Video: Partisanship and the search for engaging news
In the video, Natalie Stroud discusses two studies. In the first she examines how different buttons affect people’s response to comments in a comment section. She introduces buttons with three different labels: like, respect and recommend. Button “like” I believe forces people to have extreme views especially when it comes to hard content like politics. You like it means you strongly agree with it. So it makes sense to have labels that help people in supporting an opposing viewpoint, basically have a tool to signal that although I don’t agree with the content, I think the argument presented is actually a strong one. These tools will ensure that a person doesn’t out rightly disregards a comment just by looking at the number of “likes”. And what I really like is that these buttons serve two purposes: business (it had significant number of clicks) and democracy (since people started interacting with information that was counter to their beliefs). The business aspect is very important and is the one that is hardly ever considered. Almost all online platforms are actually businesses. So until and unless the suggested research increases user engagement or benefits the platform in some way, why would a platform implement it in the first place?
I can think of another label for a button which might have similar affects like respect does. News platforms can promote “share” button if they don’t already. If a person sees an article of opposing view being shared multiple number of times, curiosity might make him click that article and check what’s so special about it.
So conclusion from this study is that a design component should not make people to have extreme views (agree/like or dislike/disagree). Rather it should make people want to listen to others especially the ones who are singing a different tune. There is one aspect of this research that we should further extend. While people can be more receptive to opposing comments when it comes to topics like “gay rights” or “favorite music genre and pop artists” are they equally receptive to political content or do they end up “respecting” the politically like-minded comments. Study should include more topics to quantify the effects of changing button labels on different topics. Another line of research could be to study the after effects of “respecting” an opposing viewpoint. After reading and reacting to a strong opposing comment/post, will the user click/google search more on opposing viewpoints? Or will he return to reading like-minded news.
In the second study, Natalie studies effects of punishment (flagging a comment) and incentives (recommendations and top news picks) on partisanship. Swearing/Profanity increases the chances of a comment to get rejected and flagged while decreases the chances of getting selected as NYT Pick. But on the other hand, partisanship and incivility also increases recommendations. Natalie suspects that this makes the news room moderators to treat partisan incivility differently. While a comment containing profane language and swear words might get out-rightly rejected, content containing partisan incivility might get accepted. Does that mean one should extensively start moderating uncivil comments? Won’t that make the comment section bland and highly uninteresting and will probably decrease user engagement. It would be interesting to see how user behavior varies with varying strictness in moderation.
Can constructive comments always promote user engagement? Will a platform where everyone is nice and right attract diverse opinions in the first place? I believe a little incivility can add the essential flavor to the discussions. Probably we can measure the extent of incivility in comments that would promote healthy discussions & debates and develop automated tools that can detect and predict when uncivil comments will deteriorate the quality of discussion and promote heated exchanges among readers. To detect uncivil comments that do not contain swear words is further challenging.
