The approach taken here is to identify as much as possible all public opinion polls that were released in Canada during the respective time period. For ease of capturing this information, to be included the poll must have the following characteristics:
- Been publicly released or discussed within the month of measurement regardless of when the field period took place.
- Must not have been released (to the extent possible to be identified) to the public before. Media references to previously released polls are not new or unique public polls.
Polls are identified through careful review of the webpages of public opinion firms in Canada. This is a major source of information since polls that make it into the public domain, even if they are done for an association or particular sponsor, are often posted for public consumption on the research firm website. In addition, google searches for public opinion polls and for news about polls is used to identify additional publicly released polls. We focus here on polls that were released/ conducted specifically for the public domain and ignore general mentions of polls or public opinion. In some cases, there may have been a poll that was undertaken but we error on the side of caution to focus on clear surveys.
Finally, our focus here is on the poll, not its life in the media. A poll can only appear once regardless of how many mentions it gets in the media.