How do people use social media?
It turns out, a lot of Americans turn to these platforms for news. The invaluable Pew Research Center just updated us on their latest findings, and the results should be very useful for anyone thinking about digital news or media in general.
To begin with, Pew found a lot of people find news content across social media. “About half (53%) [of U.S. adults] say they get news from social media…” And yet at the same time more than half of that population don’t trust the stuff: “six-in-ten (59%) of those who at least rarely get news on social media say they expect that news to be largely inaccurate…”
Which platforms do we prefer? Here’s the breakdown:
More than one third of Americans turn to Facebook for news, which is a big chunk of the population. Around one fourth fire up YouTube for this purpose. Following those is Twitter, then smaller and smaller numbers for the rest.
(There’s an interesting twist to this point: “These lower percentages for news use are in some cases related to the fact that fewer Americans report using them at all, compared with the shares who use Facebook and YouTube.
“)
Next, authors Amy Mitchell and Elisa Shearer slice the results by demographics, and I found these results utterly fascinating. Take a minute to work through this:
The gender breakdowns are very clear at times, with women massively preferring Facebook and Instagram compared with men, and the reverse for Reddit.
Age reveals the persistent truth in the much derided age and tech correlation – i.
e.., across the board, folks over 50 and especially beyond 65 are still less likely to use these technologies, at least for the purpose of finding news. Note, too, the preference of Reddit, Twitter, and Instagram among the under 30s.
Increased education seems to drive Twitter, Reddit, and above all LinkedIn news hounding.
Lower amounts of education point users towards Facebook and YouTube. I am very curious how this distinction maps onto which social media platforms higher education teaches.
Racial differences are also interesting.
Among people of color, Black and Hispanic populations prefer Instagram, while LinkedIn leads for Asians. White people tend towards Facebook and Reddit, with a sharp dropoff for Instagram.
Political parties are intriguing, with Democrats simply more likely than Republicans to use social media for use across all platforms. The gap is most clear with Reddit, yet also very visible with Instagram and Twitter.
What can we take away from this?
To begin with, there are practical lessons for academic communication. To the extent a college or university is concerned with its population’s news diet, the Pew findings come in very handy. If a campus wants to reach out to specific populations under the aegis of news, it now has helpful targeting information. And campuses seeking to trace how they are talked about can similarly prioritize.
I wonder how this impacts information and digital literacy programs. Should it nudge instructors to foreground certain social media platforms for extra attention?
Will the increasing social divide by educational attainment continue to play out through our social media news gathering habits? That is, will we see low information voters (for example) sticking to YouTube, while those with graduate degrees hew to LinkedIn and Twitter, creating another pair of information bubbles?
I do wonder as well about the differences by race. Is Instagram the closest thing we have to a common space in this topic? Conversely, will Reddit and Facebook become the digital news equivalent of white flight suburbs?
Naturally I would like to see more data and analysis. I’d love to see more about use behavior – how they use social media (following hashtags? making lists?), how often they make or share content versus sucking it down. Jon Lebkowsky told me he wanted to learn if people obtained news from social media content, or from professional news media publishing through social media. And I wonder about platforms missing from this report, namely blogs.
On a personal level, I find some of this strange. I do not look for news on YouTube, LinkedIn, or Instagram. I do get some slices of information from a couple of Reddit boards. On the other hand, I follow gender and education (yet not age) in using Twitter for some news. And I apparently fly in the face of my gender, age, and education in using several good Facebook groups for some bits of news. As you might suspect from the previous paragraph, I get much more news from my RSS reader.
What do you think of the Mitchell and Shearer report? I’d love to hear your thoughts. The comments box stands open.
useful data, great implications explication. Thanks for sharing, Bryan
None of the online news outlets that I depend on were listed, which is why I wonder if this survey (I hate surveys — remember the election polling in 2016?) is subtle pillorying.
Well, to be clear, none of these are actually news outlets. They’re social media platforms that play an important role in news aggregation and filtering for those who use them, which especially means Generation X and younger.
