« September 2008 | Main | November 2008 »

New Direction vs. Experience (and an Early Voting Update)

October 31, 2008 7:41 AM

With the economy proving far scarier to voters this Halloween than the threat of terrorism or the wars in Iraq and Afghanistan, likely voters are looking more for new ideas and a new direction than an experienced hand – though it is a close call.

Just fewer than half, 48 percent in the latest ABC News/Washington Post tracking poll, say new ideas and a new direction are more important in their choice for president than strength and experience. Slightly fewer, 43 percent, say the opposite.

That tilts the advantage to Barack Obama, given the sharp difference in vote choice. He wins 90 percent of “new ideas” voters, while 86 percent of “experience” voters go to John McCain. 

The preferences split largely along party lines. Seventy-seven percent of Republicans – the party that sent a war hero and longtime senator into battle for the presidency – say they prefer a strong, experienced candidate. About as many Democrats – whose party nominated a relative newcomer largely on his promise of change – focus on a new direction and new ideas. And independents divide down the middle.

Even with this division, Obama passes muster on experience with a majority of likely voters, his greatest vulnerability earlier in the campaign, and a point on which McCain has tried hard to make progress. Fifty-five percent say Obama’s experienced enough to serve effectively as president, essentially steady since Oct. 11. It cracked 50 percent for the first time on Sept. 29, just after the first presidential debate.

Nonetheless, it’s still a close call in some key groups. Independents, for example, divide by 51-46 percent on whether Obama’s experienced enough for the job.

…and an Early Voting Update

Separately, our latest tracking poll data find that one in six likely voters has done the deed, and as many say they will between now and Election Day. Their preference: Obama by a substantial margin.

Seventeen percent now say they’ve already voted, favoring Obama over John McCain by 59-40 percent. An additional 18 percent say they’ll do so before Tuesday, for a possible total of 35 percent voting early or by absentee ballot. That would be a record by far, well above its levels in 2004 and 2000 alike, 22 and 15 percent, respectively.

Obama’s wide margin among early voters reflects his campaign’s push; likely voters who plan instead to hold off until Election Day divide much more closely, 50-45 percent. And while state-level reports match the high numbers reflected in this poll, it remains to be seen whether long lines at early-voting locations depress some of that anticipated turnout.

Obama’s edge among early voters is particularly broad – better than a 2-1 margin – in the 16 battleground states and eight toss-up states as designated by our Political Unit. And intended early voting, that 35 percent total, peaks in some pro-Obama groups – among blacks and Hispanics, city dwellers and in the West. It’s also high among seniors, a more divided group.

October 31, 2008 | Permalink | User Comments (402) | TrackBack (0)

Leadership and Change

October 30, 2008 8:00 AM

After a year positioning himself as the change candidate, Barack Obama owns the issue. On leadership overall, though, it’s a far closer call.

Likely voters by 60-34 percent in the latest ABC News/Washington Post tracking poll say Obama would do more than McCain to bring needed change to Washington. That’s roughly where they’ve been since March, with the exception of a 51-39 percent result just after the Republican convention, a gain for McCain that didn’t hold.

The result is another of several that conjure up the 1992 election. Bill Clinton held a 25-point lead over George H. W. Bush on who’d do the best job “bringing needed changes,” almost identical to Obama’s 26-point lead today. (As we’ve reported previously, Obama’s also the first Democrat since Clinton to lead on taxes, and the first since Clinton to hold a clear lead in trust to handle the economy – the dominant issue now as then.)

McCain is more competitive on leadership; likely voters divide, 49-46 percent, Obama-McCain, on who’s the stronger leader. That’s better for McCain than Obama’s 56-39 percent lead on this question Oct. 11; but worse than McCain’s best, a 53-40 percent lead back in March. For a candidate who’s campaigned heavily on experience and judgment, McCain must have been looking to do better on leadership in the final week.

Beyond Republicans and conservatives, McCain does best on leadership with evangelical white Protestants (72 percent pick him as the stronger leader than Obama) and rural voters (59 percent). But it’s closer in some other groups, such as whites (53 percent) and men (51 percent).

Views on change are more lopsided; just Republicans, conservatives and evangelical white Protestants pick McCain over Obama as best to bring needed change. Rural voters divide by 47-41 percent, Obama-McCain. Obama preferred on change by 54 percent of whites, and, in the political center, by 58 percent of independents.

…and Hispanics

Separately, we’ve taken another look at Hispanic voters in our tracking poll. As in past years they’re somewhat less engaged politically: While 83 percent of whites and 82 percent of blacks report being registered to vote, that declines to 67 percent of Hispanics. (Some, of course, are not citizens, and therefore ineligible to register.) Moreover, among registereds, 63 percent of blacks and 59 percent of whites are following the election “very closely”; among Hispanics that declines to 45 percent.

Whites account for about three-quarters of likely voters, blacks for about one in 10 and Hispanics for 6 percent – each group essentially matching its share of 2004 turnout, per the network exit poll. In one interesting difference, though, Hispanics are far more likely to identify themselves as first-time voters – 31 percent do so, compared with 19 percent of blacks and 9 percent of whites.

As far as vote preference, using our aggregate data over the past 14 nights for a good sample size, Hispanics favor Obama over McCain by a wide 69-28 percent, much like their Democratic vs. Republican vote for House seats in 2006 (69-30 percent) and similar to Clinton’s 72 percent support from Hispanics in 1996. Al Gore won 62 percent of Hispanics in 2000, John Kerry 58 percent in 2004, though a less-reliable 2004 exit poll figure of 53 percent often is reported.

October 30, 2008 | Permalink | User Comments (33) | TrackBack (0)

Do You Know This Candidate?

