Fact checking the US Elections

TL;DR: Based on politifact data, it appears that lies and false statements are depressingly common in the run-up to this years US election, and furthermore that the TV talking heads are as bad as the politicians.


I was looking at the latest updates on http://www.politifact.com and thought it may be interesting to pull some comparisons of how truthful the claims being made by different people are. As an arbitrary choice point, I decided to look at what percentage of overall claims are “Mostly False” or worse. I’m in two minds whether “Half True” and worse would be a better measure…

Some caveats:

  • I am just basing this on the per-person summaries on politifact
  • I am unsure whether these summaries are only valid for a specific timeframe (e.g. there may be comments which are a decade old), although the earliest statement I could find appeared to be from May 2007.
  • There may be multiple instances of the same lie, although I think politifact tries to stop that happening
  • This doesn’t take into account how often a specific lie/statement is made – a statement repeated internationally ad nauseam could be argued to be worse than something quoted in a local paper
  • I’m sure there are lots of other issues in my analysis
  • The results are only based on what politifact has actually analysed – there may be many lies or truths which have not been analysed which could massively skew the results
  • I am aware of the accusations of bias and controversies with politifact – it would be interesting to see a comparison between that site and others. For the moment, I’m willing to trust the politifact data, albeit with a pinch of salt

The Results

Remember, the smaller the number the better – this is a measure of how much of a liar people are… I’ve ordered them from most truthful to least.

First of all, how do the current presidential (and VP) candidates look:

  1. Barack Obama 27% (112/402)
  2. Joe Biden 34% (19/56)
  3. Mitt Romney 43% (67/157)
  4. Paul Ryan 44% (8/18)


And what about the Republican runners-up?

  1. Ron Paul 39% (15/38)
  2. Rick Santorum 52% (24/46)
  3. Newt Gingrich 59% 36/61

Interestingly, Ron Paul ends up looking more truthful than the current Republican candidates…


And the congressional leadership?

  1. John Boehner (R) 53% 28/53
  2. Mitch McConnel (R) 40% 4/10
  3. Nancy Pelosi (D) 40% 8/20
  4. Harry Reid (D) 53% 8/15


And finally, what about the press? Note that these numbers aren’t especially useful due to the very small sample size for many of them.

  1. Arianna Huffington (Huffington Post) 0% 0/4
  2. Donna Brazile (D) 11% 1/9
  3. Paul Krugman (New York Times) 14% 2/14
  4. Sean Hannity (Fox News) 17% 1/6
  5. Gail Collins (New York Times) 20% 1/5
  6. George Will (Washington Post) 25% 4/16
  7. Rachel Maddow (MSNBC) 47% 7/15
  8. Mike Huckabee (R) 48% 14/29
  9. Ann Coulter (R) 60% (3/5)
  10. Bill O’Reilly (Fox News) 60% 6/10
  11. Glenn Beck (Fox News) 65% 15/23
  12. Jon Stewart (Comedy Central) 67% 2/3
  13. Lawrence O’Donnell (MSNBC) 67% 2/3
  14. Rush Limbaugh 80% 12/15
  15. Ed Schultz (MSNBC) 80% 4/5
  16. Steve Doocy (Fox News) 100% 2/2
  17. Bill Maher (HBO) 100% 3/3

These numbers were pretty depressing given that the press, in theory, are supposed to be informing the public. Generally speaking only the written press come out well here – TV doesn’t fare well at all. Note though that the above doesn’t include their own reporting of other people’s lies (which the press may represent as truth), which I think could actually make things worse, and would be an interesting analysis for someone to do.


What does the above show you? Well, that’s for you to decide, but to me it appears the US political engine is full of falsehoods and, for want of a better word, liars. Some may be due to bad research, some may be accidental slips, but the large numbers point to a generally laissez-faire approach to the truth.

What I find even more disconcerting is the role of the press itself. The analysis unfortunately doesn’t cover what the press are reporting, or how, which would be very interesting to see, and may make all the difference. But when you look at just the data above, it appears that the TV press cannot be relied upon. Written journalism comes out much better, albeit with a tiny sample size of journalists.

I suppose the main point to be taken away is don’t believe anything you see/read, as a large proportion of statements being made in US politics turn out to be at least mostly false.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s