« All in All, Kerry Is in a Pretty Good Position | Main | The Political Landscape on the Eve of the Convention (Continued) »

The Political Landscape on the Eve of the Convention

A boatload of interesting polls have just been released which, considered together, give us a sense of how the political landscape lies on the eve of the Democratic convention. Based on both the national picture and on the leanings of key constituencies, the Democrats appear to be in excellent shape, even if much work remains to be done in converting the Democrats' many advantages into a large, durable lead for John Kerry.

Start with the new Pew Research Center poll. Perhaps the most striking findings in the poll concern the dramatically improved issue advantages and image of the Democratic party. Here are the Democrats' leads on which party can a better job on a range of issues: dealing with the economy (+27, up from +7 in mid-2003); protecting the environment (+27); improving the educational system (+16, up from -3 in early 2002); dealing with the economy (+12, up from +3 in fall, 2002); making wise decisions about Iraq (+2, up from -9 in fall, 2002); making wise decisions about foreign policy (+2, up from -10 in fall, 2002); reflecting your views on gun control (+2); improving morality in this country (-2, up from -11 in early 2002 and -23 in early 2001!); coming closest to your views on homosexuality (-2); and dealing with the terrorist threat at home (-15).

Of just as much significance are the results on party image. Five of the six image questions were on positive attributes and the Democrats lead on each one of them: is concerned with the disadvantaged (57-23); is concerned with people like me (50-30); can bring about needed changes (46-35); is able to manage the federal government well (40-37, the first lead Democrats have had on this attribute since mid-1992); and governs in an honest and ethical way (37-34). Only on "is concerned with with business and powerful groups" do the Republicans have an advantage--and a wide one (61-22).

In terms of approval ratings, Bush fares poorly in this poll. His overall approval rating is 46 percent approval/46 percent disapproval, slightly down from their June poll. Slightly down as well in the last month is his rating on the economy, now at 42/52, more evidence that Bush's happy talk on job creation and the allegedly robust economy is convincing no one. His rating on Iraq is a nearly identical 42/53, a slight increase in disapproval over last month. And his rating on "the nation's foreign policy" is actually a bit lower at 40/48 (as recently as January of this year, his rating in this area was a comparatively strong 53/36). Only on terrorist threats does his job rating break into net postive territory (54/40) but this rating too is down from last month and way down from the end of last year.

Pew's trial heat question (which includes Nader-Camejo) gives Kerry-Edwards a small 2 point lead (46-44) over Bush-Cheney among RVs. That includes a 12 point lead for Kerry-Edwards among independents and a 6 point lead in the battleground states.

The Pew data also show that voter interest is running high in this election--signficantly above interest levels in 2000 and 1996 and comparable with 1992--suggesting this will be a relatively high turnout election. And their data indicate that voters are now split on who is going to win the 2004 election, whereas before, regardless of who they personally supported, voters believed by wide margins (40 points in January, by 19 points in May and by 15 points in June) that Bush would prevail.

The wind is shifting and the voters know it!

The new Gallup poll gives Kerry-Edwards a slightly larger lead (4 points) over Bush-Cheney among RVs, with or without Nader-Camejo in the mix. Internals of the horse race question show Kerry-Edwards with a whopping 21 point lead among independents. And, just as in Gallup's last poll, Democrats are now supporting their ticket even more strongly (91-8) than the Republicans are supporting theirs (87-8).

Kerry-Edwards also have a wide 23 point lead in the solid blue states (59-36) and continue to lead in the purple, up-for-grabs states, though by smaller margin (48-44) than in Gallup's last poll.

In addition, the Kerry-Edwards ticket continues to enjoy a substantial advantage in favorability ratings over the Bush-Cheney ticket, though slightly diminished from Gallup's last poll. Kerry's favorability rating is 55 percent favorable/37 percent unfavorable (a +18 net rating), while Bush's is 52/46 (+6). Similarly, Edwards' favorability rating is 52/26 (+26), while Cheney's is 47/43 (+4).

