Participant characteristics

Participant sociodemographic information for each survey are shown in (Table 1). The sample predominantly consisted of individuals aged 25 years or older (97.0–98.2%), were living in urban areas (39.2–40.5%), were from Ontario (34.4–38.7%), did not identify as a minority (63.8–68.3%), and had completed some sort of postsecondary education such as a bachelor’s degree (24.4–27.9%) or college/non-university degree (21.3–23.0%). Approximately half of our sample had annual household incomes of $60,000–$219,999. There was a relatively even balance of males (45.8–50.1%) and females (47.1–52.3%) in our sample. The response rate ranged from 10.2–12.6% across the survey waves. See (Table 1) for the response rate of each individual survey.

Table 1 General sample characteristicsCOVID-19 vaccination trends over time

We observed a relationship where over time as the number of individuals who got vaccinated increased, individuals who were willing to get vaccinated decreased. This suggest that individuals who were wanted a vaccine were able to acquire one. From W25(Capacity Limits) to W32 (Mask Mandate Removed), willingness to receive a third dose of the COVID-19 vaccine decreased by 65.6 percentage-points, which is similar in magnitude to the 70.3 percentage-point increase in prevalence of those who received a third dose. This shift was largest between W25 (Capacity Limits), −W27 (Fig. 1E, F) (Table S1). The prevalence of those who were unwilling to receive a third dose of the COVID-19 vaccine did not vary considerably over time, ranging from 6.1–9.9% (Fig. 1E, F) (Table S1).

Fig. 1

Time trends for recurring survey questions

Over time people were willing to vaccinate decreased, and people who were vaccinated increased suggesting that people who wanted to receive a vaccine were able to get one.

Public attitudes towards the COVID-19 pandemic over time

Participants’ outlook on the COVID-19 pandemic generally improved over time. From W26 (Border Measures) to W32 (Mask Mandate Removed), the prevalence of those who perceived that “the worst is behind us” increased by 26.5 percentage-points. In contrast, the prevalence of those who perceived that “the worst is yet to come” decreased by 18.9 percentage-points (Fig. 1A) (Table S1). Similarly, from W25 (Capacity Limits), to W29 (Freedom Convoy Ended), the prevalence of those who expected that society will “return to normal” in the short-term (within 6 months or less) increased slightly by 2.9–3.9 percentage-points, whereas the prevalence of those who expected that society will “return to normal” in the long-term (within 1 year or more) decreased 5.4–7.6 percentage-points (Fig. 1B) (Table S1).

Participants’ self-reported stress levels were relatively similar to one another and remained fairly stable over time, such that participants most commonly reported a persistent “moderate” level of stress experienced during the COVID-19 pandemic (Fig. 1C) (Table S1). There appeared to be a convergence of self-reported stress levels over the study period.

Public attitudes towards COVID-19 vaccine mandates over time

With regards to public attitudes towards wearing masks in stores and while using public transit, participants most commonly reported that they “strongly agree” with these measures, although this decreased from 65.7 to 38.6% between W25 (Capacity Limits), and W32 (Mask Mandate Removed) (Fig. 1D) (Table S1). In contrast, those who “strongly disagree” with these measures was the second most common response, which increased from 11.6 to 19.4% during this same time period (Fig. 1D) (Table S1). Similarly, with regards to public attitudes towards COVID-19 vaccine certificates and proof of vaccination measures, “strongly support” was the most common response and fell from 66.0 to 43.1% between W25(Capacity Limits), −W32 (Mask Mandate Removed), whereas “strongly oppose” was the second most common response and rose from 15.9 to 20.6% during this same time period (Fig. 1G) (Table S1).

The majority of participants believed that their province was reopening at “about the right pace”, which remained relatively stable over time (33.0–35.4%) between W28 (Emergency Act)–W32 (Mask Mandate Removed) (Fig. 1H) (Table S1). Amongst participants who did not believe that their province was reopening at the right pace, they increasingly believed that the reopening was at a faster pace than desired. From W28 (Emergency Act) to W32 (Mask Mandate Removed), the prevalence of participants who believed that their province was reopening “too slowly” decreased by 10.9%, whereas the prevalence of those who believed that the reopening “quicker than desired” or “too quickly” increased by 8.8 and 8.4%, respectively (Fig. 1H) (Table S1). Notably, the intersection between those that responded “too slowly” vs. those that responded “quicker than desired” or “too quickly” occurred around W31 (Statement to Remove All COVID Restrictions) at a prevalence of approximately 19% (Fig. 1H) (Table S1). A slight majority of participants believed that “we should keep COVID-19 related restrictions in place for the time being”, which remained fairly constant over time, ranging between 52.3–55.8% (Fig. 1M) (Table S1). In contrast, 39.8–41.4% of participants believed that “it is time to bring an end to COVID-19 related restrictions”, which also remained fairly constant over time (Fig. 1M) (Table S1).

Trust and approval towards government agencies over time

Over time, participants trust in the various levels of government and the Public Health Agency of Canada varied. In general, trust and approval towards the three different levels of government and PHAC improved.

With respect to PHAC, participants “approve” response rate increased 5.4% over time (from W28 (Emergency Act) to W32 (Mask Mandate Removed)) and the “strongly approve” response rate increased by 3.2% during the same time period (Fig. 1J) (Table S1). The “disapproval” response rate remained steady. However, the percentage of participants who stated they “strongly disapprove” decreased 5.5% over time (W28 (Emergency Act) to W32 (Mask Mandate Removed)) (Fig. 1J) (Table S1).

