TRUTH IN DETAIL
THE SMALL NUMBERS MATTER MOST
Backroom politicos, pollsters and journalists are all grappling with a flood of campaign intel,
Polling and journalism, particularly political journalism, have a lot in common.
Both enjoyed huge surges in prestige, as well as income, in the 1970s and 1980s, when the trades began to be viewed as professions.
And both, thanks largely to leaps in technology, have been facing many serious and similar challenges in recent years.
Just as the Internet has given us do-it-yourself journalism, for instance, it has also given us do-it-yourself polling.
Media outlets that used to pay thousands of dollars for comprehensive public-opinion polls can now throw up an online questionnaire and call it an instant survey of the political landscape.
Seasoned pollsters complain there’s no science or rigour to these methods — “that’s not a poll, it’s a suggestion box,” Darrel Bricker, chief executive officer of Ipsos, is fond of saying.
Reporters make similar laments when bloggers claim they are practicing journalism. In the age of I-made-it-myself columns or opinion surveys, success is measured more by the buzz created and less by the methods used.
Given the similar threats to their businesses, then, you’d expect pollsters and journalists to be hanging together. BUT IN CANADA, the polling-journalism relationship has hit a bumpy patch, particularly in the form of way-off election predictions at the federal and provincial level.
What’s gone wrong? Here’s one explanation: we journalists have been paying attention to the wrong polls.
Surveys and data are just as important as they were in the heyday of the 1970s and 1980s, but it’s not the big, horse-race numbers we should be watching — it’s the small, micro-data now forming the backbone of winning campaign strategies.
Few people know this better than U.S. journalist Sasha Issenberg, author of The Victory Lab, a best-selling book on how American political campaigns are mining mountains of data to build support.
Issenberg, recently named by Politico as one of the top 10 journalists who broke away from the pack in the presidential campaign, was in Toronto this past week to share some of the things he has learned by keeping his eye on the microcampaign.
The Samara organization, dedicated to improving publicpolicy debate in Canada, hosted two speaking events for Issenberg. Judging by the 400-plus crowd that filled the Isabel Bader Theatre on Tuesday night, many of them seasoned political pros, there’s clearly an appetite in this country to learn about this much-overlooked aspect of modern electioneering.
When I got my chance to ask a question on Tuesday (I was lucky enough to be drafted as moderator), I asked Issenberg an important piece he wrote for the New York Times this fall.
Headlined: “Why Campaign Reporters are Behind the Curve,” Issenberg warned: “Over the last decade, almost entirely out of view, campaigns have modernized their techniques in such a way that nearly every member of the political press now lacks the specialized expertise to interpret what’s going on . . . It’s as if restaurant critics remained oblivious to a generation’s worth of new chefs’ tools and techniques and persisted in describing every dish that came out of the kitchen as either ‘grilled’ or ‘broiled.’”
So, I wondered, if we are behind the curve in political reporting, how do we get out ahead of it? How do we tell our readers or our audiences about campaigns that are created out of thousands, even millions, of “data points” guiding the political professionals?
Issenberg replied that a little humility was in order. Political journalists don’t have to stop covering the horse-race numbers or the big opinion trends — they’re still important, he said.
But they have to stop pretending that the big picture is the only picture — that the campaign is being decided on the basis of what people see on television or on the artificial stage sets craft- ed by the politicians. “Fundamentally, good political coverage needs to acknowledge that we cannot write with (any) sort of confidence about the entirety of the enterprise,” he said.
“We need to be respectful enough of our readers to acknowledge how much of this is out of our reach and find a new knowledge of campaigns to engage that doubt.”
So there’s the challenge. If we want to repair the rift between polling and journalism, first we have to tell our audiences what we don’t know — what those horse-race numbers aren’t telling us. A five-point rise or fall in the polls may be far less important, in other words, than the data informing us what is motivating (or turning off ) voters at the individual level.
Next, we should start trying to find a way to know and report more on this micro-data.
We have nearly three years to go before the next federal election in Canada. That’s enough time, you’d think, to put journalism back in touch with the polling numbers that really count. Susan Delacourt is a member of the Star’s Ottawa bureau.