October 29, 2008 7:20 AM

Barack Obama and John McCain alike have made strides since June in acquainting likely voters with their positions. The difference: For Obama’s supporters, it looks much more likely to matter.

More than three-quarters of likely voters in out ABC News/Washington Post tracking poll feel they know at least a good amount about the candidates’ positions on the issues, up 18 points for McCain and 21 points for Obama since the general election campaign got underway. That’s part of what campaigns are for.

But the importance likely voters place on those issue positions is not the same. Forty-five percent say the candidates’ positions on the issues are more important than their personal qualities – and they’re a very broadly pro-Obama group, favoring him by 67-30 percent. 

Fifty-three percent, however, say either that the candidates personal qualities are more important, or that the two are equally important (42 and 11 percent, respectively) – and this group favors McCain by 14 points, 55-41 percent.

That may be what McCain’s campaign manager, Rick Davis, had in mind when he told The Washington Post in early September that what mattered in the race wasn’t issues but the voters’ “composite view” of the candidates. From McCain’s perspective that makes sense. From Obama’s, not so much.

Another approach to the data reinforces that conclusion. Among likely voters who feel they know at least a good amount about Obama’s positions, he leads by 56-40 percent. Among those who know at least that much about McCain, it’s a 51-46 percent McCain-Obama race.

That effect also appears among the subset of likely voters who feel they know a “great deal” about each candidate. Those who feel they know a great deal about Obama’s positions support him by a 2-1 margin, 64-31 percent. Those who know a great deal about McCain’s favor him by a smaller 8-point margin, 52-44 percent.

These results point to two conclusions about current strategy. One is that the McCain campaign portrayal of Obama as an unknown quantity may be aimed at breaking down the link between feeling informed about Obama’s positions and supporting him. The other is that Obama’s heavy advertising campaign, including his half-hour network television ads tonight, seems aimed at building up that very same link.

…and a Look at Movables

Another subject of interest, since we’re swimming in data these days, is what we call movable voters. Twelve percent of likely voters overall (using all 12 nights of our tracking poll), they’re those who are either outright undecided (just 2 percent) or who have a preference but haven’t definitely made up their minds (10 percent more).

Who are they? How movable? Here’s a sketch:

- Movables split about evenly between being highly movable - either undecided or saying there’s a “good chance” they’ll change their minds (7 percent) - and those who say it’s “pretty unlikely” they’ll change, 5 percent.

- They split by 37-34 percent between McCain and Obama, with 7 percent saying “neither” and 20 percent no opinion (based on the four nights of tracking data we released last night).

- Highly movables go 31-25 percent McCain-Obama, 13 percent neither, 29 percent no opinion.

- Unlikely movables, by contrast, split 50-49 percent Obama-McCain - more decisive, as you can see.

- Movability peaks in groups less rooted in partisanship: independents (17 percent movable overall, 11 percent highly), moderates (15 percent overall, 9 percent highly), white Catholics (15 percent overall, 8 percent highly) and white mainline Protestants (15 percent overall, 9 percent highly). 

- They’re also less engaged. Just 45 percent of movables are following the campaign very closely, compared with 68 percent of likely voters who’ve definitely decided on a candidate. So if these are folks Obama’s aiming at in his TV specials tonight – well, it’s not at all clear how apt they’ll be to tune in.

October 29, 2008 | Permalink | User Comments (42) | TrackBack (0)

Ready for 2012

October 28, 2008 7:30 AM

For nearly one in 10 likely voters, it’s not a week from Election Day, it’s four years and a week. Their work in 2008 is done.

Those are the 9 percent in the latest ABC News/Washington Post tracking poll who say they’ve already voted, either by early in-person voting or absentee ballot. Their preference: Barack Obama over John McCain, by 60-39 percent.

That leaves 91 percent yet to vote, but more are coming. A total of 34 percent of likely voters intend to vote early, including those who’ve already done so and those who say they will in the next week. This overall early voting group favors Obama over McCain by 59-39 percent, essentially the same as it is among those who’ve gotten it done already.

It’s even more lopsided in the 16 battleground and eight toss-up states, as identified by our Political Unit. In the battlegrounds, those who say they’ve already voted report a 69-30 percent preference for Obama over McCain; it’s about the same, 66-32 percent, when we add in those who intend to vote in the next week.

In the eight toss-ups (Colorado, Florida, Indiana, Missouri, Nevada, North Carolina, Ohio and Virginia) these preferences are 74-25 percent (already voted) and 69-29 percent (including those who plan to do so). (Note, given our sample size there’s a 10-point error margin on this estimate.)

Preferences are far more lopsided than in the 2004 ABC/Post tracking poll: Fifteen percent reported voting early, splitting 52-46 percent in favor of George W. Bush over John Kerry (and at this stage, a week out, 4 percent said they’d already done so). Obama’s mounted a major early-voting effort this year, and it looks to be bearing fruit. (Actual early voting in 2004 was 22 percent, higher than the poll estimate.)

Early voting to date this year peaks among senior citizens – 17 percent say they’ve voted – and Westerners, 14 percent. These also are prominent when we include those who intend to vote early or absentee in the next week: Fifty-seven percent of Westerners, 47 percent of seniors, 41 percent of blacks, 41 percent of urbanites and 39 percent of single women. Apart from seniors, those are particularly strong groups for Obama.

The sharpest differences are regional. Just 8 percent in the Northeast say they either have voted early or plan to do so, jumping to 25 percent in the Midwest, 40 percent in the South and, as noted, 57 percent in the West, where voting rules in many states encourage early participation.

October 28, 2008 | Permalink | User Comments (22) | TrackBack (0)

Politics and Prejudice

October 27, 2008 9:01 AM

With a personally popular African-American candidate leading in the presidential race, our ABC News/Washington Post tracking poll finds a significant drop in the number of likely voters who report feelings of racial prejudice.

