Release coincides with the launch of the federal elections to assist journalists and the public

Polls strengthen democracy by giving voice and influence to Canadians. In advance of the federal elections, the Canadian Research Insights Council (CRIC) issued today Canadian-specific standards on the conduct and release of public opinion research.

The new Public Opinion Research Standards and Disclosure Requirements will serve as an important tool for members of the public and the media to assess the quality and validity of survey results that are publicly released. This is particularly relevant in advance of the federal elections on October 21 given the important role that election surveys play in engaging Canadians and informing them as to how their vote can count.

“In today’s era of misinformation, fake news, click bait and foreign interference, fact checking outlier or bad polls is more important than ever,” stated Barry Watson, Chair of CRIC. “No poll is better than a bad poll, as bad or misleading polls misinform voters and hinder democracy.”

With the release of CRIC’s world-leading standards on public opinion polling, Canadians and the media can be confident that when they see a poll by a CRIC member firm, they know the findings are accurate, honest and grounded in statistical science.

The new standards outline the responsibilities of CRIC member companies when conducting public opinion research and the details that must be included when releasing research into the public domain. Member companies must comply with the rigorous standards and make available sufficient details to help the media and members of the public verify the quality of their research. This includes:

• The exact wording and presentation of questions and response options whose results are reported.
• A definition of the population under study
• The dates data were collected
• A description of the sample design with sufficient details to determine whether respondents were selected using a probability or non-probability sample
• A description of the sampling error in a survey’s results if scientifically applicable
• Whether weighting was used to adjust results to make them more reflective of the intended population
• And for research on public policy topics and election voting, detailed tables by key demographic that show both weighted and actual number of respondents.