Tuesday, May 31, 2011

Engineering Thinking: "Baloney Alert: More PolitiFact Nonsense" (Updated)

The lights are on but the facts aren't home.

Ed Walker over at Engineering Thinking points out a common tactic employed by PolitiFact. Just as they claim ObamaCare is not a government takeover of healthcare because the government doesn't "own the hospitals and the doctors are [not] public employees.", PolitiFact has determined that the Energy Independence and Security Act of 2007 doesn't ban incandescent light bulbs.

Walker writes-
In this case, here are the facts: the government has not literally banned incandescent light bulbs, true. But it has passed regulations requiring light bulbs to have efficiencies that are impossible for them to achieve. There is no technology on the horizon that will allow incandescent bulbs to achieve that efficiency. Manufacturers of incandescent light bulbs have reacted accordingly by shutting down production. Therefore — bottom line — the government has indeed, in essence, banned the use of incandescent light bulbs.
This type of semantic hairsplitting is common for PolitiFact. As Walker notes-
PolitiFact often likes to twist and distort the context of statements, in effect gerrymandering them into one of their preferred liberal themes.
Kudos to Walker for pointing it out. You can read our previous review of Engineering Thinking here. And as always, check out Walker's blog for more.

Bryan adds:

I have one reservation about Walker's analysis, stemming from the fact that at least one of PolitiFact's stories--one for which I had a review in the works--provides an example of an incandescent bulb that meets the new efficiency standards.  The new bulb apparently borrows some halogen bulb technology to produce a bulb similar in significant ways to the traditional incandescent bulb.  Therefore it may overstate matters to say that incandescent bulbs cannot be made to meet the new standards.  On the other hand, the fact is that the traditional halogen bulb cannot meet the standard without significantly increasing the cost of the bulbs.  I solicited  his view of the Philips EcoVantage bulb, the one given as an example in the PolitiFact story:
A halogen lamp is a form of incandescent lamp that has been modified to achieve higher efficiency. The efficiency is achieved by various tradeoffs, including higher cost, more intense output, and much much higher temperature. I do not consider them suitable for standard incandescent bulb replacements (they are better for spot lighting, such as to illuminate artwork), and will not use them in standard fixtures because of the increased fire hazard. Suggesting that they are drop-in replacements is, in my opinion, irresponsible.
Walker's suggestion that PolitiFact's logic is amiss is accurate.  The term "incandescent bulb" has at least two senses in play.  One stands for the traditional bulb.  The other stands for any bulb making use of a glowing filament.  PolitiFact equivocates by ruling on a statement justified under the former meaning by applying the latter meaning.  I appreciate Jeff emphasizing that point with his comparison to PolitiFact's imperial semantic edict with respect to "government takeover."

Note how Philips handles the term "Incandescent" in the EcoVantage brochure:

The second column represents the traditional incandescent bulb.  That's how people use the word.

The EcoVantage bulb comes with a few warnings, illustrating Walker's point about tradeoffs:
Warnings Before using bulb, see operating instructions. Adherence to the operating instructions will reduce the risk of personal injury or fire. The filament capsule contained inside this glass bulb is pressurized, operates at high temperature and could unexpectedly shatter. Should the outer bulb break, particles of extremely hot glass could be discharged into the fixture and/or the surrounding environment, thereby creating a risk of personal injury or fire.

Update 6/5/2011 (better late than never):

Commenter Panta Rei left some valuable information worth appending to the body of the post:

All current incandescents, including Halogens etc, are to be banned by
2020 (USA) and 2016 (EU)

The USA 2007 legislation essentially has 2 components, based on 2012 and 2020

As Ed Walker says, it is in practice a ban, because of the eventual CFL equivalence mandated.

Monday, May 23, 2011

Caught making stuff up, PolitiFact hopes nobody else notices

On May 16, 2011 PolitiFact published a fact check of Laura Ingraham's claim that RomneyCare is "wildly unpopular" in Massachusetts.

Just one problem.  Ingraham didn't say it.

Ingraham was on the O'Reilly Factor for her regular "Week in Review" segment.  O'Reilly brought up Romney's May 12 speech defending his RomneyCare plan, a state health care reform in Massachusetts notable for its similarities to the Democrat-supported PPACA.

Look, I like Mitt Romney. I think he's a really smart guy and I think he would be a good president. I think a lot of the people who might be running would be a good president.

