J.J. (aka Rich Karlgaard from Forbes): The Bradley-effect will return this year!

[Tender Branson]

Does the Bradley Effect Overrate Obama in the Polls?

Rich Karlgaard, Forbes Staff
9/16/2012 @ 6:05PM

Despite an awful previous week for President Obama that included a “day the roof fell in” the president is up 3.1 points in the latest Real Clear Politics average of polls with 51 days to go.

Polls fool, of course, but the Real Clear Politics methodology is very good. RCP’s final poll average in 2008 was accurate. It forecast the 2008 national results within 0.7 of a point. Predict an election within a point and you’re a genius. RCP’s poll averaging methodology is genius.

Nevertheless, I suspect the 2012 polls might be missing something big and therefore be off by a few points.  So here’s a prediction: The day after the election, analysts will be talking about the Bradley Effect.

From Wikipedia:

     [The Bradley Effect] is a theory [which] proposes that some voters will tell pollsters they are undecided or likely to vote for a black candidate, while on election day they vote for the white candidate. It was named after Los Angeles Mayor Tom Bradley, an African-American who lost the 1982 California governor’s race despite being ahead in voter polls going into the elections.”

Why, you ask, would voters lie to pollsters?

     The Bradley effect theory posits that the inaccurate polls were skewed by the phenomenon of social desirability bias.[7][8] Specifically, some white voters give inaccurate polling responses for fear that, by stating their true preference, they will open themselves to criticism of racial motivation. Members of the public may feel under pressure to provide an answer that is deemed to be more publicly acceptable, or ‘politically correct‘.”

The Bradley Effect was observed in dozens of elections throughout the U.S. — from Virginia and North Carolina, to Illinois and California — for about 15 years. It started with George Deukmejian’s surprise win over Tom Bradley in 1982 and operated predictably in American elections until the mid-1990s.

Then it vanished.  Obama’s victory in 2008 bore no traces of it. Obama even slightly reversed the Bradley Effect by outperforming the poll predictions in the 2008 election. The Bradley Effect appeared to be dead in 2008 — and happily gone as America entered a new age of post-racial equality in the realm of electoral politics.

So why do I suspect the Bradley Effect will return in 2012? Two reasons:

1.  How many moderate Republicans and self-described economically conservative independents voted for Obama in 2008, against their core beliefs, in order to facilitate a genuine American moment? Think back to 2008: America elected an African-American president who himself was born when racial segregation was legal in the early 1960s. That was, and is, a huge deal.

Here in Silicon Valley, I know hundreds of red-blooded, fire-breathing, libertarian capitalist entrepreneurs who punched the Obama card in 2008. They did so with great emotion and joy. One was a billionaire entrepreneur and venture capitalist, Marc Andreessen, who invented the web browser and co-founded Netscape. So excited was Andreessen in 2008 that he changed his middle name to Hussein. Eventually, Andreessen came to see Obama, whom he still likes personally, as bad on issues like regulation and entrepreneurial capitalism. Andreessen now supports Romney.

2.  Obama’s most fervent supporters have unwittingly laid the ground for the Bradley Effect’s return. Typical is MSNBC’s Chris Matthews, who claims to hear racial dog whistles in every Republican utterance. To the keen canine ear of Matthews, even the phrase “Chicago politics” is suddenly a racist term in 2012. To the liberal comedian Bill Maher – once a refreshing libertarian – another libertarian, Matt Drudge, is a race baiter. Evidence? Drudge often runs unflattering photos of Obama on the Drudge Report. And Clint Eastwood? His comedy riff at the Republican convention was racist, or at lease racially insensitive. This is how one liberal saw it:

     I was appalled that Eastwood or someone had decided that an empty chair could be used to represent Obama. Addressing the thin air, Eastwood seemed to be talking to an invisible man. Invisible Man is the name of a 1952 novel by Ralph Ellison that is considered a canonical work of African-American literature. The unnamed protagonist, a young African-American male who grows up in the segregated deep south in the early 20th century, believes himself to be figuratively “invisible” to society. People do not see him as a person, as himself, but as all manner of stereotypes of an African-American man. In the course of the novel, the protagonist moves to Harlem in New York City and — continuing the invisibility metaphor — lives underground in an abandoned coal cellar.”

