To any young writers out there trying to get their start, I offer the most important piece of advice: Never, ever, ever read the comments.
But for too many leaders, brands, and public figures these days, absorption into the world of the “very online” leads to the decision to react to things that aren’t real, to focus on issues that aren’t genuinely important, and to distract themselves away from where their actual constituency (their voters, their supporters, their customers) stands.
We live in a moment where, if you’ve got something to say, instant feedback awaits you — for better or worse. In and of itself, getting feedback isn’t a bad thing. For someone who works in the world of opinion research as I do, getting feedback on one’s performance should be a good thing. If we can’t measure how we’re doing, how will we know what we can do better? Hearing what people think of your work, assessing which criticisms are valid, and adjusting your approach and worldview accordingly is not necessarily a bad thing.
But internet comments, or really most discussions that flourish on social media, run afoul of a key principle of opinion research. They are certainly deeply unrepresentative of who your “constituency” really is.
This week, Pew Research Center is out with fresh data confirming all of that and more. Focusing specifically on Twitter, they find the most active 25& of Twitter users produce a whopping 97% of all tweets on the platform. Only a fraction of people even use something like Twitter, and then, only a small fraction of those users produce what we see on there.
Promoting one’s work on a platform such as Twitter feels essential, and conversations that happen in online spaces can have offline consequences. But if you, as a writer or executive or leader, come to believe that what happens in comment sections or Twitter threads is reflective of how most people think, you’re likely to be very misled.
In just the last few months, two hit TV shows have showcased great examples of flawed characters turning to internet comments for validation. In HBO’s Succession, media scion Kendall Roy is obsessed with knowing how people on Twitter think about him, making his hangers-on sift through tweets to update him on what the internet is saying. And on Apple+’s Ted Lasso, hero-turned-villain Coach Nate turns to Twitter for validation of his worth as a soccer coach, leading him to become an egomaniac (from the good tweets) and perpetually angry (from the bad tweets).
When people start to believe that online discourse is representative of reality, it can lead them badly astray.
A decade ago, I was optimistic that social media might offer a new way for people like me in the polling industry to understand public opinion better. If we live in a world where more and more people are declining to pick up the phone to take a survey, but at the same time are more and more willing to post their opinions out in the open for all to see, couldn’t we just do a better job of aggregating those views? Why go to the trouble of asking people what they think if they’re eager to tell us?
But much like other optimistic predictions around social media from a decade ago (that it would lead to a flourishing of free speech, the toppling of dictators, and so on), this did not pan out like I’d hoped. Republicans are particularly pessimistic about the role social media outlets play in American life, with 60% saying Twitter is bad for democracy, compared to only 28% of Democrats who agree, according to the Pew study.
Demographically, Pew finds that a place like Twitter is very unrepresentative of the average American — or even the average highly politically engaged American. They find younger and more progressive Americans are far more likely to say they look at Twitter “too many times to count” in an average day. And even among those who are most active, few of their tweets are original content about their own views; far more of their content contains either retweets or replies.
It is easy to believe that the things you see online are real, deserving of response or reflection. No one is immune from caring about what other people think. It is human nature to want to gauge our standing, to know if others agree with us. (In a way, this explains why many people care about poll results in the first place and are more willing to get upset if a poll shows they hold a minority view.)
There are tools and strategies for regularly measuring how you’re doing when it comes to the constituencies that matter to you. Good polling is just one of those tools. If you want to be a smart executive, leader, or commentator, find better mechanisms for getting feedback than peeking at the comments.
We now have even more data to prove it: Twitter is not real life.
