This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
During the COVID-19 pandemic in Canada, Prime Minister Justin Trudeau provided updates on the novel coronavirus and the government’s responses to the pandemic in his daily briefings from March 13 to May 22, 2020, delivered on the official Canadian Broadcasting Corporation (CBC) YouTube channel.
The aim of this study was to examine comments on Canadian Prime Minister Trudeau’s COVID-19 daily briefings by YouTube users and track these comments to extract the changing dynamics of the opinions and concerns of the public over time.
We used machine learning techniques to longitudinally analyze a total of 46,732 English YouTube comments that were retrieved from 57 videos of Prime Minister Trudeau’s COVID-19 daily briefings from March 13 to May 22, 2020. A natural language processing model, latent Dirichlet allocation, was used to choose salient topics among the sampled comments for each of the 57 videos. Thematic analysis was used to classify and summarize these salient topics into different prominent themes.
We found 11 prominent themes, including strict border measures, public responses to Prime Minister Trudeau’s policies, essential work and frontline workers, individuals’ financial challenges, rental and mortgage subsidies, quarantine, government financial aid for enterprises and individuals, personal protective equipment, Canada and China’s relationship, vaccines, and reopening.
This study is the first to longitudinally investigate public discourse and concerns related to Prime Minister Trudeau’s daily COVID-19 briefings in Canada. This study contributes to establishing a real-time feedback loop between the public and public health officials on social media. Hearing and reacting to real concerns from the public can enhance trust between the government and the public to prepare for future health emergencies.
The World Health Organization declared COVID-19 to be a global public health emergency on March 11, 2020 [
Since the COVID-19 outbreak, social media has become the most accessible source for obtaining news and health information and for exchanging opinions. Given their self-isolation, people extensively post opinions, express emotions, and exchange ideas about COVID-19–related policies on the web [
This study has two goals. The first is to examine whether YouTube comments are a useful source of public opinions and priorities on COVID-19; the second is to identify public responses and discourses related to Prime Minister Trudeau’s COVID-19 policies in Canada over time. According to Bora et al [
We followed the text mining pipeline, including data preparation and data analysis [
We used a purposive sampling approach. Our sampling frame was Prime Minister Trudeau’s COVID-19 daily briefing videos, which were published on the official CBC channel on YouTube [
We present the characteristics of these videos in
The data set of the study consisted of all the comments posted by YouTube users under each of the 57 videos. We used YouTube’s open application programming interface to collect all the comments under each of Prime Minister Trudeau’s daily briefing videos on May 25, 2020. We coded the information retrieval script in Python, version 3 (Python Software Foundation). We retrieved a total of 47,149 comments from these 57 videos. After removing 1017 non-English and duplicate comments, 46,732 comments remained as our final data set, as shown in
We treated the set of comments published under each video as a separate document for the topic modeling (eg, documents 1-57). Thus, we obtained a total of 57 documents for topic modeling.
Data preprocessing chart.
We preprocessed the raw data to ensure data analysis quality. Preprocessing is an important task for analyzing YouTube comments, as it cleans the data by diminishing its complexity to provide better results [
We removed handles (@user) and their content, as they did not contribute to the analysis.
We removed all non-English characters (non–American Standard Code for Information Interchange [ASCII] characters) because this study focused on English messages.
We removed words that did not make sense, such as “#newYork sdaaawd asjdasd @!!!!” We used the Nostril evaluator [
We removed stopwords to ensure better results [
We applied a stemmer. A stemmer is a rule-based process of stripping the suffixes (eg, “ing,” “ly,” “es,” “s”) from words. A large number of conjugates were removed.
We used latent Dirichlet allocation (LDA) [
The thematic analysis enabled us to interpret the patterns, topics, and themes from the unsupervised machine learning processes [
Our sampling frame included 57 of Prime Minister Trudeau’s daily briefing videos published on the CBC YouTube channel between March 13 and May 22, 2020. We describe the statistics of the videos in
Descriptive statistics of the collected videos (N=57).
