PolitiFact is a hot freakin’ mess
Editor’s note: The following post is from Ben Kirby, author of the Spencerian. Please visit his site here.
I have had my words for PolitiFact. Whether it’s their riduculous rulings on how many donuts the Governor of Florida sold – helpful! — or their “lie of the year” which was happened to be, well, true, it’s all what the kids call a hot mess.
I don’t want to get into it too much, but they screwed up a ruling on the State of the Union last night – and it’s bad. So egregious was their error, they have been forced to backtrack:
Have private-sector jobs grown by 3 million in 22 months, with the best annual totals since 2005?
EDITOR’S NOTE: Our original Half True rating was based on an interpretation that Obama was crediting his policies for the jobs increase. But we’ve concluded that he was not making that linkage as strongly as we initially believed and have decided to change the ruling to Mostly True. The original article is archived here.
Skipping ahead to the ruling:
Our ruling
Obama is correct on the numbers. By mentioning his policies, he’s making a modest linkage that they deserve credit for the improvement when economists say they are just one factor. On balance, we rate the claim Mostly True.
Great, Mostly True versus the original Half True.
That right there is at the heart of my problem with PolitiFact. What the fuck is the difference? Half True? ”Mostly” true? Did private-sector jobs growby 3 million in 22 months or not? It’s a binary equation: yes or no. True or false.
It probably won’t be, but this could be the end for the Pulitzer Prize-winning PolitiFact. See, the thing about checking facts is you have to at least appear like you don’t have an agenda (and between naming the true Democratic ads discussing Medicare as the “lie of the year”, and one of the Democratic president’s lines about the recovering economy in the State of the Union as only “half true” when it was entirely true, it appears as though PolitiFact may have a bias, here). You also have to appear like you know what you’re talking about.
And that’s more than half true.
PolitiFact has a long history of rating technically true claims as less than true, and the rating for Obama falls in line with that, particularly in light of editor Bill Adair’s pledge over the summer to take credit and blame more closely to account in ruling on statistical claims.
While it’s true that PolitiFact is a mess, Kirby’s example is a poor one aside from the ripples it caused in the liberal blogosphere. A much better example of PolitiFact’s ineptitude comes from another SOTU claim: that the president ended a rule that treated spilled milk the same as spilled oil. PolitiFact rated that claim “true.” Yet prior to the date PolitiFact determined the president ended that rule, PolitiFact Virginia ruled “Pants on Fire” the claim of a Virginia politician that an EPA rule requires the same treatment for spilled milk as for spilled oil.
The two ratings are irreconcilable. Moreover, both contain monstrous journalistic “holes.” For example, neither PolitiFact story notes that President Bush left behind an EPA directive that would have prevented milk produces from having to comply with the same rules as oil producers wrt spills. President Obama put a hold on that directive (common with incoming presidents) then diddled around with it for two years before finalizing the exemption for milk producers in early 2011–in time to take credit in his SOTU speech for an exemption Bush would have pretty much accomplished in 2009 if not for Obama’s actions (treatment of raw milk spills constituting the lone caveat, from what I can tell).
The revised rating for Obama’s job creation claim features no clear backing rationale (as in why exactly PF was initially wrong in gauging the degree of credit Obama was claiming). It reinforces the impression that the “Truth-O-Meter” functions as the fig leaf of objectivity for the opinions of the PolitiFact staff.