A nice piece of targeted empirical research with implications for news website UI and UX design, the study showed how site traffic could increase massively based on the aesthetic/functional qualities of the site — 90 percent in some cases. The study compared a contemporary “cleaner, photo-heavy scheme” versus a “more classic print-style layout.” The researchers also found that contemporary design could increase audience recall of the news content: “Layout matters, and it is consequential in terms of pageviews and what people recall from the news…Broadly, these results support news organizations experimenting with changes to their homepage, and considering a move from a more classic to a more contemporary design.” For more, see the Lab’s more detailed review.
The paper is meant as both a reality check for local news organizations and as a how-to for dealing with certain realities, namely: “The typical local newspaper gets about five minutes per capita per month in Web user attention, less than a local TV station earns in a single hour. Local newspaper traffic is just a rounding error on the larger Web.” Hindman notes that the “bottom line is that any successful strategy for digital local news requires sites to grow their audience…Audience growth is just as essential for plans that rely on selling subscriptions.” His recommendations including focusing on load times and personalized recommendation engines, as well as practicing A/B testing and optimizing content for social media. “The plight of newspapers is far worse than many journalists and editors realize,” Hindman concludes. Overall, his prescription is to focus on how to build consistent, repeat visitors, the idea of compounding “stickiness”: “Newspapers…need to rethink what they are optimizing for: not raw traffic, but audience growth. Small gains in stickiness can compound enormously over time.”
“Beyond Memorability: Visualization Recognition and Recall”: From Harvard, MIT and the University of Michigan, published in IEEE Transactions on Visualization and Computer Graphics. By Michelle A. Borkin, Zoya Bylinskii, Nam Wook Kim, Constance May Bainbridge, Chelsea S. Yeh, Daniel Borkin, Hanspeter Pfister, and Aude Oliva.
The study provides some useful insights for news data visualization, and it serves as a good reminder that audiences need some help when interpreting visual information. The researchers conduct lab experiments with a variety of real-world graphics, including many from news organizations, and find that titles and text really help viewers interpret visuals and then recall information afterwards. Like them or not, pictograms — when done well — also facilitate recognition and recall. The keys to good data viz, according to the study, are: “having a good and clear presentation, making effective use of text and annotations, drawing a viewer’s attention to the important details, providing effective visual hooks for recall, and guiding the viewer through a visualization using effective composition and visual narrative.”
“Exposure to Ideologically Diverse News and Opinion on Facebook”: From Facebook and the University of Michigan, published in Science. By Eytan Bakshy, Solomon Messing, and Lada Adamic.
This paper speaks to the ongoing debate over the power of algorithms and audience “filter bubbles.” It might seem strange for a company to need to study its own algorithm, but it’s a dynamic, complex software system. The researchers find that, although the algorithm does tailor news based on liberal or conservative leanings (and prior behavioral patterns), the filtering problem is minimal: “After [algorithmic] ranking, there is on average slightly less cross-cutting content: conservatives see approximately 5% less cross-cutting content compared to what friends share, while liberals see about 8% less ideologically diverse content.” The study took some heat for its methodological design and approach. See here and here for critiques. But overall, it stood as an important contribution that set the agenda for more discussion of algorithms and impacts.
“Tweeting From Left to Right: Is Online Political Communication More Than an Echo Chamber?”: From New York University, published in Psychological Science. By Pablo Barberá, John T. Jost, Jonathan Nagler, Joshua A. Tucker, and Richard Bonneau.
This big data study joins a conversation about ideological segregation on Twitter that has taken place for several years now (see this column last year.) The sample size is enormous: almost 4 million Twitter users and 150 million tweets. The researchers find that ideological segregation — the proverbial “birds of a feather” phenomenon — is much more visible with explicitly political issues, and that on other national events, left and right often speak with one another. Two important other findings: “With respect to both political and nonpolitical issues, liberals were more likely than conservatives to engage in cross-ideological dissemination,” and “previous work may have overestimated the degree of ideological segregation in social-media usage.”
“Interacting Is Believing: Interactivity, Social Cue, and Perceptions of Journalistic Credibility on Twitter”: From Hope College and Lehigh University, published in Journalism & Mass Communication Quarterly. By Mi Rosie Jahng and Jeremy Littau.
This study, which is based on an experiment involving about 150 students, suggests that journalists who engage more with audiences on Twitter increase their perceived credibility. Obviously, the results are limited by the experiment’s sample demographic. But it’s intriguing to contemplate how the very act of replying to the audience itself may bolster the standing of journalists. However, another 2015 study of journalist interactivity on Facebook — by Jayeon Lee of Lehigh, published in the Journal of Computer-Mediated Communication — produced somewhat contradictory findings and finds it can be a “double-edged sword”: In terms of professional dimensions, audience engagement diminished perceptions of journalists and associated news products.
“Changing deliberative norms on news organizations’ Facebook sites”: From the University of Texas at Austin, Purdue University, and the University of Wyoming, published in the Journal of Computer-Mediated Communication. By Natalie Jomini Stroud, Joshua M. Scacco, Ashley Muddiman, and Alexander L. Curry.
This study looks at how journalists might promote better, richer, and more reasoned civic discourse about news by playing a stronger role in comment threads. The strength of the research is that it leverages a real-world field study and randomization to pinpoint effects. The researchers found that “reporter involvement was related to lower levels of incivility and greater use of evidence from commenters.” Overall, the study “provides evidence that an individual can affect norms in online comment spaces. And to a goal of promoting deliberative discussion online, this study offers support for a practice that can be enacted — engaging with commenters.” The research is part of the Engaging News Project’s important ongoing investigation of comment threads and the effects of journalistic engagement on civics and democracy.