- First, national TV programming was measured, then local programming was measured using a subset of the national panel.
- TV news consumption was defined as the amount of time devoted to watching any of the ~400 programs defined as "news" (including categories such as morning shows, late-night comedy shows, and hard news).
- Next, desktop and mobile media consumption were measured using Comscore's panel, which breaks out the total amount of time spent on different types of media sites by demographic bucket.
- Desktop and mobile media consumption was defined as the amount of time spent on any article that had been published on one of more than 800 websites. Correspondingly, fake news consumption was defined as the time spent on 1 of 98 websites identified as sources of fake news.
- Third, Nielsen's nationally representative desktop-only web panel was used to measure news consumption on the top four social media sites (Facebook, YouTube, Twitter, and Reddit) and on the top three search engines (Google, Bing, and Yahoo).

Materials and Methods in "Evaluating the fake news problem at the scale of the information ecosystem" Study

- Certain types of misinformation could have been mischaracterized due to the nature of this study.
- News-relevant social media content that was not tied to a specific URL, as well as false or misleading information circulating on generally reliable news sources were labeled as areas of uncertainty and therefore were not classified as fake news in this study.
- Therefore, estimate of fake news consumption in this study may not be entirely comprehensive.

Limitations of "Evaluating the fake news problem at the scale of the information ecosystem" Study

- Even though the low prevalence of fake news was demonstrated in this study, fake news is still a problem worthy of attention because of the disproportionate impact of fake news compared with regular news.
- However, the findings of this study that fake news is extremely rare suggests that concerns of possible threats to democracy should adapt to become more broader instead of solely focusing on the circulation of fake news.
- Misunderstanding of political matters could originate from a combination of "ordinary bias and agenda setting in mainstream media," as well as the overall small amount of exposure Americans have to news content.
- Future work on misinformation and its potential impacts on democracy should address all potential sources of problematic content, as well as the lack of relevant content, instead of solely focusing on the content that is the most easily identifiable and least associated with conventional media interests.

Discussion of Findings of "Evaluating the fake news problem at the scale of the information ecosystem" Study

- To address the first two of the three aforementioned challenges.
- This was done by drawing on three different sources to measure consumption across the two main modes of news production (TV and online). Total consumption was integrated across the modes by demographic buckets.

Johns Hopkins University

- Because there is no single source of data that covers all three modes of consuming news, researchers select data sources on "the basis of their availability," which doesn't necessarily correspond with representativeness.
- The studies done on fake news often fail to consider how much of the fake news is consumed relative to other types of news or non-news related content.
- Even when the prevalence of fake news is low compared to other content, fake news can still be considered important because of its disproportionate impact on certain populations. Therefore comprehensive questions of prevalence on their own cannot resolve questions about influence.

Reasons Why It Is Difficult to Empirically Evaluate Disagreements Over the Prevalence and Importance of Misinformation

Allen, Jennifer & Howland, Baird & Mobius, Markus & Rothschild, David & Watts, Duncan. (2020). Evaluating the fake news problem at the scale of the information ecosystem. Science Advances. 6. eaay3539. 10.1126/sciadv.aay3539. 

Learn Before

Related

Learn After