Routledge 2008 Election Review

I’ve been thinking a lot about the Bradley Effect lately, mainly because commentators won’t shut up about it.  The Bradley Effect, named for the late Los Angeles mayor Tom Bradley, describes the phenomenon of a black candidate underperforming his or her poll predictions on Election Day.  Presumably, non-black voters tell pollsters that they will vote for the black candidate, and when they actually vote, they don’t.

This summer, Harvard post-doctoral fellow Daniel Hopkins released a paper declaring the Bradley effect dead.  Hopkins compared poll prediction and actual results for every U.S. Senate and gubernatorial contest featuring one black candidate from 1989 to 2006.  He finds a Bradley Effect through 1995.  After that point, he argues that the Bradley effect all but disappeared.

For the most part, I agree with Hopkins, and cite his paper whenever asked.  Two years ago, after Harold Ford’s performance in the Tennessee Senate race was within the margin of his internal polls, I began to think that the Bradley Effect might be a relic.  I was glad to have proof that my hunch might be correct.

In the past few days, though, I have been thinking about hedging my bets.  I still think that this race is Obama’s to lose, and I doubt that there will be a Bradley Effect.  However, the more I think about Hopkins’ argument, the less I’m satisfied with it.

Hopkins is constrained in his analysis by a small sample size, though he controls for that in his statistical analysis.  In the past 20 years, there have been only 20 general election contests featuring a black candidate.  One of those races had no available polling data, and the other race pitted two black candidates against each other (Alan Keyes and Barack Obama).  That left only 18 gubernatorial and senate elections for him to study:  7 in the early period (pre-1996) and 11 in the latter period (1996 and later).

When you have a sample size that small, it is easy to look at each race individually, as well as in the aggregate.  This helps to put flesh on the story, and it raises important new research questions.

Of the eighteen races examined, eight of them turned out to be blowouts, with the non-black candidate winning by margins of 30 points or more.  The black candidate won in three of these races (Doug Wilder, Carol Moseley Braun and Deval Patrick).  The other seven were within 20 points (Three of the seven were within 10 points).  That means that only a third of the races were even marginally competitive.

Now, I am not suggesting that Hopkins should have excluded uncompetitive races from his sample.  He rightly included studied universe of available races.  However, it may be that the Bradley Effect is an artifact of competitive elections only.  It seems as though the clear losers were almost equally as likely to underperform as overperform (Incidentally, there are roughly proportional numbers of competitive and uncompetitve elections in the pre- and post-1996 period).  When you think about people feeling pressured to indicate support for a black candidate, it makes sense that people would feel greater pressure if that candidate could actually win, not if they are a long shot.  That raises new questions about how to operationalize competitiveness, but it would not be unreasonable to consider media coverage or fundraising as covariates as proxies for competitiveness, among others.

The thing that bugs me most about Hopkins’ analysis is his explanation for why 1996 is so pivotal.  He contends that because of the passage of the 1996 Welfare Reform Act (One could throw in the Crime Bill for good measure, too), racial issues became less salient on a national level. With racial issues no longer being as prominent, Hopkins contends that whites were more willing to follow through on professed support for black candidates.

That is an unsatisfactory argument.  Almost all of the races in Hopkins sample, whether they were running before or after 1996, featured deracialized black candidates, or candidates who deliberately did not run on explicit civil rights platforms or use racial appeals to blacks and who ran away from race wherever possible.  For instance, Doug Wilder ran as a pro-law and order, pro-death penalty, fiscal conservative.  These candidates that are in this sample weren’t wrapping themselves in the cloak of black nationalism to try to get elected and certainly weren’t making racial issues salient to their campaigns, so to suggest that those macro occurrences were so pivotal may not be accurate.

Moreover, the passage of the Welfare Reform Act did not prevent these candidates’ opponents from attempting to flame racial tensions after 1996.  There are vivid examples of race-baiting, both successful and unsuccessful, before and after 1996.  For every Jesse Helms “hands” ad, there’s an RNC bimbo ad.

