The Dark Side of Reader Comments
Interactivity has been the name of the game for websites for the last few years, and user-generated content (UGC) has been a mainstay of building content and boosting engagement. Indeed, “content marketing” is the latest web marketing buzzword and reader engagement is high on the list of goals. In contrast to a few years ago, today just about every content site, from tiny blogs to the largest news sites, allows readers to add comments. It seems like a complete win-win – readers can get heard, and the sites benefit from more engagement and additional content. It turns out that in some cases that “free” content comes with a price.
A new study, described by two of the authors in the NYTimes, looked at the effect of comments on how other readers interpreted an article, and the results were surprising. Ordinary comments didn’t change opinions of the readers – a sad thing, perhaps, for thoughtful comment writers who expect to inform and persuade those who read the article.
The comments that DID affect the opinions of the readers were those that used insulting language. All readers read the same comment content, but half of the readers saw comments which included gratuitous epithets, like, “You’re an idiot,” and “You’re stupid if you…” instead of more neutral language. The readers exposed to the nastier language became more opinionated and actually drew different conclusions from the story:
Uncivil comments not only polarized readers, but they often changed a participant’s interpretation of the news story itself… Those exposed to rude comments, however, ended up with a much more polarized understanding of the risks connected with the technology. Simply including an ad hominem attack in a reader comment was enough to make study participants think the downside of the reported technology was greater than they’d previously thought.
I’ve run and administered large online communities, and have always tried to enforce civil discussion. Sometimes, this comes with the price of removing member content and banning members who are consistently rude or disrespectful to other members. That strategy has been based more on community building experience and my own preference for friendly, welcoming communities. Despite the occasional complaint about censorship (or, even more amusingly, that I was trampling on their “First Amendment Rights”), I’ve found that communities where civility reigns can prosper and grow. Communities where new arrivals are routinely “flamed” for asking dumb questions have more difficulty getting new members to stay engaged, and even long-time members quit when exposed to hostility.
With this research, I can see why civil discussions work and less civil discussions inevitably head downhill. The study didn’t look at any post-reading interaction, i.e., the subjects couldn’t add their own comments, but it’s easy to extrapolate. Reading rude comments polarized the opinion of the reader – given the opportunity to comment, that reader would almost certainly post a more polarizing and perhaps rude comment himself!
As more rudely expressed opinions build in a discussion, the nastiness feeds on itself. This is a phenomenon well-known to community moderators, and now this study gives us some background to go with real-world experience.
What’s a Site Operator to Do?
If you run a news site or a blog, it’s a bit demoralizing to think that a few comments from barely-literate morons can change the way readers view your own professionally written content. And, in case you missed it, the last sentence included a gratuitous epithet to help stir the passions of my own readers!
I stand by my belief in moderating rude comments. Yes, it’s labor-intensive. Yes, some outspoken people don’t like it when they can’t scream at others to make their point. But, it works.
With a little more hesitancy than usual, I’ll invite you to weigh in on this topic. What has your experience been as a commenter, community participant, blogger, or community operator? Does it bear out the researchers’ findings? Have you come up with any creative solutions? Leave a comment! (Remember, be nice…)
I agree. I know I focus on the negative when I read comments and especially reviews. I can also say that at that point I am completely distracted from the article and if the users are allowed to go crazy I am unlikely to comment or go back.
Brian, I think it’s human nature to respond to hostility in words with an uptick in one’s own “threat level,” perhaps dating to our hunter-gatherer days. Hence, we tend to view things from that perspective and, given the opportunity, respond by signaling that we aren’t intimidated.
It does seem that the negative stuff draws attention far more than the positive. Perhaps people just see it as the comment equivalent of a car wreck, where they can’t turn their eyes away. As the negativity gets progressively worse they just want to see where it ends up going before it fizzles out.
I definitely agree that it’s important to consider the type of community you want and moderate based on that. If you want quality engagement and a good atmosphere then it’s well worth the fairly minimal time invested to keep things clean.
Agree, Cody – fighting does attract eyeballs for a while. And maybe you can build a community that way, though it wouldn’t be my favorite kind of environment.
So glad you brought this up. I manage a few blogs and while I haven’t had any particularly rude posts, I do see a lot of self-serving, promotional posts that aren’t even close to the topic. Those get killed immediately, but others are sort of a fine line decision. Would love some guidelines for those situations.
However, I once waded into a conversation about female vs male behaviors in financial investments. I thought my comment was fair and not in any way sexist. Guess I was wrong. Got a terse “F… you!” reply within minutes, and I chose not to pursue the conversation. But, it soon triggered the proverbial (pardon the expression) “s…storm” — with posts both attacking and defending me in language I wouldn’t repeat to a drunken sailor.
I evolved into both a symbol of evil sexism for one side and a martyr to radical feminism for the other — without saying another word! Another example of what you write in text being perceived so much differently than what you say out loud.
I see that kind of misundertanding in forums all the time, Bob. Sometimes attempts at humor go awry. Emoticons can help communicate intent, but they aren’t always enough. In other cases, what someone intended to say is completely misinterpreted by someone else, and people start talking past each other.
