Skeptophilia (skep-to-fil-i-a) (n.) - the love of logical thought, skepticism, and thinking critically. Being an exploration of the applications of skeptical thinking to the world at large, with periodic excursions into linguistics, music, politics, cryptozoology, and why people keep seeing the face of Jesus on grilled cheese sandwiches.

Tuesday, November 2, 2010

Nine out of ten doctors recommend this blog!

According to a new study, alcohol is worse than heroin.

At least that's what the headline says.  (Read the original story here.)  Of course, when you read the actual story, you find out that that's in fact not what the study says -- or at least, is far enough away from what the study says to create a distinctly false impression.

What the study (originally published in the highly respected journal Lancet) did was to rank various legal and illegal drugs (and it correctly classified alcohol and tobacco as drugs) in order of the overall harm done to society.  In terms of sheer numbers, alcohol came out the clear favorite -- considering all of the deaths due to drunk driving, to heart and liver disease from overconsumption of alcohol, to loss of careers, marriages and friendships from alcoholism, no drug has the impact on society that alcohol does.

I'm not disputing the above facts at all.  However, what I object to -- as usual -- is the way the media, in their typical sound-byte fashion, has created a mistaken impression.

Re-read the headline.  If you read that -- and nothing else -- what would you think?  I don't know about you, but what I would take away from that is that to an individual it is worse to drink alcohol than to take heroin.

I know that the article's contents then go on to correct that impression, and I also agree with the statement that anyone who only reads the headline of an article deserves, on some level, to be misinformed.  But what I object to is that in order to sell subscriptions, or obtain readers, the members of the media will craft eyecatching, and often misleading, headlines -- and the public leaves with seriously erroneous information.

Consider, for example, the headlines when it was discovered that altering the levels of a single enzyme, telomerase, could extend the lives of roundworms by a factor of ten.  This is pretty impressive -- equivalent to a human living to be 800.  Of course, the article detailed that (1) the effect had yet to be demonstrated in humans, and (2) elevated levels of telomerase are thought to predispose tissue to becoming cancerous (cancer cells being, for all intents and purposes, immortal).  But what did the headline read?  "FOUNTAIN OF YOUTH ENZYME DISCOVERED."

I realize that a headline that said, "We found something that makes roundworms live longer" would not sell magazines.  But still, in the interest of not taking advantage of the credulous public, it seems to me that it would be better to err on the side of caution.

Back to the alcohol vs. heroin debate.  My main problem with this one is that people, in general, don't understand the concept of risk in the first place -- witness the number of people who are afraid of flying as compared to the number of people who are afraid of driving, and then compare the actual risk (number of people killed in the activity divided by the number of people who participate in the activity).  It reminds me of the comment by economist Dan Gilbert -- he showed a photograph of an airplane crash, a terrorist attack, a burning building, and a swimming pool, and asked which didn't belong.  "It's the swimming pool," he said.  "It's the only one up there which has a significant risk of killing you."

The misassessment of risk (and its partner, the misassessment of gain) is, for example, why people play the lottery and visit casinos.  It's why people accept high but familiar risks (driving) but refuse to take small but dramatic risks (flying).

It's also why I find the alcohol/heroin headline appalling.  While the overall damage done by alcohol to society is clearly greater than that done by heroin, compare the actual risk (percent likelihood of harm) of consuming alcohol as compared to taking heroin.  The likelihood of any given individual coming to harm from drinking is actually quite small, whereas everyone who takes heroin comes to harm from it.  Let me repeat that: the risk for alcohol is small (but non-zero); the risk for heroin is 100%.  Is that the impression that this headline gives?  Of course not.

I realize that it's not the media's fault if their readers are too ignorant to understand the subtleties of a particular story; their job is to inform the already-reasonably-well-educated, not to educate the foolish.  But that said, I think it's also incumbent upon them not to take advantage of the gullibility of the public when it comes to the types of thinking that almost everyone is bad at -- e.g., understanding risk.  The danger, to me, is not just that people leave misinformed -- it's that they sense that on some level they're being lied to, and end up distrusting not the sensationalism of the media, but the source of the study itself, i.e., the scientists.  The scientists themselves were clear about what their study did and did not accomplish; if the media garbles that, either accidentally or through a deliberate desire to misrepresent, it's hardly the scientists' faults.

It's always important to read the news with a critical eye -- as I always say to my classes, there's no such thing as unbiased media.  Even what stories they decide to cover implies a bias, in that they are making the decision of what warrants coverage and what does not.  But beyond that, the misrepresentation, and misunderstanding, of statistics is so common that it behooves us all to read a little more carefully when we see the word "data" -- whatever the headline of the article might have claimed.

1 comment:

  1. This was fortunately discussed vigorously with regard to the CERN release of their anomalous neutrino transit times. Of course the pop-press was just dying to kill Einstein. I went through about 50 news piece headlines until I found the Goldilocks one I even trusted to read.
    Btw, I suffered more from alcohol's brain-function quirks than from the dozen or so bags of heroin I shot. Just an anecdote, of course, but it knocks the 100% off the chart. Barring late-onset damage...