Canadians are paying an extraordinary amount of attention to the US presidential race, particularly the Democratic contest between Hillary Clinton and Barack Obama. It’s not primarily because people are concerned about the policy implications. Rather, people are caught up in this race because it offers the best of human drama that can be offered β€” candidates are up, then they are down; the historic possibility that a woman or an African American may be the president; lifelong dreams and well-honed strategies lie in public ruin; the pollsters are crashing; the pundits are invariably wrong; Bad Bill. It’s a novel, and each primary adds a new chapter.

However, for those who study the craft of politics, each of these offers more than just human interest. There are serious implications to be drawn and lessons to be learned. Here are some.

As a professional pollster, I feel morally compelled to begin with a discussion of the problems the polling industry has had with this primary season. The extent of the problem is not in any dispute. In fact, in the wake of the New Hampshire pre- election polls, the American Association for Public Opinion Research announced the formation of an ad hoc committee to evaluate pre-election primary poll methodology and the sponsorship of a public forum on the issue. That’s how serious the credibility crisis is. By way of example, one of the more prominent political polling companies in this cycle has been Rasmussen Reports. On its Web site it claims to have been the most accurate pollsters in the 2004 election. So far it has been wrong on the California Democratic race by 11 percentage points, the Alabama Democratic race by 19, the Georgia Democratic race by 20. It is not a singular example. John Zogby is one of America’s pre-eminent political pollsters and his firm, Zogby International, is large and successful. He was merely wrong by 16 points on the Democratic race in New Hampshire, by 13 points on the Democrats in South Carolina and by a shocking 23 points in California. And these problems are not unique to the public pollsters. The Clinton camp β€” known for its heavy reliance on internal polling β€” clearly believed it was going to lose New Hampshire based on its own polling. It goes without saying that when the polls are off by that margin, they are worse than useless β€” they are providing voters, campaign decision-makers and analysts with incorrect information on which to make decisions.

There are a wide variety of explanations of the challenges facing pollsters in this primary season. (For those who wish to learn in greater detail, I refer you to the excellent Web site pollster.com). Adding to the complexity is the fact that the mistakes do not all go in one direction. In Iowa and South Carolina, pollsters thought Obama would likely win, but by nowhere near the margins he won by. In New Hampshire and California, some pollsters picked Obama by large margins and Clinton actually won β€” in California she won by a large margin. After the shock of New Hampshire, where polls showed Obama building a significant lead following his win in Iowa, many people attributed the discrepancy between the polls and the actual results to the so-called Wilder effect. It is defined by Wikipedia as follows:

The term Bradley effect or Wilder effect refers to an explanation advanced as the possible cause of a phenomenon which has led to inaccurate voter opinion polls in some American political campaigns between a white candidate and a non-white candidate. Specifically, there have been instances in which such elections have seen the non-white candidate significantly underperform with respect to the results predicted by pre-election polls. Researchers who have studied the issue theorize that some white voters gave inaccurate polling answers because of a fear that by stating their true preference, they might appear to others to be racially prejudiced.

Douglas Wilder is an African American who, when running for governor of Virginia, was picked by polls to win by 9 percent and actually won by less than 1 percent.

The phenomenon of “socially correct answers” is well known to the Canadian polling industry. It has historically manifested itself consistently among Quebecers when they are asked about their support for the separatist option. Separatist support is consistently overstated in polls because it is seen by some in Quebec to be more acceptable to be a separatist than to be a federalist. It has been referred to as the Liberal “ballot box bonus” in Quebec.

The Wilder effect seemed like a plausible, if not definitive, explanation for the New Hampshire Democratic primary. However, events soon outstripped it. Subsequent primaries revealed that pollsters were underestimating Obama’s support, not overestimating it. It is now likely that the biggest underlying problem β€” driven by the unusual appeal of both the Clinton and Obama candidacies β€” is in determining who will actually vote in these primaries. Because most members of the general public do not vote in presidential primaries, one of the hidden methodologies applied by all pollsters is determining the “likely voter.” All organizations have “screens” they develop to project likely voters. None are the same. Few are publicly available and transparent. The problem all organizations are having is that history is no guide to this primary season. Most states will set turnout records this year. In Iowa, fewer than 125,000 Democrats participated in the 2004 caucuses. In 2008, more than 230,000 did. In South Carolina fewer than 300,000 people participated in the Democratic primary in 2004. This year almost 550,000 did. In Virginia, turnout for the Democratic primary jumped from just under 400,000 in 2004 to just under a million in 2008. Given the polarization around demographics in this election, knowing who those additional voters are is critical to getting the poll numbers right. Are they women? Are they African American? Are they Hispanic? Are they well-off or low-income? Are they highly educated or do they have minimal education? A surge in turnout by any of these demographic segments has the potential to dramatically turn the result β€” especially given that overall turnout rates are still low compared to a general election. Most public polls do not have a large enough sample size or a comprehensive enough questionnaire to gauge these micro-markets accurately.

