Mark Blumenthal | August 24, 2007
Topics: 2008 , Measurement , The 2008 Race
Are the national front-runners in the race for president - Hillary Clinton and Rudy Giuliani - "coasting on high name ID?" That's the question that Gallup's Lydia Saad attempts to answer in a must-read analysis based on data from three Gallup polls conducted in July and August. Saad's answer appears to be yes for Giuliani, but no for Clinton. That is, Giuliani's front runner status does appear to depend, at least for now, on relatively high name recognition, while Clinton leads even among voters who can rate her two best known rivals. Though before assuming that the Democratic race is over, however, we need to consider what "name recognition" really means.
Saad's analysis is well worth reading in full, but here is the gist: Slightly less than half (46%) of Republican's nationally know Rudy Giuliani, John McCain, Mitt Romney and Fred Thompson well enough to rate all four. Among these voters, Giuliani trails Fred Thompson by eight points (33% to 25%). Giuliani's double digit national lead in Gallup's polling comes entirely from the 54% of Republicans who are unfamiliar with one of the top four candidates, as the Gallup graphic shows:
Among Democrats, the pattern is different. Less than one in four Democrats (23%) is not yet familiar with each of the three best known candidates, Hillary Clinton, Barack Obama and John Edwards. While Clinton holds a very wide lead over Barack Obama (53% to 17%) among those who are unfamiliar with one of the candidates, she still leads by a comfortable 13 point margin (43% to 30%) with Edwards finishing a distant third (with 13%) even among those who know all three candidates.
Even as Obama and Edwards build their name identification among Democrats, it would appear unlikely that this increasing public familiarity with Clinton's rivals alone would upset her lead...
On the other hand:
Giuliani is at greater risk than Clinton of losing support as the campaign progresses and his opponents become better known.
Results from other polls, particularly the recent up-tick in Clinton's national totals, support the conclusion that Clinton's lead depends on more than mere name ID. Even Obama's pollsters apparently concede the point, as reported by Ryan Lizza's in his recent GQ cover story:
When [Obama's pollsters] compared the percentage of Democrats who said they strongly approved of Obama with the percentage who said they would vote for him, they found that the latter number was significantly lower than the former. Inside the campaign, aides dubbed this "the Gap." It was a sobering, hard number that quantified the difference between vague enthusiasm and actual votes. For Hillary Clinton, the gap is much smaller. The majority of voters who strongly approve of her also say they will vote for her.
In fact, Hillary was collecting about two-thirds of Democrats who liked her, while Obama was collecting less than half. The numbers suggested that the calculus for Hillary voters was much simpler: Democrats who liked her knew all they needed to know about her. But for Obama voters, there were questions. Was he tough enough? Did he have enough experience? Could he actually win in the general election?
Of course, "strong approval" is a much tougher measure of name recognition than the percentage who are simply able to rate a candidate. Still, this report leads me to a thought that may have some bearing on whether the Democratic standings will change as voters continue to learn more about the candidates: "Name recognition" alone is a grey concept. Voters may recognize a name but still know little or nothing about the person behind it.
Most pollsters measure name recognition by counting those able to rate a public figure favorably or unfavorably. Gallup's question, for example, asks respondents to report "if you have a favorable or unfavorable opinion of these people -- or if you have never heard of them." However, minor differences in the answer categories presented can sometimes produce different results. Consider the percentages able to rate the three Democrats on recent surveys by Gallup, the Pew Research Center and CBS News (here and here; I've used the results among all adults to keep the populations comparable):
CBS News is the only pollster of the three to offer respondents the option to say they are either "undecided" or that they "haven't heard enough yet" about a candidate "to have an opinion." When I exclude the "undecided" respondents, the percentage able to rate each candidates drops considerably, with the biggest drops for Edwards and Obama. My educated guess is that much of the uncertainty comes from a lack of information; voters know the name, but not much about it.
How do I know that? Aside from many years of listening to voters offer what they know about vaguely familiar politicians in focus groups, we have some direct evidence. The same CBS News survey also includes open-ended questions asking respondents for "the first thing that comes to mind" about each candidate.
CBS News Polling Director Kathy Frankovic describes the results for Obama in a recent column:
Ten percent of registered voters said the first thing that came to their mind when they heard the name was the foreignness of it, and/or a perceived association with "Osama bin Laden." And those voters who did NOT have a favorable opinion of Obama were twice as likely as those with a favorable opinion to mention this. Those who thought his name would affect others also were more likely to be thinking of it themselves. Nearly one in five of those who said that they thought many voters would have a problem voting for Obama because of his name said in the earlier poll question that the name's non-American associations were the first thing they thought of when they heard his name....
Obama could overcome this problem, because the foreign connection is much more likely to be made by those who are paying less attention to the campaign than by those paying closer attention: Fourteen percent of those paying little or no attention to the campaign say it's the first thing that comes to mind when they hear Obama's name, compared with just 7 percent of those paying at least some attention.
As Obama keeps campaigning, and as voters start paying more attention, pollsters like me will continue to try and find out what voters see and whether their focus moves beyond his name, to the candidate himself and his platform.
Of course, Frankovic's observation has less bearing on the gap the Obama pollsters observed between "strong approval" of their candidate and vote preference, though that too may change as voters learn more about the candidates, particularly in the early states. Either way, Frankovic's column, like the Saad/Gallup analysis, is worth reading in full.