The Press-Politics of the Presidential Primary Process
By Andrew R. Cline, Ph.D.
26 May 2003
Note: This essay was first published as a series of short articles on the Rhetorica: Press-Politics Journal weblog in May 2003.
Note: I added additional entries from Rhetorica in October 2003.
Did the Democrats' first debate of the pre-primary season matter?
There is ample evidence that voters, especially the uncommitted, find such debates politically useful, i.e. these debates help voters make choices. The formats of televised debates, however, do not allow for much depth of presentation. Instead, these debates are a platform to demonstrate message and image control.
Such debates, then, give voters a "feel" for the candidates' abilities to perform under pressure--to look and sound presidential. I watched the re-broadcast of the first debate from South Carolina on C-SPAN. I saw three potential presidents: 1) Dick Gephardt, 2) John Kerry, and 3) Joe Lieberman.
Does this mean Gephardt won? Hardly. The formats used in these debates are not structured to create a specific win-lose outcome. The concept of winning here is entirely a matter of individual interpretation. I liked the way Gephardt looked and sounded relative to the others.
Should we be choosing a President based on image? Do you have a choice?
A predictive model developed by Professor William G. Mayer1, of Northeastern University, suggests that the coming months are the most critical for the Democrats--long before many Americans begin paying close attention to the campaign. He contends that the nomination process will be over before the Iowa caucuses.
The model is simple and has two indicators. The first indicator: The candidate who leads in the last poll before the Iowa caucuses wins the nomination (Mayer confined his research to data gathered by Gallup). The second indicator: The most effective fundraiser wins the nomination (this does not necessarily mean the candidate with the most money).
According to a CNN/USA Today/Gallup Poll taken in late April, Joe Lieberman leads (22%) followed by John Kerry (18%) and Dick Gephardt (16%). The rest of the challengers register single digits.
Mayer draws these conclusions from his model:
1- The current nomination/primary system favors frontrunners.
2- Momentum is an over-rated factor in the process.
3- The frontrunner often stumbles, but these incidents are rarely fatal.
4- The longer we live with this system, the more we learn about it--and this works to the advantage of frontrunners.
5- Money is important, but it's not "the whole ball game."
These conclusions suggest that "campaign strategy" is an oxymoron. The candidate's ethos and past political performance have already made the argument, the message, and the image. As long as the frontrunner avoids disaster, he/she simply must maintain ethos. Mayer says:
And if campaign strategy doesn't matter, presidential nomination races will be decided more and more on the basis of such fundamental factors as popularity and money--resources that frontrunners, almost by definition, will have in greater supply than their competitors.
The race is not quite over, but, if Mayer is correct, it soon will be. That makes these debates extra important. The lack of live TV coverage, then, becomes not simply an anoyance to political junkies but a detriment to our democratic process.
Long, hot summer...
According to a recent FOX News poll, the top three Democrats contending for the nomination are: Joe Lieberman (19%), Dick Gephardt (14%), and John Kerry (12%). Everyone else is pulling up the rear in single digits. Some may argue this is the order of finish in the recent debate in South Carolina, although I gave Gephardt the edge.
If Mayer's predictive model is correct, then this poll, and others to follow, are far more important than most people understand.
One wonders what information respondents are using to make up their minds. Very little press coverage of a political campaign focuses on the creation, implementation, and critique of policy. At this point it's a popularity contest based on name recognition and image. Mayer's model suggests that the contest will be over long before we hear much about policy.
(Don't be fooled by coverage of the dueling healthcare plans, for example. Very little reporting of the various plans covers anything more than the surface features.)
Because most Americans get their news from TV, the nomination will go to the candidate who does the best job of creating a pleasing TV image while avoiding the ire of the press. Winning the nomination, which means winning this summer's popularity contest, means employing good kairos.
The search for drama...
Among the structural biases of journalism, the narrative bias is the tendency of the press to understand issues in terms of "stories," i.e. human drama. Further, this bias tends to promote the creation of master narratives, which are set stories that frame an issue.