Characteristic | Range | Mean (SD) |
Number of views | 74,924-839,921 | 364,487 (205,984) |
Likes | 492-4405 | 1826 (1059) |
Dislikes | 290-2750 | 676 (344) |
Comments | 197-2653 | 830 (544) |
The number of views for the sampled daily briefing videos over time (n=56).
Like counts and dislike counts for the sampled daily briefing videos over time.
We retrieved and downloaded the associated 46,732 comments for each of the 57 videos as our final data set. We identified 1 to 3 salient topics for each of the documents (#1 to #57) associated with each daily briefing video, resulting in 168 topics. We present the topic modelling results for each of the 57 documents in
Occurrences of each theme from March 13 to May 22, 2020. PM: Prime Minister; PPE: personal protective equipment.
In addition to counting the number of occurrences, we also tracked the prominent themes over time, as shown in
Prominence of each theme from March 13 to May 22, 2020. Early, day 1-10; mid-month, day 11-20; late, day 21 and after. PM: Prime Minister; PPE: personal protective equipment.
We selected the 50 comments with the highest numbers of likes and the 50 comments with the highest numbers of replies. We present 19 comments that ranked in the top 50 most liked and replied in
Most popular comments, origins (daily briefing dates), assigned themes, and numbers of replies and likes.
Comment | Date of origin (2020) | Theme | Replies, n | Likes, n |
“Why are we still not checking people at the airports?” | March 28 | Strict border measures | 104 | 488 |
“They should be pausing all mortgages and rent payments, the CRA can’t give you money by the 1st” | March 22 | Rental and mortgage subsidies | 99 | 525 |
“I really hope when this is defeated that our society doesn’t go more Orwellian” | May 04 | Good wishes | 88 | 230 |
“MR TRUDEAU: All you need to do is issue a moratorium stating that all Canadians with a rental lease or a mortgage is immediately suspended for 3 to 6 months. After which their existing monthly payments will restart and continue as usual. This will put at least 30% more money in every person’s account fairly. These funds can be used to pay cred it cards, car loans, food etc. It is a brilliant idea already being implemented in the USA by a number of Banks and institutions. This is simple and would help all Canadians. Obviously no penalty would by involved and this would be a Government approved legal postponement of all rents and mortgages for the period stated.” | March 24 | Individuals’ financial challenges | 78 | 246 |
“Its time to come together and support each other and question everything.” | May 04 | Good wishes | 68 | 802 |
“I don’t care about staying home. I’m a homebody anyways. I care about being poor and ending up completely homeless after the whole Covid19 crisis ends. My landlord already wants me to move out by August, and no one knows if it will have died down by then or if I’ll have a job by then so I can safely move.” | March 29 | Individuals’ financial challenges | 67 | 165 |
“Meanwhile, Canadian Tire, Lowes, Home Depot, Costco all the checkout attendants facing hundreds of customers per day and not one of them was wearing a mask and that was yesterday March 21st” | March 22 | Quarantine | 67 | 356 |
“Comments are allowed? Interesting” | March 26 | Open government | 66 | 335 |
“The scariest YouTube title ever” | April 09 | Emotion | 63 | 340 |
“I just want to know if these people are secretly getting haircuts. The rest of us will look like sheep dogs if this continues.” | March 26 | Emotion | 54 | 178 |
“I wish you didnt do the voice over, he is already doing it bilingually. we dont need to hear it two times in English besides the translation is badly done.” | April 14 | Not related | 54 | 615 |
“Anybody coming into Canada from other countries should have mandatory isolation for 14 days.” | March 28 | Quarantine | 53 | 222 |
“I dont understand why people still went on cruises even tho they were specifically told not to a longggggg time ago” | March 28 | Emotion | 51 | 269 |
“Lets have open questions from the public.” | April 09 | Open government | 40 | 203 |
“Close the border dont exempt American citizens. Close it down.” | March 17 | Quarantine | 38 | 304 |
“We should not rely on the US, like we are the 51st state!” | March 05 | Emotion | 34 | 180 |
“Dont have a social gettering of more than 10 people. Thats the guideline in South Korea. I think 50 is too much^^” | March 17 | Quarantine | 33 | 415 |
