Politi-Score: 2,788 Reasons Part 3

Doing calculations on PolitiFact (PF) rulings seems to be gaining in popularity. I noted a comment on the Facebook PolitiFact page the other day with regard to Ron Paul being the highest rated person on the PolitiFact’s truth-o-meter; however, I did offer a response that it was not Ron Paul but losing candidate for governor of Florida, Alex Sink. In fact, between Ron Paul and Alex Sink was the liberal public advocacy group moveon.org.

Just so you don’t have to use the link, the Politi-Score is calculated by assigning points to each ruling as follows:

Ruling	Number of points
True	100
Mostly True	75
Half True	50
Barely True	25
False	0
Pants on Fire!	-25

So, for example, if a politician has one True and one False ruling, that would be 100 for True and zero for False, totaling 100 points. Divided by two rulings, the Politi-Score averages to 50.

Getting back to Ron Paul and Alex Sink, here is a listing of everyone who had ten or more PolitiFact rulings from PF’s inception through the end of 2010, listed by highest to lowest Polit-Score. And yes, Michele Bachmann bottomed out the truthometer as the only one with a negative score (some Half Trues and Barely Trues from factchecks her 2011 SOTU response did put her into positive territory).

Nancy Pelosi and Sarah Palin are about equally truthful according to the Politi-Score.

Of note to PolitiFact detractors, it appears that there is no relationship between being more "truthful" according to PF factchecks and being an election winner.  PolitiFact Florida may be a possible suspect in the hunt for liberal bias: they had the second highest Politi-Score favoring Democrats during the 2010 election cycle, their paper (St. Petersburg Times) endorsed for Florida governor Alex Sink, who was fact-checked by PolitiFact Florida (who also do PolitiFact National), and who *just happened* to emerge with the Numero Uno PF score of 75 (a perfect "Mostly True") while her Republican opponent and winner Rick Scott mustered 50% more fact-check rulings and a Politi-Score of 45, less than "Half True".  On the other hand, Kendrick Meek, Democratic candidate for Senate from Florida, scored lower (46) than the winning Republican (tea party) candidate Marco Rubio (54).

It also appears that those candidates who sort of stand outside their parties--like Ron Paul of the Republicans, who is actually an anti-Federal Reserve, semi-gold-standard Libertarian, and Dennis Kucinich, who is the ultimate  Liberal--scored higher than other "tow the line" Republicans and Democrats...which may say something about our two-party system and may support why the tea partiers have risen to such prominence.

I had initially published a graph showing the total PolitiFact rulings by count: here is a matrix breaking down the count of 1,362 for those with over 10 rulings each, the last column showing the rating breakdown for the grand total 2,788 for comparison purposes.  The second matrix takes those numbers and breaks them down by percentage to make them relative.

NOTE:  The Republican Politi-Score has dropped by two tenths of a point (from 38.9 to 38.7) due to a miscalculation on my part, and which also caused a delay in publishing this post to re-check the rest of my database for accuracy.  Michele Bachman had six Falses and seven Pants on Fire which I had inadvertently counted as six Barely Trues and seven Falses.   Bachmann's lower numbers were of higher magnitude on the Republicans' overall average than I had expected, which just goes to show how much inflammatory rhetoric can have a negative effect.  My apologies for the error, must be my sub-conscious Bachmann bias (ha ha).

Here I have put them into a line graph and compared the over ten rulings in total, by party affiliation, as well as ALL rulings. The result was similar to a “low roller versus high roller” comparison I had done in an earlier post for 13 rulings and over. It seems those with “double-digit” PF fact check rulings tend to be on average, more truthful.

The percentages correlate rather closely except for "Other" in "Pants on Fire".

The legend indicates the Polit-Score for each "line" depicted.  They are pretty consistent except for the "Other" which took a steep turn upwards in the False and Pants on Fire categories; most of this was due to "chain e-mails," a large number of which were anti-Obama but published early in 2008 during the primaries, meaning it was hard to tell if they were from his Republican or Democratic opposition.  So to be on the safe side, they were counted "unknown" and not attributed to either party.  And again, conclusions are premature, although it appears based on PolitiFact's database of rulings, that on average, Democrats make more truthful statements than Republicans.  But, as noted in a similar exercise done in 2009 by Daily Kos:

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.

....Better known in certain blog circles as "selection bias."  Happy Valentine's Day!