On this, I don't get it though Bill. I mean costs have gone up. It's wildly unpopular.
PolitiFact assumed Ingraham was talking about the program's popularity in Massachusetts.  But that doesn't fit the context.  Romney's concern is RomneyCare's popularity with Republican primary voters.  If Romney doesn't win the Republican nomination then he has no realistic shot at the presidency, period.

Not only did PolitiFact fail to figure that out from the context, PolitiFact has failed to respond to the call to revisit the issue.

It isn't a close issue.  Does the need to protect the brand from admissions of error so easily supersede the desire to tell the truth?  Or is ideological bias the better explanation for the reluctance to correct or clarify?

Jeff adds:

It's important to note how subtle PolitiFact's tomfoolery is. In this case, simply adding two words ("in Massachusetts") changes the entire standard used to rate Ingraham's statement. (Actually PolitiFact used four-"in the Bay State"). Sleights-of-hand like this give PolitiFact an appearance of honest and authoritative fact checking when in reality they're creating a statement out of thin air. It's unlikely a casual reader would catch these additional words that incorrectly frame Ingraham's comment. They've created an amazingly lifelike straw man.

Of course, it's not the first time that PolitiFact has rated a statement that was never said. And while adding words to reframe a fact check is misleading, the same goes for ignoring words that actually were said.

Update: 5/25/11

Thursday, May 19, 2011

PolitiFact's push poll

Crossposted from Sublime Bloviations.

Push polls are not really polls at all; their object is not to measure public opinion, but to manipulate it by providing as many “respondents” as possible with hypothetical, sometimes blatantly false information, about candidates, political parties or initiatives.
Election Law Journal
The third-rate fact checkers at PolitiFact published a push poll yesterday.

Are they trying to make it easy for me to prove they're biased?  I'm literally astounded by some of the things they do.  I couldn't have been more surprised if a magical plesiosaur pointed out the way to Candy Mountain.

Okay, so it was set up as a quiz rather than a poll, but other than that the structure was the same.  Writer Angie Drobnic Holan (who ought to know better) reviewed for readers a couple of (skewed) PolitiFact ratings noting the similarities between RomneyCare and PPACA, the latter also known as ObamaCare.  No mention is made of differences between the two through that point unless we count the different names.

Drobnic continues from there:
Both leave in place the major insurance systems: employer-provided insurance, Medicare for seniors and Medicaid for the poor. They seek to reduce the number of uninsured by expanding Medicaid and by offering tax breaks to help moderate income people buy insurance. People are required to buy insurance or pay a penalty, a mechanism called the "individual mandate." And companies that don't offer insurance have to pay fines, with exceptions for small business and a few other cases.
She moved on from talking about the similarities to talking more about the similarities.  Apparently RomneyCare and ObamaCare are so similar that there's not much point in mentioning differences, if any.

After having the similarities impressed on us, we're supposed to take a quiz to see if we can even tell the difference between the two.  We get ten descriptions which apply either to RomneyCare or ObamaCare.  One is a poll result rather than a feature of the bill, leaving us with nine relevant points of comparison.

The first concerns the individual mandate, an obvious and well-known similarity between the two plans.  The second concerns employer penalties for failing to carry insurance--another fairly well-known feature the two plans have in common.  The third mentions tax credits for those who have trouble paying for health insurance.

Notice a trend?  We're a third through the nine relevant points and the focus is strongly on the similarities.

The fourth and fifth touch on two different ways the plans expand Medicaid coverage.  The first mention of a difference of any significance comes in terms of a similarity (expanded Medicaid coverage).

The sixth mentioned a poll showing people split on favoring one of the health care plans.  The plan was ObamaCare, as it turned out, and Drobnic helpfully achieved an even 41 percent in favor with 41 percent against without mentioning the substantially higher percentage within the latter figure who "strongly oppose" ObamaCare.  The later explanation does make clear that RomneyCare is relatively popular in Massachusetts, but the overall message remains one of equivalency.  Massachusetts is, after all, a predominantly liberal state.

The seventh was a tax credit for small businesses designed to provide an incentive for the purchase of employee health plans.  This represents the second difference of some significance, as apparently no such provision exists in RomneyCare.