“Dr Tam waiting for ‘evidence’ of human to human transmission is disgraceful especially considering entire cities in China had already been locked down and the Three early German cases traced back to a Chinese business woman had already been reported. The precautionary principle should have been followed and Canadians deserve answers. We deserve better and Dr. Tam let us all down” | April 14 | Quarantine | 32 | 162 |
“Never in my life have I heard so many answers that dont answer a question. Ever. Lol” | April 14 | Open government | 32 | 317 |
This study is the first to investigate and track public opinions and reactions to Canadian Prime Minister Trudeau’s daily COVID-19 briefings. This study uses a machine learning approach and longitudinally examines public discourse about Prime Minister Trudeau’s daily briefings over time. The results reveal several prominent themes over time, including strict border measures, public responses to Prime Minister Trudeau’s policies, individuals’ financial challenges, government financial aid for enterprises and individuals, and essential work and frontline workers. In addition, the themes of vaccines and reopening were highly discussed from late April. The study demonstrates that YouTube comments are a useful source of public opinions and priorities during global health emergencies.
This theme refers to comments that discuss topics such as border entry policy, international travel, and border screening measures. The majority of the comments in this theme requested that the government implement stricter border measures to prevent transmission from carriers who have or do not have COVID-19 symptoms, exemplified by the comment “CLOSE THE BORDER... Stop all flights from Italy and Europe NOW.” Public discourse on this topic started in mid-March and remained a hot topic until mid-April. The results correspond with a policy announced by the Canadian government on March 18, 2020, that implemented a ban on all foreign nationals entering Canada, including restrictions on all nonessential travel at the Canada–United States border. Discussions of border measures were not prominent in late April and early May but reappeared in mid-May, consistent with the border closure agreement between Canada and the United States, which was valid until May 21.
Our results suggest that the “border policy issue” has continuously been a concern for Canadians, who support strict border measures to prevent the spread of COVID-19 in Canada. Previous studies demonstrated that air transportation has the potential to spread influenza, such as A/H1N1 [
Public responses to Prime Minister Trudeau’s policies refer to public opinion comments on the Canadian government, parties, or the Prime Minister himself. The majority of comments include opinions regarding the government’s performance and policies during the COVID-19 pandemic. One representative comment is “Respect and listen the Prime Minister, he is trying his best with the Government to keep everyone in our Country safe from Covid19 with emergency funds for those that need it.” Compared to other themes, this theme is more popular, and it continuously received public attention from mid-March to late May. The popularity of this theme is due to the positive interaction dynamic between the general public and the government of Canada, which encourages people to communicate with government organizations on social networks. This interaction can be traced back to 2011, when Canada acknowledged the need to use social media to interact with the public for the first time [
Government Health Officials WHAT? They are months behind! Recommended NO MASKS only because they did not prepare, and they hadn't enough masks.
Previous studies have examined mixed sentiments toward governments’ policies in emergencies. During the 2009 H1N1 influenza pandemic, more than half the US population felt positive emotions toward the government’s policy responses, while a certain number of people did not [
Financial challenges refer to comments that include people’s discussion, concerns, or policy proposals related to financial loss during the pandemic, such as “having no income for essential living costs due to lost jobs.” One representative comment is:
I’m more worried about the financial issue than the health one right now. We are doing everything we can health-wise in this family, but my son is now out of work, and we know hubby will be soon.