Of course, liberals are correct when they say racism has not yet vanished in America. It never will. There will always be some Americans who would never consider voting for an African American of any party — and that’s shameful. But liberals also tend to vastly overestimate racism. If their fears were true, David Duke would be as big as George Wallace.

story continues ...

http://www.forbes.com/sites/richkarlgaard/2012/09/16/does-the-bradley-effect-overrate-obama-in-the-polls

[Likely Voter]

In coming weeks, we can look forward to all the classics:

- DNC sugar high convention bounce is fading

- Bradley Effect is back

- Pollsters are over-estimating minority vote

- All the undecideds are going to vote for the challenger

- The polls are part of the liberal media bias against Republicans

- The next jobs report will wipe out Obama's lead

- Karl Rove/Koch Bros. have a "secret weapon" commercial that will destroy Obama


And then of course there will be the fine art of cherry-picking. If you look at just the Rasmussen national poll and various other pollsters here and there, then Romney is really winning right now!

[Lief 🗽]

Oh man I hope they finally release the "Whitey" tape!

[wan]

2008 would have been the yr for the Bradley effect. It didn't happen then so why now? Most voters already voted for him 2008 and know him. I'm sure a small percentage exist but not much. Same for the Mormon effect. I personally know people that won't vote for Romney because he's Mormon.  

[technical support]

In coming weeks, we can look forward to all the classics:

- DNC sugar high convention bounce is fading

- Bradley Effect is back

- Pollsters are over-estimating minority vote

- All the undecideds are going to vote for the challenger

- The polls are part of the liberal media bias against Republicans

- The next jobs report will wipe out Obama's lead

- Karl Rove/Koch Bros. have a "secret weapon" commercial that will destroy Obama


And then of course there will be the fine art of cherry-picking. If you look at just the Rasmussen national poll and various other pollsters here and there, then Romney is really winning right now!

this is what the repubs will say when they down and desperate

[Sbane]

2008 would have been the yr for the Bradley effect. It didn't happen then so why now? Most voters already voted for him 2008 and know him. I'm sure a small percentage exist but not much. Same for the Mormon effect. I personally know people that won't vote for Romney because he's Mormon. 

Yeah, if the bradley effect is even real to begin with, we would have seen it in 2008. Now people can genuinely say that they don't approve of his job approval and thus vote against him. There was no bradley effect in 2008 and there won't be one in 2012.

[Franzl]

Can't be J.J., he's still convinced there was a Bradley Effect in 2008. (Or rather, he can't stomach admitting he was wrong.)

[ChrisFromNJ]

I think the opposition to President Obama by the same people who voted for him in 2008 is already baked into the cake and reflected in his lower approval ratings, likability and poll numbers. Anybody who says that there will be a significant Bradley effect this time around, with no evidence to support their conclusion, is probably engaging in wishful thinking.

[J. J.]

First, I think was a Bradley Effect, i.e. people lying to pollster because of race, in the 1980's and 1990's, but it was dwindling over time.  It arguably was there in some statewide races in the 2000's but it was not strong.

Second, the states where I though it would be pronounced, PA, OH, it was not there.  The PA numbers were right on.

Third, there were states where it looked like it happened, IA, where I never would have expected it.  There also seem to be cases where voters lied to the pollsters, but they said they were voting for McCain and voted for Obama.  That seemed to be in AZ and NM, and may be tied to voters of Mexican ancestry.  (A study from Harvard indicated that during the primaries, Obama underpolled with African-American voters.  Basically, they didn't want to look like they were voting for Obama just because he was black.)

Fourth, in trying to see if there were more states where Obama underpolled, there is the problem that a several large states were not polled.  I had found one poll within a month of the election in NY.  Obama overpolled and overpolled outside of the MOE.  The poll was not in a week of the election, and it was not a major polling firm.  Was that because of the electorate shifting toward the end?  Was it because the methodology of poll was bad?  Was it because a significant number of people didn't want the pollster to think they were voting for McCain because Obama was black?

