Previous Month | RSS/XML | Current | Next Month

May 29th, 2012 (Permalink)

Is British science reporting worse than American?

I don't know the answer to this question, but Daniel Engber claims that it is in a Slate article published today:

I'm not the first to notice this trend…but it has started to worry me. A great garbage patch of science journalism has been forming across the Atlantic, and bits of flotsam are washing up on our shores. What makes the Brits so susceptible to these ginned-up studies and publicity stunts? And what happens when their faux research starts drifting across the Internet?

I tend to agree with Vince Kiernan:

Vince Kiernan, a veteran reporter who now studies the history and practice of science journalism at Georgetown, doesn't see much difference between the coverage in Britain and America.

Even so, it probably won't hurt to keep our eyes open for "bits of flotsam…washing up on our shores" from across the pond.

Source: Daniel Engber, "Dodgy Boffins", Slate, 5/29/2012

I still believe in Global Warming.  Do you?
May 25th, 2012 (Permalink)


The Heartland Institute is a midwestern conservative think tank that "question[s] claims that man-made global warming is a crisis". Earlier this month, it briefly displayed a billboard in the Chicago area showing a mugshot of Ted Kaczynski, the Unabomber, with the legend "I still believe in Global Warming? Do you?" Apparently, the billboard was intended as the first of a series, with additional ones showing Charles Manson, Fidel Castro, and Osama bin Laden.

If there is any doubt that this is an attempt to argue guilt by association, the institute's president made it clear in an accompanying statement:

“The most prominent advocates of global warming aren’t scientists,” said Heartland’s president, Joseph Bast. “They are Charles Manson, a mass murderer; Fidel Castro, a tyrant; and Ted Kaczynski, the Unabomber. Global warming alarmists include Osama bin Laden and James J. Lee (who took hostages inside the headquarters of the Discovery Channel in 2010)." …

Are you saying anyone who believes in global warming is a mass murderer, tyrant, or terrorist?

Of course not. But we are saying that the ethics of many advocates of global warming are very suspect. [Here, they go on for four paragraphs to talk about a couple of scandals involving scientists who don't question claims that man-made global warming is a crisis.] Poor judgement…believing the ends justify the means…desperation. Now do you see why we really shouldn’t be surprised to learn that Charles Manson, Fidel Castro, Ted Kaczynski, and other famous criminals believe in global warming?

No, sorry, I don't see it. If anything is the result of poor judgment, believing the ends justify the means, and an act of desperation, it's this billboard campaign. In a statement made after the billboard was ended, Bast tried to explain the reasoning behind it:

This billboard was deliberately provocative, an attempt to turn the tables on the climate alarmists by using their own tactics but with the opposite message. We found it interesting that the ad seemed to evoke reactions more passionate than when leading alarmists compare climate realists to Nazis or declare they are imposing on our children a mass death sentence.

In addition to guilt by association, Bast seems to believe that two wrongs make a right. Apparently, some of Heartland's supporters were rightly upset about the billboard and it was pulled after only a day. The institute then claimed that the "experiment" with the billboard was over, which sounds like an ex post facto excuse. The word "experiment" does not occur in the original statement about the billboard campaign, which said that more billboards were planned. It seems that the "experiment" was abandoned only after the supporters objected.



May 24th, 2012 (Permalink)

New Book: Unnatural Acts

Robert Todd Carroll, of The Skeptic's Dictionary fame, has a new book out called Unnatural Acts: Critical Thinking, Skepticism, and Science Exposed! Unfortunately, it's only available as an e-book. You can read a sample online via Amazon's "Look inside!" feature.

May 23rd, 2012 (Permalink)


Doctor Who Helped Capture Bin Laden
Jailed for Treason

I can't wait to see that episode.

May 21st, 2012 (Permalink)

Java Jive

Two cups of coffee a day cuts overall risk of dying by 10 percent, research shows

Before you sing "Waiter, waiter, percolator!", read down to the third paragraph from the end:

Be careful - this doesn't prove that coffee makes people live longer, only that the two seem related. Like most studies on diet and health, this one was based strictly on observing people's habits and resulting health, so it doesn't prove cause and effect.