In mid-June, asked to “honestly assess yourself,” 32 percent of likely voters said they had at least some feelings of racial prejudice, including roughly equal number of blacks and whites alike, and much like a similar result in 1999. But now it’s down to 15 percent.

Explaining the change is difficult because the results show no real differentiation. Self-reported feelings of prejudice are down across the board, among blacks and whites, among Obama and McCain supporters, and across other groups – partisan, ideological, age, education and locale (urban, suburban or rural) among them.

In June, for instance, 32 percent of whites reported some feelings of racial prejudice, as did 35 percent of blacks. Now it’s 15 percent among whites, 21 percent among blacks. Among Obama supporters self-reported feelings of prejudice have gone from 33 percent  to 16 percent; among McCain’s, from 31 percent to 13 percent. It's down among high- and low-income and high- and low-education whites alike. And in both polls whites who reported no prejudice have had essentially the same vote preference as whites overall.

While any influence of Obama’s candidacy is speculative, he is broadly popular: In a tracking result last week, 64 percent of likely voters expressed an overall favorable opinion of him. That was the same in June. But then Obama and McCain were running evenly, and much has transpired since, from the debates to the deterioration of the economy, both advantages for Obama.

One possibility for the change is a context effect; the question in June was asked after four others on the subject of race relations and discrimination; the question in 1999 was asked after a single question on racial discrimination. With no such lead-in in the tracking poll, it could simply be that respondents had given less thought to the issue. Any way you cut it, it's an interesting result, if also an inconclusive one.

October 27, 2008 in Favorite Posts | Permalink | User Comments (45) | TrackBack (0)

Divided Gov't and McCain's Closing Case

October 26, 2008 8:06 AM

One of John McCain’s closing arguments – as reported by George Stephanopoulos last week – has some resonance among swing-voting independents: resistance to Democratic control of the presidency and Congress alike.

Likely voters overall favor Democratic control of Congress by 8 points, and half in principle support single-party control of both branches. But that masks sharp partisan differences, with independents tilted slightly toward McCain’s view.

Independents by 45-37 percent in the latest ABC News/Washington Post tracking poll say they’d rather see the Republicans than the Democrats win control of Congress. And by 43-34 percent they're a bit more apt to say divided government is the better way to go.

In addition to independents it’s possible McCain’s argument could hold some appeal to movable voters, who likewise are less rooted in partisanship.

In October 2006, more likely voters – 58 percent overall, including 58 percent of independents – preferred to see the Democrats control Congress, which was the outcome of that election, driven by frustration with the war in Iraq.

Preference for control of Congress 
        In 2008      In 2006
       Dem   Rep    Dem   Rep
All    50%   42     58%   38
Dems   94     3     96     3
Inds   37    45     58    32
Reps   10    86     11    87

There’s a sharp gender gap today; women by 56-36 percent prefer Democratic control of Congress while men by 50-43 percent would like to see it go to the Republicans. Whites split evenly; blacks, the Democratic Party’s single most loyal group, almost unanimously favor its continued control.

On the question of whether divided or single-party control is better in general, 50 percent overall side with single-party control – Democrats most, independents least. Thirty percent favor divided government – independents and Republicans most. And 14 percent have no preference either way.

Preference for Congress/presidency 
       Same   Different  No diff.
All     50%      30        14
Dems    63       13        16  
Inds    34       43        17 
Reps    49       37         8

October 26, 2008 | Permalink | User Comments (84) | TrackBack (0)

The Ground War Rages

October 24, 2008 7:06 AM

Three in 10 likely voters say they’ve been contacted by phone, e-mail or in person by Barack Obama’s campaign, rising to four in 10 in the battleground states – in both cases an advantage for Obama over John McCain in the ground game.

Nationally 29 percent report being contacted by phone, in person or via e-mail or text message on behalf of the Obama campaign, vs. 21 percent who report a contact from the McCain camp. In the 16 battleground states, 38 percent report an Obama contact vs. 27 percent from the McCain campaign. And in the 8 tossups states it’s 42 vs. 29 percent.

These are – for both campaigns – enormous numbers of direct contacts. With turnout of perhaps 130 million voters, the equivalent of 38 million report having been contacted to date by the Obama campaign, 27 million by McCain’s – with a week a half yet to go.

Traditionally, contacts closer to Election Day could matter more. But this does not look like a traditional election: As reported in the latest ABC News/Washington Post tracking poll last night, 31 percent of likely voters plan to vote early or by absentee ballot. That might lend more clout to early contacts.

Obama’s efforts seem either better targeted or more effective. He holds a wide 75-23 percent lead in vote choice among likely voters who’ve been contacted by his campaign; among those he’s not contacted, by contrast, it’s a 50-44 percent McCain-Obama race.

McCain’s efforts do not show this differentiation. He trails by 10 points, 52-42 percent, among people he’s contacted, and by about the same, 55-42 percent, among those he has not.

                    Vote preference
                     Obama  McCain
Contacted by Obama
  Yes                 75%     23
  No                  44      50
Contacted by McCain
  Yes                 52%     42
  No                  55      42

In another indication of targeting, Obama has an especially sharp advantage among likely voters under age 40: in this group 30 percent report an Obama contact, while just 11 percent have heard from McCain. And 44 percent of Democrats report an Obama contact, vs. 25 percent of Republicans who’ve been contacted by McCain. It’s closer among independents; 26 percent have heard from Obama, 20 percent from McCain.

Nationally, 21 percent say they’ve heard from the Obama campaign by phone or in person, 16 percent by e-mail or text message (mailings weren’t measured in this poll).  Combining those who reported both kinds of contacts produces the net of 29 percent.