Conflict among politicians, especially in election campaigns, is one of the definitions of news in politics. This is the source of the so-called horse-race reporting we see so often.
According to Mayer's recent scholarship, the primary season is a highly stable political event. Yet the press covers the primary season as a volatile, or unstable, political event. The master narrative dictates that these races are up for grabs among several political actors. The data, however, do support this view.
This suggests to me that the most important period of the campaign is the summer and fall before the Iowa caucuses. During this period, a candidate must create a presidential image and a winning story. That image and story must be adopted by the press as the master narrative for that candidate.
But, like all great fiction, the resolution of the story cannot be easily attained. Hence, the press looks for drama, and often creates drama (the illusion of instability), in an otherwise stable system. The drama they seek is human: people versus people. This search for human drama dictates that policy analysis is boring, undramatic, and unworthy of serious scrutiny.
Stable process = lack of choice...
A stable process is a predictable process. And, as Mayer has demonstrated, the leader in the last poll before the Iowa caucuses will be the winner of the nomination.
The press covers this process as if it were unstable, i.e. as if the outcome were in doubt. I maintain this is an example of the narrative bias of journalism, a bias that dictates that "stories" be covered as human dramas and master narratives be constructed to frame issues, situations, and characters.
On the surface, it seems the press should not portray a stable process as an unstable process merely because such framing does not conform to the facts. But narrative bias blinds the press to the bigger issue: The stability of the process robs voters of any real choice.
How did the process become stable? There are many "correct" answers to that question, but I want to focus on the role of the press. Despite coverage that portrays instability, the press stabilizes the process by: 1) deciding which candidates are viable, 2) creating or adopting master narratives for each candidate, 3) conducting polls and writing "stories" that encourage voters to adopt numbers 1 and 2.
Poll position is crucial...
The press contributes to the stability of the primary process in three ways. First, I want to consider the the contention that the press decides which candidates are viable, i.e. which have a realistic chance of winning.
The concept of viability is important for two reasons. First, according to many studies of voter behavior, most people are loathe to vote for candidates that they believe cannot win. Common sense tells us this is "wasting" a vote. Second, viable candidates are more successful fundraisers. Nobody wants to waste money on a loser.
As politicians announce their candidacies, the press begins making evaluations of viability. These evaluations are based on numerous criteria--except voter choice. Until the primaries happen, no voter choice is recorded except in opinion polls (and many of these do not weed out non-voters). And such opinions, so early in the campaign, are most often based on name recognition, momentum, and fundraising ability--only the latter not under the control of the press.
It is more than seven months until the Iowa caucuses, and we already have a frontrunner according to the polls and the reporting. While Lieberman apparently trails in the polls in Iowa and New Hampshire, he leads a current national poll. If we accept Mayer's predictive model, the primary-state polls do not matter. Whoever leads in the national polls before Iowa wins the nomination. Consider this from the cited reporting:
And there is still nervousness in the campaign that unless Lieberman does well in fund-raising during this quarter, he could be all but written off by party pros.
Does it matter what party pros think? The concept of viability suggests, and the Mayer model predicts, that Lieberman will do well in fundraising (well enough?) because he leads in the polls. This does not, however, suggest that he will remain the frontrunner. In seven months, another among a limited number of candidates (Kerry? Gephardt?) could, and almost certainly will, emerge. The press will evaluate these performances and have the last word on their viability.
The question is: Emerge based on what?
A good story...
The second of the three ways the press stabilizes the primary process offers an answer to this question: Frontrunners emerge based on the level of viability created for each. The press creates viability by constructing distinct master narratives (from the structural biases of journalism) for each candidate.
In horse-race coverage, each candidate is framed by a story of their candidacy. These stories may be positive or negative depending upon how the press interprets each candidate's actions.
These stories are built up over time, typically between the time of announcement and the beginning of the primary season. Some candidates, usually those who have previously run for President or are otherwise well-known nationally, may begin the race with a set narrative.
