Astrology is the name given to a whole wide range of belief systems which supposedly connect the positions of the planets with human affairs. Natal astrologers claim that they can read important characteristics about a person from seeing their natal chart, showing the places of the sun, moon, and planets when they were born. This is a testable claim. One's location and time of birth are objective facts available to the outside world. Astrology is therefore open to the methods of science to judge its validity. In contrast, the claim that the universe was made and is run by a Supreme Being is not testable, because the nature of the Supreme Being is left so vague as to not be falsifiable by any objective test. After all, one can always say "It just looks like cause and effect...really, it was God's decision to make this happen". The claim is neither provable nor falsifiable. Any claim which is addressable by objective tests, will likely be tested by someone. And a field so popular as astrology has drawn interest from a number of scientists to test. So let's take a look at what such tests have been done and what they say.
A UC physicist and MacArthur Award winner Shawn Carlson performed what many regard as the most careful set of tests of natal astrology to date, back in 1985 - double-blind tests taking great care to incorporate the suggestions of 30 prominent American and European astrologers in the conduct of the tests. It was well done enough to satisfy the requirements for publication in Nature - one of the most respected of all scientific journals. Here is a link to an article discussing these tests and the results. Unsurprisingly, the tests showed that astrologers did no better than random chance in reading their subjects. What was interesting is that despite the care Carlson took in communicating well with the astrology community, at a gathering after the publication of the results the astrologers showed so little respect, objectivity, or professionalism..."I attended an NCGR party - I was the only non-astrologer in the house - to discuss the research shortly after it was published. The discussion was, to put it politely, energetic. I have not yet received a serious scientific challenge to the paper." The newsletter of the American Federation of Astrologers Network published a response in January (1986). "I was very disappointed to see that it largely consists of personal attacks," Carlson said. He said its few substantive criticisms are attributable to ignorance of his experiment, of the CPI, and of basic scientific methodology.
This article from Skeptico describes Carlson's tests in more detail, and also many others and links to published results. And another in Skeptico shows how abysmally poor were the high-sounding criticisms of Carlson's work by S. Ertle. Note that subjects were given 3 astrological chart readings; one based on their actual birth data, and two based on other people's birth data - and to choose the chart which best described themselves. So this was not an outside clinician judging the personality of someone, but the person themselves making that intimate judgement. Yet still, they chose their own chart no better than random chance, while the astrologers themselves claimed that when analyzed, the test would show more than 50% of them would get it right. It is also interesting that the astrologers were given the opportunity to attach a number proportional to their confidence in the accuracy of their horoscope reading, and the results showed a noticable tendency for the highest-confidence readings to actually be the most INaccurate. There is another very interesting point that wasn't made in these articles - I'm impressed that the astrologers consented to this test. That strongly suggests that they really genuinely believe in astrology (vs. being simple con artists). Many other attempted studies don't get this far, evidently because the astrologers fear they'll fail. Regardless, astrology fails no matter how much or little the astrologers believe it.
An entire book has been written (Culver and Ianna 1988) on the scientific tests of astrology and the attempts by astrologers to provide some physical basis for why astrology should work, and it too shows astrology utterly fails in all its claims. The Astronomical Society of the Pacific has also organized a resource webpage on pseudoscience, including astrology which has many great studies linked. Some have tried to criticize astrology because orbital precession has shifted the sun signs relative to the calendar by fully one month since the ancient origins of astrology. But this criticism is naive - since astrologers have provided no agreed mechanism by which astrology could work (e.g. tidal effects on body fluids, some hypthetical 5th force, or whatever). Therefore, the fact that the sun is actually in Ophiuchus in December when I was born, not Sagittarius as the astrologers say, makes no difference on the claims of astrology. No. What kills astrology is that there is no actual correlation between birth place/time and personality, when real people and astrological readings are actually done in an unbiased way. In other words - hypthesized mechanisms to explain how astrology works miss the point - the point being - there is NOTHING to EXPLAIN. To be clear: Astrology claims there is a cause / effect relation between the planets positions, a person's birth place and time, and their personality. ANY claim of cause and effect MUST show evidence that effect is indeed correlated with cause! If in fact the evidence shows there is NO correlation (as these tests show), then there is no basis for ANY claim of a causation by ANY imaginable cause. It does not matter what cause the believers want to speculate might be true (the "cosmic unconscious" or "spirits" or some undiscovered vector field from undiscovered physics... it does not matter). If effect is uncorrelated with cause, there is no basis for claiming causation, regardless of what cause one might try to imagine.
I find that my pointing out astrology's failures has little effect on the astrology believers in my classrooms. Psychologists have described two psychological tendencies which help explain why individuals can feel so strongly that astrology really works when in fact it doesn't. Confirmation bias is one, whereby people tend to better remember instances when their prejudices are confirmed than when they are not. The other is the Forer effect (also called the P.T. Barnum Effect), which refers to a tendency for individuals to give high accuracy ratings to descriptions of their personality that are presented to them as tailored specifically for them, but are in fact vague and general enough to apply to a wide range of people. Another clear problem is that in commercial astrology, the astrologer has the person right in front of them, and we're all very tuned by Mother Nature to gather a ton of information about someone just by having them in front of you. Clearly it's an evolutionary advantage to do so with some accuracy, so that's how natural selection has bred us. To test the actual claims involving the planets, we need to remove that source of personality information, which is what a careful test will begin with as a bare minimum.
I often have students tell me after my astrology discussion that they do read their horoscopes but just because its "fun". We all respond positively when we think someone is genuinely seeing our best qualities, and astrologers are clearly practiced at writing nice things with regard to personalities. But this is a slippery slope. Here's an excellent essay on the dangers from those who know better, of being simply "nice" to astrology-believers and letting it pass uncommented. Realize that there was a day in the past when these believers welcomed astrology into their minds. They can welcome it right back out the door just as easily, if their commitment is instead to "whatever the Truth is". It doesn't mean they're deeply "wrong". It in fact means, their self-concept is secure, they're courageous and commited to intellectual honesty rather than perpetuating and compounding past mistakes.
There's more links you can find for yourself using the wonders of Google, but you get the picture: Astrology makes objective and testable predictions which utterly fail, and astrologers and their fans show very little interest in acknowledging these failures. Perhaps another truism is relevant...