For McCain, 14 percent have heard by phone or in person and 10 percent report receiving campaign e-mails or text messages, 7 and 6 points behind Obama, respectively.

In the 16 states identified by the ABC News Political Unit as battlegrounds, 32 percent report a call or in-person contact from the Obama campaign, 17 percent an e-mail or text message. In these states the comparable numbers for McCain are 23 and 10 percent.

And in the eight toss-up states – Colorado, Florida, Indiana, Missouri, Nevada, North Carolina, Ohio and Virginia – 34 percent report an Obama call, an additional 18 percent an e-mail or text. From McCain, it’s 25 and 10 percent, respectively.

              Total contacts by:
                 Obama  McCain 
National          29%    21%
16 battlegrounds  38     27
8 toss-ups        42     29


                In person/phone
                 Obama  McCain
National          21%    14%
16 battlegrounds  32     23
8 toss-ups        34     25


                  E-mail/text 
                 Obama  McCain
National          16%    10
16 battlegrounds  17     10
8 toss-ups        18     10

October 24, 2008 | Permalink | User Comments (30) | TrackBack (0)

Ideology and Ethics

October 23, 2008 7:22 AM

Amidst the rocks in the road for John McCain is this: His criticisms of Barack Obama in terms of ideology aren’t showing progress, and on ethics Obama’s actually improved.

"Obama – born of the corrupt Chicago political machine” says one recent McCain ad. Nonetheless likely voters by 49-39 percent pick Obama over McCain as having “higher personal and ethical standards.” That’s improved from an even split in an ABC/Post poll at the start of their race last May.

Higher personal and ethical standards   
          Obama   McCain
  Now      49%      39 
  May      41       41

“Obama and congressional liberals - too risky for America," another McCain ad says. In the latest ABC News/Washington Post tracking poll 40 percent of likely voters call Obama too liberal. But that’s essentially unchanged since summer, and a majority, 55 percent, instead says he’s “about right” ideologically. Fewer, meantime, say McCain’s where they’d like him to be – 42 percent.

Part of McCain’s challenge is that he comes in for some criticism on both sides of the ideological spectrum. Thirty-eight percent of likely voters see him as too conservative, much like the 40 percent who call Obama too liberal. But an additional 15 percent see McCain as too liberal, while just 2 percent call Obama too conservative.

         Too      Too     About 
       liberal  conserv.  right
Obama    40%       2       55
McCain   15       38       42

In a can’t-buy-a-break result, 19 percent of conservatives call McCain too liberal, and an additional 11 percent of conservatives say he’s too conservative; that leaves 65 percent who say he’s about right. By contrast, 90 percent of liberals say Obama’s about right ideologically, as do 64 percent of moderates. Just 37 percent of moderates say McCain’s in the right place ideologically.

Given their importance as swing voters, it’s worth noting where independents go on these questions. On who has higher standards, independents divide fairly closely, 45-38 percent, Obama-McCain. And while 56 percent of independents call Obama “about right” ideologically, that slips to 43 percent for McCain.

October 23, 2008 | Permalink | User Comments (44) | TrackBack (0)

Battlegrounds, Reds and Blues

October 22, 2008 7:09 AM

Our latest tracking poll results underscore Barack Obama’s advantage on the electoral map – both in comparison with past years and in terms of this year’s battleground states.

In the 16 states identified by the ABC News Political Unit as battlegrounds, Obama leads John McCain by 54-43 percent among likely voters in our ABC News/Washington Post tracking poll. That’s a party switch from the last two elections: George W. Bush won these same 16 states in 2004 and 2000 alike. And in 1992 and 1996 Bill Clinton won these 16, and the White House.

Cutting it down to the eight states rated by our Political Unit as tossups this year (Colorado, Florida, Indiana, Missouri, Nevada, North Carolina, Ohio and Virginia) Obama has a 52-45 percent advantage.

The past is not predictive, and history is littered with alleged bellwethers that, like all statistical bubbles, eventually popped. But Obama’s aggregate lead in these highly contested states underscores his advantage in national polls and in states alike. (It’s noteworthy, as well, because likely voters in these battleground states are disproportionately white, compared with those in the less highly contested states - 83 percent vs. 72 percent.)

Indeed, Obama’s current 11-point advantage in the 16 battleground states is larger than the margins by which they ultimately were won in any of the last four elections – Bush by 5 points in 2004, Bush by 4 in 2000, Clinton by 6 in 1996 and Clinton by 2 in 1992.

Another comparison to past years also may be instructive. In 2004 George W. Bush won 30 “red” states by a 14-point margin overall, 57-43 percent. Today Obama and McCain are dead even – 49-49 – in those same 30 states. (This excludes Alaska, where we don’t have tracking data.) In the 2004 “blue” states, meanwhile, John Kerry won by 9 points – but Obama’s now ahead by 18 (excluding Hawaii).

It’s a similar story if we jump back to 2000: That year Bush won his red states by 12 points; now it’s essentially tied (McCain +1) in those same states. And while Al Gore won his 2000 blue states by 12 points, Obama now leads in those states – again, by an 18-point margin.

                  2008       Exit polls
                 tracking      Dem-Rep
               Obama-McCain  2004   2000  
08 battleground   54-43      47-52  46-50 

04 red states     49-49      43-57
04 blue states    57-39      54-45

00 red states     48-49             43-55
00 blue states    57-39             54-42 

       

October 22, 2008 | Permalink | User Comments (38) | TrackBack (0)

The Economy... and those Clinton Democrats

October 21, 2008 8:16 AM

Two factors that weren’t fully anticipated early this summer are boosting Barack Obama in the presidential race. One, most powerfully, is the economy; another – surprise – are Clinton Democrats.

