Mark's AstroDatabank Dream
When Mark McDonough established the AstroDatabank Company with Lois Rodden in 1996, he had a dream. Mark hoped that through assembling a large research database that one could prove with statistics which of astrology's many claims had validity. In fact, Mark even offered $1,000 for a research project demonstrating a probability as small as .001 and up to a $5,000 cash prize for a research project with a probability of .0002. Mark did not give out this prize before he sold AstroDatabank in 2005.
AstroDatabank's final attempt to achieve those significance levels lay in complicated multi-factor assemblages called AstroSignatures, which tested for the presence of many traditional astrological factors in a themed data set, as compared to an unthemed control group. Since AstroSignatures could incorporate up to 300,000+ factors, they always struck me as being unwieldy and useless to the everyday practitioner.
Halloran Software's New Approach
I had no alternative of my own for AstroSignatures until the autumn of 2017. That is when I started working with large chart collections that were collected years ago by Michel Gauquelin, ranging in size from 622 to 3646 charts, as well as some categories that my friend Linden Leisge had me export from AstroDatabank into the AstrolDeluxe Platinum format so that he could research them. I turned to these collections for help in understanding the nature of each of the declination parallel and contraparallel aspects between the planets, as I committed myself in October 2017 to writing declination aspect interpretations. I found plenty of common themes in the biographies of famous men and women who had close versions of these aspects, but I wanted to supplement that information with big picture overviews that would be true of everyday people in general, not just celebrities. AstrolDeluxe Platinum already had the ability to quickly find all the people in a chart collection who have a particular aspect between two planets - all I had to do was tell it to record the number found divided by the number searched to get the percentage frequency of that aspect for each collection. When I compared these percentages, it could easily happen that the same aspect might occur in 5% of one collection but only occur in 2% of another collection. With such large collections, significant aspect variations between them were unlikely to be purely random. That means that occupations with a high frequency of a particular aspect have a positive correlation with that aspect, such that the occupation should be a congenial one for people with that aspect, people with the qualities that the aspect confers. If organized correctly, these correlations can be used to predict suitable occupations just from the birth chart.
To organize the aspect frequency data, first I automated searching for all the possible aspects in a collection and recording those aspect frequencies. Then I automated merging each collection's results so that the names of the collections would appear in order from high to low frequency for each possible aspect. Given that I used 11 zodiacal aspects and 2 declination aspects and a total of 16 chart planet points, that gave a total of 120 unique combinations per each aspect, so that the final table has sorted vocational or category frequencies for 1,560 planetary aspects. There are 22 collection categories. From Gauquelin, I had Actor, Alcohol, Dementia, Military, Murder, Musician, Painter, Politico, Science, and Writers. I tried to balance out these 10 chart collections with 12 additional collections extracted from version 4 of Lois Rodden's AstroDatabank - Astrologers, Athletes, Businessfemale, Businessmale, Computer, Engimech, Healers, Lawyers, Life-80s [Octogenarians], Pilots, Psychol, and Teacher. The final master table ranking these 22 vocations for all 1,560 aspects runs to over 1100 pages.
Applied to Human Beings
The average person has about 60 or 70 of the possible 1,560 planetary aspects. Just as every person has a different individuality, every person has a different selection or concatenation of aspects. To determine the relative ranking of the 22 categories in a person's chart and life, we can go to the master table and look up those approximately 65 aspects and add up the rankings found for each of the 22 categories. If many of the 65 aspects have Computer ranked high, then this could be the top-ranked category for the individual. For example, Computer is the strongest vocational category for Bill Gates (top aspect is Jupiter conjunct Pluto), Pilots is number two (top aspect is Jupiter sextile Neptune), Psychol is number three (top aspect is Sun parallel Chiron), and Murder is number four (top aspect is Pluto biquintile Midheaven).
At least in the one case known to me, this method gives the same results as a popular career assessment test called the Strong Interest Inventory, named after its inventor Edward Strong. When I was in college, I took this test. It gave me an A+ rating in two categories - Computer Programmer and Psychologist, based on my having a personality similar to that of individuals who had succeeded in those two fields. Computer Programmer and Psychologist are the same two occupations that statistical astrological research assigns to the top of my Vocational Ranking list. The thing to appreciate about this technique is that unlike the 300,000+ factors of AstroSignatures, the technique only depends upon aspects between the planets. The technique transfers a set of aspect frequency numbers that have been found to be true at the level of a large group down to the level of the individual. It achieves the astrologer's goal of chart factor synthesis while being free from subjective human judgment. Both Eastern and Western astrology agree upon the importance of planetary aspects.
The 1977 book Recent Advances in Natal Astrology by Geoffrey Dean has a section starting on page 307 devoted to Statistical Studies of aspects. On pp. 313-15 the book discusses the work of Dieschbourg of Luxembourg, who "investigated the occurrence of aspects among groups of eminent professionals vs a control group. The study involved nearly 12,000 cases (about 300,000 aspects) and took 8 years." On page 314 is a table of aspects observed versus expected for different planet combinations and occupations, where the one for Mars-Jupiter and a collection of 6158 in science/art/literature had a p-value from Chi-squared at <.0001. The observed was 3243 where the expected was 3079. So this was already written about 40 years ago, but it did not change the world. I think that it is not enough to report something, it has to be incorporated into a useful tool.
Two areas of vulnerability involve the collection of birth data for a category. There are articles on the Internet that touch upon Mark McDonough's attempts to have AstroDatabank arrive at a signature for Alcoholism and a potential flaw with the AstroDatabank data for alcoholics. The flaw has to do with about half of the charts marked Alcoholics also being celebrities, which could confuse factors for being a celebrity with factors for being an alcoholic. I think that Gauquelin's large collection of 1793 Alcoholics avoids this pitfall. His Alcoholic charts date from 1880 to 1940, with the number of births peaking around the year 1910, and based on their longitudes and latitudes of birth, they are all from France. The other area of vulnerability does relate to birth dates in a collection not being evenly spread out over the years but peaking in number at a time when certain outer planet aspects were in effect, causing the frequency for those aspects to be over-represented for that collection category. When I added the Vocational Ranking table to the Display menu inside AstrolDeluxe Platinum, I had to deliberately exclude using data from over 50 outer planet aspects in order to counter this vulnerability.
I am aware that human experience cannot be divided up into just the 22 categories used for this research project. This is a very crude initial mapping. Consider it to be a proof of concept project, to be expanded upon by future data collectors. There should be many more diverse categories, which could include singers, firemen, real estate agents, and mystics.
Description of Software for Automated Statistical Research and Vocational Ranking.Updated astrology software news.
Return to Halloran Astrology Software Home Page.
Copyright © 2018 Halloran Software, Los Angeles, California
Last modified on March 19, 2018.