Friday, June 24, 2011

More Politi than Fact

The fact-checkers at PolitiFact usually do a pretty good job of analysis. Their ultimate verdicts, though, on their cutesy “Truth-o-Meter”—True, Mostly True, Half True, Barely True, False and Pants on Fire—seem rather too whimsically applied… not to mention that they are described on the site as “rulings”, rather than opinions or decisions: these people take themselves very seriously. One could but wish they were as serious about the work.

A few days ago, they analyzed Jon Stewart’s rhetorical question of Fox News’s Chris Wallace, “Who are the most consistently misinformed media viewers? … Fox viewers, consistently, every poll.” PolitiFact gave that statement a “False,” and then proceeded to describe how Fox viewers, though pretty damned ignorant, aren’t quite as uninformed (a different thing, not in fact tested by PolitiFact’s purported evidentiary polls) as some others. So Stewart’s claim regarding Fox’s peculiar brand of mendacity is not, in fact, demonstrated by “every poll”: just by the all the relevant ones.

Chris Mooney makes this point specifically and emphatically at
My research, and my recent post, most emphatically supports this statement. Indeed, I cited five (1, 2, 3, 4, 5) separate public opinion studies in support of it—although I carefully noted that these studies do not prove causation (e.g., that watching Fox News causes one to be more misinformed). The causal arrow could very well run the other way—believing wrong things could make one more likely to watch Fox News in the first place.

But the fundamental point is, when it comes to believing political misinformation and watching Fox News, I know of no other studies than these five—though I’d be glad to see additional studies produced. Until then, these five all point in one obvious direction.

“Every poll,” to quote Stewart.
PolitiFact readers tend to be a pretty savvy lot—not universally, of course, but generally—and they pretty well let the site have it for its sloppiness. The key, as both Mooney and a plethora of commenters on PolitiFact site and Facebook page noted, is that Stewart’s claim can’t be measured by whether a respondent can name the Secretary of State, but by whether s/he thinks, contrary to all evidence, that Barack Obama was born in Kenya, that climate change is a fraud, or that Saddam Hussein was involved in the 9/11 attacks.

I cite here a small handful of the earliest comments, all of which have more to offer than whoever came up with PolitiFact’s irrelevant analysis:
• Kona Lowell: “Politifact seems to be intentionally misunderstanding. Fox viewers, according to Pew, tend to believe things that are untrue by wide margins. It doesn't mean they don't know who is president.”

• Andrew Bacon: “‎‎91 percent believe the stimulus legislation lost jobs. 72 percent believe the health reform law will increase the deficit. 72 percent believe the economy is getting worse. 60 percent believe climate change is not occurring. 49 percent believe income taxes have gone up. 63 percent believe the stimulus legislation did not include any tax cuts. 56 percent believe Obama initiated the GM/Chrysler bailout. 38 percent believe that most Republicans opposed TARP. 63 percent believe that Obama was not born in the U.S. (or that it is unclear).

• Jeff Ferguson: “I think you guys kind of missed the point of his comment. When he says constantly misinformed, he isn't referring to the type of historical or even current trivia that the Pew studies requested during the polls, but instead things like ‘will the new health care laws require death panels’ and so forth. He wasn’t shooting for a measure of intelligence, because just watching Fox doesn’t necessarily make you stupid, but instead, that the viewers of that show would simply be misinformed about key issues.”

• Jimmy Rumple: “Politifact screwed this one up by not knowing, or ignoring, basic definitions. Stewart said ‘misinformed,’ not ‘uninformed.’ Politifact didn't check for the right thing.”

• Manuel Somohano: “Seriously PolitiFact? Seriously? Are you trying to be ‘fair and balanced now’?......”
In fact, PolitiFact took such a pounding from its own readers that staff writer Louis Jacobson had to be trotted out to… well, I’m not sure exactly what he was doing. Acknowledging the fact that the readers really thought the evidence adduced by PF was insufficient to claim Stewart’s remarks are false, yes. Making the case for the relevance of the evidence, no. Taking responsibility for the laziness of the research and the shoddiness of the analysis (N.B., PolitiFact cites not a single one of the actually relevant polls referenced by Mooney), absolutely not. Mostly… well, mostly it was repeating what had already been written as if that made a difference.

The Stewart fiasco, however, was simply a warm-up act. The real fun came yesterday, when Obama advisor David Axelrod is quoted as saying, “If you're Governor [Mitt] Romney and you say I'm going to turn this economy around, I've got the answers. You don't offer them. Then people have a right to say, why is it that your state was 47th in the country in job creation when you were governor?” PolitiFact then checked out the claim, verified it, linked to a confirming article from that socialist rag, Rupert Murdoch’s Wall Street Journal, and then… get this… declared Axelrod’s claim only “Half True,” anyway.

