I assume that most people here know about PolitiFact, the non-partisan fact checking organization run by the St. Petersburg Times. In a time when rampant dishonesty and factual distortions muddle the political rhetoric on all sides, this resource is invaluable.
The other day, an idea popped into my head: how often, according to PolitiFact, have politicians, advocacy groups, and media figures wandered into the realm of lies and unreality, and which political persuasion is more likely to do so?
PolitiFact doesn't provide aggregate tallies that would answer these questions - so I decided to break the data down myself. Check out the results below the fold.
Limitations:
* The data is limited to the statements chosen out of the totality of American political discourse by PolitiFact for evaluation. Therefore, conclusions cannot extend to groups in their entirety; rather, all conclusions are based on the limited, subjective sample chosen by PolitiFact.
* The sample sizes (number of evaluations) for each group vary, though not too significantly.
* Certain politicians were excluded from the analysis because of the high number of evaluations PolitiFact made for them (the cut off was 13 or more evaluations). Inculsion of these individuals would overwhelmingly weight their honesty over the collective honesty of the group. They are included in the data table for reference.
* A couple of people were eliminated from the analysis because of they were unclassifiable (accounting for only 10 evaluations).
With these limitations in mind, I tallied the following results:
Adjusted for number of evaluations, the distribution looks like this:
I devised and implemented the following system to determine an overall "Honesty Rating":
- As I don't believe that making a correct, truthful statement should positively effect the score (such should be expected from elected officials), I immediately dropped all "True" evaluations.
- I assigned to simple point value to each of the evaluation types: Mostly True – 1, Half True – 2, Barely True – 3, False – 4, Pants-on-Fire – 5.
- I calculated the point total for each group according to the number of evaluations of each type listed by PolitiFact and by the point system above.
- Finally, I divided the point totals by the total number of evaluations to ensure equally weighted averages amongst all groups.
The final weighted Honesty Ratings (where higher numbers indicate less honesty) are:
Republican/Conservative Politicians and Political Groups: 3.02
Democratic/Liberal Politicians and Political Groups: 2.45
Conservative Non-Politicians: 3.36
Liberal Non-Politicians: 2.78
These values are, of course, dependent on the limitations mentioned earlier. They are also limited by the subjective point system described above – it could be argued that a "False" or "Pants-on-Fire" evaluation should receive much higher point ratings than the ones assigned in this analysis. However, a simple escalation of points is simpler.
The conclusion of this analysis is unavoidable: based on the PolitiFact evaluations, Republicans and Conservatives are significantly more likely to make false or dishonest statements than Democrats and liberals, both as politicians and non-politicians. If I were to add in the "True" scores with a point value of 0, the conclusion would remain essentially the same – indeed, it would exaggerate the difference slightly:
Republican/Conservative Politicians and Political Groups: 2.46
Democratic/Liberal Politicians and Political Groups: 1.77
Conservative Non-Politicians: 2.75
Liberal Non-Politicians: 2.02
All evaluations are as of 10/11/09.