Just in time for Hillary Clinton’s appearance with Obama in Orlando last night – the first time they’ve shared a stage since their show of unity in Unity, N.H. back on June 27 - our tracking poll shows Obama with his best support to date among Clinton Democrats, 84 percent.

That’s Obama’s best yet in this group, up from 72 percent in mid-June. At that time one in four Clinton supporters either was going to McCain or undecided; now it’s down to one in 6. That may be as good as it gets for Obama; after all, these are people whose preferred candidate he defeated in a long and hard-fought primary campaign.

We define Clinton Democrats as Democrats and Democratic-leaning independents who wanted her, not Obama, to win the nomination. It’s a substantial group, 20 percent of all likely voters. Some of the most heated debate heading into the Democratic National Convention focused on whether or not they’d come around. Looks like.

On the economy, meanwhile, our latest results show continued high levels of economic unease – with a dramatic effect on vote preferences. Eighty-five percent of likely voters are worried about the economy’s direction the next few years, 44 percent “very” worried; two-thirds are worried about their own family’s finances, one in four very much so.

And boy does it matter. Among people who are “very” worried about the economy, Obama leads McCain by a huge 65-29 percent. Among everyone else, though – be they somewhat worried, or not too worried or not worried at all – McCain leads, 54-43 percent. With no economic worry, in theory at least, there’d be no Obama lead.

The differences aren’t quite so stark on personal finances; here Obama leads big among the very worrieds (by 66 to 27 percent) and less big among somewhat worrieds (by 10 points). Those who aren’t too worried split evenly; McCain leads only among those who aren’t worried about their family finances at all. His problem: They account for just 12 percent of likely voters.

                    Obama   McCain
Clinton Democrats    84%     15

Economy:
Very worried         65      29
Not “very” worried   43      54

Family finances:
Very worried         66      27
Somewhat worried     54      44
No too worried       48      48
Not at all worried   34      63

Check here for more on our latest ABC/Post tracking poll.

October 21, 2008 | Permalink | User Comments (59) | TrackBack (0)

The Powell Endorsement

October 19, 2008 11:38 AM

Endorsements tend to reinforce predispositions rather than change them. Nonetheless Colin Powell's is unusual, in that it both crosses the aisle and comes from a particularly well-liked quasi-political figure - one, as a bonus, who's steeped in the military experience Barack Obama lacks.

A few data points:

-In a Fox News poll in August, registered voters by nearly 2-1 said a Powell endorsement would make them more likely rather than less likely to vote for Obama – 35 percent more likely, 19 percent less so. (“No difference” was not offered as a choice; 43 percent volunteered it anyway.)

-Same poll, 76 percent reported an overall favorable opinion of Powell, 13 percent unfavorable.

-In an ABC/Post poll back in October 1995, 64 percent of Americans said Powell should run for president in 1996. In a head-to-head matchup Powell led Bill Clinton by 10 points, 52 to 42 percent, among registered voters. (Among Republicans and Republican-leaning independents, though, Powell was not the frontrunner in preference for their party's nomination; he ran second to Bob Dole.)

-Same October 1995 poll, 70 percent viewed Powell favorably. And 54 percent rated his leadership abilities as “outstanding” or “above average,” well above his contemporaries (Bush 41 got above-average leadership ratings from 39 percent, Bob Dole 26, Bill Clinton 25). Powell also bettered Ronald Reagan’s leadership rating (43 percent outstanding/above average).

Note, part of Powell’s popularity stems from the fact that he’s never waded as deeply into the political fray as those who battle it out to win and hold elective office. It’s when they engage at that level that public figures are at greatest risk of losing some of their appeal, or at least putting it on the line.

Regardless, beyond his popularity and Republican credentials, Powell's endorsement may resonate for another reason: In making it he criticized John McCain for negative campaigning - a concern that, as I reported here a week ago, the public clearly shares.

October 19, 2008 | Permalink | User Comments (332) | TrackBack (0)

Dissecting the "Bradley Effect"

October 15, 2008 8:20 AM

Barack Obama’s lead over John McCain has reignited debate about the alleged “Bradley effect” – the notion that, in polls, white people lie about voting for a black candidate. It remains, at best, what I called it nine months ago: a theory in search of data.

The Bradley effect and its variants (Wilder effect, Dinkins effect), attempt to explain why the level of support for white candidates was understated, or for black candidates was overstated, in pre-election polls in six biracial contests from 1982 to 1992. That's right - anywhere from 16 to 26 years ago.

Two problems: There are, in fact, beaucoups reasons beyond lying that those polls could have been wrong. And there have been plenty of accurate polls in such contests since.

Consider 2006: Good-quality polls were accurate in five U.S. Senate or gubernatorial elections in which white and African-American candidates faced off. (There was no good, publicly released polling in a sixth). One, in Massachusetts, understated the white candidate by 5 points (not that it mattered in a 55-35 percent blowout); another, in Pennsylvania, understated the black candidate by 5 points (in a 60-40 rout). But as for a consistent Bradley or Wilder effect: It didn’t happen.

Or take the 2008 Democratic primaries: We find 25 good-quality pre-election polls in which Hillary Clinton was understated beyond sampling error, as the Bradley effect posits. But we also see 24 polls in which Obama’s support was understated – the reverse of the Bradley effect. And there were three in which Clinton was overstated, also contrary to the effect. In none was Obama overstated. An additional 21, moreover, were within sampling error on Clinton, as were 23 on Obama.

We also looked at this another way, via the margin between the candidates. In final-week, airworthy polls, Clinton was understated on the margin in five states – but Obama was understated in 10. The gap was close to the final result in seven more.