This article closely adheres to the media template for a health news story: first, put the good news in the headline and the main part of the story; then, in the last paragraph or few, include the bad news, which is: "Made you look! The good news is wrong." Fortunately, there were a couple of articles in major news sources that did not adhere to this template, and actually did a good job explaining the study to a lay audience so that I don't have to. For instance, here's one from Maclean's in Canada:

…[B]ecause of the nature of this “observational study”―where no intervention is introduced, where subjects aren’t randomized, where researchers just look at the link between an exposure to something and a certain outcome―the authors of the article were careful to acknowledge that, “Whether this was a causal or associational finding cannot be determined from our data.” Yet, this brief but crucial note seemed to be lost in some of the reporting on the subject or referenced only several paragraphs after hyperbolic headlines and opening sentences. … But this kind of science has its limits. With observational studies―as opposed to controlled experiments where interventions are introduced―scientists look at phenomena “out there.” All they can show is an association between two factors, such as coffee and mortality, not that one factor caused the other, or vice versa.

So, the headline above is unjustified in saying that coffee "cuts" the risk of dying, since that means that coffee causes the decline. Instead, the most that can be justifiably stated is that coffee is "linked to" or "associated with" a lower risk. Such a statistical association does not justify a change in behavior, such as starting to drink coffee if you don't already do so, or drinking more than you do now, as is explained in The Boston Globe:

Going for that extra cup of coffee based on this study finding, however, would be a foolish thing to do since an outside expert, the journal’s editor, and the study investigator all told me they didn’t believe the modest finding should push Americans to drink even more coffee than they already do.

It's a good thing that not all news publications report on health as well as Maclean's and The Boston Globe do here: it would just about put me out of business!


May 18th, 2012 (Permalink)

Blurb Watch: Dark Shadows

The latest Johnny Depp/Tim Burton movie is a remake or parody―I'm not sure which―of the old Dark Shadows gothic soap opera. Here are two blurbs from an ad for the new movie, together with context from the reviews from which they were taken:

Blurb Context Source
Manohla Dargis, THE NEW YORK TIMES
Barnabas’s liberation does the same for Mr. Depp’s performance, and it’s delightful to watch how the actor handles the vampire’s readjustment to the world of the living.… “Dark Shadows” isn’t among Mr. Burton’s most richly realized works, but it’s very enjoyable, visually sumptuous and, despite its lugubrious source material and a sporadic tremor of violence, surprisingly effervescent. Manohla Dargis, "A Vampire Thirsty and Bewildered", The New York Times, 5/10/2012
We know we can expect a pitch-perfect performance by Depp, who plays Barnabas with a lasered intensity, and we know Burton's sets and art direction will be spectacular. … Here Depp seems to inhabit a world of his own, perhaps in self-defense. The others seem to be performing parodies of their characters. "Dark Shadows" begins with great promise, but then the energy drains out. Roger Ebert, "Dark Shadows", Chicago Sun-Times, 5/9/2012

The Ebert quote is a typical blurb in which a positive statement about an aspect of a movie is taken from a fairly negative review―Ebert gives the movie only two-and-a-half stars out of a possible four. The Dargis contextomy, however, is odd: she actually wrote that “'Dark Shadows' isn’t among Mr. Burton’s most richly realized works" rather than that it is "richly realized". In other words, the blurb almost reverses Dargis' meaning. Now, maybe she thinks that the movie is richly realized, though other Burton works are more so, but that's not what she wrote. What makes this blurb especially bizarre is that, unlike the Ebert review, Dargis' review is mostly favorable and the adwriter could have easily found less misleading things to quote.


May 16th, 2012 (Permalink)

How to Hoax Wikipedia

The Atlantic has an interesting article on a historian whose university classes have perpetrated hoaxes on Wikipedia. I think it's a breach of academic ethics to require or even allow students to engage in hoaxes for a class. That the hoax is about unimportant matters and is soon admitted diminishes the wrongness, but doesn't eliminate it. I'm surprised his university doesn't put a stop to it.

