The Globe’s Frank Phillips recently called the ongoing race for the corner office in Massachusetts “one of the least-energized statewide races in years.” His Globe colleague Jim O’Sullivan speculates that this might have something to do with “a candidate lineup that has not, to put it politely, exactly set the electorate on fire.” While I agree that there have been few fireworks to date, I wonder if the Globe’s own coverage of the race hasn’t played a role.
The Boston Globe’s new “Capital” section, launched in June and devoted entirely to politics, includes a poll. Every week, the races for statewide offices are characterized and analyzed with the help of candidate-preference polling data. The poll not only forms the basis for the Globe’s analysis, it also shapes the coverage of the rest of the state’s political media. Bay State voters have never before had weekly polling snap shots of the major party nomination contests for governor in June, July, and August. So far these snap shots, which began on June 6th, have shown very little change in the major party nomination contests, with Martha Coakley and Charlie Baker seemingly coasting to their respective party nominations. However, if the media narrative wasn’t driven by these very frequent reminders that Coakley and Baker are way ahead, would Coakley and Baker be so far ahead in the polls? Has this weekly poll exacerbated what political scientists call a “bandwagon effect,” which is when “information about majority opinion itself causes some people to adopt the majority view.” The Globe poll has been published every Friday since June 6th and the public’s attention to politics in the summer is notoriously unfocused, making the likelihood of a bandwagon effect greater.
In Charlie Baker’s race for the Republican nod, the impact of this bandwagon effect is virtually irrelevant. Baker would be killing Mark Fisher no matter what the circumstances and we don’t need a poll to tell us why that it. Mark Fisher is a nut! Steve Grossman, however, is no long-shot extremist. On paper, he stacks up very well with Martha Coakley. So, why is Martha Coakley still so far ahead of Steve Grossman in the polls? To what degree is her persistently large polling advantage a function of a bandwagon effect?
There is little doubt that the Globe’s political coverage drives the political coverage of reporters and media analysts from across the state. This year the weekly Globe poll has been part of or the main focus of most of the media coverage of the Democratic gubernatorial primary contest since it began appearing in the paper in early June. This weekly “snap shot” has almost certainly had an unintentional impact on both the media coverage and the public’s perception of the ongoing Democratic nomination contest. If there were no media polls in the race for the Democratic gubernatorial nomination so far there is little chance that Martha Coakley would be considered the lone frontrunner. Media coverage of the race between the Attorney-General and State Treasurer Steve Grossman would almost certainly reflect a very close and competitive fight between two Democrats who have already won statewide office. If anything, Grossman would be the perceived frontrunner having earned the party endorsement at the State Democratic Convention in June. With little real policy differences between Coakley and Grossman, media coverage would be dominated by distinctions between the Coakley and Grossman candidacies. The relative electability of Coakley and Grossman would certainly be a much greater part of the ongoing media narrative of the race.
Unfortunately for Steve Grossman, a weekly poll question with a margin of error that has ranged from 5 to nearly 7 percentage points has convinced the political media that he is a “second-tier” candidate; that the story of the Grossman campaign is about his failure to reduce Coakley’s “big lead.” The problem is that other than name recognition, which depends quite a bit on …you guessed it… press coverage, it’s very hard to explain why “likely Democratic primary voters” would prefer Coakley to Grossman by such large margins. The reality is that they don’t, but because Coakley’s persistent lead is so large, political journalists feel safe in assuming that her lead in the nomination fight is real enough to take for granted. This assumption might be reasonable, but it is definitely exerting a subtle (and I think pernicious) influence on both the media coverage and the public’s perception of the contest.
What if Bay State voters (and political journalists) only had the benefit of media polls released every 4-6 weeks during the summer months? What would be written and broadcast about the contest between Coakley and Grossman in weeks two through six if journalists had to rely on other ways of measuring the weekly progress of the race? The publication by the state’s flagship media outlet of a weekly reminder that Coakley is coasting to victory during a time period when the candidates’ campaigns know that paid media is a waste of money due to public inattention to politics has rendered attention grabbing media coverage (i.e. the kind of coverage that moves poll numbers) of the race all but mute.
When I wrote in June about the pitfalls of assuming that Coakley’s huge polling lead over Grossman was real and deep a couple of prominent members of the Boston political media establishment slapped my argument down implying that I was either naive or trying to hype the competitiveness of an uncompetitive race. Nonetheless, you’d have to look long and hard to find many media stories about the substantive reasons for “likely Democratic primary voters” to prefer the AG to the Treasurer by 30 percentage points. Ironically, there were more stories about the differences between the Democratic activists who dominated the party convention and Democratic voters who will turnout in the September 9th primary.
It’s very hard not to wonder if the Globe’s introduction of a weekly poll, along with the media’s (over)reliance on this poll, hasn’t actually contributed to the low intensity of the Democratic race, and thereby to the lack of movement in the polls. I am fairly confident that because the poll is dictating the tenor of the Globe’s coverage, and because the Globe’s political coverage drives the coverage of the rest of the state’s political media, Steve Grossman has been much less able to get the kind of useful free media attention afforded to candidates perceived by the media to be serious contenders, and that without such attention it has been near impossible to overcome a significant statewide name recognition disadvantage during the dog days of summer.
The degree to which media analysts have swallowed the polling numbers whole is well illustrated by the conclusion of Boston Magazine editor David Bernstein that the Grossman campaign’s latest ad buy signals the Treasurer’s realization that he can’t catch Coakley by September 9th. Why does Bernstein, one of the most widely read political commentators in the state, come to this conclusion? Because the ad is a positive one containing no sharp attacks on the AG. My own discussions with folks in Grossman’s camp suggest Bernstein is at best wrong and at worst actively abetting the self fulfilling prophesy sustained by the weekly polling numbers.
Guess what? Around Labor Day the public’s attention to the election will pick up and the margin of the AGs polling advantage will narrow. How much it will narrow is hard to say, but one of the reasons for it narrowing is that more actual Democratic primary voters will turn their attention to the race, which will improve the accuracy of media polls because the composition of the Democratic primary electorate will become easier to measure. In other words, Coakley’s lead will be measured much more realistically because pollsters will have a much easier time identifying “likely Democratic primary voters.” Most media analysts will claim that the AG was either able to hold off her advancing rival, or that he overtook her in the final weeks of the campaign. The very likely reality, however, is that the measurement of the “race” is what actually will have changed, not the relative position of the candidates in the minds of primary voters.
I don’t suppose we’ll ever know how the race would look at present if the political media hadn’t let the polling numbers dictate the story line and instead covered the Coakley-Grossman contest with an emphasis on substantive differences between the candidates. I wish we had seen more efforts by media analysts to try to explain in substantive terms what might be driving the preferences of Democratic primary voters. One reason this may have been useful is that the folks who will actually show up at the polls on September 9th to choose a Democratic gubernatorial nominee are what political scientists call “high information” voters. The low intensity of the race to date actually increases the likelihood that these voters will be disproportionally of the “high information” variety, and high information Democratic voters do not, and have not over the last three months, prefer(red) Martha Coakley to Steve Grossman by more than 25 percentage points.