It seems fair to suggest an effect, to be an effect, should occur with some consistency.

Though not limiting himself to good-quality polls, a Harvard researcher, Daniel J. Hopkins, wrote a paper in August assessing polling data from 133 biracial elections from 1989 to 2006 – “the first large-sample test of the Wilder effect.” Hopkins concludes that there once was an effect; blacks running before 1996 averaged 2.7 percentage points below their poll numbers. “Yet this effect subsequently disappeared,” Hopkins reports. His paper’s title begins: “No More Wilder Effect.”

Beyond arguing general “social desirability” in whites saying they’d vote for a black candidate (and I’m not sure I get that), some theories more specifically suggest that some poll respondents will tailor their stated vote preference to their perception of the interviewer’s race – that whites, for instance, won’t tell a black interviewer they support McCain. Our own, current data tell a different story: We’ve been comparing the race of interviewers and respondents in ABC News/Washington Post polls all year, and we see no such effect.

Among registered voters in a dozen national ABC/Post polls this year, 53 percent of white respondents told white interviewers they supported McCain – as did 52 percent of white respondents speaking with black interviewers. Forty-one percent of whites told white interviewers they supported Obama; an identical 41 percent said the same to black interviewers. And 93 or 94 percent of back respondents backed Obama, regardless of the interviewer’s race. (It’s unclear, moreover, how well respondents can even tell the interviewer’s race in a telephone survey. In one of our recent polls 55 percent identified it correctly, but 23 percent were wrong and 22 percent declined to hazard a guess.)

There is an academic study from 1991 with a contrary finding, saying that whites, when speaking with a white interviewer, were 8 to 11 points more apt to support the white candidate in Doug Wilder’s 1989 Virginia governor’s race. But that was one poll, with just 172 white respondents, carried out by student interviewers – and the report itself footnotes a context effect in the study "that may heighten the race-of-interviewer effect." Our own aggregate data this year, showing no effect, include 7,261 white respondents.

Another concern focuses not on directly lying, but simply ducking the question by declining to state a vote preference at all. In our polls this year, however, the number of respondents who've either declined to answer the horse-race question, or say they have no opinion, is no higher than in past elections, nor do they look different demographically in any way that would suggest that their absence creates bias.

The notion of a Bradley effect is undermined by contemporaneous data as well. As a Pew Research Center memo points out, at the same time that two polls clearly overstated Wilder’s lead in the 1989 Virginia governor’s race, a third gave him a closer, 4-point edge. And while polls in Carol Moseley Braun’s 1992 Senate race understated her white opponent’s support, another, in her primary race, understated her own support, not her white opponents’. Further, while the final WABC/Daily News poll in 1989 overstated David Dinkins’ lead for New York mayor, two weeks earlier it had Dinkins ahead by just 4 points. (He won by 2.) Did New Yorkers suddenly decide to prevaricate?

Consider also a Real Clear Politics piece by Republican pollster Lance Tarrance, who polled in the very 1982 California governor’s race that gave the Bradley effect its name. He calls the concept “a pernicious canard,” and says his own data for Republican George Deukmejian didn’t reflect it. And in a February 1983 report on the Bradley race, California pollster Mervin Field speculated that “racial bias” was a factor, but also offered more concrete reasons for his poll’s error: a heavier than usual and one-sided absentee vote, the presence of a handgun initiative on the ballot and a lower turnout of minority voters than he’d expected.

The turnout issue is critical. When a pollster produces a bad estimate in a pre-election survey, it’s an easy out to blame the respondent. (Visions of New Hampshire are dancing in my head.) But in fact there’s a boatload of ways pre-election polls can go wrong – being done too far from the Election Day, employing poor sampling techniques and, especially, misestimating who’s actually going to vote. It could well be that the absence of a Bradley effect since the mid-’90s isn’t about fewer people lying, but about polls simply producing better likely voter models.

All this leads to the current election. The role of race is a fascinating element, and no one can confidently predict how it’ll play out. One study says polls may be understating Obama by 3 to 4 percentage points nationally. Another, not specific to any reputed polling effect, suggests Obama could be doing as many as 6 points better if it weren’t for latent racism. (I’m skeptical, as usual, and awaiting details on the model.)

Our own study this summer found that there are some whites who are ill-disposed toward a black candidate, but others who are well-disposed to one, and on balance Obama’s support from whites has been around the average for Democratic candidates in the last eight presidential elections. In our latest poll, moreover, 91 percent of registered voters called themselves comfortable with the idea of Obama becoming the first African-American president; 74 percent were "entirely" comfortable with it.

One thing we know for sure is that today’s polls do not predict the election outcome; they simply show where the contest stands today. If the final polls are accurate – as generally they’ve been in the past – a lot of ink will have been spilled on the Bradley effect in vain. If not, there’ll be many places to look. Blaming respondents should not be the first.

October 15, 2008 in Favorite Posts | Permalink | User Comments (69) | TrackBack (0)

Attack Blowback

October 12, 2008 6:02 PM

The McCain campaign’s more aggressive tone is prompting pushback from the public: Registered voters by a broad margin now believe John McCain is more focused on attacking his opponent than on addressing the issues in the 2008 presidential election.

Barack Obama, by contrast, is perceived even more widely as sticking to the issues, this new ABC News/Washington Post poll finds – a striking point of differentiation between the two. More differences will be reported in the full release of this ABC/Post poll on ABCNews.com at 12:01 a.m. and Good Morning America on Monday morning.

While McCain’s image as the more negative of the two is not new, it’s sharpened considerably, coinciding with his campaign’s more pointed criticisms of Obama in the last few weeks, including Sarah Palin's accusation that Obama’s been "palling around with terrorists."

Registered voters by a 24-point margin, 59-35 percent, now say McCain is more focused on attacking his opponent rather than addressing the issues. That’s grown from a roughly even 48-45 percent split on this question in late August.

There's far less criticism of the tone of Obama's campaign: Registered voters by 68-26 percent say he's mainly addressing the issues, not attacking his opponent, a slightly more positive rating than in August.

              Candidate is mainly:
            Addressing    Attacking 
            the issues   his opponent 
Now: 
 McCain        35%           59 
 Obama         68            26 

8/22: 
 McCain        45%           48 
 Obama         64            29 

PARTISAN – It’s noteworthy that Republicans, despite their general antipathy toward Obama, don't broadly see him as running a negative campaign; they divide essentially evenly on the question, 44-46 percent. Democrats, by contrast, overwhelmingly say McCain's going negative, 80-16 percent.

The deciding factor, as ever in presidential politics, is independents. They see McCain as mainly attacking his opponent, by 61-33 percent, but Obama as mainly addressing the issues, by 68 -26 percent.

                 Obama is mainly:       
             Addressing    Attacking   
             the issues   his opponent 
Democrats        87%          12
Independents     68           26
Republicans      46           44

                 McCain is mainly:       
             Addressing    Attacking   
             the issues   his opponent 
Democrats        16%          80
Independents     33           61
Republicans      62           32

GROUPS – Nearly 9 in 10 African-Americans say McCain’s been attacking Obama more than addressing the issues, but likely of greater concern to McCain’s campaign is that 54 percent of whites say so as well. So do a majority of men (57 percent), as well as six in 10 women and nearly two-thirds of moderates.

Among swing voter groups, beyond independents, 58 percent of white Catholics and 59 percent of married women alike say McCain’s been mainly attacking his opponent rather than addressing the issues. So do 56 percent of non-evangelical white Protestants and 57 percent of veterans, a natural affinity group for McCain.

Far fewer in any of these groups say Obama’s been mainly on the attack – a quarter of men and women alike, including married women; fewer than three in 10 whites, white Catholics and non-evangelical white Protestants; two in 10 moderates; and a quarter of veterans. Even among core McCain groups such as conservatives and evangelical white Protestants, more say Obama is mainly addressing the issues than attacking his opponent.

For McCain, finally, among the biggest changes is the result among onetime supporters of Hillary Clinton against Obama – the sought-after Clinton Democrats. In August 59 percent in this group said McCain was mainly attacking Obama, not addressing the issues. Today, 74 percent of Clinton Democrats say so.

Click here for a pdf with details. These results are from a detailed new ABC/Post poll on the election and the mood of the country; for full results check back online at 12:01 a.m., and tomorrow morning on Good Morning America.

October 12, 2008 in Favorite Posts | Permalink | User Comments (373) | TrackBack (0)

Of Markets and Marriages

October 10, 2008 1:25 PM

Polling notes follow on two very different issues in the news: The market and gay marriage.

MARKET - In the past the public has reacted with fortitude to steep declines in the stock market. There are several reasons: While a lot own stocks or stock funds, plenty don’t; those who do tend to be long-term, buy-and-hold investors, many with an investment timeline long enough to wait out a crash; and stock and stock-fund owners tend to be better-off financially, and thus better positioned to absorb a hit.

To some extent, also, a sense of the market’s inherent risk is built into public attitudes. In a poll we did in March 2005, 69 percent called the stock market a risky place to put their money. In July 2002, after a one-week, 10 percent drop in the Dow (and a two-year, 30-percent drop), even more, 80 percent, called it risky. Yet investors invest.

None of this mitigates the real impact of the current turmoil on millions of individual investors, particularly those ready to call on long-built savings for tuition or retirement. Nor does it anticipate the broader economic impact yet to be felt. And while the past adds perspective, it’s not predictive; the current turmoil is virtually unprecedented.

Recent polling indicates that about 45 percent of adults own stocks directly or through mutual funds, excluding retirement accounts. More, 57 percent in a Pew poll in February, reported having retirement accounts such as an IRA or 401(k). In a January poll, 40 percent reported having a traditional pension plan. Fewer directly own stocks, outside of mutual funds – 22 percent in that poll we did in 2002.

And among all stock and stock fund owners, in a poll we did back in 2001, 92 percent called themselves long-term, buy-and-hold investors.

As to recent data: In a Wall Street Journal/NBC poll completed Oct. 5, 49 percent of Americans said they were being personally affected by the Wall Street and mortgage crisis; considerably fewer, 20 percent, said they were being affected “a great deal.”

Broader concern nonetheless clearly is high. In an ABC/Post poll Sept. 22, with the financial crises well underway but not yet at its fever pitch, 79 percent Americans were worried about the economy’s future, 72 percent about the performance of the stock market and 60 percent about their own family’s finances. Again, fewer in each case were “very” worried, 40, 33 and 22 percent, respectively. One obvious question is whether the ongoing crisis has heightened those high-level worries.

Our ongoing ABC News consumer confidence index remained very low this week, -43 on its scale of +100 to -100 last week. It’s been stable, but part of that stability may stem from the fact that it’s computed on a four-week rolling average. We’ll see what happens when a fresh week rolls in next week. At the same time, likely for the same reasons cited above, the index has not responded dramatically to past stock market declines. (And one strongly negative factor on consumer views, gasoline prices, in fact have eased.)

As noted, previous steep market declines were accompanied by concern, but well short of despair. In the July 2002 poll I’ve mentioned, 69 percent said they were concerned by the market’s drop; it was 64 percent in March 2001, and 68 percent after the particularly nasty tumble of October 1987. But in each of these cases far fewer were “very concerned” – 28, 24 and 14 percent, respectively.