The pinhead (anonymous) reporter, you see, presumes to know what Axelrod intended to imply, deciding without benefit of evidence that Axelrod was claiming that Romney was responsible for the weak job growth: “some may interpret the Democrats' statistic that Romney is to blame….” Axelrod didn’t say it, and while that may have been what he was hinting at, I frankly doubt it.

What Axelrod, who after all is a senior advisor to President Obama, not a pundit or an elected official himself, is saying is in fact quite the opposite of what the PolitiFact Cretin interpolates. Axelrod knows very well that the economy is not subservient to the will of the Chief Executive, whether that’s the president of the country or the governor of a state. It was Romney, not Axelrod, who suggested otherwise, blaming Obama for the less-than-robust recovery and saying he could do better. Well, Mr. Romney, you had your chance and you blew it big time. So either the present economy isn’t necessarily Obama’s fault (in which case you’ve got nothing to run against) or you bear primary responsibility for the sluggishness in Massachusetts during your tenure as governor, so your allegedly brilliant economic scheme has been proven not to work. Pick one. Axelrod didn’t say it was Romney’s fault; he didn’t imply it. What he did imply, and was right to do so, is that Romney can’t have it both ways.

I’m not exactly sure when PolitiFact got into the implications business. They sometimes consider a statement based on its literal truth, sometimes (apparently) on what someone might think it implies. Sometimes they give a speaker the benefit of the doubt as to what s/he might have meant, e.g., Newt Gingrich’s claim that “The Reagan recovery, which I participated in passing, in seven years created for this current economy the equivalent of 25 million new jobs, raised federal revenue by $800 billion a year in terms of the current economy, and clearly it worked. It's a historic fact.” (Note: PolitiFact radically misinterprets Gingrich’s comment in their headline, suggesting the Reagan policy would have already created all those jobs: even Gingrich isn’t that absurd.) PolitiFact mentions in passing that Reagan’s policies might not have been responsible for the recovery (Cf. the Massachusetts economy not being Romney’s fault), but then proceeds to ignore its own statement.

More to the point, both halves of Gingrich’s assertion are wrong. PolitiFact goes through some arithmetic contortions to achieve a number close to the 25 million jobs figure cited by Gingrich. Of course, he didn’t say “nearly” or “approximately.” Moreover, PolitiFact cherry-picks the numbers they use to fact-check: notice that they start two full years into the Reagan presidency. At the same point in the Reagan presidency we’re currently at in the Obama presidency, June of the third year, unemployment was at 10% and 1.1 million jobs had been added since the low-water mark in employment. Today, unemployment is at 9.1% and about 2.1 million private sector jobs have been created since the “trough.” See how brilliant Reagan was? More to the point, recoveries need to build momentum. It is unreasonable to expect the beginning of an upturn (where we still are now) to generate the same number of new jobs as even the average months once a rebound has hit its stride. In other words, Gingrich’s figure here is mostly true and mostly irrelevant (as opposed to Axelrod's entirely true and partially irrelevant riposte).

PolitiFact itself exposes the silliness of the second half of Gingrich’s claim: coming up with a total of $188 billion, not $800 billion. (Apparently Gingrich is now claiming to have misspoken, intending to argue the $800 billion figure applied over the life of the recovery, not in a single year. Perhaps so—people do misspeak—or perhaps he was lying and got busted.)

What’s remarkable here is that Gingrich’s almost true and completely uncontextualized first half and totally false second half—both of which must be true to make the statement true—still merits a “Half True” from PolitiFact. Yes, precisely the same rating as the demonstrably true, no argument, statistic cited by Axelrod.

The upshot of this is simple: PolitiFact is still a useful tool in determining whether a candidate, office-holder, or pundit is really telling the truth. But those halcyon days when it could savor its reputation as a thorough, reliable, and objective source—a place where a glance at the final “ruling” would resolve a question of fact—those days are gone, and they ain’t coming back.


Jack Marshall said...

I think the Politifact piece was sloppy indeed, but I also think picking which groups are more or less ignorant is a sucker's game. Something like 60% of MSNBC watchers think Bush and Cheney bombed the World Trade Center---I'll rate that as worth 6 "death panels" on the ignorance/delusion scale. Meanwhile, a lot of the individual "ignorant beliefs" cited by the commenters simply are not. Believing that the Health Care Reform Bill, for example, will NOT increase the deficit is certifiable.

manjushri924 said...


Do you have a link or a reference for the MSNBC/truther stat? I've looked, and can find only a reference to an online poll on the MSNBC website: hardly scientific. I also found a more objective poll in which Democrats (as opposed to MSNBC viewers) thought there was evidence that the Bush administation "knew"--but the question was worded so that it was difficult to tell whether a positive response was a conspiracy theory or simply a recognition of the existence of that famous briefing memo.

As for health care and the deficit, I suspect the commenter was short-handing a question on the WPO poll about what most economists believe, not what is necessarily true. The CBO, for example, says the bill will reduce the deficit. They might be wrong, but that's another matter.