There are currently thousands of propaganda websites masquerading as local news websites across the United States, as the New York Times reported in October 2020 and the Columbia Journalism Review reported in August 2020.
The network of websites spells disaster for the news ecosystem on a number of levels, especially if the sites receive a lot of attention. As Renée Diresta articulated in this WIRED piece, there is an important distinction between “free speech” and “free reach.” Free speech entails Brian Timpone’s ability to write and publish “propaganda ordered up by dozens of think tanks, political operatives, corporate executives and public-relations professionals” (NYT). For example, consider this article from March.
A hotel owner by the name of Monty Bennett paid to publish the article, as part of a broader effort to lobby for hotel stimulus amidst the pandemic. Merely publishing the article constitutes free speech, but the bigger issue is free reach — how far the article spreads and how many people it misleads. In this case, Monty Bennett did not seem to get much attention when he tweeted out the article he paid to publish.
But what about the other 1,200+ websites? After reading the reports from Columbia Journalism Review and the New York Times, I was curious about how much attention they were receiving on social media platforms, especially now that we are days away from the election. I am still collecting some data about the topic, but one of my initial findings seemed important to share right away.
While Twitter has suspended many accounts associated with the propaganda websites, some accounts have amassed sizable audiences
CJR generously published a list of all 1,200+ websites from their analysis, which I scraped to check for social media profiles. 79% had Facebook accounts, 9% had Twitter accounts, and only 6% (N=74 ) were still collectable from Twitter’s API (here is the final dataset).
It may be tempting to think that if only 74 out of 1,200+ websites have active Twitter accounts, the propaganda operation was unsuccessful. But these operations are notorious for leveraging “long tail” engagement. For example, Renée Diresta and a team of analysts found that in the Russian operation from 2016, only a few pages received meaningful engagement. Here is a chart from their study that illustrates the long tail pattern:
At a high level, you can see the same pattern with the websites I analyzed:
Here is a closer look at the sources with a lot of followers:
As a Chicago resident, “Cook County Record” caught my attention. The Twitter account looks like this:
It might seem unsuspicious, but once you look at other sources, the illegitimacy becomes apparent. Here is the “Pennsylvania Record:”
And this is what the “Florida Record” looks like:
However, anyone can buy Twitter followers, so I briefly wondered if these propaganda accounts were just buying popularity.
Unfortunately, the Chicago Record account was followed by several reputable news outlets from my Twitter network (ProPublica Illinois, The Chicago Reporter, and a few individual journalists), meaning it seems to have infiltrated a network of legitimate journalism.
This is a preliminary finding, and there is a lot more data to crunch. Stay tuned if you are interested! In the next few weeks, I will continue analyzing the network of propaganda websites, with a particular focus on swing states like Pennsylvania and Florida (as the map below from CJR shows the propaganda sites are concentrated in these states).
To abruptly wrap things up, I will leave you with this memorable video from 2018, which shows just makes manipulated media palpably haunting:
As always, reach out if you have questions, ideas, or comments. Here is the full list of sources that I found with Twitter accounts, available on GitHub: