Across the English-speaking developed countries, the broad story told about journalism is much the same: as the consumption of traditional journalism sources continues to decline, those who do consume news increasingly gravitate toward partisan sources. These communities of news consumers turn into echo chambers where citizens hear only the views that they themselves already hold. One consequence is that the traditional broad-based media seems increasingly alien and out of touch. And so, people lose their trust in mainstream media and instead seek out only those media sources that reflect their increasingly polarized opinions.
The broad strokes of this narrative are well-documented in the United States. But does this general narrative apply in Canada?
It turns out the story here is not so simple. We conducted a study asking Canadians about their news consumption habits, their political preferences, their beliefs about the world, their trust in news organizations and what (if anything) they want the government to do to support Canadian media. The study was commissioned by the Max Bell School of Public Policy at McGill and executed by the Loewen Lab at the Munk School of Global Affairs and Public Policy. The survey was conducted online using a sample provided by Qualtrics between August 30 and September 24, 2018. We surveyed 2,245 English-Canadian respondents, including English-speakers in Quebec, and 452 French-speaking Quebecers. All results were weighted by age and gender within each province based on targets from the 2016 census.
We find that as in all competitive democracies, Canadians do disagree about the policies they support. And these policy preferences are related to what news sources they consume: liberals prefer left-leaning news sources, and conservatives prefer right-leaning sources. However, citizens from across the political spectrum, who support different policies and consume different news sources, nevertheless share a common knowledge of baseline “facts” about our society.
Moreover, and perhaps most surprisingly, across these groups there is widespread trust in Canadian mainstream news outlets such as the CBC, the Globe and Mail, the Toronto Star and the National Post. Finally, this relatively high level of trust is a key factor in explaining why some Canadians are willing to back greater government support for Canadian media and greater taxation of Internet platform companies that are seen as contributing to the demise of traditional media.
To investigate how Canadians’ self-reported news consumption correlates with their views, we asked individuals how frequently they read news on various websites. Our list was generally representative of both English and French media in Canada and abroad. Based on how often individuals reported visiting either the 10 English-language sites (figure 1) or the 10 French-language sites (figure 2), we divided the respondents into clusters.
Membership in a cluster is determined by two factors: first, how much news people read overall, and second, the balance of right-wing, neutral and left-wing news sources they consume. Those whose overall news diet is greater than the median are considered high use, and those with consumption below the median are considered low use. This gives us six groups: low-use left-wing readers, low-use neutral readers and low-use right-wing readers on the one hand; and high-use left-wing readers, high-use neutral readers and high-use right-wing readers on the other.
There are three main reasons for composing clusters in this fashion. First, dividing users according to their usage allows us to see if those who are the principal consumers of news sites differ systematically from those who are not. The future of Canadian media likely relies much more on the former group than on the latter. Second, by paying attention to usage we can better understand how it is correlated with political polarization and other outcomes thought to be driven, at least partially, by the volume of news that individuals consume. Third, by also paying attention to the ideological composition of a news diet we can understand how much the political content of news drives differences in the views of Canadians.
These groups differ in what news they read, but demographically, they don’t look all that different from one another. For example, in English Canada the average age in each group ranges from 42 among neutral high-use readers to 51 among neutral low-use readers. Among francophone Quebecers, neutral high-use readers’ average age is 38, and it is 53 among neutral low-use readers. Gender differences are starker: among right-wing low-use readers, just 34 percent are female. Among left-wing low-use readers, 60 percent are female. We see an even starker divide in Quebec where among right-wing high-use readers only 34 percent are female, while a surprising 88 percent of left-wing low-use readers are female.
These groups differ more substantially in what they want from government. For example, we asked respondents to agree or disagree with the following statement: “There should be more free trade with other countries, even if it hurts some industries in Canada.” Agreement varies significantly. In English Canada, just 32 percent of low-use left-wing consumers agree with this statement, compared with 47 percent and 48 percent of high-use and low-use right-wingers, respectively. Among French-speaking Quebecers, just 27 percent of low-use neutral readers agree, contrasted with 59 percent of high-use neutrals and 37 percent of high-use right-wing readers.
Similarly, when asked if Canada should increase the number of immigrants it admits each year, 56 percent of English high-use right-wing readers disagree, while 40 percent of high-use left-wing readers share that opinion. Opinions also vary among French-speaking Quebecers: 59 percent of low-use neutral readers disagree that immigration should be expanded, compared with only 34 percent of French low-use left-wing readers.
