Now that I've gotten your attention, no, this thread isn't about some recently published paper that supports astrology. It is, however, about a test that we're doing right here on the BABB. In a thread discussing astrology, gzhpcu made a statement that he's made before, that he investigated a particular type of astrology based on Jung's work, and based on the descriptions of people it provided, felt like there was something to it. He's acknowledged that that isn't exactly a scientifically supportable claim, but still feels that the personalist descriptions fit remarkably well. For myself, I think it's more likely a case of preferentially noticing the parts of such a description that match, and neglecting the parts that don't match as well, but I think we can all agree that judging how well a single description fits someone is a pretty subjective matter.

Therefore, it seemed to me that the best way to explore the issue would be to simply test whether the descriptions created match better than descriptions for someone else's birth date, so I suggested such a test here, and gzhpcu graciously agreed to do the work for such a reasonable test.

So, here's the way it works. I'm acting as a moderator for the test, and while I fully expect a null result, I'll be acting strictly as a neutral party in conducting it. My actual interaction with the participants will be minimal. The participants will send me, via private message, their birth details, including time and place of birth. Both of these should be as accurate as possible. Time of birth should be at worst within five minutes, and within a minute would be better. Place of birth should be to the arcminute or better in both latitude and longitude (I'm willing to do the looking up if you just have the name of a hospital, for example). I will randomly assign each participant to a group of four (if there happen to be two people with extremely close birth times and places, I'll make certain that they are in different groups, though). All of the birth information will be passed on to gzhpcu, without names attached. He's agreed to do a natal chart for each person, and then providing a description for that person. Each participant will then receive all four of the descriptions for the people in their group, in random order, and will then choose the description that they feel fits them best. Once everyone has made that determination, we'll see how many choose the one that was actually based on their own birth information, and I'll post the full results here.

One would expect that, just by chance, one in four people will choose the description that was actually created for them. If, however, gzhpcu's technique actually creates descriptions that match people better than average, we should see a significantly larger number of matches. I'm hoping for a total of twenty participants, to have at least a moderately reasonable sample size. So that would mean an expected five matches. Six or seven would be hard to justify as anything other than a statistical fluke, while eight or nine might be arguable (my personal recommendation would be for further tests, though that would clearly involve quite a bit of work for gzhpcu). If there were ten or more matches (half the participants), I'd acknowledge that such a result seems unlikely to be mere chance.

So, how can you participate? So far, I've had a total of four people respond with their birth details, and two more say they'd be happy to participate, though they haven't yet forwarded the information. You're welcome to participate, whether you think astrology is valid or not. The requirements are: send me a private message with your birth time and place, as specified above; once you receive a set of personality descriptions, make an honest effort to choose the one which you feel best suits you; don't make any effort to discuss the specific details with myself or gzhpcu in any way during the test (for example, don't send a private message telling gzhpcu your birthday, and suggesting that he slip the word "asparagus" into the description so you can recognize it ). I'll likewise make certain not to provide any outside information that might give a hint as to which description goes with which person, and of course gzhpcu won't actually know who the participants are until the test is complete, so he won't be able to base anything on his knowledge of people from their posts here.

If anyone has recommendations for improvements to this protocol, please let me know, and I'll do what I can to accomodate them, if everyone agrees that they seem reasonable changes.

Oh, and if we can, I'd like to keep this thread limited to discussing this test, details thereof, significance of the results, once they're in, and so forth. If you want to discuss astrology in general, and its good or bad points, I'd suggest posting them in the thread I linked to at the beginning of this post. Since this is purely an empirical study, I'd prefer not to get bogged down in a discussion of what the proposed mechanism for astrological influence might be, or anything of that nature. We can discuss whether there needs to be a mechanism after we see if there's any phenomenon that needs to be explained!