The eighth was a gimme.  It talked of "experience" showing that the plan would not lower the cost of insurance premiums.  Those who do not realize that ObamaCare has no real track record yet might consider the statement to apply to either plan.  And it's a pretty good prediction for ObamaCare, but PolitiFact insists that it applies to RomneyCare alone.  Perhaps that's based simply on the past experience angle.

The ninth referred to the establishment of an outcomes research board featured in ObamaCare.  This is the first major difference between the two, though it is described in entirely innocuous terms throughout the story despite the fact that it was roundly criticized from the right for its similarity to Great Britain's NICE.  And even NICE sounds pretty nice until you hear about some of the ways they meddle in patient care.  Conservatives think the outcomes research feature in ObamaCare is intended as a rudimentary NICE, poised to blossom into the real thing as the federal health care bureaucracy burgeons.  PolitiFact makes it sound nice, but not NICE at all.

The tenth touches the tip of the iceberg for the most significant differences between ObamaCare and RomneyCare:  the taxes.  The PolitiPush poll mentions an increase in the Medicare payroll tax applied to families with earnings over $200,000 and for individuals earning over $250,000.  The increase from 1.45 percent to 2.35 percent applies only after the aforementioned income thresholds, which isn't how PolitiFact described it.  PolitiPush also mentions a 3.8 percent hike on investment income.  Find the balance of the iceberg here.  PolitiFact couldn't mention them all because there simply wasn't room in a quiz of this kind.

The ordering of the quiz questions seems designed to establish the impression of strong similarity.  The differences may have been chosen with an eye toward public acceptance, though to be fair the taxes mentioned were among those expected to generate the greatest revenue.  The fees imposed on insurance companies warranted no mention in the PolitiPush poll.

Did we miss anything?  Well, yeah.  There's the IPAB, the continued federal government crush of unfunded mandates on the states and a set of regulations that may end up putting private insurance companies out of business in the United States.

Push polls like PolitiFact's suppress certain information on the publisher's behalf while leading the person answering the questions toward a set of desired beliefs.

What possessed PolitiFact to publish something like this?  Disgusting.

Sunday, May 15, 2011

Drawnlines Politics: "Unfair Politifact Rating for LeMieux?"

The political blog "Drawnlines Politics" serves up a sizzling observation about PolitiFact Florida's recent rating of a campaign ad by Republican candidate for Senate George LeMieux:
Politifact Florida recently rated Senator George LeMieux's claim that he never sought an earmark only "half true", while at the same time calling the individual parts of his claim "true." LeMieux maintains a spotless record of earmark requests, though Politifact chastises him for voting for bills that had earmarks already included in them.
read the rest
The post makes a good point.

PolitiFact's headline and deck make it look like it is "Half True" that LeMieux did not request earmarks and voted to ban them.  But both those things are true.

The "Half True" comes from another aspect of the ad, where it criticizes the incumbent, Sen. Bill Nelson (D), for voting for earmarks:
Again, the wording of the earmark claims in LeMieux's ad are technically true. LeMieux has not personally requested any earmarks, and he has pushed to ban them all. But we think it's a bit hypocritical for LeMieux to call out Nelson for voting for a bill that included billions of dollars worth of earmarks (and specifically the infamous Bridge to Nowhere).
It's true that the comparison drawn in the LeMieux ad falls well short of logical precision.  But the same holds true for the PolitiFact rating, as PolitiFact claims to rate a set of two claims from LeMieux but instead bases the final rating on an implicit claim contained in the comparison between LeMieux and Nelson.

In the end, PolitiFact's story is about as misleading as the LeMieux ad.

Friday, May 13, 2011

Associated Content: "The Misuse of PolitiFact's Rankings"

Anonymous contributor "Thales" of Yahoo's Associated Content provides us yet another reminder of how PolitiFact's collected ratings are pretty much useless without a check on selection bias:
There's no way they could cover every political statement made by even a couple politicians, let alone dozens of them. But if PolitiFact hasn't given a comprehensive evaluation of all of the statements of these politicians -- or at least equally representative samples from each of them -- then how can they be used to rank these people in terms of veracity or ignorance? They can't.
Thales also notes PolitiFact's tacit approval of the misuse of its rankings:
PolitiFact has said nothing about this use -- I would argue, misuse -- by The Daily Beast of their Truth-O-Meter rankings. Although The Daily Beast is accurately reporting PolitiFact's evaluations, they're doing it in a way that leads to false (or at least questionable) implications. This kind of situation -- accurate statement with questionable implication -- has been variously rated by PolitiFact in the past as being anywhere from "True" to "Half True" (again, they're inconsistent in these rankings). But, to date, PolitiFact and the Truth-O-Meter remain silent on The Daily Beast's "10 Most Ignorant Politicians" list.
As to Thales' first point, PolitiFact would not need to rate every political statement in order to negate selection bias if the selection process were random.  But that process is not random, unless we disbelieve PolitiFact's description.