Let me be clear:  I do not dispute Hopkins’ general findings:  the pre-1996 candidate underperformed their polls, and the post-1996 candidates did not.  There is such a thing as front-runner decline, and it affects black and white candidates alike.  People are more racially tolerant now than ever before.  They are also more forthright in surveys about all kinds of things, like infidelity.  Being honest about race may be less of a stretch now.  This all suggests the end of the Bradley Effect.  However, given the small sample size, it is reasonable to warn that this paper will not be the last word on the subject.  As more blacks run for high office and we can collect more data, we may be forced to come to different conclusions about the role race plays in predicting black candidates’ chances for success.

Andra Gillespie

As we draw closer to November 4, campaigns, journalists and pundits alike are rightfully paying more attention to voter registration and GOTV (get-out-the-vote) efforts.  There is little argument that interest in this historic election has generated more new voters than ever before.  Along with the record voter registration levels, though, comes greater scrutiny about whether those voters are in fact qualified to vote.

So far this election season, there have been reports of plans to use foreclosure lists to disqualify voters evicted from their homes who did not update their voter registration records.  And this week, the activist group ACORN has come under fire for alleged voter fraud across the country.  ACORN workers allegedly registered fictitious voters; in Nevada, for instance, officials are investigating whether ACORN works registered the starting lineup of the Dallas Cowboys.

The most partisan among us will use an investigation like this to cry foul, this time against the Democrats, who would likely benefit from ACORN’s registration efforts.  However, before we rush to judgment that ACORN deliberately orchestrated a fraudulent voter registration effort, we must consider the role that casual labor still plays in doing GOTV and how casual labor can lead to systemic vulnerabilities that enable registration fraud.

Given all of the interest in this election, it is hard to believe that there could potentially be a dearth of volunteers, either in the campaigns or among surrogate groups and 527 organizations, to register voters and to knock on doors to remind people to vote.  But there very well could be a dearth of volunteers for some of these efforts.  As such, it is not unusual at all for organizations to rely on paid labor.

When paid workers do their jobs, the fact that they’re being paid is irrelevant.  However, the risk with casual labor is that they may be more committed to getting paid than to doing their job.  If this is true (and we don’t know yet if that is applicable in this case), then it is not out of the realm of possibility that these workers could have pretended to work by fabricating voter registration cards.

I first started observing and participating in voter outreach efforts for research in 2000.  Since then, I have had the privilege of witnessing GOTV efforts at the local and national level, with partisan and non-partisan groups, in eight states.  I have seen seamless GOTV efforts where the workers were committed to reaching voters; and I have witnessed haphazard outreach efforts that were plagued by apathetic staffers or outright fraud.

The types of fraud that I have witnessed provide insight into what may have (allegedly) happened to ACORN this year.  It is not inconceivable for a worker to report to work, hide out for a few hours, and then report back at the end of the day with a stack of forged registration cards and demands to be paid. I very distinctly remember working on a campaign where I was asked to make follow-up phone calls to voters who were canvassed one Election Day.  My task was to call everyone who had not been home when the paid canvassers visited.  I prepared to call voters on one stack of walk lists when I noticed that the canvassers had reported that every voter had supposedly been home—in the middle of a work day!  I pointed this out to a supervisor, who confided that he had been sending drunk canvassers home all day and wasn’t at all surprised that some people had misrepresented their work.

I wish I could say that that story was unusual.  Unfortunately, it is not.  In my doctoral dissertation, I reported the results of a voting experiment my colleague David Nickerson conducted with ACORN, YouthVote, and the Pew Charitable Trusts in St. Louis during the 2002 general election (I should note that I was not a part of the data collection team).  During this election, ACORN and YouthVote recruited canvassers to remind voters to vote in the weeks before the election.  As the canvassing operation got underway, ACORN and YouthVote officials discovered that many workers were just inappropriate.  They had to fire most of the workers they initially hired.  Not surprisingly, this effort did little to increase voter turnout.

Moreover, in her book Activism Inc. (Stanford University Press 2006), Dana Fisher tracks the efforts of college students hired by third party organizations to canvass for interest groups during summer vacation.  While this type of canvassing is different from electoral canvassing, Fisher’s findings have applications for explicitly political canvassing.  She found that these students often knew very little about the causes for which they were canvassing.  Turnover was high.  Moreover, workers were under pressure to produce results (in this case, usually petition signatures or donations).

