Grading PolitiFact *Liberal*-Style: Silence is Golden like Fine Wisconsin Cheddar

Most of the PolitiFact rulings reviewed for the purposes of Grading PolitiFact (PF) Liberal Style are not that far off on how PolitiFact actually ruled on them; they are usually argued a notch or two apart from the Truth-o-Meter ranking (unfavorably to Liberals). Rarely do I come across rulings that are way off the mark. And rarely do I encounter a ruling where it appears the “rulee” is being penalized by several Truth-o-Meter grades.

When I looked at this ruling in PolitiFact’s banner listing, on the State Democrat Party of Wisconsin who said (here is the youtube video of one of the times it made this claim, graphically depicted below)  that they had collected “over 1 million” signatures for the Governor Scott Walker recall, and the by-line under the “False” Truth-o-Meter of “Short by 70,000,” I recoiled as my brain computed the percentage, and I immediately went to a Politi-Psy post on 13 PolitiFact rulings from the first quarter of 2011 where I was attempting to ascertain just how far off statements fact-checked with “numbers” could be to qualify for a Mostly True. Mostly True because, as I recalled, the range to be “off”  was from 7 to as much as 15%. And yes, my Mostly True review indicated 7% off qualified for Mostly True. So here was one, I thought, that should be closer to the Mostly True side. Seventy Thousand by itself looks like a big number, but everything is relative--70,000 is 7 percent of 1 million. Seven percent off. So…why were they given a False?

According to PolitiFact Wisconsin’s Tom Kertscher:

On March 12, 2012, the Government Accountability Board announced that 931,042 signatures had been submitted to recall Walker.

That’s a lot of signatures, but not 1 million.

On the day the GAB announced its official figure, a Milwaukee Journal Sentinel reporter asked state Democratic Party spokesman Graeme Zielinski in an email why there was a difference of nearly 69,000 between the number of signatures the party announced and the number counted by the elections board.

Zielinski replied he didn’t know, then backtracked, calling the "over 1 million" claim an estimate.

Said Zielinski: "Nobody knows how many signatures were submitted because of the sheer volume and the way the signatures were counted in the closing days. That's why we estimated our total."

But the number was never presented as an estimate….

The Democratic Party of Wisconsin claimed it submitted to the state elections board "over 1 million signatures" to recall Walker -- a claim that has been repeated and repeated. The board said just over 931,000 signatures were submitted
.
We rate the party’s statement False.

The ruling also noted that the Milwaukee Journal Sentinel had reviewed “ a random sampling of 500 Walker recall signatures found that 15 percent of them could not be verified.” But as noted, 15% falls at the outer bounds of Mostly True as well.

So PF says it wasn't presented as an estimate...why doesn't it say 1,000,321 then?

With the Mostly True post in hand with its links to the rulings I had reviewed with the 7-15% variance, I posted a comment at the PolitiFact Wisconsin Facebook page with those that could be easily compared to this ruling, in particular one on Florida Governor Rick Scott, who had stated that his budget had cut $2 billion in taxes, when in fact PolitiFact Florida could only find $1.7 billion, a 15% difference, and yet handed Rick Scott a Mostly True. I imagine that Scott repeated this "accomplishment" as well. So why did he not get a False like the State Democrat Party of Wisconsin, who were not as far off as Scott? As I noted in my Facebook comment, even if for the sake of argument there was a claim of 1.05 million signatures, which is OVER one million, the official count is still about 7% off what the State Democrat Party said it was.

As for Kertscher’s claim that the State Democrat Party of Wisconsin, on his inquiry, called it an estimate when it wasn’t, I would say that many times when PF inquires with the person who made the statement, they do not respond. One good example is Michele Bachmann. If the State Democrat Party of Wisconsin had not responded to PolitiFact’s inquiries with what PF believed to be a disingenuous claim, and instead remained silent, would they have received a Half True for not appearing to mislead PolitiFact with excuses? Was it a mistake for the State Democrat party to even respond to PolitiFact? Maybe, as my title alludes, silence IS golden.  

Additionally, if it wasn't an estimate, one would think the State Democrat Party would have given the exact number instead of simply saying "over one million."

It appears PF Wisconsin “downgraded” the Mostly True to False because the story was repeated, and because when questioned called it an estimate (and that there was a comingling of signatures from another recall, which PF determined not to be true). But that is a pretty severe downgrade when compared to other fact-checks of the same nature involving fairly straight “numbers” claims, especially in this instance when it’s not finances, which are often budget and forecast numbers which are highly fluid, often based on soft or subjective assumptions, and subject to change.

As an aside, I would say that those who performed the official count were Republicans (worked for or were appointed by the Republican Governor) and may have had reason to disqualify and throw out signatures for reasons Democrat counters might have left them in. Kertscher never addressed that there might also be a bias in how the signatures were counted which may have made the total count lower.

So, is the statement that “over one million signatures were collected” when in fact 931,042 signatures were, enough to be called inaccurate? It’s not even enough “to give a different impression” for a Half True ruling to apply. The Democrats collected roughly close to one million signatures. If they had said over one million and the official count was under the required number (540,208) to be collected, I’d say that was a candidate for Pants on Fire. If they made just barely over the required number while saying they had a million, I might call it a False. If they had about 700,000 I might call it a Mostly False. But being that close to one million is not really inaccurate enough.

And comparing it to Republican rulings of the same genre, we have a very good case of “PolitiFact bias” to the conservative side. One other intriguing item of note: the writer, Tom Kertscher, is the second most “Republican-favorable” writer at PolitiFact in terms of the 2011 Truth Index (Sean Gorman at PF Virginia is Number 1). The Truth Index on rulings he’s done for Republicans “more than doubles” that of the Democrats (a negative 18.4 for Republicans but a much lower negative 43.6 for Democrats, for all 2011 rulings).

While certain conservative PolitiFact detractors are likely not going to notice this ruling, it would do them well to utilize it, because of this connection. And it makes me wonder, if this had happened in Ohio, what writer Tom Feran (PF’s most Democrat-favorable writer) would have awarded it. I bet it wouldn’t have been a False.