This is a polar graph of the number of Mars's of sports champions
divided into segments comparable to Placidus houses. It seems like there
are many more Mars's in the 9th and 12th houses for sports champions
(the solid line) than for the control group (dotted line). However, this
graph is drawn a bit out of scale. In the Mars effect, the chance of a
sports champion's Mars ending up in a Gauquelin sector is 22% vs. 17.9%
if the Mars placements were purely random. That 4% differential is a big
deal because statistically speaking the chance that this was only a
fluke is one in a thousand. This gives the scientists something that is
very difficult to explain away. But unfortunately, it gives the
consulting astrologer very little to back up their work.
Here's why. If Mars ends up in the Gauquelin sector 22% of the time
for sports champions that means that 78% of the time it is outside of
the Gauquelin sectors. Thus for all intents and purposes if you see a
client's chart with planets in a Gauquelin sector, you might as well
forget about giving it any extra weight.
This is the best we have been able to do in quantitative astrology
research. Every other statistically significant study in astrology has
fallen apart upon replication. Someone might find more Virgos to be
librarians in one study, but when another person replicates the study
the effect disappears.
I did not know this until after we shipped the AstroDatabank. I
figured we had built the ultimate research machine and now I could take
some time off to read up on astrology research. I was ready to do some
of the research that I had intended to do when I first went to Lois
looking for some good data to work with. The more I read the more
depressed I got. I just kept reading of one failed or inconclusive
experiment after another. I thought, "Oh my God, I've just invested
three years and hundreds of thousands of dollars to build an astrology
testing machine when the whole enterprise is failed from the get
go." Then I realized there was a common factor in all the research
failures they were all testing only one factor at a time. They were
all looking for a single astrological factor to correlate with a single
biographical trait Sun signs with vocations, Sun-Moon aspects with
successful marriages, Jupiter aspects with luck, and so forth.
We should know better than this. Reading a chart is not as simple as
listing the planetary positions and their meanings. We know that
astrological factors can contradict or reinforce each other. Whether
it's gathering groups of people with planets on an angle, Mars Square
Pluto, or Sun in a particular sign, it is way too simplistic to expect a
single astrological factor to correlate with a specific behavioral
trait. If we are going to test correlations between astrological factors
and personality tests or vocations, we are going to have to consider the
combined effect of multiple factors to get good results.
Why has the history of astrological research been almost exclusively
composed of single factor tests? I believe the answer lies in an adage
often quoted by Abraham Maslow, "If all you have is a hammer,
everything looks like a nail." The tools one has at one's disposal
limit the kind of research one can do. Until recently the main tools of
astrological research have been hand counting.
Who can blame the Gauquelins for choosing a simple single factor
hypothesis like "planets on an angle correlate with vocations"
when they had to hand count each of the thousands of charts. They tested
many astrological factors. In retrospect it is obvious why angularity is
the only single factor research that has paid off it's because
angularity is probably the most powerful factor in a chart.
Everyone expected that computers would be a gigantic boost to
astrological research. The astrology research programs made the counting
faster, but they all counted only one factor at a time as well. It now
takes only seconds to find out how many alcoholics have Sun in Pisces or
Moon in Pisces or Neptune on an angle. We can even do some fancy
searches like looking for how many eighth harmonic aspects are in
charts, but basically it's still single factor research.
In order for us to have a breakthrough in quantitative astrological
research we need to be able to teach computers to analyze charts in a
way that mirrors the process a human astrologer goes through. The
computer needs to be able to look at multiple factors at once, weighing
them together and balancing them off each other.
If you think I am talking about some time way in the future, think
again. I know three astrology software companies who are selling
software today that can do this trick of multi-factor analysis of charts
Cosmic Patterns, Air Software, and AstroDatabank Company. Their
products actually look for patterns in charts and score charts according
to how closely they fit an established pattern.
Most report writers look at one factor at a time planet in sign,
planet in house, or planet aspecting planet and spit out a
paragraph. Cosmic Pattern's Kepler program has a major life themes
report that looks at hundreds of factors before deciding which 5-10
major life themes apply. David Cochrane who programmed the report told
me he wanted to go beyond trait analysis (planet in sign, planet in
house, planet aspecting planets) and find sub-personalities. Underneath
the covers, the program looks for combinations of three planets that are
both related through a harmonic relationship and a midpoint
relationship. He then looks at the type of relationship among the three
planets before issuing text that describes this sub-personality. I count
this as a major step forward in taking report writing out of simple
unintegrated single factor astrology.
Alphee Levoee of AIR Software tells me he is using neural network
technology to examine the charts of days when the market did really
well. Neural network technology is designed to emulate the way humans
learn through pattern recognition. The software builds in an internal
model of what are the relevant weighted factors in up days and down days
and then tells you how likely today is going to be an up or down day in
the stock market or for a particular stock.
To give you a better sense of how these systems work, let me give you
a quick example of pattern recognition using the AstroDatabank
interface. Suppose you were looking for an AstroSignature for
alcoholism. You might look for the following:
- Sun or Moon in Pisces
- Number of aspects from Neptune to the luminaries
- Neptune on an angle
- Saturn on as angle
The real study has 50 rules and is on our web site; this is just a
simple one for instructional purposes. Here is what it would look like
in software:

