Politi-Score: More Red in Ohio

During the 2010 election cycle, a trend was noted concerning the PolitiFact (PF) State Truth-o-meter rulings: that the “redder” affiliated state newspapers (ones that, for example, endorsed more Republican candidates) tended toward rulings, when averaged out through the Politi-Score, to either be more balanced between the parties, or to even favor Republicans, especially when compared to PF National. So how did that work out now with ALL the rulings instead of two months’ worth?

Before starting, I should note that I will be changing my “About This Blog” page shortly to reflect I am going with what I call “Wide Measure” rulings only from now on. I do like the innateness of 100 points for True, 50 points for Half True and zero for False.

That being said, here are the Democrat and Republican Politi-Scores compared for all states, for “former and current office-holders” only, according to that measure.

The “50” for Half True is the first focus on this chart. The averages ranged below 50 (Half True), but only PolitiFacts National and Florida, who have some of the same writers, had scores above 50 for the Democrats. Since together they dominate over 3/4 of the rulings, it caused the “ALL” scoring in the last column to average to above 50, even though all other PF states scored 50 or under collectively (for both parties).

During the 2010 election season, the ruling spreads and Politi-Scores were given in this somewhat cluttered graph (if I say so myself), with the “Blue” PF states, “Red” PF states and swing (purple) PF states designated, as determined by their past candidate endorsements, particularly in the 2004 and 2008 presidential elections.

During the 2010 Election Cycle, PolitiFacts Texas and Virginia rated Republicans "truthier."

I thought that a simpler way to present any data disparities found between the Democrat and Republican Politi-Score would be to combine them: to put the Blues—PF’s National, Florida, Georgia and Oregon, as one collective average, against the same for the Red/Purples—PF’s Ohio, Rhode Island, Texas, Virginia and Wisconsin. This was the final Politi-Score result for Democrats and Republicans (for current and former office-holders):

The “even-handedness” between the two parties for the “Red/Purple” PF states becomes greatly apparent here: the “Blues” of PF National etc. have a 22% gap between Democrats and Republicans while it’s roughly 10% for the “Red/Purple” PF states. The “Red/Purples” have a much lower Politi-Score for both parties, but significantly lower than the “Blue” Politi-Score for Democrats; indeed, this score stands out like a “sore thumb” relative to the other scores, the difference to the other three scores ranging from 22 to 30% in favor of the PF National (etc.) Democrats.

For the Truth-o-meter range of scores, here is a line graph which represents the across the board percentages of rulings on which the Politi-Scores are based, with the blue lines for Democrats and the red dashed lines for Republicans.

What can be gleaned from this chart is the close proximity of the lines for Republican Barely True, False and Pants on Fire rulings between PF’s “Blue” and “Red/Purple.” The large disparities occur in True and Mostly True, trailing off somewhat at Half True. In other words, it appears the True and Mostly True rulings are what carries the high Politi-Score for the Democrats, but at the same time, both PF “colors” (all the states on average) find almost the same percentage of Republican statements Barely True, False and Pants on Fire.

Conclusions? Back in 2009, Cinematical, the PolitiFact rater from Daily Kos who developed his own scoring system, left out the True rulings because he didn’t “believe that making a correct, truthful statement should positively effect the score (such should be expected from elected officials)."

He might have a point, but the scores would still favor the Democrats. A potential theory arises, if there is indeed liberal bias (which the Politi-Score of other PF states appear to point to) by PolitiFact, it would be in (PF National) selecting more statements that appear to be True as made by Democrats, or those that could more easily be ruled as True. It should be noted that PF’s counterpart fact-checker Factcheck.org appears to verify far more “False” statements, and it appears that PolitiFact’s affiliates may even be doing that judging from their Politi-Scores as compared to PF National (etc.).

As to liberal bias for “certain,” that cannot be definitively proved. Neither can that one comment I hear frequently that “Truth has a liberal bias.” At least PF Ohio wouldn’t agree, since they had more red than anyone else.