The examples above only highlight the perils of casual labor.  It is impossible to run most large-scale campaigns without paying some people to register and canvass voters.  However, campaigns and interest groups must take care to monitor all their workers.  Without proper supervision (Let me be clear, I am not accusing ACORN of failing to supervise their workers), any worker is capable of making things up and doing damage to an otherwise noble election effort.

Andra Gillespie

 At this stage in a presidential campaign, with just a month to go, most lame duck presidents have quietly faded from center stage.  The focus should rightly be on the two presidential candidates (and running mates) who want to replace him.  But not this time around.  I cannot remember when we have had this unusual of a political circumstance. 

Out of the blue, just seven weeks before the election, comes a policy crisis of such momentous proportion that it cannot be ignored and could lead to the end of the careers of some lawmakers.  Perhaps this is George Bush’s lasting revenge.  The signs were there–subprime meltdown, Bear Stearns, IndyMac, anemic economic stimulus plan.  But Bush’s administration dallied while the financial foundations shook.  Then Congress is handed the nearly politically impossible task of bailing out the credit market, facing angry voters who only see Wall Street greed and manipulation.  What momentary euphoria Republicans may have had immediately following the St. Paul convention had to be dashed by the financial meltdown. 

Ironically, the Bush administration (or at least Henry Paulson) will get some grudging credit for its bold rescue plan;  it will also get deserved blame for being asleep at the regulatory switch.  This is all little help to McCain;  less help to Republicans trying to return to Congress. 

Dennis W. Johnson

As the author of an American Government text, American Government: Political Change and Institutional Development (4th ed., Routledge, 2007, 5th edition available fall 2009), designed for use by college freshmen and sophomores, I treat the “Bush Doctrine” comprehensively in the chapter of the book dealing with the place of the U.S. in the world. After all, the “Bush Doctrine,” announced in September 2002 and formally titled “The National Security Strategy of the United States,” is the official statement of how the U.S. will act in the world. At its announcement six years ago, the “Bush Doctrine” that the U.S. would act preemptively if threatened, with “coalitions of the willing” if formal institutions like the U.N. were unwilling or too slow, and that the U.S. would actively seek to maintain global military supremacy, was widely recognized as a dramatic change in U.S. national security strategy. I expect my college freshmen to know that; I certainly expect a candidate for vice president of the United States to know that. It seems clear that Sarah Palin was unaware of the Bush Doctrine, and was certainly unaware of the more formal document, “The National Security Strategy of the United States,” when Charlie Gibson of ABC News asked her about them in her first national press interview on September 11. This suggests that to say that Sarah Palin faces a steep learning curve is something of an understatement.

The great danger with a President McCain is the very real potential for war with Iran. McCain is determined that Iran will not have nuclear weapons and is willing to use military force to make sure this does not happen. Iran is determined to have nuclear weapons to balance the Israeli nuclear arsenal. There is no way to reconcile these two opposing objectives. If McCain becomes President the American people should prepare for war with Iran. In my view this would be an enormous mistake. The United States is already fighting two wars in the Middle East. If it enters into a third war, it might well find itself at war with the entire Islamic world, over a billion people. To be sure, there are some people who would like to see this. The neocons, who pushed for the war in Iraq, have not gone away. They have been quietly lobbying the Bush Administration for an attack on Iran (see the recent publications of Seymour Hersh). Osama Bin Laden and his followers would also like to see the U.S. attack Iran. He and his followers have tried, and to some degree have succeeded, in getting the United States to over extend itself, to isolate itself from its traditional allies, and to distance itself from so called “moderate Arab states.” Iran is not Iraq. It has considerably more resources in every area that matters. And, while Iran lacks the wherewithal to directly strike the United States, it has access to U.S. forces across the border in Iraq. As we have gotten closer to the election McCain’s bellicose language towards Iran has diminished, but it would be a mistake to believe his passion for this issue has gone away. McCain is an old “Cold War” warrior. His entire life has been devoted to fighting, or preparing to fight. Under a McCain presidency Americans should expect and prepare to fight, and the all volunteer military will not be enough.  Adrian R. Lewis     