One last explanation has to do with the “fluidity” of opinion in these primaries. The theory goes that primaries always have more last-minute decision-makers than elections do. This is because in the general election most people make their decision on the basis of the Party they support, and don’t need to have as much information about the specific candidates to make a decision. In the primaries, the voter does not have as easy a way to differentiate candidates and therefore has to learn more about them. For many voters, much of this information comes late in the campaign, leading to late decisions. Obama’s campaign held much of its pre-Super-Tuesday advertising budget for the week before the vote. Pollsters may have trouble capturing all these late decisions, especially if they stop polling days before the vote. This happened in the 2004 Canadian federal election, when the public pollsters largely missed the last-week move to the Liberal Party. The factor may be exacerbated in this primary season because of the lack of a dominant candidate throughout, and the fact that many Democrats find both of their major candidates very compelling and attractive.

A presidential election campaign has three key tools at its disposal to get out its message β€” the media coverage (called “earned media” in the business), largely generated by the candidate’s tour; paid advertising; and, increasingly, the Internet.

Paid advertising can often be decisive in determining an election outcome. US elections have seen many memorable advertisements β€” “Morning in America,” “Daisy” and “Willie Horton” come to mind. But in this primary season advertising is less important than normal. With so many of the primaries bunched together in a short period of time, campaigns were unable to have the kind of weight behind their advertising that really makes a difference. Even the vast amounts of money Clinton and Obama have been raising are inadequate to finance a national advertising campaign of months in duration. An additional complication is that many of the most effective political television ads are negative or comparative, and this tactic is very dangerous to employ in an internal party campaign.

Obama has essentially turned the Internet into a true second advertising vehicle, one with extraordinary reach to an untapped source of voters β€” those 25 and younger. Whether it is ubiquitous YouTube videos like “Obama Girl” or “Yes We Can” or raising $30 million a month in online donations on a regular basis, the Obama campaign has dominated the Clinton campaign in the Internet portion of the campaign.

However, for reasons that are less easily explained, Obama has also dominated the media coverage of the campaign. In late January, the Washington-based Center for Media and Public Affairs issued a report that found:

Since mid-December, when the presidential candidates turned their full attention to the Iowa caucuses, Sen. Barack Obama has led the race for good press and Sen. Hillary Clinton has lagged the farthest behind. From Dec 16 through Jan 27 five out of six on- air evaluations of Obama (84 percent) have been favorable, compared to a bare majority (51 percent) of evaluations of Mrs. Clinton. The gap in good press has widened since the New Hampshire primary, with Clinton dropping to 47 percent positive comments and Obama holding steady at 83 percent positive.

The numbers get even uglier if you include the coverage of Bill Clinton. The same study found that 75 percent of his coverage was negative.

Some might think that much of that was deserved and brought on himself, but what would explain the difference between the coverage of Hillary and Barack? Is it the novelty of Obama? Clinton fatigue? Years of feuding between elements of the press corps and the Clintons? Traditional media empathy for the underdog? Traditional media demand for a contest? Sexism? TV comedians have joked about Hillary Clinton almost as often as all the other Democratic presidential candidates combined, according to a new study by the Center for Media and Public Affairs. The main targets of their humour were Clinton’s personal appearance and her chilly personality.

Whatever the explanation, it will be of little comfort to the Clinton campaign. Even when one has the capacity to drive the message with a full advertising campaign, that kind of earned media barrage is likely insurmountable. Especially as her campaign tries to turn around the momentum to hold support in Texas, Ohio and Pennsylvania.

Some will say that people don’t change their opinions because of what the media thinks, but concede that the media tells people what to think about. They call it setting the “frame.” Even on those terms, the media coverage has favoured Obama. Because Obama’s campaign is nothing short of a phenomenon in terms of the size of the crowds he attracts and the enthusiasm he generates, the usual media bias toward horse race coverage rather than policy analysis has strongly favoured him. Clinton has no chance of shoehorning a debate over health care or economic policy into this frame.

