Reading the tea leaves

Posted by cwaddell under Election 2011, Election 2011 Campaign strategy, Election 2011 Faculty links, Political Strategy

Andre Turcotte

There are a few milestones in an election campaign. The milestones are moments when voters begin to focus on what is at stake in the election and start the process towards deciding which party will receive their support. The first few days of the campaign is one such milestone. At that time, voters begin to realize they will be called upon to cast their ballot sooner than later and they seek to catch up on what has happened since the last time they paid attention to politics. It is an important process since we know from the 2008 Canadian Election Study that about one-third of voters (33%) will make up their mind about which party to support during the election campaign – some of them very early on – while close to 20% will wait until Election Day.

In the process of getting up to speed on politics, voters will seek information from families and friends; will pay closer attention to the news and party advertisements or will be visited by local candidates who will distribute campaign literature. Others may turn to polls to get a sense of where the respective parties stand within the public opinion environment. Those voters who are turning to polls for some insights are likely confused at this point.

Depending on which polls they accessed during the last week, voters have learned that the Tories have either an historic lead over the Liberals or the same Tories are barely ahead of their rivals and heading toward another minority government. Specifically, one poll recently suggested that the Harper Conservatives were 19 points ahead of the Liberals. Two other polls had the Tories with a double-digit lead (one at 16 points, another at 14 points).  In contrast, two polling organizations described the race between the Tories and the Liberals as very close – one established the Tory lead at 6 points and the other at 7. There are no methodological contortions that can explain such variations.

When George Gallup, Archibald Crossley and Elmo Roper began conducting polls in the 1930’s, one of the objectives was to democratize politics by giving a voice to voters. Gallup’s newspaper column was entitled “America Speaks” and purported to share with voters the kind of information about the public mood that was once the sole purview of political parties and their strategists.  While party polling was always more in-depth than media polling, voters were at least given a peek into the backrooms of politics. Today, the growing gap between media and party polling means that the voters is more likely to be mystified than enlightened by the latest polling data.

In the political marketing literature, the current practice of polling in political campaign is referred to as “market intelligence,” Accordingly, the purpose is to use the tools of scientific polling to identify, segment and target the electoral marketplace. The data is then used to refine the communication strategy and more importantly, to support Voter ID and GOTV (Get-Out-The-Vote) efforts. National horse-race numbers that are so fascinating to media outlets have become largely inconsequential to party strategists. As documented in the soon-to-be-released Lees-Marshment’s Political Marketing in Canada, the Conservatives did not even conduct nation-wide polling during the 2008 campaign since they understood that their success hinged on targeting few voters in key ridings.

In contrast, media coverage continue to focus on nation-wide variations in polling results that say very little about what is actually going on inside the respective campaigns. More troubling is the media practice to let wildly fluctuating poll results guide the nature of their coverage. In the next few days, a new set of polls will be released. Inevitably, the 19-point Tory lead is unlikely to hold and any closing of the gap will be reported as the Liberal campaign gaining momentum while in fact, a simple (but far less newsworthy) regression towards the mean may be at play. Similarly, the two polling organizations with closer vote intentions may report a widening gap and in the process, suggesting that the Harper campaign is gaining steam. While one of those two scenarios may likely be true, both cannot be accurate. The discerning voter should choose to ignore all of this and rely on facts rather than fiction in deciding who should run the country.

Andre Turcotte is an assistant professor in the School of Journalism and Communciation.