And to amplify Thales' second point, PolitiFact can hardly scold the Daily Beast for its misuse of the data without undercutting its own practice of directing readers to judge individuals according to their Truth-O-Meter "report card":
(W)e thought it would be timely to look at Beck's record on the Truth-O-Meter. As you can see from the running tally in his PolitiFact file, we've rated 17 statements by the Fox News talk show host. It's fair to say that record skews toward the False end of the Truth-O-Meter.
Yes, and?

Tuesday, May 10, 2011

PFB Smackdown: "WisFactCheck" and Wisconsin job creation

We at PolitiFact Bias emphasize in part sets of anecdotes as an evidence of PolitiFact's ideological bias.  Where the combined sets of anecdotes soft on liberals and those harsh on conservatives outweigh their opposites, a solid (though short of absolute proof) evidence of bias results.

Since the comparison between the sets of anecdotes affects the case for bias at PolitiFact, the "PFB Smackdown" feature takes the attempts by liberals to show a right-leaning bias at PolitiFact and exposes the errors therein.

Which brings us to "WisFactCheck":
Another day, another outrageous claim from the right-leaning Politifact.

The latest eyerolling claim from Politifact is that Sen. Lena Taylor's claim that Walker "has not created at single job" was not accurate:
We at PFB admit that PolitiFact's work exhibits problems that harm both conservatives and liberals at times.  And, of course, we expect that the legitimate examples of the former exceed those for the latter.  So what's the evidence in this case?

Thursday, May 5, 2011

Red State: "Yet Again Politifact Shows Itself to be Leftist Propaganda Masquerading as an Agent of Truth"

Red State's Erick Erickson noticed a compelling similarity between PolitiFact and the political left:
Now Politifact is going after Senator Rob Portman for saying
As an immediate bridge, we should increase access for oil exploration and production in energy-rich areas of the country like the Outer Continental Shelf, and in parts of Alaska
Get ready. What is Politifact saying?
Not so fast, say the experts. Pretend that environmentalists dropped all objections to drilling for oil on the Outer Continental Shelf — that area that lies offshore between states’ jurisdictions and the end of United States oceanic boundaries. Also pretend that the public decided its need for oil trumped what environmentalists see as the sanctity of the Arctic National Wildlife Refuge, or ANWR. Since we’re just pretending, everyone join in: Drill, baby, drill.
Then wait.
See ya around 2021.
That is exactly what the Democrats said in 2001 when George Bush tried it. Today is the tomorrow the Democrats told us yesterday was too far away to do anything about. Politifact continues the Democrats’ talking points.
I could wish that Erickson, in the entirety of his post, provided a more thorough account of PolitiFact's flim-flam.  But his basic point is accurate enough, even if it is expressed in terms unlikely to move persons not already in agreement with political conservatism.

I'll fill in a few details to help out the liberals and progressives.

Even if we forgive PolitiFact for taking "immediate" in an absolute sense probably not intended by Portman, PolitiFact loses sight of the fact that merely demonstrating a willingness to exploit our own energy resources immediately reduces dependence on foreign energy.

"Why is that?" the progressive or mainstream journalist might wonder.

When foreign nations see the U.S. move to exploit its energy resources it immediately starts to close a window of time foreign nations can use to hold us over a political barrel.

Suppose OPEC starts an embargo.  The longer OPEC can expect the U.S. to provide an increase in its own energy exploitation, the stronger the political power of the embargo.  Anything that shrinks the window immediately shrinks the power of the embargo.

So even taken hyper-literally, Portman's claim is no less than "Barely True" using PolitiFact's grading system (Portman was graded "False" on this item).

The PolitiFact story is blind to these facts.  The story is reliant on the opinion of an official linked to the administration (U.S. Energy Information Agency) and on extrapolated information from other of PolitiFact's fact checks.