Nor has there been much attention paid by mainstream media to the contradiction between Obama’s Kennedyesque oration and the Bill Clintonesque incrementalism of his policy proposals, or the achievability of the hopes and aspirations he is creating.

None of this is to say that the things the media is covering are not real, nor that the compelling aspects of Obama’s campaign do not merit coverage. It is to say that only certain things get covered.

Between the apparent bias against Hillary Clinton in the media and a decidedly unfavourable media frame, she has been in a purely defensive position. All she can do is try to hold on to as much existing support as she can. It is a testament to the strength of her appeal that she was able to withstand the Obama surge on Super Tuesday as well as she did. Out of money and getting hammered in the media, her candidacy could have been swept away that day.

The importance of the role of media is something that should not be ignored in Canada. With the exception of thoughtful reflections in the Globe and Mail by Lawrence Martin, this issue gets little attention. Canada has a very high level of concentration of ownership in the media sector. In Vancouver, CanWest owns both of the daily newspapers and the dominant source of television news. Leave aside the fact that, as heavily regulated industries, media outlets are inherently political. Both of the dominant media organizations in the English Canada β€” CTV and CanWest Global β€” are clearly favourable to the governing Conservatives. The Toronto Star is reliably Liberal. The McGill University Observatory on Media and Public Policy studied newspaper coverage of the 2006 Canadian election β€” as reported in Policy Options (March 2006) β€” and stated, “Editorial and opinion content showed clear partisan preferences, generally in line with widely accepted beliefs about the newspapers in our study, while news content remained relatively neutral.” Canadians, no less than Americans, would be wise to seek out a variety of views when making their voting decisions. Americans have access to web sites such as RealClearPolitics, which serves as a comprehensive clearing house of articles about the election with no agenda of its own. The Huffington Post, while having a clear liberal bias, is a terrific source of both breaking news, for those who follow in real time, and original opinion pieces. Canada has no such equivalents, and we could use them.

When analyzing the remarkable Barack Obama campaign, there are so many things to consider: the fact that he is the best political orator of his generation, the fact that he has the ability to inspire such hope through his voice and bearing, the unglamorous hard work of putting together a ground organization that has crushed the Clinton campaign in the caucus states, revolutionizing political Internet usage or raising record amounts of money. The two most profound forces behind his success to date are his ability to capture the mood for change, and his ability to attract significant support from white Democrats.

The desire for change is one of the most potent forces in politics. With very few exceptions, an extended stay in power for one party will generate a desire for political change. The demand can be for different policies, different style or even just different people. However, the mood in the United States is particularly receptive to a message of change because of the massive dissatisfaction with both the Bush administration and the direction of the country. Only one in four Americans thinks the country is headed in the right direction. For context, around 60 percent of Canadians think our country is going in the right direction. While virtually every candidate for both parties’ nominations paid at least lip service to representing change, Obama has become the clear standard-bearer for change in the Democratic race and in the country overall. And when the mood for change in the country is that strong, and you represent change, and you are running for the nomination of the opposition party β€” well, your positioning is pretty good.

In part this is due to a strategic miscalculation on the part of the Clinton campaign. Clinton’s calculation at the outset of this process (more than a year ago) was clearly that being a woman was enough change, maybe even too much change, for a lot of voters. She deliberately underplayed perhaps the biggest change card available β€” the fact that she would be the first female president of the United States of America. Therefore, instead of emphasizing the change that she embodied and that her policies reflect, she emphasized the safeness of her candidacy. She could be the commander-in-chief. She could win a general election. She had been there, done that. That strategy created the vacuum into which walked Barack Obama.

But he did more than just walk into it, he exploded into it. He turned his campaign into a movement. He has made people feel empowered and part of the “change” he promises. He has made people feel that big things are possible. And he has made people believe that he represents a new way of doing politics, not just different policies. In so doing, he has unlocked the key to likely victory in the Democratic primary, and possibly to the presidency. He is attracting people to the political process who have previously not bothered to participate because they did not think it mattered to them who the president was. He is drawing these new voters primarily from the ranks of young people and African Americans. According to the respected Pew Center, “In all of the 2008 contests for which exit poll data are available, young people have constituted an average (median) of 14 percent of Democratic primary voters, up from a median of 9 percent in the set of comparable contests in 2004.” Because he has made them believe that fundamental change is possible, because he has made them believe that it does matter who wins, they are voting. Obama is the most important reason for those record turnouts bedevilling pollsters. The US has a 50 percent turnout in most Presidential elections. If Obama can motivate an additional 5 percent to come out and vote, he will break up the red state/blue state deadlock and have the Reaganesque realignment he talks about.

