Articles and Analysis


Panagakis: Race a Wild Card for Pollsters

Topics: Barack Obama , Bradley/Wilder , Race

Guest Pollster Nick Panagakis is president of Market Shares Corporation, a marketing and public opinion research firm headquartered in Arlington Heights, Ill.

The National Council on Public Polls analyzed national presidential poll accuracy In 2004. The found that eleven of fifteen pollster margins were off by 0% to only 2% from the winning margin., well within the margin of error.

This year polls are now showing Obama from +5 points to +13 ahead, more variability than four years ago this week. This year a demographic variable appears to be having an acute effect on voting estimates. That variable is race and may explain the difference in poll margins.

I did a "what if" spreadsheet analysis with hypothetically variable percentages of black composition of total voters and corresponding variable percentages of the white/other races of turnout. I set a constant result for the vote for president by African-Americans.

Typically, polls lack sufficiently large samples of black voters to reliably estimate their voting intentions. I derived my estimate from a tracking poll's pooled results: 95% of African-American would vote for Barack Obama, 3% for John McCain and 2% for other candidates. A one or two point disparity from 95% for Obama doesn't make much difference in this analysis. If anyone has better results, please respond to this column.

The key variables are the racial distribution of voters and how white/race voters will vote.

Exit polls in the last few elections have shown African-American varying from 10% to 12% (in 2004) of total voters. My analysis ranged from 10% to a hypothetical 15% to check incremental margin gain.

General voter interest in this election may be so high, black composition could remain at 12%. But if it increases to 13%, that adds a one-point margin gain for Obama. For each one percentage point increase in black composition, the overall margin for Obama increases by one percentage point.

The other variable is how whites and others would vote. The following hypothetical scenarios cover most of the overall margins current polls are showing. They range from +4.8 to +9.4 Obama winning margins.

Assuming McCain is ahead by 7 points among whites/others and with 12% African-Americans of all voters, overall results are a Obama win of +4.8%. At 13% black voter of total turnout this would result in a +6.0 point Obama win.

Assuming McCain ahead by only 5 points among whites/others and with 12% African-American sample composition. This would yield an Obama win of +6.6 points. At 13% black voter composition results are a +7.6 point Obama win.

Assuming McCain down to +3 points among whites/others and with 12% black voter sample composition. This would yield an Obama win of +8.4 points. At 13% black voter composition results in a +9.4 point win for Obama.

In closing this analysis has nothing to do with the Bradley Effect theory. And nothing to do with reverse-Bradley. (Since when can a theory have it both ways?) Pre-election poll versus exit poll or post election analysis examining such variables could have confirmed or denied the effect.



A good discussion. I would suggest that the impact of a highly charged election campaign on turnout will tend to raise turnout among whites as well as blacks. This doesn't mean a Wilder type effect per se, but does point out that one shouldn't look at turnout levels among just one sector.

Ironically, if there is a Wilder effect in this election, it may likely occur where Blacks are hesitant to say they are NOT voting for Obama.

It is interesting that if you take polls that provide adequate demographic information and extrapolate backwards to the top line using the various demographic categories you find that Obama's margin is less than the topline suggests. In the DailyKos R2K it's a three to four point difference on a consistent basis AND it is always on one side of the topline.

Theoretically it should distribute somewhat randomly on either side of the published topline, not consistently reside to one side of the white voter groups confidence levels.

With the exception of R2K/DailyKOS, none of those producing big numbers for Obama publish significant enough demographic data to do the backwards extrapolations.



You're not even addressing the likely larger gain in Hispanic margins for the Democrat this cycle. Increasing from 88% for Kerry to 95% for Obama is not a big deal unless turnout surges enormously. Hispanics may give Obama 20 more points in margin than they gave Kerry, and also surge in number. If that were the case, 1/3 the increase in Hispanic turnout compared to AA turnout would net the same gain for Obama. It's easy to see the Hispanic effect being even larger in terms of the final result.


538 had a very extensive post along these lines in May, examining separately and in combination increases in black, youth, and hispanic turnout, and looking at that impact both nationally and state-by-state using their demographic model.

The numbers are a bit outdated but conceptually well worth a look. Basically, every 10% in increased black voter turnout gained Obama a point in the national numbers; every 25% in youth vote about .9% of a point; and every 25% in hispanic vote about .4%. In what he called the 'best case' scenario the combination of the three increased the EV count by nearly 100 from the 'baseline' of a 49.5-50.5% Obama-McCain split.

Anyway, interesting post, like I said, and well worth the read.


Nick Panagakis:

IN 2004, both the Census survey of voting and the national exit poll estimated that 6% of the vote was Hispanic. But the national exit poll has begun using a broader definition "Hispanic Origin" which is 8%, still lot smaller than 12% black.