Gallup also asked a series of questions about Kerry vs. Bush on the issues and on personal attributes. It's interesting to match them up, where you can, with the Pew results comparing the Democratic and Republicans parties. By and large, Kerry's advantages on issues and attributes tend to run about 4-10 points behind the Democrats' advantage on similar issues and atributes. Clearly there's room for improvement there for Kerry.

And speaking of room for improvement, I hate to be a broken record on this, but results like these from the Gallup poll continue to bother me. At this point, just 45 percent believe Bush has a clear plan for handling the situation in Iraq, compared to 54 percent who think he does not--a net -9 on the question. But check out the result of the same question for Kerry: 33 percent think he does have a clear plan, compared to 56 percent who think he does not--a net -23 on the question.

Ouch.

Turning to key groups for the Democrats in the upcoming election, today saw the release of not one, but two, major new polls of Hispanics--one from The Washington Post/Univision/Tomas Rivera Policy Institute and the other from the Pew Hispanic Center. And if you're Matthew Dowd, leading Bush-Cheney campaign strategist, who has famously remarked that "As a realistic goal, we have to get somewhere between ... 38 [percent] to 40 percent of the Hispanic vote" in 2004 for the GOP to be successful, these polls are very bad news indeed.

Start with the horse race results. Both polls give Kerry-Edwards a 30 point lead over Bush-Cheney among Hispanic RVs. This is a wider margin than Al Gore had among Hispanics in 2000, when he carried them by 27 points (62-35).

The Washington Post (WP) poll (which was conducted in the 11 states with the highest concentrations of Hispanics) has Kerry-Edwards over Bush-Cheney by 60-30, even with Nader-Camejo included. The Pew Hispanic Center (PHC) poll, which was conducted nationally, has Kerry-Edwards over Bush-Cheney by a very similar 62-32. Note that the Bush-Cheney figures of 30-32 percent aren't anywhere near the 38-40 percent target set by Dowd. And they're not likely to get much nearer since one would expect Hispanic undecideds to break toward the Democratic challenger, not the Republican incumbent.

These results are actually worse for Bush and the Republicans than earlier polls this year by the Democracy Corps and others, which gave Kerry and the Democrats healthy leads but not quite this good. So Hispanic voters, it would appear, are trending against the Republicans.

Dowd, of course, refuses to accept this evidence, offering as a counter that a few small Hispanic subsamples in conventional national polls have showed Bush's support among Hispanics in the 40 percent range. But this doesn't pass the laugh test. These samples of Hispanic voters are not only ridiculously small (perhaps 50 voters or so), but they also suffer from the well-known problem that standard telephone polls make no special efforts (use of the Spanish language, etc.) to secure Hispanics' participation and hence tend to draw more upscale, conservative samples of Hispanics than the specialized efforts discussed here.

Looking at the views of Hispanics, as captured in these polls, it's not hard to see how Kerry-Edwards could have such a commanding lead at this point. In the WP poll, Bush's overall approval rating among Hispanics is 36 percent, with 54 percent disapproval. On the economy--by far Hispanics' top voting issue--Bush's approval rating is worse, a dismal 32 percent approval/60 percent disapproval. And his rating on Iraq is worse still, 29/62. In addition, his rating on immigration is 27/55 and his rating on education is 40/46. Only on the US campaign against terrorism (54/38) does he have a net positive rating.

But even on this issue, where Bush gets his best approval rating, Hispanics still say they prefer Kerry over Bush by 43-35. And they prefer Kerry over Bush on every other issue as well: the economy (53-28); Iraq (45-34); immigration (46-26); and education (51-27). Kerry is also viewed, by 25 points (55-30), as the candidate who would do a better job coping with the main problems the nation faces over the next few years.

In addition, Hispanics give Kerry higher ratings than Bush on "understands the problems of people like you" (Kerry, 53 yes/23 no vs. Bush, 37/55); "can be trusted in a crisis" (53/21 vs. 47/44); and "is a likable person" (69/14 vs. 61/34). And even on "is a strong leader", where Kerry and Bush get about the same number of yes votes, Kerry's net rating is quite a bit higher than Bush's (57/22 vs. 58/36).

