent reports.

    52% Tjat just seems astronomical (none / 0) (#43)
    by Cream City on Sun Mar 02, 2008 at 01:50:12 PM EST
    In a state often considered comparable to Ohio, turnout was not much above the usual, predicted 35% -- in Wisconsin. And that is usual for Wisconsin in a primary, but it is one of the "most voting" states in the country. (Of course, Ohio is getting much better weather than we had in Wisconsin, where we definitely did not have what we call "Dem weather.")

    Parent
    it's meaningful in one sense. (none / 0) (#41)
    by KevinInNYC on Sun Mar 02, 2008 at 01:32:45 PM EST
    If Ohio Dem voters think Obama is the likely nominee, but are still siding with Clinton in large numbers, it suggests that the party has yet to coalesce around one standard-bearer.  

    The assumption from the beginning had been that all Clinton had going for her was her inevitability factor.  Once that was shattered, however, she still came back to win in NH, NV, and many states on Super Tuesday.  Now, after so many consecutive losses, one would think Obama is poised to run away with things in Tuesday, and he still might.   But I doubt any state will be by a big margin other than VT, and I doubt the margin will be from actual Democrats.  The party's voters are still supporting Clinton in large numbers.

    And the s-d's know that most Dems (none / 0) (#42)
    by Cream City on Sun Mar 02, 2008 at 01:45:29 PM EST
    support Clinton, according to the polls. So the s-d's who are elected by us -- governors, Senators, etc. -- are accountable to Dems and have to consider how many of Obama's independents and moderates would be s-d's voters next time. Or not.

    Parent
    The Democratic Race (none / 0) (#44)
    by VJCMAJD on Sun Mar 02, 2008 at 02:27:16 PM EST
    Maybe the good folks are finally becoming offended at the notion that the election can be bought; and that the person with the biggest fundraising effort [Obama] would likely win.  The whole cycle should be shorter, the criteria for running higher.  He is  squandering millions, while Hillary is getting her message out with far less spending.  We should all be voting based on the candidates' prior body of work, not the flash and trash of these past several months.  Her resume is more substantive, and she's shown she's not afraid to stick her neck out on tough issues.

    OH And TX Polls: Clinton By 16 In Ohio? | 43 comments (43 topical)

    Readers Must Login or Create an Account to Comment