Exit Poll: Kerry won Hispanic Origin 53%/44%. Won Hispanics 56%/43%.

I agree. Obama will do much better than Kerry. The Democrat won the Hispanic vote 67%/31% in 2000, 72%/21% in 1996, 61%/25% in 1992, 69%/30% in 1988, and 62%/37% in 1984. None did 20 points better than Kerry. (Any ideas why Bush did so much better than past GOP candidates in 2004?)

As you can tell, there are a far more Hispanics voting Republican than Blacks voting Republicans, 9%-12% in years above. You could start with Florida Cubans. Florida Hispanics accounted for 10.9% of U.S. Hispanics in 2004 in the Census survey.


Vicente Duque:

Thanks to Mr Panagakis and other participants for wonderful discussion on Race, Blacks, Hispanics and the Percentages. Excellent!

Race, Racials and Raciality are of increasing importance but in a positive way, and not only as violence, aggression or hate.

You pollsters, commentators and bloggers, prepare yourselves for many hours analyzing Race and Racials in future elections or in marketing tasks.

Lots of People that look "perfectly" Anglo have a Hispanic last name. Many new friends "confess" to you that they have Latino ancestors or Native American Indian ancestors.

It is estimated that a Big Proportion of Americans, mainly in the South, have Black ancestors. The older your family in America the more likely to have Black genes.

So there may be an antiBradley effect lurking in the shadows. I suspect that many "perfectly looking" Anglos or "Whites" have some ancient family story of persecution, segregation, or bullies against an ancestor and are very antiBigotry and don't buy the racial fears. Things were not easy for Italians and Irish many decades ago.



Vicente Duque




I was talking about a 20 point improvement in the margin, not in the Democrat's support. If support came in like 2000, that would be about a 40 point increase. I think polls have come in around 65% to 35% in Obama's favor this cycle, so that would be a likely improvement of about 20 or more points in margin.

The AA margin is only expected to increase between 5 and 7 points (I think that 95% support is really the top of the possible range). That's only 1/4 to 1/3 the margin improvement among AA voters as Hispanic voters.

So if turnout was exactly the same this year as it was in 2004 among races, Obama would likely gain 1.5 times as much advantage from the change in Hispanics than AA's. Here's where those numbers come from:

African Americans
2004 - 12% of Electorate, Kerry: 88%, Bush: 11%
Margin = 77 points, 9.24 point Kerry AA vote advantage
2008 - 12% of Electorate, Obama: 93%, McCain: 6%
Margin = 87 points, 10.44 point Obama AA vote advantage
AA Net Gain = 1.20 points

2004 - 8% of Electorate, Kerry: 53%, Bush: 44%
Margin = 9 points, 0.72 point Kerry Hispanic vote advantage
2008 - 8% of Electorate, Obama: 65%, McCain: 34%
Margin = 31 points, 2.48 point Obama Hispanic vote advantage
Hispanic Net Gain = 1.76 points

Of course if AA turnout surges, Obama picks up votes at a pace that is almost 3 times faster than the same improvement in turnout among Hispanics.

I expect improvements in both Hispanic and AA voter turnout this cycle, and I might be a little conservative with Obama's Hispanic support.

BTW, you were wondering why Bush did so good with Hispanics in 2004. I think that comes down to three things:

1) Bush was encouraging immigration reform that would have granted citizenship to many illegal aliens.
2) Bush was specifically targeting Hispanic voters with ads and other forms of electioneering.
3) Bush was pushing religious themes that were popular among some Hispanics.

After 2004, there were some very large protests concerning immigration in several areas of the country, and there are a lot of hard-line Republicans pushing for deportation. I believe that this is the primary reason why Hispanics are moving away from supporting Republicans this cycle, plus they aren't doing much of anything to target this group.

Rove was smart enough to understand that if Hispanics went the way of AA's in terms of their virtually monolithic Democratic support, there would be virtually no way that Republicans would ever win another national election without completely retooling their party and their message. The hard-line fools in the Republican party were too caught up being divisive and lashing out at minorities that they failed to see the inevitability of the political power that Hispanics wield, and how much this will grow.


Memo to Bramster and Nick Nick Panagakis :

Thanks a lot for this wonderful discussion and the excellent data you have provided.

Reading the Los Angeles Times first web page, I get the idea that there is a Big Obama Landslide in Preparation. And that Blacks and Latinos are crucial.

See also the LA Times page in Early Voting, October 25.


The Information in this Newspaper can only encourage and enthusiasm the Obamists.

Vicente Duque


Post a comment

Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.