The only thing that we can say is that Obama overpolled on some normally good nation polls, like Gallup, but didn't on some others, Rasmussen and PPP.  (And 2008 made PPP.)

[Landslide Lyndon]

First, I think was a Bradley Effect, i.e. people lying to pollster because of race, in the 1980's and 1990's, but it was dwindling over time.  It arguably was there in some statewide races in the 2000's but it was not strong.

Second, the states where I though it would be pronounced, PA, OH, it was not there.  The PA numbers were right on.

Third, there were states where it looked like it happened, IA, where I never would have expected it.  There also seem to be cases where voters lied to the pollsters, but they said they were voting for McCain and voted for Obama.  That seemed to be in AZ and NM, and may be tied to voters of Mexican ancestry.  (A study from Harvard indicated that during the primaries, Obama underpolled with African-American voters.  Basically, they didn't want to look like they were voting for Obama just because he was black.)

Fourth, in trying to see if there were more states where Obama underpolled, there is the problem that a several large states were not polled.  I had found one poll within a month of the election in NY.  Obama overpolled and overpolled outside of the MOE.  The poll was not in a week of the election, and it was not a major polling firm.  Was that because of the electorate shifting toward the end?  Was it because the methodology of poll was bad?  Was it because a significant number of people didn't want the pollster to think they were voting for McCain because Obama was black?

The only thing that we can say is that Obama overpolled on some normally good nation polls, like Gallup, but didn't on some others, Rasmussen and PPP.  (And 2008 made PPP.)

In other words, you suck.

[milhouse24]

Liberals have been saying for months that Romney should be beating Obama because the "economy sucks" 

The Liberals are right, Romney should be winning. 

Romney's losing because he's a Mormon and the liberal media hates Mormons. 

Obama's winning because he has the "stature of the presidency" and voters will tell pollsters they lean support towards the President.  Voters also don't want to "give up on the PResident" or appear racist, just because he is black. 

Carter was completely inept with disasters in both foreign policy and national economic policy. 
GHWB was a terrible politician/campaigner who couldn't communicate with voters or appear responsive to the national economic recession. 

Obama is a good speaker who connects with voters, and he staved off a Depression, but can't deliver a recovery.  Voters just have not been quick enough to bail on him yet.  The Liberal Media is helping to keep Obama up in the polls (because he's black superman!), lol. 

[Politico]

It's not so much fear of being labeled a racist as fear of being labeled a bigot now (e.g., Obama and Co.'s clever wedge on gay marriage; fortunately, Biden The Buffoon blew up the grand plan back in May, somewhat diminishing the intended effect...if you guys lose by a 2000 margin, you ought to blame Joe Biden).

The electorate is so polarized, with the left far more whiny than the right when it comes to certain social issues (which the electorate is not overwhelmingly concerned with right now), that I suspect a lot of Romney supporters are closeted about their support. There is a sizable chunk of people who hate Romney, hate successful businessmen like Romney, hate many of Romney's positions, hate Romney's religion, but will nonetheless mark the ballot for Romney without ever telling a soul about it. Why? In a nutshell, because the next four years cannot look like the past four years. People realize there is a chance of change with Romney whereas there is nothing new with four more years of Obama. In your heart, you know I'm right...

[Okay, maybe Mike Johnson is a competent parliamentarian.]

There is indeed a chance of change with Romney. There would also be a chance of change with the reanimated corpse of J. Pierpont Morgan. There are worse courses to be on than the current one.

[I spent the winter writing songs about getting better]

First, I think was a Bradley Effect, i.e. people lying to pollster because of race, in the 1980's and 1990's, but it was dwindling over time.  It arguably was there in some statewide races in the 2000's but it was not strong.

Second, the states where I though it would be pronounced, PA, OH, it was not there.  The PA numbers were right on.

Third, there were states where it looked like it happened, IA, where I never would have expected it.  There also seem to be cases where voters lied to the pollsters, but they said they were voting for McCain and voted for Obama.  That seemed to be in AZ and NM, and may be tied to voters of Mexican ancestry.  (A study from Harvard indicated that during the primaries, Obama underpolled with African-American voters.  Basically, they didn't want to look like they were voting for Obama just because he was black.)