However, while readers from different clusters certainly disagree on political issues, interestingly, they do not disagree very much on key facts about the world. We asked respondents to evaluate 10 statements covering issues such as climate change, middle-class incomes, human rights, health care, refugees and the gender gap. There was no shortage of potential for disagreement here on the substance of the facts. Nonetheless, it was only on the issue of climate change that stark differences in views about the facts of the world differed. Even then, differences are not enormous: when asked if human activity is the main cause of climate change, 77 percent of English high-use left-wing readers say this is certainly or probably true. The share for high-use right-wing readers is just 12 points lower, at 65 percent. In French-speaking Quebec, the difference between those groups is even smaller, with just a two-point gap (75 percent and 73 percent for left-wing and right-wing high-use readers, respectively).
What is the big takeaway from this? Observers often worry that the modern media environment is polarizing citizens into different communities separated by their views on facts about the world. This does not appear to be the case in Canada. How much citizens read and what they read is simply not strongly related to how they interpret basic facts about the world.
If Canadians are not deeply polarized over facts, nor are they polarized over their views of the organizations that provide the news, at least among the mainstream organizations. For example, we asked respondents how much they trust the news provided by several outlets. Among the “Big Four” sites — CBC News, the Toronto Star, the Globe and Mail and the National Post — trust is notably high. Across all English respondents, the share indicating that they trust CBC News “somewhat” or “very much” is 92 percent; it is 85 percent for the Toronto Star, 91 percent for the Globe and Mail and 86 percent for the National Post. In sum, the overwhelming majority of respondents across all groups say they trust each of these sites somewhat or very much.
This is not true for newer and smaller outlets. For example, among English respondents, between 33 percent and 67 percent say they do not trust the Rebel at all. For the National Observer, the minimum “not at all” share is 24 percent, and it goes as high as 45 percent. Even the Toronto Sun registers “not at all” levels of trust of as much as 31 percent among some groups.
These patterns are mirrored when we ask respondents about their trust in US, British and French news outlets. The percentages of English Canadians saying they do not trust the New York Times, Wall Street Journal, CNN and the BBC at all are 13 percent, 12 percent, 16 percent and 10 percent, respectively. By contrast, they do not trust newer American outlets, whether it be Breitbart (55 percent), Vice (34 percent), Vox (42 percent) or Buzzfeed (38 percent). Similar numbers were found among French-speaking respondents. The percentages of respondents who indicate that they have no trust at all in the New York Times, the Wall Street Journal, the Washington Post and Le Monde are 15 percent, 21 percent, 20 percent and 17 percent, respectively. Rates of no trust are much higher for Vice Quebec (38 percent), for example.
Overall, this is a healthy picture. While Canadians disagree on politics, and their media consumption habits tracks this disagreement, they agree on the basic facts of the world, and they largely trust their main media outlets. This is somewhat at odds with the usual story of a group of polarized citizens consuming their own set of “facts” within insular media environments.
The reality nonetheless is that these same mainstream media outlets are struggling. Do Canadians wish to grant them some extra support? This picture is less sanguine. Just 22 percent of English respondents and 23 percent of French respondents agree that the federal government should provide news outlets with direct subsidies. Similar numbers (27 percent and 29 percent) agree that they should receive more favourable tax treatment. However, a majority are supportive of the government levying more taxes on advertisements placed on Google or Facebook and using that money to support Canadian news outlets, with 38 percent of English Canadians and 51 percent of French-speaking Quebecers agreeing.
Taken together, this is at best tepid backing for government support of the Canadian media. There is no obvious silver lining here. Even among those who indicate they have some or a lot of trust in each of the big four news outlets, support for government action is not higher: 22 percent support direct subsidies, 28 percent support more favourable tax treatments, and 39 percent support taxing Google and Facebook to support Canadian outlets.
This study provides only a snapshot of Canadian political division and perceptions of the news media. But it does offer a counterpoint to the narrative of disagreement not just over policy, but over basic truths about the world, that has taken root in the United States. While Canadians are divided, and while our political discourse is becoming harsher, filled with misinformation and vulnerable to malicious microtargeting, particularly on social media, we have two important backstops. First, we share a common set of “facts” about core issues in our contemporary political debate. There is little sense of the “alternative facts” rhetoric that has poisoned the American discourse. Second, we broadly trust the organizations that are providing us with the quality journalism that safeguards against harmful speech and false news.
But there remains a significant policy challenge: Canadians don’t really support government aid to the very organizations that they say they trust the most but that are, aside from the CBC, struggling or even on the verge of bankruptcy. It is not clear whether the low level of support is due to a belief that the industry already receives enough state aid (through, for example, the subsidy to the CBC); perhaps the reason is that news consumers don’t realize the extent to which the oxygen that fuels their social feeds comes from the dying breaths of the news dinosaurs. The one glimmer of hope on this front is the relatively high levels of support for platform taxation. This is increasingly looking like the most promising policy window, with money from the platforms redirected to journalism without looking like government aid.
© Policy Options