MARRIAGE - Vis-à-vis the Connecticut ruling, the latest national polling indicates greater acceptance of gay marriage than in the past, with anywhere from a slight majority opposed, to, in the latest result, an even split.

In a Time magazine poll in August likely voters divided 47-47 percent on gay marriage. That result was a tad odd, since it was a less even 51-42 percent opposed in a Time poll just a month earlier, and 53-44 percent opposed in a CNN poll in late June. All these, though, are closer to the 58-35 percent opposition Time measured in 2004, and our own average of 57-37 percent in polls from 2003-2006.

Support for civil unions has been higher, and also has shifted – in our own data, from 40 percent support in 2003 to a high of 55 percent support in 2007.

Another approach has been to ask which of the three options people prefer – gay marriage, civil unions or no legal recognition of gay relationships. Most nationally favor one of the two forms of legal recognition. In the latest, a Quinnipiac poll in July, 32 percent favored gay marriage, 33 percent civil unions, 29 percent no recognition. Newsweek and CBS polls earlier in the summer had very similar results.

In Connecticut itself, Quinnipiac asked the three-way question in early 2007: Thirty-nine percent favored gay marriage, 33 percent civil unions, 22 percent neither. It hasn’t asked a straight yea/nay on gay marriage in Connecticut since spring 2005, at that time finding 53 percent opposed, 42 percent in favor.

October 10, 2008 | Permalink | User Comments (22) | TrackBack (0)

Pre-Debate: An Epic Snit

October 07, 2008 5:29 PM

A little polling background for the debate tonight:

-Americans are in an epic snit. In a Gallup poll completed Sunday only 9 percent said they’re satisfied with the way things are going in the United States, the fewest since Gallup first asked the question in 1979. Ninety percent are dissatisfied. The only election year since 1984 in which satisfaction was even close to this low was in 1992, but even then it bottomed out higher, 14 percent, and earlier, in June.

-Sixty-nine percent named some aspect of the economy when asked the most important problem facing the country. In our own national ABC/Post poll last week, 53 percent called it the single most important issue in their vote choice, with all other mentions in the single digits.

-Also in Gallup’s data, 25 percent approved of the president’s job performance, a new Bush low for Gallup, while 70 percent disapproved. That’s very much like the 26-70 percent in our ABC/Post poll last week.

-This is not a hold-your-nose election; majorities of Americans have a favorable opinion of McCain and Obama alike – 55 percent and 62 percent respectively in the Gallup poll. That may make it a hard decision for swing voters not rooted in partisanship. (McCain’s favorable rating is the same as Bush’s at about this time in 2004; Obama’s is 10 points better than John Kerry’s.)

-As reported in our last two national polls and our Ohio poll yesterday, economic concerns have turned the dynamic in Obama’s favor; he leads McCain in better understanding the economic problems Americans are having and, by extension, in trust to deal with the issue. Obama’s also built back to a 2-1 lead in trust to “bring needed change” to Washington. But he’s still dogged by the experience question, a continued opening for McCain. McCain, on the other hand, is even more dogged by the suspicion he’d be a Bush retread.

-This dynamic has pushed Obama into leading status in most (not all) national polls, and competitive or leading status in many battleground states. In airworthy state polls just out today, he’s +10 in Pennsylvania, +8 in New Hampshire, +7 in Nevada, +5 in Wisconsin and dead even in North Carolina. Yesterday we ourselves had him +6 in Ohio.

Advantage Obama - but with 19 percent of likely voters still movable.

October 7, 2008 | Permalink | User Comments (39) | TrackBack (0)

Palin's Chance

October 02, 2008 5:18 PM

Our poll report today underscores the world of doubt facing Sarah Palin in advance of tonight’s debate. The table below adds some historical context.

It’s clear that the top of the ticket drives the vote. Nonetheless we do ask people if a vice-presidential pick makes them more or less likely to support the team. In our latest poll, 23 percent say Palin makes them more likely to vote for McCain, 32 percent, less likely – a net negative of 9 points.

There are a couple of ways to look at this. One is that none of the VP candidates we’ve previously measured ever approached a 32 percent “less likely” number. It’s partly mitigated by Palin’s “more likely” result, but it means she’s unusually divisive.

The other is that only one previous vice-presidential nominee in our data ran a net negative in vote effect – Dan Quayle in 1988. Clearly it wasn’t decisive, since George H.W. Bush won that election. But Quayle does not look to have helped. (And Barack Obama certainly hopes he's no Mike Dukakis.)

Other comparisons with Quayle are mixed. He did substantially worse than Palin in personal favorability, but better than in terms of being seen as qualified for the presidency. Not good – just better.

A challenge in all this is that Quayle, in four years as vice president, never turned around those initial doubts about his competence. Tonight is probably Palin's single best chance.

                             More    Less    No 
                            likely  likely  diff.    Net  
9/29/08 
 Obama's choice of Biden      25      13     61      +12
 McCain's choice of Palin     23      32     45       -9 

9/4/08 
 Obama's choice of Biden      22      11     66      +11
 McCain's choice of Palin     25      19     55       +6

7/25/04
 Kerry's choice of Edwards    24       9     66      +15

8/7/00
 Gore's choice of Lieberman   15      10     73       +5

7/29/00
 Bush's choice of Cheney      14       6     78       +8

8/15/96
 Dole's choice of Kemp        18       6     75      +12

9/19/88
 Bush's choice of Quayle       7      19     68      -12

9/18/88
 Dukakis' choice of Bentsen   20       6     71      +14

7/12/84
 Mondale's choice of Ferraro  22      19     49       +3

(Among registered voters except 8/7/00, gen pop)

October 2, 2008 | Permalink | User Comments (70) | TrackBack (0)