I agree with Bob that misinterpretation runs rampant. I wrote an editorial, basically citing two sides of an issue and saying I don’t have an answer. One scathing commenter called me a “liar from the pit of hell.” Um, okay.
I think people read pieces how they want to read them, bringing their suitcases of preconceived notions along for the ride.
What frustrates me is how those barely-literate morons (we call them trolls in my town) discourage intelligent commenters from joining what could be productive conversations.
Yep, Carrie, I’ve seen long-term, very productive members leave a community after one thread got out of hand and nasty. Sad the impact that a few bad members can have. That’s why it’s key to nip these problems in the bud.
I like what Bob Boucher writes in that too easily people become caricatured into a one-dimensional cartoon-like representation of only one thing. Form there, the conversation only degenerates in ever-decreasing circles.
I don’t mind a firm exchange of views, but civility and decency go a long way.
Hi Roger. This is fascinating research and I can relate only too well to the research findings- aggressive language polarises the audience’s responses and can even change their understanding of what the article was about.
I am also very interested to notice my own behaviour when I comment. A blogger once replied to a comment but clearly had not taken the time to really read what I wrote. I found my reaction was far more extreme than normal- I simply did not go back to the site!
Can you throw any insight into what causes the oversensitivity you described and what I experienced? I would be interested to understand more.
If I had to guess, Jane, it is that hostile language is a threat indicator and may kick us into “fight or flight” mode. You chose flight, but you could also have responded in an even nastier way (fight).
Or, there’s always the third path – trying to let our conscious, rational mind take control and continuing the conversation in a calm, friendly manner. In my experience as a community builder, most people don’t choose that approach. Those members that can stay engaged in the conversation but fail to respond to attacks and name-calling and try to dial down the rhetoric make great moderators!
Thanks- that really helps.
As it happens the blogger I mentioned did me a massive favour as I am now only too aware how careful and empathetic I need to be on my own site…..
Dave Wimer mentioned in a post (alas, wish I could find it again to link to it) a system developed at Harvard where an opinion was posted, commenters then submitted their replies but, and here’s the key, the replies were all posted at the same time a few days later. The process repeated as often as replies kept flowing in. As Dave observed, a very lawyer-like system.
Replies were grouped according to the content submitted, forming sub-threads spawned by the original post. It was noted that the delay in publishing allowed time for reasoned replies to be formed and written. The publishing format also made it difficult for active (reactive?), vocal individuals to dominate the conversation.
Seemed like an interesting and democratic method of handling replies. What do you think?
Interesting concept, Glenn, but it sure takes the immediacy out of the equation. I think by the time the stuff gets posted, a major loss of momentum will occur.
There is an up-side to this down-side. Polarized individuals are more likely to engage, if only a little. I read somewhere that the best way to get engagement on Facebook is to rant about someone or something, or lament one’s life…
This same phenomena can be leveraged constructively if the polarizing comments are used in the body of the text itself. But the trick is to leverage the ad homonym attacks at a universally despised “them” … (barely-literate morons) … so that people’s polarized emotions are siding with you. Deliberately create and foment the “us vs. them” division, and make it obvious that any sane person is guaranteed to be on the same side as the author.
The more I think about it, the more this is sounding like a formula for a sunday sermon… 🙂
John, you are absolutely correct that individuals with a polarized view are more likely to engage. Post something like, “5 bad things about Apple” and the fanboys will descend like a horde of locusts! Similarly, political topics bring out both sides. The problem is when the rhetoric gets out of hand and ugly.
I think the rule is never to read the comments on news articles, and when you do read or comment act as civil as possible. There’s a big difference between reading a thread of entirely hateful comments, vs reading a few not so intelligent responses in a longer thread of good comments. Thankfully I haven’t received too many negative comments on my own writing so I haven’t had to deal with that, but it seems like as a writer the best thing you can do is either ignore it, or acknowledge it and move on rather than baiting for a response.
This is fascinating analysis and that i will relate all too well to the analysis findings- aggressive language polarities the audience’s responses and may even amendment their understanding of what the article was concerning.
I am additionally terribly interested to note my very own behavior once I comment. A blogger once replied to a comment however clearly had not taken the time to actually browse what I wrote. I found my reaction was much more extreme than normal- I merely failed to return to the site.
Comments are like those radio shows that allow you to call in. They want your opinion whether its positive or negative amd well often I find myself listening to how the different points of view are defended by either side before forming my own opinion of the topic.
Come to think of it, the only time I’ve ever commented on my local TV facebook page is when the media personality made a callous remark obviously meant to inflame the discussion (and I said as much). On my own blog I don’t allow certain comments. After all, it is MY blog.
These rants seem to “shut people down” or “fire them up”. Within social media it is not uncommon to see people either “mute” or delete the ranter, or jump in with them, creating a world of group “confirmation bias” that leads to “group think” and often lead to some form of intolerance and out right hatred.
There is a classical conditioning effect that is created by the ranter, and that is that if the ranter is associated with a business and continues to rant, people do not want to associate themselves with their business either. As we are well aware if someone dislikes a business they will tell everyone they know, however if they like a business they may just tell a few.
And that is why I emphasize on moderating all comments readers post. I dont tolerate rude comments and that is why I have set up a few comment rules and any failure to follow the, a trash awaits.