Looking back on the lessons of the 2012 campaign season, we clearly learned two things about polls:
- Despite all of the criticisms and accusations of partisan bias, many pollsters had it right.
- Future elections will see more polling, not less.
For someone who has urged journalists to focus less on handicapping elections and more on exploring issues, these lessons suggest that maybe it’s time to for me to fine-tune the role of polling in election coverage.
As poll analyst Nate Silver said on “Morning Joe” this morning, “If you’re gonna do horse race then do it the right way.”
I still don’t think the focus of our daily election coverage should be limited to who’s ahead in Iowa, Ohio or Dade County (not that we’d ever know that for sure). But here’s the reality we need to acknowledge.
America is engaging, in effect, in an iterative election process — one in which we vote for our public officials almost daily until that one date on which we make a final decision.
Put another way, polling is allowing us to record the thought process by which Americans choose a president.
We need to cover that thought process. It’s a great story, and one that — told well — can help us better understand what matters to us as a nation. But we can’t stop there. We also need to cover the issues in the context of that evolving vote. We need to explain why candidates are taking — or changing — positions; why campaigns are targeting certain voter communities; why entire states are excluded from a campaign’s attention.
Together, the two uses of poll data can add up to important coverage.
Now let’s remember: Not every poll is credible. But an increasing number are (and Silver, an aggregator and analyst of polls, certainly is). We need to identify credible polls and ignore the others — unless, of course, we learn that a campaign is basing decisions on that data.
And let’s also acknowledge that the voter as blank slate is a myth. We can pretend that campaign season is a time when all voters are open-minded as they weigh the positions of the candidates, and that polls intrude, artificially, into the process.
But that’s not how it works, and it never has worked that way.
Many, if not most, voters hear the candidates’ positions having decided — at least at that moment — which candidate they support. That being the case, part of a journalist’s responsibility in reporting on campaigns includes explaining how events are being influenced by which candidate is leading, and by how much.
How many times during this campaign was it necessary for journalists to explore why President Obama or Mitt Romney shifted a position? And how often was it possible to explain their shifts by looking at the candidates’ standing in national, state or other polls?
Again, this is not a suggestion that journalists focus solely on the question of which candidate is winning on any given day. It is a suggestion that journalists with access to credible polling data include that data in their analysis of the candidates’ positions and the reasons for those positions.
Some suggestions for getting ready for the next election season (which probably already started somewhere):
Review how your newsroom used poll results during the 2012 campaign. Did you simply report who’s winning and trailing, like baseball scores? Or did you use polling data as context that would help a reader or viewer understand developments in the campaign? Use the review to craft a plan for when to use polls as stories and when to use them as context.
Identify someone on your staff who understands or wants to learn about polling. Someone who speaks the language and is comfortable with numbers. You’re not looking to create a staff pollster; you’re looking for someone who can look at a poll, help determine its credibility and spot areas for reporters to explore in their stories.
As the next election season approaches, create a calendar that lists dates on which the results of specific polls will be released. Use those events to preview upcoming campaign decisions.
Practice in your own backyard. Identify credible polling sources in your community that measure public sentiment on issues between elections. In the time before the next election cycle begins, use polls on certain local issues to create context for your community’s discussion of those issues.