How Skewed Is My Blogging?

All Generalizations Are False


About a year ago, Monica had a post on liberal bias in the media. Drawing on this Business Insider article (amongst others),1 she pointed out that those “who work in newspapers and print media are almost exclusively liberal.”


Last month, Monica had a follow-up post in which she introduced Reading these posts piqued my interest in analyzing my own media consumption and, therefore, my own biases. So, in the name of transparency, I have combed through every single one of my blog posts here at Difficult Run. I checked every single link, ran that link by MediaBiasFactCheck, and properly categorized it.2 I then counted how many times a source was linked.

Now, as fun and insightful as this may be, it’s highly unscientific. Here’s why:

  • is not a verified, academic source: It’s simply a useful internet tool with its own biases. For example, it labels the BioLogos Foundation–a Christian group established by Francis Collins, the former Director of the Human Genome Project–as “Conspiracy-Pseudoscience” because “they ascribe evolution to the hand and workings of God, which is not known or provable[.]” The problem with this label is that BioLogos isn’t doing science. It’s doing theology (which is why you have so many articles dedicated to scriptural interpretation by biblical scholars). Or, more accurately, it’s thinking about science theologically. If someone is citing it as a scientific source, they misunderstand the nature of the group.
  • Every link is counted equally: I could be trashing a Washington Post article, praising one by Vox, and having a good-natured debate with the National Review. The content could also be very different. One could be reporting a new scientific study, while the other provides an opinion on the current political climate. Doesn’t matter. Every link counts the same.
  • Not every link is registered at Several sources I link to multiple times were not included on the site. This ranges from institutions like the World Bank, academic centers like Greater Good Science Center, think-tanks like the Mercatus Center or the Center for Economic and Policy Research, or libertarian blogs like Bleeding-Heart Libertarians or Marginal Revolution. Not to mention the numerous studies I cite. The absence of these sources will obviously skew the results a bit.
  • Oft-cited sources tend to be a handful of authors: For example, the conservative think-tank American Enterprise Institute (AEI) makes it into my top 10 cited sources (see below). However, this consists of largely three authors: James Pethokoukis, Mark Perry, and W. Braford Wilcox. While not always the case with Reason, the bulk of my citations come from their science correspondent Ronald Bailey. This means a good chunk of the Reason citations are summaries of new scientific studies rather than libertarian screeds. Finally, some citations of partisan sources feature their token other partisan (e.g., Megan McArdle or Tyler Cowen in Bloomberg). So the partisanship of some sources may only be surface level.
  • Links/quotations are incestuous: I use a lot of quotes in my posts. These quotations are often self-referential. Washington Post articles link to other Washington Post articles. Same with the New York Times. Same with Reason. Even though I may be quoting only one article, those quotations may have five other links to the same source. And those links get counted (see above).
  • The Left/Right dichotomy is flawed: I read a lot of what would be considered libertarian material (e.g., Reason). Libertarians are often seen as conservatives, yet libertarians tend to oppose conservatives on issues like immigration, trade, same-sex marriage, drug laws, military intervention, etc. This makes the lumping of libertarian sources with “the Right” problematic.

So, after all those caveats, here is what my blogging at Difficult Run looks like:

List of Sources3

Left Bias

Left-Center Bias

Least Biased

Right-Center Bias

Right Bias

Extreme Right




Number of Citations by Partisanship

Extreme Left Bias Left Bias Left-Center Bias Least Bias Right-Center Bias Right Bias Extreme Right Bias Total Pro-Science Pseudoscience Total
0 114 454 269 284 120 4 1245 71 2 73
0% 9.2% 36.50% 21.61% 22.81% 9.64% 0.32% 97.30% 2.74%

If Left-Center, Least, and Right-Center are considered “Centrist,” then 80.88% of my links are Centrist with only 9.2% being Leftist and 10% being to the Right. However, if I consider Left-Center as part of the Left and Right-Center as part of the Right, then 45.62% of my links are Leftist, 21.61% are Centrist, and 32.77% are to the Right.

Top 10 Most Cited

  1. The New York Times (10%)
  2. The Washington Post (6.35%)
  3. Reason (6.02%)
  4. The Economist (5.54%)
  5. Brookings Institute (4.58%)
  6. National Bureau of Economic Research (4.34%)
  7. American Enterprise Institute (4.26%)
  8. Vox (3.45%)
  9. The Atlantic (3.37%; tie)
  10. The Wall Street Journal (3.37%; tie)



Overall, not a bad mix.