Senator McCain deserves credit for pushing for the “surge” in U.S. combat forces in Iraq. It has, without doubt, contributed to the reduction in violence. However, it is a mistake to conclude that the “surge” alone is the reason for the improved conditions in Iraq. Ironically, Iran deserves considerable credit for the reduction in violence. It possesses the geographic proximity, the cultural affinity with the Shia, the Islamic fundamentalist leadership, and the resources to provide the Shia in Iraq with the same type of resouces China provided North Vietnam in the 1960s. Iran possesses the wherewithal to make life for U.S. forces considerably more difficult, violent, and deadly. A third factor that has contributed to the reduction in violence is the new segregation in the country. Shia who lived in Sunni areas have been killed or run-out of their homes. Sunnis who lived in Shia areas have been killed or run-out of their homes. The Kurds have long been separated. In addition, the Kurds, Sunnis, and Shia have all formed there own militias to provide security. The government of Iraq has accepted these divisions and, for the most part, the existence of these well armed militia forces. A fourth factor is the existence of a new, effective Iraqi Army. The U.S. Army, Marine Corps, and coalition forces have invested enormous time and resources into building this army. Their efforts have started to pay off. Finally, it should be noted that if you pump eight billion dollars a month into any region of the Earth in time you should see some results. Other factors too have contributed to the reduction in violence. It is almost impossible to say which factor was the most important; however, the role of Iran is significant, and the agreements between the Shia and Sunnis are also signficant. The point is that it is wrong for McCain, or Bush, to claim too much credit for the reduction in violence in Iraq.  The issues are much more complex.

Adrian R. Lewis         

John McCain makes a strong argument that he is a maverick, an independent-minded man who marches to the beat of his own drum. There is a lot of evidence for this going back decades; including McCain’s opposition to Reagan’s sending troops into Lebanon, his work with Russ Feingold on campaign finance reform and Ted Kennedy on education, his challenge to the Bush administration on torture, and much more.

But riddle me this; why would a voter choose McCain’s promise of change over Obama’s promise of broader change? If Obama is elected president, he will be looking at a Democratic Congress with expanded majorities, probably about 246 to 189 in the House and about 55 to 43 in the Senate (there are 2 independents who currently caucus with the D’s, but who knows where Leiberman will be come January 2009). Given these expanded Democratic majorities, Obama should be able to move his proposals for change, both in regard to national security and domestic economic policy, efficiently through the Congress, especially in his first year.

McCain is, to be sure, a fighter. But why would voters worried both about the economy and troublesome wars in Afghanistan and Iraq elect a Republican president at regular odds with his own party and facing big Democratic majorities in Congress. In normal times one might say that voters feel more comfortable with divided government because it lets one party check the other. But these do not feel like normal times. What then would lead voters to pick the paler version of change (better execution of similar domestic and foreign policies) over the more vibrant (different policies)?

It seems to me that there are three general possibilities. One is that voters believe that McCain’s tinkering with the Bush policies will actually produce the change they desire, but this seems implausible except perhaps for core Republicans. Second is that McCain’s experience is so much greater than Obama’s that significant numbers of voters will see a choice of some change with McCain as safer than rolling the dice with Obama. Worrying about Obama’s experience is not unreasonable and lots of thoughtful people are doing it. Obama must use the remaining weeks of the campaign, and especially the debates, to resolve these doubts in as many minds as he can.

Finally, Rick Davis, McCain’s campaign manager, has said this election will turn on character rather than issues. This is certainly the McCain campaign’s hope as the issues are cutting hard against them. Unfortunately, this idea of character, not overtly with the McCain campaign but certainly on the web, easily morphs into the question of with which candidate are you most comfortable. In this broad and opaque area of “comfort,” there is room for much mischief. Obama will have to continue working the “shares our values” theme hard to insure that web-based innuendo and attack do not take root with the moderates and independents who he will need to reassure if they are to feel confident enough to plump for thoroughgoing change. What are the chances?

The outcome of the 2006 elections raised questions about what factors contributed to the Democratic sweep of Congress and more importantly whether the mid-term results signaled a broader trend that will continue this November. Many of these questions deserve revisiting as the 2008 elections approach. In 2006 many observers speculated that voters may have been swayed by the issues, particularly the war in Iraq. Iraq continues to occupy a prominent place on the political scene as the 2008 campaigns unfold. A recent CNN/Opinion Research Corporation Poll revealed that roughly two-thirds of Americans continue to oppose the war in Iraq. Will Iraq be a major determinant of how citizens cast their ballots this November or will other issues such as the economy, the rising price of gas, and health care emerge as more influential? We must monitor the 2008 campaign developments to determine which issues candidates and voters attend to throughout the coming months.

