Astrology: The woo-woo that works

French psychologist Michel Gauquelin gained notoriety in the 50’s after publishing data showing that sportsmen were born in a non random fashion when considering the movement of planet Mars, the nicknamed Mars Effect has been the core of passionate discussions over its statistical validity since then but, beyond whether this effect truly holds or not, there are many reasons why genuine statistically significant data can be found in the astrology world, so we’d do better not to ignore planets and stars entirely.

mars_effect
Number of sportsmen born given the position of planet Mars according to Gauquelin’s data

We humans develop efficient strategies in our daily life that are useful for most situations, for example, if we see dark clouds and a few moments later it rains we associate dark clouds with rain and, voilà, next time we see dark clouds we take measures. The problem begins when we break a leg right after seeing a black cat, our association machine, a.k.a brain, does its magic and next time we see a black cat we take measures too. But you know what? The brain is right!

What is not so right is human difficulties to removed associations once they are set in our brains, a.k.a stubbornness. We humans develop all sort of strategies too keep our associations alive and demand extraordinary amounts of evidences to break them yet, even when those evidences are presented, we keep fighting them by doubting the methodology or the honesty of the persons bringing them up. There might be evolutionary advantages explaining why we create associations so easily but cannot break them with the same ease, but whatever the reasons are the problem only worsens when in some cases our stubbornness makes the associations come true! And that is what astrology is all about. These are a few examples of how astrology makes spurious associations come true:

  • “Our daughter is Gemini, let’s take her to a dance school”. This increases the number of Gemini dancers artificially.
  • “The planets says that the new company X is going to do very well”. If enough believers in astrology play in the market the price for X will rise.
  • “Aries are very successful businessman”. It will be easier for them to be promoted if their bosses believe in astrology.
  • “We have incompatible signs, this relationship can’t work”. Self suggestion and negativity will increases the chances for a couple to break up when problems come or even not to initiate a relationship at all.

I have met women that planned their pregnancies for their babies to have a particular astrological sign, thousands practice abortions in order to avoid their babies being born in a non auspicious alignment. Crazy? Well, if you live in a society with a strong believe in astrology maybe you should consider to do the same since, regardless whether you believe in astrology or not, the way that society might treat your son/daughter might make a difference in his/her future.

All the phenomena explained above will produce real statistical evidences for astrology, but we can even find more non real statistical evidences when we engage in data dredging and check alignment after alignment until we have data showing exotic associations like “when Jupiter aligns with Venus there is an inordinate amount of poker players being born”.

Gauquelin’s Data

The real statistics that astrology produces are still interesting from a sociological point of view. It is still interesting to know things like if sociological and psychological pressure make people from some signs to favor some profession. So let’s play a bit with the astrological data published by Monsieur Gauquelin retrieved from the Centre Universitaire de Recherche en Astrologie, and we will see that not only psychological and sociological pressure might make astrology come true, but also seasonal phenomena might play a role.

Assuming Mr. Gauquelin didn’t mess with the data (at least intentionally) we can check for evidences of Sun Zodiacal signs favoring some professions. The following table shows the p-values for a Chi Square test where the null hypothesis is that all signs have the same chance for the profession tested.

 Chi-Square test p-values for professions (Ho no differences among signs)
Sports Science Army Painting Music Acting Politics Writing Journalism
0.0015 0.1216 0.0421 0.0130 0.1756 0.0368 0.1161 0.2864 0.2519

We can see how activities like Sports and Painting have really low p-values that might suggest evidences for some signs favoring these professions yet, since we are analyzing nine professions at once, having an unusually low p-value within the collection of p-values becomes something not so unusual. To measure how unusual are these p-values within the context of multiple comparisons we can calculate new adjusted p-values using different methods, the following table shows the adjusted p-values using the rather conservative Holm-Bonferroni method.

 Adjusted p-values using the Holm-Bonferroni method
Sports Science Army Painting Music Acting Politics Writing Journalism
0.013 0.581 0.258 0.104 0.581 0.258 0.581 0.581 0.581

After the adjustment we see that the p-value for Sports is still very small and well below the classic significance level of 0.05 for rejecting Ho.

From an astrological point of view these results are actually a bit disappointing since, though we found statistical significance for Sports, no other professions show any promising p-value. But since the study still have the psychological / sociological value let’s have a look at how the astrological signs favor Sports:

barplot.astrological.sign.sportsLooking at the bar diagram one might think that being Pisces is going to give us better chances to succeed in Sports whereas being Scorpio guarantees poor chances. A non magical explanation for these results might be the different rate of physical strength between older and younger children within the same classroom in school. A few month difference in an early age might mean the difference between failure and success in school games and, consequently, the difference between feeling good about sports and feeling sports is not for you.

The smooth increase in sportsmen from Scorpio to Piscis seems to support this theory. After the sudden drop in Aries the numbers initiate again another smooth increase which would indicate that a good number of Aries are one year class ahead and Piscis one year behind. Disproving this theory would be very easy for anyone having access to students birthdays in a school by simply checking if Piscis are significantly older than Aries.

If this theory holds, a possible way to improve our sons and daughters chances to succeed in sports, and for identical reasons in their academic records, would be to enroll them in school one year behind so that they have a physical and intellectual advantage over the rest of students and, this way, build up self-confidence in their abilities and brighten up their hopes for the future.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s