Master narratives are human dramas, not policy dramas. The press--with the help of the candidates' own attempts at image control--frames articles and TV coverage to match these narratives.
To see how this works, and to check the early narratives, read "The Crowded Bus," from the American Journalism Review. This article offers an excellent overview of the process of news coverage and the press' role in the creation of viability and master narratives (although I think author Rachel Smolkin might disagree with my use of "creation" in this context).
Another article from earlier this year is worth a second look now, "The Invisible Primary" by journalism professor Christopher Hanson for the Columbia Journalism Review. You'll notice he's unaware of Mayer's predictive model of the nomination process. Be that as it may, he comes to a conclusion I heartily endorse:
So here is a modest proposal: news media should front-load their own schedules and start full-throttle coverage of candidates’ policies and characters today.
Writing the plot...
A master narrative, in this context, is a set characterization of a candidate that leads to a set plot line in the "story" of that candidate's campaign.
Master narratives have many sources. Most of these set characterizations spring from the candidates themselves as normal image control. Sometimes, circumstances of a campaign create such narratives. And the press has been known to create them, too.
No matter the source, a master narrative is generally constructed this way:
1- A pattern of behavior is noticed.
2- The behavior is characterized, i.e. given a name.
3- The character is portrayed as part of a plot, i.e. a set course of actions, consistent with the character, beginning with a central tension and leading to a climax and denouement.
4- The candidates words and actions are analyzed by comparing them to the character and the plot.
Thus the narrative defines the candidate.
Let's consider the narrative created by candidate Howard Dean. The character he's created for himself I'll call the "electable liberal."
Over the last few months, Dean has openly countered the war, chided his opponents for moderate and conservative views, pushed an extensive healthcare package (and compared it favorably to such systems in Canada, England, and France), and claimed he represents the "Democratic wing of the Democratic Party." And he backs it all up with a generally successful record as governor of a notably liberal state.
This character may take full advantage of the conventional-wisdom plot of presidential campaign politics: run on the wings for the nomination and then switch to run in the middle (central tension of the plot) to win the national campaign (climax).
Dean created the narrative by acting consistently based on the electable liberal character. And the press accepts and transmits this character and the conventional-wisdom plot to news consumers.
Okay, but isn't this just standard image building in campaign politics? Yes, it is. The problem is that the press solidifies master narratives so that reporters have trouble writing stories with any characterization or plot other than the one allowed by the master narrative. And this structural bias of journalism can have detrimental effects for both citizens and candidates.
Primary instability paradox...
There's an interesting paradox to press coverage of presidential primary campaigns. Let's call it the "primary instability paradox."
As Mayer's predictive model demonstrates, the process is stable, i.e. the outcome is in little doubt. As the public record shows, the leader in the last national (Gallup) poll before the Iowa caucuses wins the nomination. This suggests that the real race for the nomination happens long before the primary season.
Because of the narrative bias of journalism, newspapers and TV cover the primary as an unstable process, i.e. as if the outcome were in doubt. Each candidate plays a role in his/her own master narrative. These personal-political stories clash with the stories of other candidates. The press focuses on these personal-political dramas--what we call horse-race coverage. This dramatic presentation of events overshadows the more crucial pre-primary season in which coverage of policy could make a real difference in the outcome.
By portraying the primary process as unstable, the press thus plays a role in the actual stability of the process. And a stable primary processes robs citizens of choice. Citizens believe they are choosing the nominee during the primaries. The choice, however, has already been made.
The choice is theirs...
Mayer's predictive model for nomination campaigns demonstrates that, since 1980, the leader of the last (Gallup) poll before the Iowa caucuses usually wins the nomination.
What interests me is what this model suggests about the pre-primary process in which the nominees are essentially chosen. If the winners are known before the primaries, then some other process and some other people are involved in the selection.
My conclusion (theory): The nominees for President of the United States are chosen largely by the press through the construction of master narratives and voters' reactions to these narratives.