On Iraq, contrary to early media reports that Hispanics were especially supportive of the war, the reverse is clearly now true. Hispanics believe that the US is losing the war on terrorism (40-37) and that the war hasn't contributed to the long-term security of the United States (48-44), while the general public still has modest pluralities in the other direction. And Hispanics overwhelmingly believe (63-21) that, considering the costs and benefits to the US, the war with Iraq wasn't worth fighting (the general public is only 53-45 that the war wasn't worth fighting).

Finally, Hispanics in the WP poll give the Democrats a 36 point advantage as the party that has more concern for the Latino community (50-14) and a huge 41 point lead on party ID (66-23).

The results of the PHC poll are generally consistent with the WP poll, though they give the Democrats a smaller (26 point) lead on party ID. The horse race results I've already discussed and Bush's overall approval rating is similar to the WP poll (35/55), as is his rating on Iraq (32/58). The PHC poll also finds that Latinos believe the Bush administration deliberately misled the American public about how big a threat Iraq was to the US (51-35) and the US made the wrong decision, not the right decision, in using military force against Iraq (48-39).

On the Bush tax cuts, the PHC poll finds that only 17 percent believe they have been good for the economy. On health care, 86 percent believe the government should provide health insurance for those who don't have it and, by 59-32, they'd be willing to pay more--either in higher health insurance premiums or taxes--to increase the number of Americans who have health insurance. In fact, Latinos say, by 55-37, they'd be willing to pay higher taxes to support a larger government that provides more services, rather than pay lower taxes and have a smaller government with fewer services.

Sorry, Matthew Dowd, these just don't seem like the kind of voters who are eager to drink the GOP Kool-Aid. In fact, maybe you should take that 30 percent support you're getting right now and be happy you're getting that much.

Tomorrow: black voters and young voters

Comments

Did I miss something? I just checked Real Clear Politics cumulative poll averages and Bush is up slightly in the three way horse race. I guess there really was no bounce or it has disappeared altogether.

Although Kerry leads in purple states, we have roughly a dead-heat between Kerry and Bush (with a modest Bush lead). I remember reading a NY Times article suggesting that Kerry could win the electoral college and Bush could win the popular vote. One of their suggestions was that while California and New York are solid blue, Schwarzennegger and 9/11 could mean Kerry would do worse in those states than Gore.

Personally, I don't care about the horse race score. The numbers that matter are the individual toss up states polling and these issues/feeling questions.

The fact that Bush is under 50% means he is in trouble, period

I believe with the new L.A. Times Poll that Kerry has led in 10 of the last 11 national polls released. The other poll is a tie. The lead is narrow to be sure, but if Kerry can build on that during the convention he is in good shape.

In response to Mr. Bartlett, I think
Real Clear actually has mis-entered the results of the new NBC/WSJ poll, as you'll see if you
click on their link to get the info directly
from NBC/WSJ.

Some of the poll results discussed by Ruy remind me of something that I have noticed before. Ruy discussed two polls that show: (1) a small Kerry lead nationally; (2) both candidates doing equally well within their party; and (3) a big Kerry lead among independants. My thought is, if Kerry has a wide (20+) lead among independants, and has about the same support among Democrats as Bush does among Republicans. shouldnt he be leading by a margin of more then 2 or 3 points? My understanding, gained from reading this site I might add, is that the Democratic Party still enjoys about an 8 point margin in voter identification. Therefore, I would expect a small Kerry lead without giving any consideration to independant voters if the two candidates have equal support among their own party. With independants showing a 20+ lean to Kerry, I would expect a margin of about 7-8 points.

So, my final thought is that polls that show Kerry and Bush enjoying about the same level of support in their respective parties and Kerry with a large lead among independants, yet only a 2-3 point margin on the overall horse race, are either over-counting Republicans, using a very small sampling of independants or both. Or am I wrong?

So, having seen all these polls today and in past weeks, what do all of you EDM readers expect the convention bounce will look like? What does it h a v e to look like to be convincing (and comforting) enough? Matthew Dowd announced this 15 point lead couple weeks ago, no doubt in an expectation game effort. I personally would love a 15 point lead but I don't expect one of these proportions. Rather 11 or 10. Am I too pessimistic?