Our results show that the public has been frequently and continuously discussing finance-related topics since Prime Minister Trudeau’s first daily briefing. The assessment of the impact of the COVID-19 pandemic on the labor market in Canada shows that the pandemic drove a 32% reduction in aggregate weekly hours worked from February to April 2020, along with a 15% reduction in employment [
We found a positive correlation between the government’s policies supporting individuals and public attention to financial challenges. CERB was launched on April 6, 2020; nearly 3.5 million Canadians applied for this benefit in the first week, and this number increased to 7.12 million on April 24, 2020 [
Discussion about enterprises and individual financial aid plans was prominent from March to May 2020. This theme includes discussions of government financial aid plans and programs supporting either enterprises or individuals, such as CERB, CESB, the Canada Emergency Wage Subsidy, and the Canada Emergency Business Account. This theme is different from theme 3, individuals’ financial challenges, which focuses on individuals’ concerns and discussions of their financial obstacles during the outbreak. In late March, the theme of government financial aid became prominent when the government administration firstly announced a number of economic stimulation programs, exemplified by this comment: “If your business/source of income was interrupted due to COVID-19, whether you lost your business or not, you could still try to apply, as you would be in true need of help at this point.” Our results confirm those of previous research on emergency individual financial aid during the pandemic and the importance of financial aid programs for businesses [
Essential work and frontline workers was a prominent theme in the comments related to Prime Minister Trudeau’s daily briefings, such as the welfare system in Canada, incentive pay for essential work, and protection of frontline workers. One representative comment stated, “The government should be incentivizing those who have to work during a pandemic, even a $500/month incentive.” This was an active public discussion topic in mid-March; it lost public attention for 20 days, and then became prominent again in mid-April. Our results are consistent with an earlier study that shows that frontline workers should be supported, and public support of frontline workers has been related to a speech by the Prime Minster of India [
Public discussion of vaccines is another prominent theme that emerged in April. We classified all relevant topics in this theme, such as the development of vaccines, exemplified by this comment: “It could take years to come up with a vaccine, you’ve been doing a good job, Trudeau. Don’t it up by making statements you can’t back up.” Previous studies analyzing Tweets related to COVID-19 also found that vaccines were a highly discussed topic [
There are several limitations to this study. First, this paper is not based on random sampling. We purposively collected comments that were posted by YouTube users under Prime Minister Trudeau’s briefing videos. They only represent a portion of the public opinions regarding Prime Minister Trudeau’s daily briefings during the COVID-19 pandemic, which causes biases. Second, our sample size is relatively small compared to the large amount of data on Twitter. Third, we treated secondary comments (ie, comment to comment) in the same manner as first-level comments for the study aims. Thus, this study has limitations in presenting the structural relationship between first-level and second-level comments. Finally, we retrieved all videos, comments, and associated data on May 25, 2020, three days after the last day of the Prime Minister’s daily briefing. It is noted that the evaluations could be different due to the increased number of comments added over time.
This study is the first to analyze and track the public discourse of Canadian Prime Minister Justin Trudeau’s daily briefing on COVID-19 on social media. Machine learning analysis was applied to 46,732 English YouTube comments from 57 videos from March 13 to May 22, 2020, and the results suggest that several prominent themes should be noted by public health agencies and policy makers, such as strict border measures, public responses to Prime Minister Trudeau’s policies, individuals’ financial challenges, and government financial aid for enterprises and individuals. Recommendations for future work include (1) further validating YouTube comments as a source of information on COVID-19 and (2) strategies that can be considered by government and public health agencies to strengthen the real-time feedback loop between those agencies and the public on YouTube. Hearing and reacting to real concerns from the public can enhance trust between the government and the public to prepare for future health emergencies.
Descriptive characteristics of the sampling frame (N=57 videos).
Modeling of prominent topics for YouTube comments on each daily briefing (video dates, topics, bigrams, and representative examples).
Salient themes, topic keywords, and representative comments.
American Standard Code for Information Interchange
Canadian Broadcasting Corporation
Canada Emergency Response Benefit
Canada Emergency Student Benefit
latent Dirichlet allocation
CZ and JX contributed equally as coauthors.
None declared.