Previous Month | RSS/XML | Current | Next Month

September 30th, 2011 (Permalink)

Blurb Watch: Killer Elite

The new movie Killer Elite should not be confused with Sam Peckinpah's forty-year old The Killer Elite. Apparently, the new one is just a killer elite. At any rate, an ad for the new movie includes the following blurbs:

Blurb Context
Rolling Stone
Peter Travers
The fun, such as it is, is all in the hardass action, and newbie director Gary McKendry delivers the goods. Statham is dynamite at this mayhem, though The Bank Job showed he could do better. … But there's nothing elite about this disposable time-killer.
The story: De Niro plays Hunter, the mentor of Danny (Jason Statham). … The sheik wants revenge against the killers of his sons, he knows Danny is the best in the world, and he correctly calculates that only the need to save his beloved teacher would lure him back into action. The sons, it turns out, were murdered by four SAS men. Danny's assignment is tricky: He is to kill them, but make it look like each death is accidental, so no one will suspect the sheik. Diabolically clever. … Meanwhile, Spike (Clive Owen) leads a shadowy group…. Their task is to shield the four targets from Danny and his boys. Got that? … The movie is a first feature by Gary McKendry…. This is an impressive debut.

The worst contextomy here is, of course, the dropping of the proviso "such as it is" from the Travers quote. Travers rates the movie only a two out of four stars, which I suppose is about a "C" grade.

Ebert, in contrast, gives it three out of four stars, which would be a "B", I guess. In context, however, his sentence "diabolically clever"―sans exclamation point, of course―sounds sarcastic rather than serious. Similarly, a movie that is an impressive debut may not be impressive simpliciter.


September 28th, 2011 (Permalink)

Head Line

Flaming Toilet Seat Causes Evacuation at High School

Source: Richard Lederer, The Bride of Anguished English (2000), p. 108

September 14th, 2011 (Permalink)

Caveat Elector

In an interview on The Today Show, Matt Lauer asked Republican presidential candidate Michele Bachmann about her criticisms of Texas Governor Perry in the previous night's debate:

Lauer: …I want to stick on this controversy in Texas, mandating vaccinations for HPV for girls as young as 12. …

Bachmann: …[G]overnor Perry chose by himself, unilaterally, to sign an executive order and put through the requirement that all innocent little 12-year-old girls or 11-year-old girls in the state of Texas would be forced by the government to take an injection of what could potentially be a very dangerous drug. …I had a mother last night come up to me here in Tampa, Florida, after the debate. She told me that her little daughter took that vaccine, that injection, and she suffered from mental retardation thereafter. It can have very dangerous side effects. The mother was crying what she came up to me last night. I didn't know who she was before the debate. This is a very real concern and people have to draw their own conclusions.

This is an obvious, even glaring, post hoc. The daughter in Bachmann's anecdote is vaccinated and "thereafter" is mentally retarded, and the girl's mother jumps to the conclusion that the vaccine caused the retardation. Then, Bachmann commits the same fallacy by endorsing the mother's hasty conclusion. Less obviously, the anecdotal fallacy lurks here, at least as a boobytrap, since a moving anecdote about a weeping mother is more vivid and memorable than a scientific study showing no connection between a vaccine and mental retardation.

Source: Alexander Burns, "Bachmann: 'Crying' mother shared HPV story", Politico, 9/13/2011

Via: Jonathan H. Adler, "Bachmann Embraces Irresponsible Anti-Vaccine Views", The Volokh Conspiracy, 9/13/2011

September 3rd, 2011 (Permalink)

Caveat Lector

A headline on the WebMD site reads:

Chocolate Good for the Heart

A subhead continues:

Regularly Eating Chocolate Cuts Risk of Heart Disease by About One-Third

That sounds like good news! Of course, those headlines were probably written by an editor, so before you run out to the candy store, here's the first sentence of the article itself:

Chocoholics have reason to celebrate today: A large new study confirms that chocolate may be good for the heart and brain. Regularly eating chocolate could cut the risk of heart disease and stroke by about one-third….

Uh-oh. Notice the "may" and the "could", and here's the next subhead:

Chocolate Linked to 37% Lower Risk of Heart Disease

Notice the word "linked", and the article goes on to say that the study "pooled the results of seven published studies involving more than 100,000 people that explored the association between chocolate and heart disease and strokes." Notice the word "association" instead of "causation".

Finally, you get to the bad news near the end of the article:

…[T]he study doesn't prove chocolate lowers the rate of heart disease. The people who ate the most chocolate in the studies could share some other characteristics that explain their better heart and brain health.

That's right. All that the study found was a statistical "link" or "association" between two variables, which does not mean that one causes the other. So, the article's headline was misleading, since the study did not show that chocolate is good for the heart, nor did it show that regularly eating chocolate "cuts" the risk of heart disease.

The study in question was a meta-analysis of observational studies that compared people's heart health with how much chocolate they said they ate. All it showed was that people who claimed to eat the most chocolate had fewer heart problems than those that said they ate less. This result could certainly be explained by something in chocolate that promotes a healthy heart, but there's an alternative hypothesis: obese people, who often have heart problems, may tend to under-report how much chocolate they consume. There's apparently already evidence that obese people understate how much they eat in general, and why wouldn't this apply to chocolate in particular? Of course, the way to test this alternative hypothesis is to do a study that doesn't rely upon self-reporting of chocolate consumption.

By their nature, observational studies can't establish that a correlation between two variables is causal. Moreover, the study itself doesn't claim to establish a causal relationship, but suggests that this hypothesis be tested by a randomized trial. The problem is not with the study, but with misleading reporting of it. So, when you see a headline that seems too good to be true, read the article before believing it; and be sure to read the whole article, because sometimes the truth doesn't come out until near the end.


Previous Month | RSS/XML | Current | Next Month