What will it need for the first "Bush is toast" stories in the press?

Bounce from the convention? I expect nearly zero.

Sidebar: I don't think Dems should even mention the ludicrous number that Matthew Dowd gave for the purported bounce number. The electorate is way too tight for it this year and Dowd is mentioning this two digit "bounce" number purely to raise expectations for the Dems which will be certainly dashed in reality.

Why contribute to the Right Wing Echo Machine and distribute their talking points wider? It's a politically motivated lie and Dowd does not even expect that bounce really, he just wants people (including the media) to repeat it.

Assessing things realistically, there will be a tiny convention bounce, if there is any at all.

Keef.

Easy, Keef. Since this site isn't part of the Right Wing Echo Machine I repeat whatever number I like. And, hey, I even found out myself that Dowd is upping the odds, go figure.

But you're making the exact same thing: deliberately lowering expectations. Of course there will be a convention bounce. And we, including you, will be very disappointed if it is tiny.

Don't worry, we're all grown ups here and won't be scared by Dowd and the Right Wing Echo Machine and any other mean bad pundit.

As far as I understand it, the convention bounce should be to convince the undecideds to go Kerry's way. There are a lot of people who know little about Kerry, and they are waiting to see if he has good positions on the issues they care about. A good convention should answer this type of question -- Kerry should erase that defict he has in the question about a clear plan for Iraq.
Additionally, this appears to be a more hotly contested election than normal (good news for Dems) and the dynamics of a race against an incumbent, and a race of nonincumbents (2000) are totaly different and much less elastic (at least on the part of the incumbent).
The keys for Democrats at this point is to drive up enthusiasm for Kerry/Edwards, drive up the youth vote, make the Blue states completely secure so that we can take the race to the battlegrounds, and even roll back some red states into battleground states, and to continue to press the truth about George W. Bush as an incompetent President -- (R)s might like Bush's 'values' better than Kerry's, but they will still have little enthusiasm in voting for someone who has been such a miserable failure.

Andy:

One possible answer to your question is that most polls are using models of likely voters that favor Republicans. For instance, although more American voters identify with the Democratic Party, polling models may predict a smaller percentage of Democrats showing up to vote than of Republicans. I believe that was the case with most polling models in 2000, when almost everyone except Zogby was predicting a clear Bush victory in their final pre-election polls. If I remember correctly, Zogby explained that his model called for a higher turnout among African Americans than was generally expected. The easiest way to detect this influence is to look at the differences in support among registered and likely voters.

While there's plenty of room for abuse (and self-deception) in setting these percentages, it's not necessarily a nefarious act. To pick an obvious contrast: elderly voters are far more likely to vote than twenty-somethings. So who's more likely to be voting this November? Religious conservatives or African Americans? Rural voters or urban voters?

As for the Democratic Party's 8-point advantage in party identification that shows up in numerous polls, freepers have an explanation for that: The damned pollsters call during the middle of the day, when Republicans are at work and Democratic welfare queens are at home. It's a subtle, nuanced argument, and I wouldn't want to attempt to recapitulate it in full here, for fear of getting a key facet wrong.

USA Today Poll also has some favorable numbers for Kerry.
http://www.usatoday.com/news/politicselections/nation/polls/usatodaypolls.htm

Andy, you are asking the same question that I asked awhile back. I've not seen an explanation. Thank you, deminva, for giving it a shot.

Ruy, sorry to be slightly OT, but has there been any polling done to show who would win if, God forbid, we were to suffer another terrorist attack in the U.S. just before the election? And what are your thoughts? Do you think people would instinctively rally around Bush, as they did in 9/11, or would people conclude he failed to keep us safe and vote him out, as Spain did back in March?

Thanks for the good work!

Is it possible, and that's just a guess, that pollers are trying to adjust any Democratic tilt polls did have in the past (or may have had)? That's perhaps the reason why Zogby was the only one realising the last minute surge for Gore in 2000. He had the courage to drop this artificial weighting.

AMEN Ruy!

JFK has to do more than just tell Americans what we already know - that Bush lied.

He has to outline a plan for getting us out of Bush's Mess-O-Potamian quagmire.