Moreover, one of the lessons that the students should learn is the tenacity of misinformation: we've seen repeated examples on this weblog of how misleading information in the form of contextomies, wildly false statistics, and conspiracy theories persist in the media despite repeated debunkings. There's even some research that tends to show that debunkings may actually reinforce false beliefs in the minds of people exposed to them. So, once a falsehood is let loose into the media world, it takes on a life of its own. Even the hoaxer's confession may not be enough to squelch it.

Also, it isn't necessary to commit a hoax to teach skepticism about Wikipedia and other online sources of information. I've been doing that here on this weblog for years without once committing a hoax. Moreover, this isn't the first time that someone has hoaxed Wikipedia (Wikipedia itself has a helpful list of occasions it's been hoaxed, see the Source below)―and it won't be the last, either―so a teacher can make the needed point by looking at such past hoaxes without creating new ones. How many times do you need to hoax Wikipedia to make the point that it's vulnerable?

I'm skeptical about the author's claim that differences in the social organization of Wikipedia and Reddit account for differences in how fast the two hoaxes discussed were exposed. Maybe so, but one data point is not sufficient to generalize from (see the Fallacy, below), and it's possible that the difference is the result of luck. Also, it seems clear from the account in the article that the hoax on Reddit was undermined by some easily fixed mistakes, and anyone who reads the article ought to be able to create a more successful hoax. No doubt some of them will; I just hope that it's not a college student doing it for credit.


Wikipedia Watch: 6/30/2008, 10/22/2008, 1/25/2009, 3/22/2009, 7/21/2009


Fallacy: Hasty Generalization

May 13th, 2012 (Permalink)

A Mother's Day Puzzle

A woman was driving her son to soccer practice when their car stalled on some railroad tracks. She tried to restart the car, but only managed to flood the engine, and the onrushing train struck the car. The woman died instantly, but her son was rushed to the hospital in an ambulance, his condition critical. When the ambulance arrived at the hospital, the boy was taken into the emergency room. The nurse on duty came running up to provide first aid but, upon seeing the boy, turned white and exclaimed: "Oh my God, it's my son!" How is this possible?


May 9th, 2012 (Permalink)


Obama’s Evolution on Same-Sex Marriage

Mitt Romney's flip-flop on stay-at-home moms: Will it matter?

Brendan Nyhan, of the much-missed Spinsanity website, has an article in the Columbia Journalism Review on a double standard in the media. According to Nyhan, when President Obama changes his position on an issue, such as same-sex marriage, he "evolves"; whereas, his presumptive challenger, Mitt Romney, "flip-flops" when he changes his. "Evolution", in this context, has a positive emotional charge to it, since it suggests growth and progress. In contrast, "flip-flop" is a pejorative term that suggests an unprincipled effort to gain political advantage. Check it out.

Source: Brendan Nyhan, "Obama ‘evolves,’ Romney ‘flip-flops’", Columbia Journalism Review, 5/8/2012

May 3rd, 2012 (Permalink)

The Zimmerman Contextomy: The Sequel

Goldfinger said, "Mr Bond, they have a saying in Chicago: 'Once is happenstance, twice is coincidence, the third time it's enemy action.'"―Ian Fleming, Goldfinger

The New York Times is reporting that another NBC employee has left, apparently as a result of the Zimmerman 911 call editing scandal. The third casualty is Lilia Luciano, an on-air reporter who worked on The Today Show and was apparently based at the troubled Miami bureau. The Times reporter is not sure whether she resigned or was let go, but other news outlets claim that she was fired. It's unclear when she left NBC, though it could have been about the same time as the still unidentified producer was let go and we're just hearing about it now.

