Campaign Research Inc. has conducted 2 polls using different methodologies over a two-night period (September 18th & 19th, 2019).
In the first week of the federal election campaign there had been a lot of discussion around polling results and the differences between results/findings of various research firms. Questions had been raised regarding the accuracy of data collection between Online Panels and Telephone Survey (landlines, cell and RDD).
Our objective was to compare the results from our Online Panel with our partner, Maru Blue, against the results of data collected using the telephone with our Live Agents.
Analysis of Online Panel over 4-month period
Over the last 3 months, support for the Conservative Party of Canada (CPC) has slowly been climbing in Ontario, while support for the Liberal Party of Canada (LPC) has been declining. Support for the New Democratic Party of Canada (NDP) has been steady at ~15%, while support for the Green Party of Canada (GPC), has remained steady at ~11%.
On September 7th, we had found that the CPC and LPC were tied in Ontario, both at 36% using our Online Panel. Since then, we’ve conducted another n=1030 with Online Surveys, and now find that the CPC is up 2% (38%), while the LPC has slipped 2% (34%). This is consistent with the trend line over the last 4 months in Ontario and is not necessarily due to recent events (Brownface/ Blackface scandal(s) recently uncovered and reported in the media).
Decided Voter Intent (Online Panel) over last 3 months
With respect to the 2-night online survey of n=1030 conducted on September the 18th and 19th, the CPC is at 38%, while the LPC is at 34%. The NDP is at 14% and the GPC is at 11%. The PPC continues to garner 2%. There seems to be very tight races in most regions, but not in the Halton/ Peel and York/ Simcoe and Durham Regions (which is considered the 905).
Live Agent Telephone Surveys (n=1005)
As stated, Campaign Research Inc. also conducted a telephone survey using Live Agents on September the 18th and 19th. This sample produced similar overall result for Ontario, with respect to the CPC having a 4% lead over the LPC, but significantly different regional results and significantly different results with respect to the support levels for the NDP (9%) and the GPC (8%).
The differences in support for the NDP and GPC with telephone and online samples has been noted by many other research firms in the past 2-week period.
With respect to regions, sample sizes are very small and produce some significantly different results than the online sample.
Merging both the Online and the Telephone Surveys
Merging both samples (Online and Telephone) gives a combined snapshot that levels out some of differences between the 2 methodologies.
“It is not certain to say that the recent Brownface/ Blackface scandal(s) that Prime Minister Trudeau is facing had any impact on our latest public opinion polling. What’s certain is that the online panel surveys and live telephone agent surveys are producing very similar numbers when considering a large enough sample size (Ontario-wide n=1000). It is also clear that the CPC has regained the lead in Ontario. Everyone wants to know the political impact that these scandals may be having – but it would be wise to wait at least another week before polling and reporting directly on that” – said Nick Kouvalis, Principal at Campaign Research Inc. Nick can be reached at firstname.lastname@example.org 519.791.9663
These studies were conducted by Campaign Research Inc between September 18th and 19th, 2019 through an online panel (n=1030) and telephone interviews (n=1005). Randomly selected Ontario adults who are members of Maru/Blue’s online panel, Maru Voice Canada, were provided with various incentives to respond. Telephone interview Respondents were randomly selected through landline/cell phone & RDD. The respondents were selected to reflect Canada’s age, gender and regional distributions in line with 2016 Statistics Canada census data. For comparison purposes, a probability sample of this size has an estimated margin of error (which measures sampling variability) of +/- 2.17%, 19 times out of 20.
The results have been weighted by education, age, gender, and region to match the population according to 2016 Census data. Certain areas or groups may be oversampled but have been weighted to reflect their proportion of Ontario’s population. This is to ensure the sample is representative of the entire adult population of Ontario. Discrepancies in or between totals are due to rounding.
The following screening question was asked in order to determine eligibility for participation in the study
"Are you 18 years of age or older and eligible to vote in federal elections?"