With respect to the federal government, the percentage of participants who responded that they “approve” increased 6.9% from W28 (Emergency Act) to W32 (Mask Mandate Removed) and the percentage who responded that they “strongly disapprove” decreased by 8.6% during the same time period (Fig. 1I) (Table S1).

With respect to the municipal government, the percentage of participants who stated that they “approve” initially decreased by 3.4% from W27 (Freedom Convoy Began) to W29 (Freedom Convoy Ended) but increased by 7.5% from W29 (Freedom Convoy Ended) to W32 (Mask Mandate Removed) (Fig. 1L) (Table S1). The percentage of participants who stated that they “neither approve nor disapprove” decreased by 5.1% from W29 (Freedom Convoy Ended) to W32 (Mask Mandate Removed) (Fig. 1L) (Table S1). Disapproval remained steady, but the percentage of participants who stated that they “strongly disapprove” decreased slightly by 2.9% from W28 (Emergency Act) to W32 (Mask Mandate Removed). The percentage of participants who stated that they “strongly approved” rose steadily by 3.7% between W27 (Freedom Convoy Began) and W32 (Mask Mandate Removed) (Fig. 1L) (Table S1).

With respect to the provincial government, the percentage of participants who stated that they “strongly disapprove,” was the most common response and this remained constant with a slight decrease of 0.6% from W27 (Freedom Convoy Began) to W32 (Mask Mandate Removed) (Fig. 1K) (Table S1). The percentage of participants who stated that they “approve” among participants initially dropped by 2.6% from W27 (Freedom Convoy Began) to W29 (Freedom Convoy Ended) but then increased by 6.2% from W29 (Freedom Convoy Ended) to W32 (Mask Mandate Removed). The percentage of participants who stated that they “strongly approve” increased by 2.7%, and those who stated that they “disapprove” decreased by 3.0% during this period (Fig. 1K) (Table S1).

Public attitudes towards freedom convoy protestors

Participants’ support for the Freedom Convoy protests was highly polarized. The vast majority of respondents were aware of this protest, with 48.3 and 44.1% following this protest “very closely” and “somewhat closely”, respectively (Fig. 2B) (Table S2). The majority of respondents “strongly oppose” (51.9%) the Freedom Convoy protests, whereas 20.3% “strongly support” the protests (Fig. 2C) (Table S2). Similarly, the majority of respondents “strongly disagree” (58.7%) with supporting a similar Freedom Convoy protest in their own community, whereas 17.5% “strongly agree” with this proposition (Fig. 2D) (Table S2). Further, 24.5% of respondents responded that they “strongly support” mandating proof of vaccination for truckers entering Canada from the United States, whereas 14.1% responded that they “strongly oppose” these measures. However, caution should be taken when interpreting these findings due to the substantial proportion of missing data (50.1%) (Fig. 2A) (Table S2). The majority of participants responded that they “strongly oppose” (60.8%) the Freedom Convoy organizers’ “memorandum of understanding” that called upon the Governor General to override vaccine mandates, whereas 14.2% responded that they “strongly support” this memorandum (Fig. 2G) (Table S2).

Fig. 2

Timeline of survey administration and coinciding key social events

Fig. 3

Bar plots for Freedom Convoy questions

We also observed polarization in participants’ identity congruence with the Freedom Convoy protestors. The majority of respondents responded that they “strongly disagree” (57.7%) with having a sense of shared identity with the Freedom Convoy protestors, whereas 14.3% responded that they “strongly agree” with this belief (Fig. 2E) (Table S2). With regards to authority, participants reported that the Freedom Convoy protests reflected their own views on “questioning authority” (38.3%) nearly two-fold more than “obedience” (19.3%) (Fig. 2I) (Table S2). Participants rated “reason and evidence” (57.4%) nearly three-fold more commonly than “morality” (22.1%) as the most important factor when evaluating the protests (Fig. 2J) (Table S2). Interestingly, with regards to self-expression, there was a similar prevalence of participants that reported that these protests reminded them of their own perspectives on “creativity” (37.5%) and on “good behavior” (32.8%) (Fig. 2K) (Table S2). With regards to sense of order, participants reported that these protests reflected their own perspectives on “order” (56.8%) more-than-three-fold more than on “openness” (17.2%) (Fig. 2L) (Table S2). Finally, participants rated “trust in science and experts” (63.9%) more-than-five-fold more than “skepticism of science and experts” (12.5%) as the most important factor when evaluating the protests (Fig. 2M) (Table S2).

Similarly, participants’ perceptions of the ideological principles of the Freedom Convoy protests was also highly polarized. Further, participants most commonly responded that they “strongly agree” (44.4%) with the belief that the Freedom Convoy protests reflect authoritarian populism in a similar manner to recent social and political events in the United States, whereas 20.8% of respondents responded that they “strongly disagree” with this belief (Fig. 2F) (Table S2). Similarly, the majority of participants believe that the protests reflect the views of a fringe minority (54.5%), whereas 33.7% of respondents believed that they reflect the broader public anger and concern over vaccine mandates during the COVID-19 pandemic (Fig. 2F) (Table S2).

Multivariable regression analyses

A multivariable logistic regression analysis was performed to examine the relationship between sociodemographic characteristics and public approval of proof of COVID-19 vaccination mandates in Canada (Table S3). The odds ratios produced by the multivariate logistic regression revealed that none of the independent variables were statistically significant at a α = 0.05.