That's not a role the press seeks. Instead, the structure of the nomination process and the structural biases of journalism dictate this outsome thus creating a process by which the press essentially chooses who will win the nomination. Voters in the primaries merely rubber stamp the choice.
Here's how it works:
Mayer suggested that campaign strategy doesn't matter much during the primary process. I would suggest that campaign strategy is most important now--during the pre-primary months.
The ultimate test of any theory is its ability to predict. So let's put mine to the test. I have begun gathering data for this project, and I will periodically report what I discover on The Rhetorica Network and Presidential Campaign Rhetoric 2004.
If Mayer and I are correct, then Joe Lieberman, John Kerry, and Richard Gephardt will show much better fundraising figures at the next reporting. Edwards will begin to fade because his national poll numbers are low. Dean, sitting in fifth place, may gain ground if his campaign strategy interests voters, but at this point, considering the Mayer's model (and despite some pundit commentary to the contrary), his candidacy appears to be a hopeless long shot.
UPDATE (1 July 2003): Howard Dean, with grassroots support from initiatives on the Internet, exceeded expectations. While this challenges the assertions above, his position in the polls remains the crucial point. How will his fundraising success affect his polls? The answer depends upon how the press adjusts and evaluates the narrative--from insurgency to "top-tier."
UPDATE (2 July 2003): Joe Lieberman reports raising $5 million for the second quarter, a $2 million increase over last quarter.
UPDATE (8 August 2003): Dean tied for second with Gephardt in Gallup poll.
UPDATE (14 August 2003): Dean takes the lead in a poll by InsiderAdvantage.
UPDATE (13 October 2003): Race is too close to call.
UPDATE (12 December 2003): With the endorsement of Al Gore, Dean takes a lead in the national polls beyond the margin of error.
Tell a different story...
I saw an interesting moment on Hardball with Chris Matthews on Friday, 23 May. MSNBC pollster Frank Luntz was questioning a panel of voters. His opening question: "Regardless of who you’re voting for, what characteristic do you want in a Democratic nominee?" After several people responded, Luntz said (with my clarifying remarks):
We’ll [the press] talk about personalities for the Democrats and you [the panel] all keep bringing it back to policy. That’s an interesting dynamic. Up until now, people [who?] were looking for, as you used, bold leadership, honesty, a vision for the future. [Luntz turns to the camera] And yet they’re all talking policy. [To the panel] Is that where the Democratic nominee is going to go, rather than focusing on attributes, they’re going to focus on policy?
Luntz continues to mention, with a sense of wonder, the panel's interest in policy. Matthews and his guests ignore it. Here is Luntz's concluding remark that Matthews cuts off to return to his guests:
I asked them to talk about candidates, talk about attributes and they kept coming back to issues. That says to me that there’s no Democrat out there that’s really captured the hearts and mind of the public as an alternative to George Bush. It is early, but there’s no one out there that’s got a clear...
In other words, the panel's interest in policy, the day-to-day stuff of governance that affects peoples' lives, is proof that no candidate has a convincing presidential image or master narrative. And the logic in that would be what? I would say this is proof that, at the moment, no narrative created by the campaigns or the press has completely usurped the voters' abilities to comprehend their own political interests.
Luntz wants them bow to the press' master narratives. But these citizens realize there is another narrative to be told, a narrative largely ignored during campaign coverage: The story of how policy affects the lives of average Americans.
During the pre-primary campaign, candidates raise money and hone their images in preparation for what many believe is the process in which the candidate will be chosen. But, as Mayer has demonstrated, the choice will be made before the primaries begin because the leader of the last (Gallup) poll before the Iowa caucuses wins the nomination. That means the most important part of the nomination campaign is happening right now.
The polls reflect choice. How are those choices made? If one relies solely on the press, especially television, that choice may be influenced by the press' assessments of political viability based on fundraising, master narratives, and poll numbers. Policy, and its affects on the lives of average Americans, plays very little role.
