You can't judge books by their covers, but you can judge scientific papers by their titles. Consider: "The planetary positions and relationships at the dates of birth of a cohort of Nigerian schizophrenics." Yeah, this is going to get weird.
The always great NCBI ROFL offers this wonderfully strange paper from the University College Hospital in Ibadan, Nigeria. Here's the basic premise:
"Some astrological hypotheses related to predisposition to severe mental illness were tested by analysing the zodiacal signs, the interactions between planetary qualities (aspects), and the occurrence of full and new moon dates, on the dates of birth of 221 schizophrenics, compared with 112 normal subjects."
So then, schizophrenia is supposed to be predicted by certain astrological signs, and so the researchers examined how well these signs lined up with schizophrenics vs. non-schizophrenics. Credit where it's due - I've definitely heard worse ways to prove a "scientific" basis for astrology. So then, what did they discover? Well, quite a bit really:
The sun signs of the schizophrenics were significantly more likely to be in the signs associated with introversion, while those of the control population were significantly more likely to be in the outgoing signs. A significantly higher proportion of schizophrenics had their Mars (i.e., symbol of aggressiveness) in the outgoing signs than the normal population. A significantly higher proportion of control subjects fulfilled operational criteria for adequacy of number of aspects between the sun and the other planets. The tendency for a higher proportion of schizophrenics to have "difficult" aspects just failed to reach significance. A significantly higher proportion of control subjects had aspects between the sun and mars; and also a significantly higher proportion of control subjects had "soft" (helpful) aspects between the sun and mars.
I'll be honest and admit I barely understood any of that. But the researchers are confident that these findings fit well with known astrological assumptions about schizophrenia...at least, until they spectacularly undercut themselves in the final sentence of their abstract:
These findings are in keeping with the well-known oddity of schizophrenia (schiz = split; phren = mind); such that, a group which collectively is characterised by an "introverted" self (i.e. sun sign), has a coexisting aggressive tendency (i.e. strong mars) and poor integration between the elements of the psyche and the self (i.e. inadequacy of aspects between Sun and other planets).
However, the findings give only partial support to key astrological postulates because there was a non-significant trend for more schizophrenics to be born in "water" signs and on full moon dates.
So then, astrology seems to be a statistically significant indicator of schizophrenia...except that it also just produces a lot of random statistical noise, so it might all be meaningless anyway. I suspect this wasn't the researchers' intention, but that actually sounds about right. And no, before you ask, I don't think the researchers took into account the possibly shifting Zodiac.