[HOROSCOPES: FACT, FICTION, OR FUN?]
A group of students conducted an experiment to test the accuracy of horoscopes. Here's what they found.
"I read my horoscope all the time," said Casey Perkins, an 18-year-old student at DeSoto (Texas) High School. So when she heard that her science teacher, Karen Samfield, was looking for students to help research the accuracy of astrological forecasts, Casey jumped at the chance. "I thought it would be fun to see how accurate they would be," she said.
Samfield and 64 of her students—including Casey—joined forces with an editor from Jr. Skeptic magazine. The students examined newspaper horoscopes every day for two weeks to see whether the predictions made by the horoscopes matched the experiences the students actually had.
Scientists have maintained for years that horoscope predictions are no more accurate than pure guesswork. Would the students' observations support that position? And if so, what would their results say about astrology?
SCIENCE OR SUPERSTITION?
Astrology is a set of ancient beliefs about how events on Earth are related to the positions and movements of the sun, planets, and stars. For instance, astrologers say they can predict a person's future from the position of the sun, moon, and stars at the exact moment of the person's birth. A horoscope is a diagram of that position.
Scientists say predictions made from horoscopes are statements so vague they fit just about anyone at any time. Ivan Kelly, a psychologist at the University of Saskatchewan in Saskatoon, Canada, said that a horoscope might say that a person has great intelligence or is highly creative. Yet just about everyone, he said, thinks of himself or herself as being intelligent and creative. Said Kelly: "Have you ever met anybody who describes himself as stupider than usual or less creative than average?"
SCIENCE, STEP-BY-STEP
The DeSoto students started their research by collecting data—information obtained during an investigation. Each day, their teacher showed them 12 horoscope readings from a newspaper, one for each astrological sign: Aquarius, Aries, Cancer, Capricorn, Gemini, Leo, Libra, Pisces, Sagittarius, Scorpio, Taurus, and Virgo. All references to the signs were removed from the readings, and a number was assigned instead to each reading.
The students examined all of the readings from the previous day and decided which one fit the events that had occurred in their own lives. When a reading seemed to fit, students considered it a "hit." If it didn't, it was a "miss."
The students didn't know which horoscope reading corresponded to their own sign, and their teacher didn't know which numbers were assigned to which signs. Scientists call this a double-blind study: meaning that neither the study's participants nor the person conducting the study are aware of the basic details of the study. Many new drugs are tested in double-blind experiments. Both doctors and patients in the studies are unaware of which pills are real and which are simply sugar pills.
Why are studies made double-blind? To remove bias, a personal preference that can unknowingly influence a decision. For instance, if the DeSoto students had known that Horoscope 4 was the Aries reading, kids whose sign was Aries might have been biased toward selecting that particular reading. If the teacher had known which readings matched which signs, she might have communicated such information unconsciously to the students. Scientists say removing bias when collecting data allows more accurate conclusions to be made about the results.
After two weeks, Gilmore compiled all the data gathered by the students. In evaluating the data, Gilmore took into consideration the number of people participating in the study, a number scientists call the sample size. Some studies involve thousands of participants; others involve just a few.
SAMPLE SIZE MATTERS
According to Professor Kelly, the number of participants who should be involved in a study depends in part on how powerful the phenomenon being studied is. For example, horoscopes are supposed to be extremely accurate predictors of the day's events. So even a fairly small sample size, said Dr. Kelly, should be enough to determine whether horoscopes are accurate.
Other studies, such as those examining drugs designed to prevent cancer, require large sample sizes, perhaps hundreds or thousands of participants.
How did the sample size of the DeSoto study stack up? "The more students you have, the better," Kelly said. "But 64 isn't bad. Actually, their study was a pretty good exercise in critical thinking."
The students' critical-thinking exercise came up with 21 hits out of 246 tries, or about 1 hit for every 12 tries. That's about what the students would have come up with purely by chance. The results fit those of dozens of other studies conducted on horoscopes, supporting the conclusion that horoscopes are inaccurate—or, in plain words, a lot of flapdoodle.
What did Casey think of the results? Horoscopes, she said, "aren't as accurate as I thought. They're fun, but it's not good to take them too far."
By A. T. McPhee
JSF
[Login to view]