What did he do?

He was the first investigator who carried out research designed to test whether astrology is true according to good scientific standards, being a professional statistician.

First of all, he tested well defined samples of people (e.g. the Biographical Index of members of the Academy of Medicine, 1939) in order to exclude selection bias and ensured that they were sufficiently large to achieve statistical significance i.e. so that the probability that any findings would be very unlikely to be due to chance. 

Secondly, he adjusted his findings for a number of confounding factors e.g. that more children are born at night than in the afternoon and that the time that a sign of the zodiac is at the horizon in the East (which corresponds to an important factor in astrology i.e. the rising sign or Ascendant differs according to the sign.    

He drew up the natal chart related to each birth duly noting the positions of the sun signs, the houses and the planets on a circle with a horizontal axis representing the horizon: the top half of the circle represents the sky during the day, the bottom half the sky during the night, with the left extremity of the horizontal axis corresponding to the East where the sun rises and the right extremity corresponding to the West where the sun sets.

Below you can see an example of a natal chart, the one of Muhammad Ali (Cassius Clay), a famous boxer.  You will notice that he was born at 06:35 PM, shortly after the sun had set in the West. 

Gauquelin assessed differences in terms of all the main astrological factors that make up a natal chart namely:

-        Signs of the zodiac

-        The Houses (according to astrological tradition the sky is subdivided into 12 equal parts, starting from the Ascendant which indicates house no 1; each house corresponds to a domain of life, such as career, family, children, health, ect)

-        The Position of the planets

-        The aspects i.e. the relative position of one planet versus another

In order to assess potential differences in the position of the planets he subdivided the circle into 36 equal sectors obtaining a circle that resembled a roulette wheel.  He then allocated the position of the planets in each chart to one of them.  In this way he obtained distribution frequencies related to each planet and could see whether the distributions throughout the circle were homogenous or not.  In the event of deviations, he calculated the probability that they were due to chance alone. 

His “roulette wheel” is shown below.

At first he assessed the charts only of Frenchmen.  This was feasible since at that time anyone could has for the birth certificate of anyone else and the French certificate included the birth time.

After his first findings related to profession were published one of the main criticisms was that his data was related only to French births.  He therefore decided to collect cases from other countries as well.  Unfortunately, he discovered that birthtimes in America were unreliable, that birth certificates do not provide birth time (UK, Ireland), that there were no birth registries (Scandinavia) or that the authorities were not willing to provide access (Eastern Europe and Spain).  That left few countries for his investigations, which in the end included 27,000 birth records:  12,000 from France, 7,000 from Italy, 3,000 from Germany, 3,000 from Belgium and 2,000 from Holland.  The selection was always related to well defined groups of professionals, such as the Index mentioned earlier.