In other words, horse-race coverage (especially on TV) forces voters to deal with image. Because the concept of viability plays such a crucial role in voters' decisions, the press effectively chooses the nominee by deciding who is viable based image, and voters react to these mediated images and narratives.
The press applies a narrative structure to ambiguous events in order to create a coherent and causal sense of events. I assert that the press, to be politically useful now, must give more attention to another narrative--the story of how the candidates' former and proposed policies might affect the lives of average Americans.
What the numbers say... (21 October 2003)
I've tried twice (a low figure) to interest reporters in the Mayer predictive model of primary campaigns and my use of it to suggest that reporters spend more time covering policy during the pre-primary period.
I suspect the reason is the reporters think I'm crazy. After all, everyone knows that the primaries are competitive races based on the candidates' performances and strategies, and each race builds on the other leading to a winner. The model, however, says this isn't true.
Like any predictive model, this one could fail. But, considering its consistency (correct since 1980), I'm sticking with it. The question arises: Why is the model so consistent? I think we can find the answer in the Central Limit theorem: "the average (or sum) of a large bunch of measurements follows a normal bell-shaped curve even if the individual measurements themselves do not" (re: John Allen Paulos). That's pretty much the same as saying: you can't see the forest for the trees.
The forest in this case is the national Gallup poll charting the preferences of likely voters. The trees are the individual state polls. Despite plenty of evidence to the contrary, many journalists continue to think that winning in Iowa, New Hampshire, or any other given state is necessary to win a nomination. They continue to believe that charting such wins and losses indicates momentum.
(What would it mean, in terms of journalistic practice, if these things turned out to be misleading or false?)
I think the Mayer model demonstrates that the Central Limit theorem works even in a complex political system. Voters in the aggregate confirm the national polls with their votes because the national polls accurately represent the normal distribution of voter preference. The individual state polls, then, constitute too much detail, i.e. detail that is not predictive of the larger primary system.
The tricky part this year: statistically, there is no front runner yet. For the model to work, someone needs to emerge, i.e. take a lead that exceeds the margin of error.
If this doesn't happen, we will technically have (in a non-pejorative sense) chaos.
More on Mayer from Manifold... (22 October 2003)
Jay Manifold sent e-mail about yesterday's post on the Mayer predictive model of primary campaigns. Here's what he says:
I think the flavor of your suggestion about the Central Limit Theorem is good, but it doesn't explain why the Mayer model has worked for 6 elections in a row. Applying the CLT suggests that Mayer should be right most of the time, say for 4 out of the 6; then the other 2 elections would have had a major-party nominee who didn't lead in the last national poll before the caucuses and primaries started. This would, in turn, be caused by many small, unrelated random effects (see a related comment here).
But he's nailed it every time. Some things to watch for, then, are:
1- Applicability of other kinds of distributions, or of bounded distributions; unless caucus/primary processes are flawlessly proportional in the allocation of delegates based on their returns, the distribution of delegates is likely to be bounded by some lower cutoff. An interesting cautionary tale is here.
2- I note that election returns where many candidates are on the ballot are likely to exhibit log-normal (power law) distributions. [Ed. Note: The recent California recall.]
3- A complete miss next year. An effective tie among several candidates could produce a situation where not only would no one have a majority, no two could form a majority. Now that would be some fun!
I was struck by Jay's reminder that Mayer has "nailed it every time." This is significant. A miss this year would (should) not raise many questions about the model itself. Rather, statistically, it would properly be seen as a normal fluctuation of random events.
As I said in my original essay (link above), if journalists accept Mayer then they have to accept that the primary process is not the thing that we think it is--a dramatic race with an uncertain outcome. If we take away the drama, if we assert--accurately!--the nominee months before the convention, what then is their to report?
Hint: the P-word.
1 Mayer, William G. "Forecasting Presidential Nominations or, My Model Worked Just Fine, Thank you." PS: Political Science & Politics. APSA. 36(2) 2003: 153-57.