Where there´s smoke, there is always also fire?
In the last few years, in our thinking and in the culture it seems that the “scientific method” is starting to get very mixed with “statistical method”. There is a lot of mistakes in this thinking, of course. And as people feel that “only what is scientific is truth” we start to see things like “the science explains why we do love animals”.
As any animal lover can testify, we don´t need any “scientific” (or in this case, pseudo-scientific) reason to love animals. They are just cute.
My dog, Blanca, looking crazy
We also see every day more and more explanations based on pseudo-evolutionary reasoning. My favorite is that nerds have better sperm.
So I think is unavoidable that astrologers will use statistics more and more. Let´s just hope they will use it properly.
Association and Correlation
When two categorical variables are statistically related (e.g, eye color and hair color) they are said to be “associated”.
When two quantitative variables (e.g. weight and height) are statistically related (with a linear relationship), they are said to be “correlated”.
The word correlation is often used wrongly to express association.
Association is easy. You prove that a certain atribute (e.g. eye color) has more probability to happen when the person has also another atribute (e.g. hair color). An example of correlation is below.
This example is from Intermediate Statistics for Dummies. A good book, better than a lot of “academic” books that you see around. In this example the poins measure the number of coffe cups sold at football games in Buffallo, New York, at several Temperatures. The straight line that was plotted show that we have a negative slope, meaning that when the temperature rises, the number of coffee cups decreases. The correlation is -0.74, which is pretty hight for this kind of study. The number is negative because, when temperatures rises, cups decrease, and when temperature lows, number of cups sold increases.
Correlation is not causation
The biggest mistake that people make (even the ones who should know better) are to confuse correlation with causation (or association with causation). Statistics can´t show relationships of cause-effect.
Let´s see the two examples above. It is obvious and clear that, when the temperature drops, people like to drink more hot beverages. In hot days, people are glad to avoid coffe and drink beer. So, we know that it can´t be the other way around: we don´t expect that people drinking coffee help to lower temperatures. If this was so, we would have a quick fix to global warming.
But this is not an statistical information. This comes from our knowledge and theories about how the World works. For the correlation, it could be just as easily the other way around. Or that the two variables are been affected by a third invisible one.
For instance, we all know that cigarettes cause cancer. We know this because of statical studies. But from the statistical standpoint, you can´t say if the cigarettes are “causing” the cancer, of if people who will develop cancer in their lives happens to love cigarettes! You could even theorize that there is some kind of strange virus (third invisible cause) that creates cancer and a loving for smoke!
In cases where we know before hand that there is only one possibility, it is all plain and easy. But let´s analyse the correlation between market share and publicity. Or between years of study and domestic violence and things get very complicate and ideology becomes very important.
You could say that “the lack of study makes people who have less economical opportunities to avoid domestic violence”
or that “people who suffer domestic violence has less social and psychological support to engage in study and drop out”
or that “people who have less emotional stability and cognitive habilities are likely to drop school and to have abbusive relationships”
You see that any number of causations are always possible. A can cause B, B can cause A, and A and B can be caused by C, etc.
As another example, studies always show that the better business have better employees (in years of education). But we don´t know if it is because the better employees really raises the profits, or because business with better profits can hire better employees.
So remember that, if you have a causal relationship, a correlation or association is likely to appear. But the correlation by itself doesn´t mean that there is causation.
And in astrology? In astrology we have an even bigger problem, because we don´t have causal relationship, period. There is no magnetic field that show when you will marriage. There is no gravitational force that leads our horaries to the right answer.
“As above so below” is not a scientific theory, it is a statement about the spiritual nature of the world, where everything is related by spiritual connection.
That´s one of the reasons why hindsight is such a horrible methodology in astrology… you will always find something happening in a natal chart at any given time!
So we see things like “oh, my mars was transiting my ascendent and then my house burst into flames”. If mars had this kind of power, no insurance housing would be available, ever.
“Directed Venus was sesquisquaring with neptune and semi-sextile quiron when I won the lottery”. Ok, but so what?
“Every time that pluto passed through mars of the US chart, a terrible event happened”. Ok, but do this mean that will happen again?
This is not the only problem with hindsight, but it one of the most obvious. As there are no causality in astrology, we can´t eliminate things as easily as we did in the example of coffee and weather.
Astrologers should at least use some simple guidelines when do study with hindsight facts:
1 – Is the simbolism correct? You can´t use venus sesquisquare quiron and say that it “caused” you winning the lottery, just because it happened at the same time. It is not enought. A lot of things at happening at every given time in a chart. The symbolism must be adequate. This is number one sin of people addicted in rectifications.
2 – Is the power of the astrological method enough? A transit of mars through an angle will happen every 6 months give or take. If this could make your house to burst into flames, we would all have buy houses under water.
3 – Can we make the same method work in foresight? Sometime ago an astrologer showed that every attack of Al Qaeda could be timed by transits of pluto to the mars of their declaration of “war”. Of course, no further transit of pluto timed anything after that.
So, always remember that just because two things happened at the same time, we have no reason to assume they are connected!