Elaborate: A browser extension for digging deeper into media bias
Catherine Yochum, Peter Cederberg, Xuehui Zhang
This intervention was created as part of our MPS/MDes I Communication Design Studio during the Fall 2020 semester at CMU. As a class, our goal was to identify elements of communication design in media that influence the way we consume news, and design an intervention to make people better readers. As a team, we focused on bias in the media. The project took place over two weeks of individual work, followed by two and a half weeks of team work synthesizing our findings and developing an intervention.
Individual Process and Study Structures
I investigated the way that news sources organized their international articles in order to identify potential sources of bias. Particularly I was interested in how America based companies reported on the world in different ways depending on their readers. If a company has less readers outside of the U.S. then how might they talk about these places? This primarily involved looking at the webpages of Fox, The Wall Street Journal and The New York Times on a desktop computer. Special attention was given to the layout of each company’s world news page and footer navigation. I did this study to see if these news companies were highlighting certain parts of the world to play to their readers interests or potentially drive a narrative. I suspected that some areas of the world that experience prejudice from American and/or are generally associated with negative press would be highlighted. This would be in connection with the idea that bad news sells.
My methods involved analyzing how each world news webpage was laid out and which regions were put in front of the reader first and why. Additionally I looked at the footer navigations of each website to see how they would suggest further navigation to their readers and the language being used. In doing so I was able to see that Fox placed a much greater emphasis on the Middle East in their webpage and their footer navigation was riddled with words that meant to inspire fear. There were also special sections of news dedicated to both global and domestic terrorism. In contrast, both NYT and WSJ lack the buzz word news categories and both list their global regions alphabetically. It should also be noted that NYT and WSJ have a much larger international readership than Fox and both have several language options for readers. Fox has just started its new international news branch so it is possible they may change their tone in the future.
I focused my investigation around cable news channels — specifically, Fox News, CNN, and MSNBC. I began by analyzing the artifact itself, finding that color (bold red, white, black, and blue), screen layout (all caps letters, widgets and headlines overlaid and alongside the main video), imagery (e.g. American flags, partial documents), motion (scrolling stories, quick cuts between video clips, branded transitions), sound (dramatic music) and spin (e.g. “Democracy 2020” and “Fight for the White House”) seem intended to create a sense of urgency and encourage extensive news consumption.
Influenced by my classmates’ explorations, I then took a closer look at their websites, at the history and management of these news channels (CNN was the first 24 hour news network; Fox has the top five most-watched cable news programs in the US), and at how they compete with one another. These investigations confirmed the networks’ incentive to hold viewers’ attention as much as possible, since ratings are tied to viewership and most viewers can only watch one station at a time. As I analyzed the wording and messages of news segments themselves, not only did I find sensationalized language, but it struck me how hard it was to differentiate between programs intended to be objective reporting versus opinion and commentary, blurring an already confusing line between fact and opinion through the overlap of news and entertainment.
The investigation I did initially was the comparison between The New York Times and TechCrunch. Unlike NYT, TechCrunch only focuses on technology news, and it is not a mainstream news venue. By choosing these two media, my intention was to compare how their communication design reflected their perspectives and impacted audiences’ perceptions.
First, I explored their typefaces, visual styles and layout structures. I analyzed the information density and hierarchy conveyed from their design. The visual elements of the New York Times communicates an image of traditional journalism, while TechCrunch conveys a digital and informal journalism vibe. Then, I looked into the actual content, such as the language tone of news titles, the perspectives of certain topics from different news venues, etc. In terms of content, I noticed that news titles tend to be extreme and conclusive, which could potentially push audiences to take sides, and read passively. On the other hand, the employment composition of TechCrunch and the NYT can lead to limited perspectives in voices in terms of racial groups under a superficially diverse team.
After this, inspired by our classmates’ investigation of the impact of social media’s features on users’ reading habits, I started to analyze news aggregators. Feedly stood out for me because its feature of personalization of news sources in users’ feeds potentially guide users to compare different opinions, think critically, and eventually form their own opinions.
