Astrology research in R?

Maybe it's a bit of a long shot, but I'm wondering if anyone has any ideas about conducting astrological research in the statistics language R. The gold standard for astrology data is the Swiss Ephemeris, available downloadable online for many years now. However, as an amateur programmer I've been unable to find a way to actually access this data in R.

Some for-fee astrology sites and apps have features for conducting research, but serious research needs to be done in a platform where you have full access to statistical tools and can write scripts to scan through thousands of data points. Although the field of astrological research is growing, it's irritatingly difficult to find information on how to set up a research platform in R. A really competent programmer should be able to write a script, API, or package that would allow for function calls on the Swiss Eph data for private research purposes. (Unfortunately I'm not at that level.) Maybe there is already something like this that has just escaped my notice, but if not, this is a serious empty hole in astrological research efforts that needs to be filled!

I can't help you, other than to recommend that you PM Alois Treindl, site owner of Astrodienst at www.astro.com , which is based in Switzerland.

Indeed, I was just thinking of emailing him today! I will do my best with some fixes I'm working on for the time being, and as a next step I will reach out to Dr. Treindl.

Also might I add that it's really helping all of us interested in astrological research, not just me. I am curious to know how others are conducting their research if they don't have ways of integrating ephemeris data with high-level statistics software. Actually, since I am new to this board, maybe that's a good question to ask. If not with R, how are people conducting research?

Alas, I think that few astrologers have much knowledge of either programming or statistics beyond Chi-square. I would categorize most of the research I've seen as a convenience sample-based frequency count.

I have a big interest in the ancient origins of horoscopic astrology, but I don't suppose that sleuthing around in that area counts as research.

I'd be curious to learn what you've done so far.

Are you familiar with the research of astrologer David Cochrane? He's got a whole series of YouTube lectures based on his work with astrological software and the Astrodienst Astro-DataBank.

https://www.isarastrology.org/news/...introduction-to-astrological-research-methods

https://www.youtube.com/watch?v=rb9tLPn5Dcw

https://www.youtube.com/watch?v=GNhprGytCZo

Yes, I've seen quite a few of David Cochrane's videos. I think he's doing excellent work for this whole area. I have considered getting Sirius (that's his software I believe) since I know he talks about the research features. Without having tried Sirius myself, I would guess that it's still going to be hard to automate and script a workflow in order to go through lots of numbers. I've opted to go the DIY route for maximum flexibility. (And to save some money.)

A lot of David's research, from what I've seen, involves looking for extreme case sampling. So he wants to take the fastest runners, the most beautiful actresses, and look for patterns in just these charts. While I see the merit in this approach, I see some potential pitfalls as well. I think we all need to follow our hunches (something astrologers tend to like doing anyway), following leads Sherlock-Holmes-style but while keeping in mind the confines of statistical methods. After all, it's not about using a bunch of fancy math. The importance of statistics is to tell us "how do we know that X is significant?"

I haven't been at this too long, but I'm getting close to having a workflow set up in R to do the kind of research I wanted. (As far as the Swiss Ephemeris stuff, it was actually much easier to figure out than I realized .) One thing that astrologers with any bent towards research should keep in mind is that the vast majority of people putting out meaningful research aren't statisticians. They are biologists, geologists, economists (well, maybe not this last one), etc. Astrologers can also add themselves to that list.

I think you've nailed the potential pitfall of David Cochrane's line of research. I was very interested to see his video on Ceres, because I don't think this dwarf planet is well understood. Cochrane's approach was to find people with a partile sun-Ceres conjunction on the Astro-DataBank, and then describe the individual with the closest conjunction by way of saying what horoscopic Ceres means.

Given that one has to start somewhere, it's hard to say that Cochrane's opening gambit is better or worse than anything else. But it's all highly problematic.

In modern astrology I'd say that a conjunction to the sun indicates what the person identifies with. But in traditional western natal and horary astrology, that planet is combust and thereby weakened, unless it is cazimi (within 17' of exact.)

So right away we have to ask, "Which astrology?"

