Journalism exists to tell people what’s happening and give them what they need to understand it. That mission does not bend to budget or format. A story about a school board vote and a story about housing prices are both doing an essential job, just at different depths. The assumption underlying both types of stories is the same: that telling people more, and giving them precise information, helps them form their own judgments and take part in the conversation that follows
We wanted to know how this plays out in comments sections. In a paper recently published in Journalism Studies, my coauthor Sheizaf Rafaeli and I analyzed nearly a million comments on 6,400 New York Times stories published between 2014 and 2022. We compared pieces from The New York Times’ data journalism section, The Upshot, to more traditional pieces published during the same period. (I am a computational social scientist who studies how information environments shape who participates in online public discourse. Rafaeli is the president of Shenkar College in Israel.)
Data journalists have sometimes wondered whether loading a story with data can be overwhelming rather than clarifying. We found that giving readers more information changes who comments, and how.
More information, more engagement
For each story, we measured several things: How information-dense it was; the numbers, sources, and charts it used; and how analytical the writing was. Then we looked at the comments on those stories. We tracked how many people commented, whether they were interacting with each other or only reacting to the story, and how much their comments leaned on sources and language that read as analytical versus emotional.
We found that stories combining statistics, external sources, and data visualizations produced a different kind of comment section. Regardless of the topic of these posts, readers tended to reply to each other. That raised another question: What were readers actually saying to each other, and how many of them were showing up to say it?
The answer turned out to hinge on something researchers call “affective style” — how much a comment is driven by feeling rather than reasoning. For example, in a comment on a 2014 Upshot piece about the World Cup, one reader wrote: “Typically, people hate good teams because they win too much and laugh at bad teams. So, if this is true, soccer fans are truly unique.” That’s affective — a feeling dressed up as an observation.
Under a different Upshot piece, this one on the shrinking American middle class, a reader wrote: “From 1950 to 1980 we had thirty years of demand side economy. From 1980 on we’ve had thirty years of supply side economy. Take your pick.” That’s not affective — it’s an argument, laid out for the reader to weigh. Denser, more analytical stories produced far more of this second type of comment.
The catch
The same stories that produced sharper, more analytic conversation also produced less conversation. The more analytically a story was written, and the more statistics it contained, the fewer people commented on it. Emotional stories did the opposite, drawing bigger crowds.
The presence of charts mattered less than you might expect. Stories with more visualizations tended to draw comments that were more analytic and better-sourced, but once we accounted for everything else a dense story was doing, the chart alone was not what moved a reader from reacting to reasoning; rather, that was the statistics and writing style.
Comments on denser stories were more likely to lean on outside sourcing and structured arguments. Researchers have connected this pattern to the digital divide, identifying a “data elite” as the subset of an audience that keeps engaging when others opt out. That tracks with something recent Pew research found about civic participation generally: People sort into distinct groups. As a comments section grows more reasoned, it also becomes more exclusive.
Most people who visit a comment section never post in it, and a section full of careful, evidence-based argument is a different thing to encounter from one that isn’t, whether or not you participate. Producing that kind of journalism costs real money, statisticians, data cleaning, chart-checking, extra editing, and that expense pays for what happens underneath the story, not just the story itself.
That is worth keeping in mind as newsrooms reopen comment sections after years of shutting them down. If the payoff is a more loyal, more engaged readership, it is not spread evenly across a newsroom’s output. Start with the hardest stories, not the easiest ones that will always draw a crowd on their own. That is where the more reasoned conversation was hiding.
Avner Kantor is a researcher at the Israel Central Bureau of Statistics.