Fourth, in trying to see if there were more states where Obama underpolled, there is the problem that a several large states were not polled.  I had found one poll within a month of the election in NY.  Obama overpolled and overpolled outside of the MOE.  The poll was not in a week of the election, and it was not a major polling firm.  Was that because of the electorate shifting toward the end?  Was it because the methodology of poll was bad?  Was it because a significant number of people didn't want the pollster to think they were voting for McCain because Obama was black?

The only thing that we can say is that Obama overpolled on some normally good nation polls, like Gallup, but didn't on some others, Rasmussen and PPP.  (And 2008 made PPP.)

Summarized: If a black candidate ever overpolls, it must be assumed that the reason is the Bradley Effect and not any other plausible explanation (even it simply being a bad poll), even if there is plenty of evidence toward a different explanation.

[Comrade Funk]

Liberals have been saying for months that Romney should be beating Obama because the "economy sucks" 

The Liberals are right, Romney should be winning. 

Romney's losing because he's a Mormon and the liberal media hates Mormons. 

Obama's winning because he has the "stature of the presidency" and voters will tell pollsters they lean support towards the President.  Voters also don't want to "give up on the PResident" or appear racist, just because he is black. 

Carter was completely inept with disasters in both foreign policy and national economic policy. 
GHWB was a terrible politician/campaigner who couldn't communicate with voters or appear responsive to the national economic recession. 

Obama is a good speaker who connects with voters, and he staved off a Depression, but can't deliver a recovery.  Voters just have not been quick enough to bail on him yet.  The Liberal Media is helping to keep Obama up in the polls (because he's black superman!), lol. 

lol

[SUSAN CRUSHBONE]

[bullinks]

Romney's losing because he's a Mormon and the liberal media hates Mormons. 

[some other bullinks]

Give me just one example of the "liberal media" attacking Romney because of his religion.

[J. J.]

First, I think was a Bradley Effect, i.e. people lying to pollster because of race, in the 1980's and 1990's, but it was dwindling over time.  It arguably was there in some statewide races in the 2000's but it was not strong.

Second, the states where I though it would be pronounced, PA, OH, it was not there.  The PA numbers were right on.

Third, there were states where it looked like it happened, IA, where I never would have expected it.  There also seem to be cases where voters lied to the pollsters, but they said they were voting for McCain and voted for Obama.  That seemed to be in AZ and NM, and may be tied to voters of Mexican ancestry.  (A study from Harvard indicated that during the primaries, Obama underpolled with African-American voters.  Basically, they didn't want to look like they were voting for Obama just because he was black.)

Fourth, in trying to see if there were more states where Obama underpolled, there is the problem that a several large states were not polled.  I had found one poll within a month of the election in NY.  Obama overpolled and overpolled outside of the MOE.  The poll was not in a week of the election, and it was not a major polling firm.  Was that because of the electorate shifting toward the end?  Was it because the methodology of poll was bad?  Was it because a significant number of people didn't want the pollster to think they were voting for McCain because Obama was black?

The only thing that we can say is that Obama overpolled on some normally good nation polls, like Gallup, but didn't on some others, Rasmussen and PPP.  (And 2008 made PPP.)

In other words, you suck.

No, in other words Obama sucks.

[J. J.]

First, I think was a Bradley Effect, i.e. people lying to pollster because of race, in the 1980's and 1990's, but it was dwindling over time.  It arguably was there in some statewide races in the 2000's but it was not strong.

Second, the states where I though it would be pronounced, PA, OH, it was not there.  The PA numbers were right on.

Third, there were states where it looked like it happened, IA, where I never would have expected it.  There also seem to be cases where voters lied to the pollsters, but they said they were voting for McCain and voted for Obama.  That seemed to be in AZ and NM, and may be tied to voters of Mexican ancestry.  (A study from Harvard indicated that during the primaries, Obama underpolled with African-American voters.  Basically, they didn't want to look like they were voting for Obama just because he was black.)