And of course, a horoscope is highly complex with hundreds if not thousands of interacting data bytes.

My background in statistics is limited to a long-ago and far-away university course, but I had a subsequent academic career where I had to interact with all kinds of scholars who used statistics in their research.

I really think astrologers have to step up and learn multivariate statistical analyses. We have to learn about complex systems. Most scientists, engineers, and a lot of social scientists do learn statistics, whether through their own departments or through the math department. They use stats routinely in their research.

I'm not sure stats are going to give us the answers, but if they do, astrologers are going to have to move well beyond simple chi-square tests and two-variable regression analyses.

I hesitate to recommend it because when last I checked this thread it was into its 51st page, but there is a lot of discussion of research in astrology during the middle pages of: https://forum.astro.com/cgi/forum.cgi?num=1495746510/0

Thanks for that link, I will check it out when I get the chance.

Which astrology, indeed. Well, before you even get to that question, there is the simple matter of qualitative vs. quantitative. I personally think it's jumping off the deep end to try and find statistical proof using purely qualitative things. I think the steady route involves looking at objective measures like date of marriage, lifespan, number of children, body metrics (height, weight), perhaps some health conditions (yes/no), etc. If you are creative and resourceful, there should be mountains of data to sift through. Then there are things like survey results and questionnaires-- not quantitative per se, but also objective and hopefully repeatable. These should also work as long as the sample size is big enough, which is a big "if".

Looking at the qualitative stuff, though-- like what represents ego, what happens when Mars squares Uranus, who is likely to feel happy in life or sad-- this stuff, from my point of view, is still in its infancy. Yes astrology has been considering these variables for a while now, but they aren't differentiated, and there's just no way you're going to mold them into anything statistical and repeatable because it's all in the eye of the beholder. That's why I think this is a "wish list" type of scenario.

Back to the astrological schools that you mention, I myself have moved towards Vedic. I think the Indian tradition has amazing "bones" as far as an underlying tradition that's both subtle and complex. There is even a long tradition of varga charts, what Cochrane calls harmonic or vibrational charts. (Actually I wonder how much he knows about the Vedic use of these charts, since I think I've only heard him mention their existence in Vedic astrology in passing.) I really like the work of Ernst Wilhelm, who is really a Sanskrit scholar and lifelong researcher. He has pored over ancient texts and interpreted them in a fairly novel way, as a result of new fragments that have come to light and a painstaking approach to detail. He claims to get predictive rates of better than 80% with some of the methods he has narrowed down. (The vast majority of astrology techniques, Western and Vedic, he more or less consigns to garbage.) Anyway, he has a Youtube channel and gives interviews in case you haven't heard of him, such as this one http://www.youtube.com/watch?v=KtbGNHEdgsU.

I see your point about the advanced topics in statistics, however I think the bigger issue is just thinking statistically. It doesn't take anything super fancy to get from point A to point B; a simple correlation over a large enough sample size, with a good and intuitive metric, would work wonders. Maybe this has already been tried and failed-- I'm not at the stage yet to know. If so, it would certainly be sobering...

Good points. Thanks for initiating an intelligent discussion.

I don't know whether David Cochrane looked at the Vedic system of harmonic charts or not. My favourite book on western harmonic charts is Harding and Harvey, Working With Astrology, but I don't think it's still in print.

I think it really matters which astrology we use, in the sense that traditional western and Vedic astrology have a long and serious history of predictive methods. Modern western astrology, in contrast, began ca. 1900 primarily as a tool for personal enlightenment. Whether or not we feel more enlightened by taking a spiritual evolutionary approach is a subjective assessment.

But then we have the problem that these 3 main systems work so differently. The sidereal and tropical zodiacs are 24 or 27 degrees off now, depending upon whom you ask. Trads work with essential dignities and mods work with modern outers if not asteroids.

From a research perspective, so many parameters have to be specified. Among them are the variables in a person's life that have no relation to astrology, like your DNA inheritance, your nationality, cultural milieu, and gender.

Simple frequency counts based on convenience samples aren't going to help us here, unfortunately.