According to TVNewser, which seems to have broken this story, Luciano got in trouble over a Today Show segment that had a somewhat different contextomy than the one we've seen before:

Contextomy Context
Zimmerman, a neighborhood watch volunteer called 911 reporting, quote, a suspicious-looking guy. This guy looks like he's up to no good or he's on drugs or something. He's got his hand in his waistband. And he's a black male. Zimmerman: Hey we've had some break-ins in my neighborhood, and there's a real suspicious guy, uh, [near] Retreat View Circle, um, the best address I can give you is 111 Retreat View Circle. This guy looks like he's up to no good, or he's on drugs or something. It's raining and he's just walking around, looking about.

Dispatcher: OK, and this guy is he white, black, or Hispanic?

Zimmerman: He looks black.

Dispatcher: Did you see what he was wearing?

Zimmerman: Yeah. A dark hoodie, like a grey hoodie, and either jeans or sweatpants and white tennis shoes. He's [unintelligible], he was just staring…

Dispatcher: OK, he's just walking around the area…

Zimmerman: …looking at all the houses.

Dispatcher: OK…

Zimmerman: Now he's just staring at me.

Dispatcher: OK, you said it's 1111 Retreat View? Or 111?

Zimmerman: That's the clubhouse…

Dispatcher: That's the clubhouse, do you know what the--he's near the clubhouse right now?

Zimmerman: Yeah, now he's coming towards me.

Dispatcher: OK.

Zimmerman: He's got his hand in his waistband. And he's a black male.

In a way, this is a worse contextomy than the first, and in another way it's better. It's worse in that, comparing it to the transcript, you can see how much was omitted between the first part and last part of the quote. In the first example, only the bit between "this guy looks like he’s up to no good" and "he looks black" was removed. It's possible that a careless editor, trying to edit out the address, might have accidently gone on to edit out everything up to "he looks black". However, in this case, the editor took a machete and started hacking, and it appears to be the result of something worse than carelessness. On the other hand, this quote looks less damning to Zimmerman than the first, since it includes some reasons why Martin looked suspicious other than being black.

However, it's getting harder and harder to believe NBC's explanation of how these contextomies came about, given that there were multiple occurrences of the first on more than one NBC outlet, and at least one occurrence of a different one. When will NBC get on top of this situation instead of letting damaging news leak out drop by drop from other sources?


Update (5/5/2012): Jeralyn Merritt (see the Source, below) points out that an earlier occurrence of the above contextomy was actually reported by veteran reporter Pete Williams on NBC's Nightly News flagship news program. In fact, this contextomy was the first broadcast, unless there are earlier occurrences that we don't yet know about. Of course, this raises the question of why Luciano was fired―if she was―whereas Williams was not. Was there, indeed, an earlier report by Luciano that we don't know about from which Williams' report was taken?

In any case, in addition to the The Today Show, NBC Nightly News broadcast at least one of the two misleading edits. So, the problem was not restricted to the morning show and the Miami affiliate. In fact, judging from the chronology that Merritt put together, it looks as though the misleading editing began first on the Nightly News and then spread to the morning show. According to Merritt's chronology, the above contextomy was first broadcast on the nightly news show by Pete Williams, and then rebroadcast the next morning by Luciano. So, it's plausible that the morning show just reused the butchered quote from the previous evening's news broadcast.

Clearly, NBC's explanation of how the original contextomy came about―it was carelessly edited for time to fit into the morning show's hectic, four-hour slot―cannot apply here, unless there's an earlier occurrence that we don't know about. NBC's original internal investigation never even mentioned, let alone explained the second contextomy, so that it appears to have been too hurried or insufficiently thorough. As a result, NBC needs an independent investigation done by a reputable team of outsiders if it wants to repair the damage done to its credibility.

Source: Jeralyn E. Merritt, "NBC Fires Lilia Luciano Over George Zimmerman Tape Edit", TalkLeft, 5/4/2012

Solution to a Mother's Day Puzzle: The nurse was the boy's father.

This is a revised version of a puzzle that I first presented here several years ago. I reworded it to remove one problem with the earlier version, and for style. Check the Source, below, if you'd like to know more about the puzzle's origin and why it's puzzling.

Source: Puzzle-of-the-Month Club, 3/1/2005

Previous Month | RSS/XML | Current | Next Month