New rules aim to make election polling more transparent

Members of Canada’s Marketing and Research Intelligence Association (MRIA) will have a new set of rules to follow when it comes to disclosing polling results to the public, especially when it comes to political polls.

MRIA says the new rules are meant to provide “more consistent and rigorous transparency” for those issuing polling on any subject matter, but especially to those who provide “opinion polls during election periods.”

It says the media and public need to be able to vet the reliability of the polling firms at all times.

“The new MRIA requirements reflect the recommendations put forward by the British House of Lords Select Committee on political polling and digital media, those made by the Canadian Association of Public Opinion Research (which are folded into the MRIA) and those from its own standards committee.”

Some of the new requirements include disclosing who sponsored the survey if it is different from who conducted it, the method used to conduct the survey, and the exact wording, presentation of questions and response options used in the survey.

A definition of the population studied, the methods used to recruit the panel or participants and whether weighting was used to adjust the results will also be required to include in a poll.

The full list of disclosure requirements can be found in the MRIA’s code of conduct.

John Wright, a pollster who sits on the standard committee for Canada’s MRIA, said the standards, a first for Canada, will place polling firms on a level playing field.

“They have to show all the same stuff so they can actually compare apples to apples,” said Wright. “It may not mean a lot to people who are just the public, but what it does mean is how everybody does their homework.”

If a polling firm fails to meet all the new disclosure requirements, the MRIA said, “the firm and its Principals may be subject to advice or a warning at one end of the scale to penalties ranging from censure or termination of its membership in the MRIA at the other end.”

Ian Koenigfest, the president of the Radio Television Digital News Association (RTDNA) of Canada, says his organization is fully on-board with the new rules as it will reduce “misreporting and misrepresentation” of polls. He adds that it’s a “positive attempt for polling companies and the media to collaborate on the dissemination of verified data.”

MRIA chair Mark Wood says that while “some discrepancies in output and predictions may still occur because of different methodologies, interpretation or voter volatility, the report draws a roadmap to address mutual areas of interest and responsibility between polling and media leaders and we will actively explore those in the months ahead.”

There are firms that are not part of the MRIA, like Mainstreet Research, who argue that they do adhere to the rules, but don’t think they should be strong-armed by an association to share some information they deem proprietary.

To that point, Wright adds,

“If you’re going to be releasing polls during an election time-frame particularly in a democracy there is nothing proprietary about transparency. And I think that anything if you can get more transparent especially during fake news and accountability is better for democracy.”

Earlier this week, CityNews compiled polling data from four previous elections. Click here for interactive graphs presenting polls taken two weeks before election day and the actual results from the election.