Before the South Carolina primary former president Bill Clinton, in a very unsubtle attempt to marginalize Obama’s campaign, compared it to Jesse Jackson’s run for the presidency in 1984 and 1988. The implication was that Obama should be expected to do well in states that have significant African-American participation in Democratic primaries. Unfortunately for the Clintons, the comparison is not backed up by facts. While Obama is, indeed, getting huge support from black Americans, key to Obama’s success has been his ability to appeal strongly outside the African-American community. A recent survey of potential voters in the Texas primary indicates that Obama is getting 65 percent of the African American vote, but also almost 40 percent of the white vote. Among white men, he leads Clinton. (Note that the problems pollsters are facing in these primaries do not have to do with collecting accurate data β€” they have to do with faulty “likely voter” segmentations, as stated earlier. Therefore, one can look at opinions and preferences among potential voters with confidence. It is when predicting primary results that one gets into trouble.)

Gender is the most important predicter of voting behaviour in the Democratic race. Women are more likely to support Clinton; men are more likely to support Obama. These trends are true regardless of other factors such as ideology, strength of partisan affiliation or education. The Texas Credit Union survey shows Clinton losing among white men, but leading by 30 points among white women. She is trailing among liberal men and conservative men, but leading by large margins among liberal women and conservative women. She is trailing Obama among “soft” Democratic men by 30 points and leading him among “soft” Democratic women by 35 points. It is likely that the female turnout in the coming primaries will be decisive. If Clinton wins, it will be in large part female voters that have delivered the victory. If she loses, it will be because men choose Obama. Superdelegates will be interesting to watch β€” the Web site Politico reports that 64 percent of them are men.

The huge support for Clinton among women has been relatively discounted. Clapping soccer moms do not make for as exciting television as cheering and chanting college students. However, it reflects a huge divide in politics β€” both in the US and in Canada.

In my years of survey research in Canada, nothing has been more consistent than the very different ways in which most women and most men think about policies and politics. Women tend to have very different policy priorities than men do. If women were running either country, the policy agenda would be very different. They also view the practice of politics differently, tending to be more suspicious of the “great leader” phenomenon that Obama represents and more comfortable with inclusive and transparent processes for decision-making. In this way, the Clinton candidacy is a potentially greater challenge to the status quo in politics than is the Obama candidacy.

Neither Obama’s race nor Clinton’s gender has proven to be an insurmountable liability in the Democratic primary. However, winning the general election is another story.

Despite the most favourable conditions possible for a Democratic victory β€” an intractable, unpopular war, an economic recession, a surly public and a Republican nominee with lots of flaws β€” head-to-head matchups show McCain running competitively with Obama and ahead of Clinton. As Bill Clinton was alluding to, it is one thing to win the Democratic primary in southern states where African Americans dominate the Democratic primary. It is another thing altogether to win the general election in those states when white voters will dominate. In fairness, he might have added that it is one thing to win primaries in the Democratic Party, where having a female president would be seen as a long-awaited breakthrough and where Gloria Steinem is an opinion leader, and quite another to win those states in the general election.

Rasmussen Reports says that around 70 percent of voters say they would be “willing” to vote for an African American or a woman for president. Accepting that at face value, it immediately shrinks the possible voting pool by one-third. However, given the propensity of people to give socially acceptable answers, it may be more telling that in the same survey only 56 percent of voters said they felt their family or friends would be willing to vote for an African American or a woman. Many traditional American voters are resistant to the charms of both of them. When wrapped in the nostalgic gauze through which the 1960s are remembered, it is difficult to remember that Richard Nixon won the presidency in 1968 β€” hard on the heels of the Summer of Love! β€” by promising to speak for the “silent majority.” Defeating McCain by either Clinton or Obama will depend on the Democratic nominee expanding the total pool of voters by motivating more young people or women to the polls than would normally be the case.

For many Canadians this is the most exciting political contest they have witnessed in some time. For political professionals, it is one for the ages.

David Herle
Contributing Writer David Herle, former pollster and chief campaign strategist for the Liberals under Paul Martin, is a principal of the Gandalf Group in Toronto.

You are welcome to republish this Policy Options article online or in print periodicals, under a Creative Commons/No Derivatives licence.

Creative Commons License

More like this