Notice that we have given a weight of 2 to Neptune on an angle and a
weight of -2 to Saturn on an angle. That means that if the chart has
Neptune on an angle it has twice the importance of an inner planet in
Pisces and if Saturn is on an angle we expect it to have a strong
sobering influence so we give it a weight of negative 2.
We then score all the alcoholics using this formula to get a group
score that is higher for alcoholics than the control group of
non-alcoholics. The greater the disparity between the score for the
alcoholic group vs. the control group, the higher the likelihood that we
have found a signature for Alcoholism. We found eleven factors that show
a differential of over 6% between the Alcoholic and control group.
I need some help in interpreting these results. We need the
human-meaning-based meaning analysis to make sense of these results. The
results are also on our website and I invite you two add your comments
to those of the other astrologers around the world who are examining
this problem. Twenty years ago when they were first building the wide
area networks that have become today's Internet, their intention was to
build a network to facilitate collaboration between researchers. We
astrologers can do this too.
Before I close I would like to share with you one quick study I did
on hemisphere emphasis, which actually proved to be ten times more
statistically significant than the Gauquelin research. One afternoon it
occurred to me that I could very easily test the hypothesis that public
figures have more planets above the horizon because everyone in the
AstroDatabank is designated as either a "public figure" or
"private person." So I counted how many of the public figures
had ten planets above the horizon and how many had ten planets below the
horizon.

I did indeed see a 28% differential in the expected direction but the
numbers were so small 55 vs 43 there weren't enough observations for
statistical significance. So I tried 9 and 1, nine planets above and one
below. Then 8 and 2, 7 and 3, and so forth. As you can see the effect is
very strong through 8 planets and starts to fall apart at 7 planets
above the horizon and 3 below. The practical significance of this study
is that unless the client has at least eight planets above or below the
horizon the hemisphere effect should not be mentioned.
If you look at this table closely, you may notice an anomaly that
gave me trouble for a week, but is now my favorite part of the study.
One would think that these numbers should be a straight graph.

The more planets above the horizon the greater that chance of being a
public figure. But 9 above and 1 below is actually less likely to
produce a public figure than 8 and 2. After wrestling with this anomaly
in the back of my head for a week it dawned on me that this result is
absolutely predicted by astrological theory. It is the power of the
singleton to draw the energy to the other side of the chart. So what
started out an anomaly has ended up as the finishing touch that nails
the case closed.
There are three points I want to make about quantitative astrological
research:
- It's much easier than you think. I did the hemisphere study in one
afternoon.
- Quantitative research can be use improve astrological theory by
calibrating when it works and when it falls apart. As we just saw
the hemisphere hypothesis falls apart after 8 planets above.
- We are on the cusp of a new age of discovery with easy to use
multiple factor techniques for chart pattern analysis.
I do not expect everyone in this room to feel comfortable doing
quantitative astrological research. Next year I'll be singing a
different tune after I have built some of the automation tools I have
designed for later versions of the AstroDatabank. But I can say that
everyone in this room can make a contribution to astrological research
through qualitative research.
Here's that pyramid of research again.
Each level feeds the one above it. The qualitative research generates
the hypotheses tested by quantitative research which is then examined
and replicated by the six people in the world who understand both
astrology and statistics.
When I say that each one of you can have a hand in the advancement of
astrological knowledge, let me be very specific. There are 660
categories in Lois Rodden's AstroDatabank and each one is a
potential research study. What is the astrological recipe for cancer,
arthritis, peak performance, career choices, etc.? Inquiring minds want
to know. What are the highest and lowest expressions of the least
favorite aspect in your chart? Wouldn't you like to know?
Take a moment to recall what it was like for you the first time you
started to study your own chart and the charts of your loved ones. Do
you remember the excitement of finding underlying patterns that made you
tick? That same sense of adventure and exhilaration of discovery is
still available by probing the basic mysteries of astrology. The only
difference between the standard chart interpretation you already do and
qualitative research is that you will be looking at a group of charts
that share a common trait and looking for other themes in common. The
same curiosity that drove you to learn astrology can be harnessed to
further the reaches of astrological knowledge.
From magazines, to local club newsletters to web sites, publishers
are hungry for your research articles. There are new astrology web sites
popping up every day, looking for new content to put on their site. We
publish a new astrology research article every other week on our site
and StarIQ.com's editorial calendar calls for at least seven new
articles per week.
So I invite everyone here to join in the discovery process. We have a
window of about ten years when all the easy discoveries will be made. We
might compare it to being an astronomer soon after the discovery of the
telescope. The software tools that are coming out today make it easy to
see distinctions that were formerly lost in the complex clusters of
data. There is so much uncharted territory to map. The seminal work will
be done now. Twenty years from now it will just be fine tuning; the
process will become as complex as contemporary astrophysics. Now is the
time for you to participate in the bold new discoveries that are just
ready to fall into your lap, as they did for Galileo. After he invented
the telescope he pointed it at Jupiter and saw four moons that had never
been seen before. Now they are called the Galilean moons.
I'd like to thank you for having the courage to come to a lecture on
such a formerly boring topic as astrology research. I hope I have made
my point that you do not have to be a triple Virgo statistician to have
fun and make a significant contribution to unlocking the mysteries of
how planetary patterns and behavior patterns are linked.
Thank you for coming and I look forward to seeing your contributions
on the web.
|