Articles and Analysis


Three Campaigns

Topics: 2008 , The 2008 Race

As someone who has spent much of his career as a paid political professional, there are moments in most campaigns when the horserace coverage of the campaign makes me want to scream. Not that I dislike horserace coverage -- I enjoy it as much as any political junkie. No, the problem is that it so often misses one fundamental aspect of the way American politics works. All too often, news and commentary about public opinion makes the implicit assumption that most Americans are attentive to politics. If all Americans followed every campaign story closely, it would make sense to watch national polls for evidence of a immediate reactions. But it just isn't so.

The reality is that politically attentive, well-informed Americans constitute what U.C.L.A political scientist John Zaller, in his highly respected text The Nature and Origins of Mass Opinion, calls a "small but important minority:"

Members of this minority can recognize important senators on sight, accurately recount each day's leading news stories and keep track of the major events in Washington and other world capitals. They are, thus, heavily exposed to elite discourse about politics.

Any attempt to gauge the absolute size of this highly informed minority is essentially arbitrary, though see Bennett, 1989; Smith 1989, Delli-Carpini and Keeter, [1997]). Nonetheless, one indication of size is when respondents to a National Election Study were asked to name as many members of the U.S. Supreme Court as they could remember, about 1.9 percent of the public could mention as many as half the members [p. 16, links added].

Or consider another measure. According to the Nielsen ratings published by TVNewser, last week's MSNBC debate garnered just 2.4 million viewers, ranking it sixth among the sixteen debates this year. The biggest debate audience so far was 3.1 million for the September 5 Fox News Republican debate. As Jay Cost points out, that audience amounts to "about 4% of the first Bush-Kerry debate in 2004."

So at one extreme are those of us who closely follow politics. At the other are the roughly 25% that have been telling the Pew Research Center they follow campaign news "not closely at all." Of course, many of these totally inattentive Americans do not bother to vote.

So the vast majority of Americans fall into a critical middle group that pays sporadic attention to politics, mostly through television. Again, quoting Zaller:

Probably from some combination of civic obligation and the entertainment value of politics, a majority pays enough attention to public affairs to learn something about it. But even so, it is easy to underestimate just how little typical Americans know about even the most prominent political events - and also how quickly they forget what for a time they do understand [p. 16].

Consider some recent findings from the political knowledge study fielded by the Pew Research Center last February. At that time, only 15% of adults could identify Harry Reid as the Majority Leader of the Senate. Only 21% knew that Robert Gates is the Secretary of Defense (without prompting, though 37% could identify his title when offered four possible choices). Only 24% knew that that both houses of Congress had recently passed legislation to raise the minimum wage.

And yet time and again, some event - like last week's debate and the coverage that followed - captures the attention of political junkies. Inevitably, journalists turn to national public opinion polls of all adults anticipating major shifts in opinion and surprised to see little or no change.

We should note that the current upward trend in support for Hillary Clinton kicked off just after the July 23 CNN/YouTube debate during which Clinton and Obama sparred over whether to meet with the leaders of Syria, Cuba, Venezuela and North Korea. Clinton's support now is 5 to 6 points higher on our trend estimates than in early July. Both a Pew Research Center survey in late July and other analysis based on more recent Gallup surveys shows Clinton's gains coming mostly from college educated, who tend to pay more attention to political news to those without college degrees.

Keep in mind, however, that the statistically meaningful six-point gain did not occur in a week's time. It has been spread out over nearly four months, amounting to a percentage point or two a month since late July. At that rate, the odds are that any significant shift since last week will be lost in the usual random variation (i.e. "margin of error") inherent in opinion surveys, especially if we focus on only one or two polls.

Going forward, it may help to thing about three different campaigns that are now underway nationally. The first is among the tiny but influential group that follows politics obsessively (and includes virtually everyone that reads this site). These voters know all about last week's debate and the coverage that followed, and could respond accordingly.

The second campaign includes most of the other voters in the United States. Most did not follow last week's debate or the ensuing story, and most remain unengaged in the campaign. Those that watched the morning news shows last week or glanced at the front page of a newspaper may have seen a story or two about Senator Clinton having a tough time in the debate, but little more. For months the main story of the Democratic race has been about Clinton's success and dominance. If that narrative changes in a way that persists beyond a week, we may see a small shift in national trial-heat polls, but expect any such change to be slow and gradual at best between now and the end of the year.

The third and most important campaign, however, is occurring right now among voters in Iowa and New Hampshire. The same division exists there between well informed voters and everyone else, with one critical difference: In those two states, the candidates are spending millions of dollars to push their messages at less attentive voters through television advertising, direct mail and other forms of voter contact. And since candidates are constantly campaigning in person in those states, the local news in Iowa and New Hampshire is also covering the race much more heavily than elsewhere.

The third campaign is the most important. It is worth watching trends there more closely, both because voters there are now tuning into politics, and because voters nationally typically start to pay more attention to politics as those two states render their decisions. For that reason, large and dramatic shifts in the national polls are far more likely in January than they are now.


Prantha Trivedi:


Excellent analysis. I thought I would ask you (as the expert that you are) what you think the impact is - of appearances on, say Saturday Night Live (a most recent example); satire on, say, Jon Stewart Show; and B.S. (passing for news) on Faux News?

Would those viewers be in (what you refer to as) the "second campaign?"



Prantha Trivedi:

Follow-up question:

Do you think that some pollsters have more of a bias toward a given candidate than others?

(I have always felt that Rasmussen is biased conservative. So, since I know that Rove wants Hillary to be the Dem leader; I automatically assume she will have higher positives than others on Rasmussen. There is another polling organization -the name of which escapes me at the moment- with a Clinton campaign insider at the helm. I also assume that that poll will have higher positives for Hillary.)



I agree. In many blogs, liberal and conservatives, it's not unusually to see a headline that screams "Poll shows x-candidate won last week's debate" The evidence they cite is any given poll that shows a small variation favorable to X candidate.
Politics are complex and nothing can be explained by a single factor.



I, too, agree. I've been out of the field professionally for a number of years, but I've wondered what work has been done to apply "contagion" theory to the events you're talking about. It seems likely that the effects of any single event can be measured in terms of direct effects on the viewers plus the indirect effects that stem from repetition of the information either in the media or by word of mouth.


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.