Evem a secret one would be better than what he's done thus far.

I suspect he will.

I hope

To me the key is the Hispanic numbers. That is the entire race right there. Arizona, New Mexico, Nevada. If those three fall it's over for Bush. If the Hispanic vote keeps to these new numbers it won't matter that Kerry is not a very good candidate. Demographics are destiny.

OT: The Pakistan connection
http://www.guardian.co.uk/comment/story/0,3604,1266317,00.html

There is evidence of foreign intelligence backing for the 9/11 hijackers. Why is the US government so keen to cover it up?

Too bad the US media is not picking up on this.

I wonder if the polls have readjusted their expected turnouts for Dems this year, in light of the record fundraising, high turnout in the primaries, etc.

If not, the polls would probably undercount Dem votes even more than they did in 2000.

Has there been any discussion by poll watchers/pollsters on this issue?

Simpka,

I would add Florida to the list of states where Hispanics will be decisive.

Looking at the math on national polls for a second, assume that Democrats are 38% of the population, Republicans are 35%, and 27% are Independent/Other. If we conclude from the polls that each candidate carries his party voyters about 91-9, that leaves Kerry with 34.58+ 3.15=37.73% of the total vote, and Bush with 35.27% (31.85+ 3.42).

That's a Kerry lead of 2.46% before allocating the Independents.

If the Independents break 58-42 for Kerry, he'll receive 15.66 of the total national vote from them, with Bush taking 11.34.

So if each party's voters support their guy with 91% and Kerry takes 58% of the Independents, he should win with 53.39% of the two-party vote, to Bush's 46.61%.

Looked at another way, given Kerry's lead based on the larger Democratic base, Bush would have to get 54.5% of the Independents to win, and I know of no poll which shows Bush with a lead among Independents, and certainly not with a 9-point lead.

What the hell? What does USA Today mean when it says "National Adults"? My initial reaction is to say they're not counting 18-25 (grrr!) but I couldn't figure it out. If not, why the hell are they counting non-voters?

Anyhow, my main point here is, if the polls are weighted for a slightly less Dem response then that's good. It keeps the heat on for Dems to turn out in November if they think everyone will be needed to end the reign of the monster.

Finally, I expect about 7-9 point lead for Kerry post convention for a bounce of about 5-8 points.

I, for one, am not expecting much of a bounce. I think this race is a lot more tightened up pre-conventions than we're accustomed to. I'm thinking 5 points, tops.

I've thought in the past that there was a possibility of Bush winning the popular election but losing the EC. Looking at the polling, however, shows that they're not really getting their base disproportionately riled up sufficiently enough for that to happen.

What's more important polling, RV or likely voters? thanks

Frenchfries, I don't expect to see much of a bounce from the convention either. Pollsters and prognosticators have been telling us all along that the voters are polarized at about 45/45. The polls, for the most part, seem to confirm this...There just doesn't seem to be much room for a large bounce if these numbers are true. I would be happy to see a 6 or 7 point bounce in all the polls. That would make me happy. If 90% of the voters have already decided how can there be much more than that?

Ruy: I'll continue to take you on on this question of Kerry's not having a "clear policy on Iraq." I think it's a very loaded, and therefore not very revealing question. In my experience, "no clear plan" is code for "I don't agree," especially when people are talking about the guy they prefer overall. So when we see Bush with a "lead" on this question, what that may signify is merely the fact (and I think we can probably agree on this) that partisan Republicans agree with their guy more than partisan Democrats agree with theirs. What this result tells me is that nearly all the voters who are planning to vote for Bush agree with his Iraq strategy. While most BUT NOT ALL of Kerry's agree with him (the remainder being mostly peace-at-all-costs types and former Dean-niks). That's not really all that significant, unless those Dems who disagree with Kerry don't vote for him, and I see no sign of that.

I remember reading a NY Times article suggesting that Kerry could win the electoral college and Bush could win the popular vote.

I remember that being bandied about in 2000, because, you know, Gore was just too unpopular to actually be the choice of the American people and would have to sneak in on a technicality.

it's funny how the media's tune changed to "support the rule of law" when the reverse happened.