Fourth, in trying to see if there were more states where Obama underpolled, there is the problem that a several large states were not polled.  I had found one poll within a month of the election in NY.  Obama overpolled and overpolled outside of the MOE.  The poll was not in a week of the election, and it was not a major polling firm.  Was that because of the electorate shifting toward the end?  Was it because the methodology of poll was bad?  Was it because a significant number of people didn't want the pollster to think they were voting for McCain because Obama was black?

The only thing that we can say is that Obama overpolled on some normally good nation polls, like Gallup, but didn't on some others, Rasmussen and PPP.  (And 2008 made PPP.)

Summarized: If a black candidate ever overpolls, it must be assumed that the reason is the Bradley Effect and not any other plausible explanation (even it simply being a bad poll), even if there is plenty of evidence toward a different explanation.

BRTD, did you even bother to read the bolded parts?

[I spent the winter writing songs about getting better]

The first bolded part is a complete myth, Bradley led by double digits...a few weeks before the election. The election was obviously tightening and he had only a single point lead in the final poll. Furthermore Bradley actually won amongst voters that day, he was defeated by absentees, and it's easy to see that absentee polling in 1982 wouldn't have been too good. It's possible in the 1982 Gubernatorial race there was nothing wrong with the polls at all, and Bradley was simply done in by a flukish surge against him caused by an NRA-backed absentee GOTV campaign against a defeated gun control measure on the ballot.

The second bolded part is an answer in search of a question, New Mexico's quirkiness and unpredictableness in elections is notorious, and it's not too hard to see why undecideds might have a swing toward McCain in the last few days in Arizona. It is absurd to assume that any polls being off is due to voters lying and ignoring all other possible explanations. Furthermore it's quite obvious if you look at the polls and actual results that once again in typical fashion J. J. is cherry-picking data to fit his narrative:

Arizona:
https://uselectionatlas.org/POLLS/PRESIDENT/2008/pollsa.php?fips=4
https://uselectionatlas.org/RESULTS/state.php?year=2008&fips=4&f=1&off=0&elect=0

New Mexico:
https://uselectionatlas.org/POLLS/PRESIDENT/2008/pollsa.php?fips=35
https://uselectionatlas.org/RESULTS/state.php?year=2008&fips=35&f=1&off=0&elect=0

Now there's a very key column in the polling data that J. J. always ignores. It's third from the right.

The last point can be easily explained by reading this: http://en.wikipedia.org/wiki/Margin_of_error

[J. J.]

The first bolded part is a complete myth, Bradley led by double digits...a few weeks before the election. The election was obviously tightening and he had only a single point lead in the final poll. Furthermore Bradley actually won amongst voters that day, he was defeated by absentees, and it's easy to see that absentee polling in 1982 wouldn't have been too good. It's possible in the 1982 Gubernatorial race there was nothing wrong with the polls at all, and Bradley was simply done in by a flukish surge against him caused by an NRA-backed absentee GOTV campaign against a defeated gun control measure on the ballot.

And it was repeated with Wilder in 1989, though it was a bit under 10 points.

You must be logged in to read this quote.

Now, I don't recall MV and CO, but I don't think there were any late polls in two states with a relatively high percentage of Mexican ancestry population, CA and TX.  It isn't that I wouldn't want to look there, but that nobody was looking there.  I'd love to see a breakdown by ethnicity, but we don't have it.

Well, we have the national poll that overcounted Obama's support, Gallup that was well outside of the MOE.  Was that a problem with Gallup (the first one 60 years) or something else, like people lying to pollsters. 

Ironically, David Axelrod agrees with me.

[Lief 🗽]

Gallup was also wildly off in 2010. Gallup just isn't a very good polling firm anymore.

[J. J.]

Gallup was also wildly off in 2010. Gallup just isn't a very good polling firm anymore.

And that is a possibility, but it was first election since 1948 where they were out of the MOE.

How were they that much off in 2010?

[Lief 🗽]

Republicans won the House popular vote by 6.6%. Gallup said they were going to win by 15%.

[J. J.]

Republicans won the House popular vote by 6.6%. Gallup said they were going to win by 15%.

I'm not too sure that would be comparable, and, even if it would, it would show a bias opposite to that of the 2008 presidential election numbers.