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.

Saturday, February 5, 2011

Stars, darts, and basketballs

For those of you who are still reeling from finding out that you are not the astrological sign you thought you were, take some comfort in a study by David McCandless.

McCandless, who evidently has the patience of a saint, analyzed 22,168 horoscopes.  His contention was that if there were anything to astrology, there should be a statistically significant difference between the content of horoscopes for the twelve (or thirteen, depending on who you believe) astrological signs.  Using computer software, he first filtered out common and relatively meaning-free words like "and" and "the," and then arranged the remaining words on a wheel-diagram.  The size of the word on the diagram represents its relative frequency.  Check it out here.

As you can see, there is no difference whatsoever between the different signs.  "Feel," "sure," "love," "keep," and "better," are the most common words on all of the signs.  In fact, McCandless has out-horoscoped the astrologers, and has come up with a generic, all-purpose horoscope that anyone, of any sign, could read every day, and accomplish much the same thing as the "real" ones:

"Whatever the situation or secret moment, enjoy everything a lot.  Feel able to absolutely care.  Expect nothing else.  Keep making love.  Family and friends matter.  The world is life, fun, and energy.  Maybe hard.  Or easy.  Taking exactly enough is best.  Help and talk to others.  Change your mind and a better mood comes along."


Wow... I feel so... enlightened.   This especially speaks to me, being that I was a Scorpio and now am a Virgo.

It puts me in mind of a now-famous demonstration James Randi did in a high school classroom.  (You can see a video of it here.)  He gave out horoscopes to the students, and told them that they were predictions based upon detailed information about the time and place of their birth.  Each student was given time to read the horoscopes, and then asked to grade them on a scale of one to five, the score being assigned based upon how accurate it was, how well it applied to each of them personally.

The results were amazing.  There was not a single student who gave their horoscope a grade of one or two; there was a single three; everyone else graded it at a four or five, with five being by far the most common score.  So on the face of it, it seems like astrology fared pretty well, in this experiment.

Until you find out that all of the horoscopes were identical.

Astrology relies on an observational phenomenon called dart-thrower's bias -- something to which we are all prone.  The name comes from a thought experiment; picture yourself in a pub, having a nice pint of Guinness with your friends, chatting about whatever.  In the corner is a dartboard, and several bar patrons, all strangers to you, are having a friendly game of darts.

The question is:  when do you notice the darts game?

The answer, of course, is:  when one of the players scores a bullseye.  Or, perhaps, misses the dartboard entirely and skewers the bartender in the forehead.  The point is, we have evolved to notice outliers -- data points that are extreme.  We tend to over count the hits (or wild misses), and simply ignore all of the average, background clutter.

This was brilliantly illustrated by an experiment performed some years ago, in which a large number of test subjects were asked to watch a video clip of a lone man shooting a basketball.  That was all it was; just a guy shooting baskets.  Sometimes he missed, sometimes he didn't.  The subjects didn't know what they'd be asked about afterwards -- they were just told to watch the clip carefully.

There was a single question after the video was shown:  what was the guy's hit rate?

The people who had made the clip had arranged it so that the guy had an exactly 50% hit rate -- not bad, for an amateur.  What blew away the researchers was that not a single person who watched the clip -- not one -- estimated his hit rate at under 50%.  Several went as high as 80%. 

The explanation is that we give more weight in our memory to the times that the ball went in than the times it missed.  The evolutionary reason for this is simple, and persuasive; if you are a proto-hominid on the African savanna, which is more dangerous -- to pay attention to a stimulus that may not be important (weighting the hits) or to ignore a stimulus that actually is important (weighting the misses)?  Clearly it's the latter, especially if the stimulus is the sound made by a hungry lion hiding in the grass.

We're programmed to notice the hits, even when they're not really very impressive.  Astrology, then, is one massive game of dart-thrower's bias.  But the fact that it has no basis whatsoever in science, or even logic, doesn't stop astrologers from fleecing the gullible public for millions of dollars annually.

Not that this will probably convince anyone, because belief in astrology also relies heavily on confirmation bias -- the acceptance of any evidence, however puny, in support of an idea you already believe to be true.  So I'm probably tilting at windmills, here.  So whatever it is that you end up believing about astrology, do take to heart David McCandless's advice:  Keep making love, and remember that family and friends matter.

No comments:

Post a Comment