A more fundamental question is how will issues influence vote choice this November. Although the idea that citizens will form their political judgments on the basis of policy considerations is intuitively appealing, this proposal is somewhat at odds with the tendency of voters to know little about politics generally and to know even less about policy specifics. A recent example of why concerns about democratic accountability are still justified involves a story aired on CNN in July. A Florida businessman, Mike Meehan, purchased three billboards that prominently display the phrase “Please Don’t Vote for a Democrat.” During the CNN interview Meehan stated “I believe 9/11 could have been prevented if we’d had a Republican president at the time.” An important point that is not made during the interview, perhaps because it is such an obvious one, is that Meehan seems to have forgotten that Bush, a Republican, was president at the time. If citizens are so poorly informed about the major features of our political system, such as which party controls the White House, then it may be wise to question how issues will influence voter decision-making in 2008. In 2006, voters used issue information much less than we might like to expect of citizens. There is little reason to believe this will change in 2008.

Dona-Gene Mitchell

University of Nebraska-Lincoln

Co-editor, Fault-Lines: Why the Republicans Lost Congress

When I heard the news this morning that John McCain has selected Alaska Governor Sarah Palin to be his vice presidential running mate, it immediately struck me as a bold choice. One challenge—perhaps the greatest challenge—for the McCain campaign is to differentiate itself from the “McSame,” “Washington insider” and “Bush-McCain-Cheney” labels. Many pundits have observed that the maverick candidate John McCain from 2000 would be dismayed by the McCain of 2008. The selection of Governor Palin represents a clear effort to return the reputation of a maverick. Palin is a self-described moose-hunting, ice-fishing hockey mom. She quite clearly is as far as possible from being a Washington insider. And Palin became governor in large part by speaking up against ethics violations and corruption within the Republican Party in her state. Her selection surely is intended to signal that McCain-Palin is not a repeat of Bush-Cheney.

Although bold, the selection of Sarah Palin also carries an element of risk. It bears remembering that any selection would bring strengths and weaknesses to the ticket. Palin holds relatively little political experience. It may be difficult for Democrats to make that argument, but Palin’s short resume does make it harder for Republicans to criticize Barack Obama as inexperienced. Ideally, this all will be to the good, and the campaigns will turn to more of a discussion of the issues and less to snippy charges about which candidate is too this or too that. In the next weeks, months and years, McCain’s selection of Palin will come to be viewed as either astute or a disaster in the mold of Spiro Agnew and Dan Quayle. For now, the selection clearly seems to have energized John McCain and his supporters.

Jeff Mondak

University of Illinois

Co-editor, Fault Lines: Why the Republicans Lost Congress

John McCain’s inability to recall how many houses his family owned brought an immediate response, counter-response, and counter-threats from the two campaigns.  Republican operatives quickly reminded anyone listening that Obama got his “mansion” with a sweetheart deal from felon Tony Rezko.  If you want to play that game, said Obama supporters, we’ll drag up Charles Keating, the Keating 5 (the five senators, including McCain, burned by the scandal) and the Savings and Loan scandal of 1989. 

But this is small potatoes compared to what is out there in cyberspace and what is to come.  Rumor-mongering, whispering campaigns, and out-and-out lies have always been a part of the great American tradition of campaigning.  But, just as on-line communications has (finally) matured into a potent force in campaigning, so too has the campaign of rumor and innuendo through e-mail, bloggers, text-messaging, twittering, and other mostly non-main-stream media sources. 

Most of it is coming from the right wing of the Republican party and aimed squarely at Obama’s character and experience. Obama’s campaign has set up a rumor-mongering hotline, www.TruthFightsBack.com to alert the campaign to rumors and to beat them back.  This election is bound to get ugly, and the attacks will come from both the right and the left.  For all the marvels of the Internet and online communication, this is a glimpse at the darkside of the new media.  Like online gambling and porn, online rumor mongering and character assassination is now a part of the new American way of elections.   

I long for the old days of Willie Horton (1988) and John McCain’s black love child (2000). 

Dennis Johnson