Design Intervention
After individual investigation, we started our team-based project. In order to find a clearer point of intervention related to media bias that corresponded with our findings, we brainstormed concepts based on our findings, including facts vs. opinions, positive news, gamification, and forms of delivery, as shown in the following image.
The initial storyboards we created for our intervention is in the following image — a daily alert on how to read news. One thing Stacie wanted us to consider was creating an intervention for audiences with no intention of addressing the news in a more critical way. What about audiences without the awareness? Our intervention also took this into consideration.
Some of the key insights that impacted our intervention are:
- Provocative journalism is affecting how people perceive the world. Mainstream news sources mostly report extreme and negative news to catch audiences’ eyes. People are more stressed out than ever by the news despite crime being lower. How might we engage and share news differently?
- Everyone is biased. One question that we keep asking ourselves is that “Are all biases bad?”, which makes us think more broadly and deeply about what is considered as bias and why. We researched different types of media bias, and wondered how to help people recognize their own bias, and see more perspectives. How might we encourage readers to identify and reflect on elements that impact the way they perceive a story?
- It is important for our audiences to form their own ways of thinking. News is not black and white. We should not provide answers or induce certain points of views, but help audiences recognize the intricacies and limitations of language and perspectives in a story. How might we help readers become critical thinkers?
Outcome
Our proposed intervention, Elaborate (Click for working prototype), is an app and browser extension intended to aid in reading news on a mobile or desktop browser. It prompts reflection on the biases present without influencing the reader towards a “correct” viewpoint.
The extension targets news readers who might be susceptible to biases without realizing it. In order to attract their attention, Elaborate advertises sensationalist fake stories on social media platforms. Clicking on such an ad takes readers to a website that samples Elaborate’s capabilities, shows them examples of bias in popular news sites.
Once downloaded and enabled, Elaborate allows the user to click on highlighted text overlays, gathering information about the overall news source’s bias, author, and type of article, and reflecting on biased phrases the intervention automatically identifies from an extensive media bias word bank. We leaned on AllSides’ outline of 11 types of media bias as the basis for what this might look like; while not all forms of bias lend themselves well to being identified through particular phrases, readers can at least begin to be alerted to instances of spin, sensationalism, mind-reading, unsupported claims, and opinion presented as fact. We also included a capability for users to highlight biased words and phrases themselves as they become savvier readers; other users will be prompted to verify these instances.
Our hope is that in both prompting reflection and creating engagement, Elaborate affords collaborative participation in identifying bias as well as enhancing critical thinking.
As far as limitations, while we aimed to reach a broad audience through social media marketing, the intervention does still require users to go the last mile of installing the Elaborate extension. However, by integrating examples of reflecting on bias into the Elaborate website before it is downloaded, we hope that the site can even serve as a mini intervention exercise in itself.
In conclusion, the intervention from Elaborate has a number of potential outcomes that can affect how users consume news. They would be able to see how the news is trying to impact their emotions and opinions through our program and community. Readers may also be able to be more aware of what it means to be objective and that everyone carries biases. These biases aren’t inherently bad but it is important to see them and recognize how it affects one’s world view. The ability to discuss bias and arguments may also allow people to become better at advocating for themselves and their views in the future.
There are possible downsides to the effects of Elaborate. Revealing the current bias in news may lead to users becoming less trusting of news in general. This would particularly be damaging with social groups that already distrust the news. Another potential downside would be internet graffiti that attempts to radicalize readers with racist or violent rhetoric. For this reason the reporting function of Elaborate will be necessary. On the other hand, thinking critically about key issues and seeing how these arguments are being framed also has the potential to de-radicalize people.
These outcomes are important because they will allow for people of many different walks of life to come together and have productive discourse about how the news is trying to influence them. This is a potential boon to anyone who is willing to open themselves up to the possibility that the news has a greater effect on them than they are willing to admit.