Glen, may I ask which ones they missed?
I noted some in my post.
What a fascinating breakdown! The difference by gender for Reddit doesn’t surprise me much, but I’ll admit that the one for Facebook does. And I wish I could see similar results in a number of other countries.
The gender breakdowns really struck me as well. Fascinating – and I hope we get data from other nations, too.
I agree with Jon Lebkowski: there is a lot of devil in the details. Just like I have a problem with hard distinctions between “in-person” and “online” learning, I have a lot of problems with characterizing news consumption in this manner. Most people get news from a lot of sources. I, myself, sometimes consume news on YouTube if I want to watch some analysis from one of the many news organizations that post there. YouTube is also a good place for news commentary. I, however, recognize the relative weight and limitations of news across the range of media available. Critical distinctions are the difference between visual and textual sources of information (and how to mix them effectively) as well as qualitative measurements of the information provided.
Television news has been decried for decades as being “information lite” with the most watched segments on local news being sports and weather and the rest of the coverage being sandwiched into sensationalism (“if it bleeds it leads”) and occasional “investigative reporting” whose quality ranges from “gotcha” reports to truly reflective pieces. That last category makes up such a vanishingly small part of the local news ecosphere that I simply refuse to invest the effort to separate the wheat from the chaff. I haven’t watched television news consistently for years.
One of the advantages of the internet to the sophisticated consumer, however, is that it allows me to selectively edit my own news feed to filter for quality over crap. The problem with the Pew Report is that YouTube contains pieces ranging in quality from the New York Times to NewsMax. It’s all “YouTube” in Pew’s eyes.
The problem with all of these kinds of reports is that they focus on the technology first, might get to certain institutions, but always seem to lose sight of the quality of information that individuals are seeking. The real crisis here is one of legitimacy, not media. Legitimacy should not be based entirely on reputation but instead on methods and standards. To the extent that “traditional” news organizations follow those methods and standards, they can maintain legitimacy in my eyes. Some “rebels” in this environment are great and reshape the information environment. To cite one example, Fivethirtyeight.com, whatever its shortcomings have significantly reshaped how we evaluate the specific information that comes from polling. The level of detailed analysis that they perform would not have been possible without the internet.
Furthermore, it matters not one whit the quality of the information that these platforms may be peddling if all of the information that gets through to users is garbage. Most importantly, it’s difficult for even a sophisticated user such as myself to navigate the quality of information available to me. It’s almost impossible to expect that of the casual consumer of information. If all I watch on YouTube is QAnon videos then my information intake will be vastly different if all I watch is the NewsHour on PBS.
What we really need here from Pew is a third-party quality rating of information sources – a guide for information seekers again based on standards and methods. This information is out there but it often fails to get foregrounded effectively. If we had a “Consumer Reports” for Digital Literacy with a transparent process and which became widely circulated that might push some of the crap to the margins. As long we don’t, we’ll still have 30-40% of Americans believing that the most transparent and rigorous election in US history was “stolen.”
This is a role for an educational establishment like Pew. Perhaps this is a useful way for them to use some of the data collected to create a better informational environment. That really needs to be the goal here.
Great points, Tom.
I agree with you on tv news and 538.
This should be broken out by usage type – which would be a fun survey to design.
Yet I think this is still a useful first step. It’s an attempt to start getting data on a practice that isn’t well understood. Watch Pew for the next.
Bryan – Thanks for publishing. I had not seen this survey, so your post was helpful. I liked your thoughts and the additional data that you would like to see in the survey. Given the power of these platforms and the data that they collect and analyze, it would not surprise me that the platforms are already aware that many people use them as a source of news. They probably know a lot more than that. I understand that Facebook allows advertisers to pinpoint their desired audience using up to 86 independent characteristics. Given that none of that is published by Facebook, interested researchers are already behind the game utilizing traditional surveys with a statistical sample when Facebook (and others) have the data for their billions or hundreds of millions of users.
Great points, president Boston.
Researching Facebook is a daunting task!