The McGrew-McFall experiment of 1990: The end of astrology?

Something went wrong

something went wrong…


Note: If there are astrological terms that you do not understand consult my glossary.



This time I try examining the McGrew-McFall experiment of 1990.

It is a rather interesting and worthy of note experiment, made in cooperation among astrologers and researchers (or, at least, I haven't heard of situations similar to those of the Carlson experiment).

The cooperation level had been so high that the two authors go to the point of affirming that, the six astrologers that took part in the experiment, declared themselves very happy of taking part in an experiment made taking in serious account the astrologers’ suggestions, seeing this as an opportunity to demonstrate, once for all, the validity of this millenary art.

Sadly things didn’t go as they expected: nobody of the participants passed the test, giving the researchers the impression of having debunked astrology instead.

Besides the justifications adduced by the team that took part in the experiment to explain the failure, even other astrologers wrote articles on the test. I found these two texts: and this, more synthetic,
Still I do not agree with many of the arguments raised, so I try to express my opinion.


The experiment


Here one can find the article issued by the two psychologists to share the test’s results:
In short they asked the “Indiana Federation of Astrologers” (the state where the researchers live) to give a set of six astrologers of proven capacity.

The astrologers received the task of linking 23 birth charts to the relative owners using a set of informations.
These informations were of various type: a whole figure photo, a vocational test, a personality test and a set of open questions (to the amount of 61) agreed by the astrologers themselves, questions on the most various topics: hobbies and interests, love life, religious beliefs, attitude toward authority, dates of significant life events, familiar situation, favourite colours and much more; many of these informations can’t be obtained by any existing personality test, but are very useful to the astrologers for a correct identification, and it is easy to understand the happiness of the astrologers for being able to work on these informations.

Some one could think that, once the astrologers have the answer to as many as 61 questions decided by them, the psychological tests are superfluous.
But they could be very useful instead, because they give an “outer” image of the examined subjects. Anyone answering to open questions strives to give an image of himself better than real, so many psychologists prefer closed questions with a range of values, often with a bipolar system (such as -5= introverted; +5 extroverted) to reduce the risk that the subject gives the answer that he believes to be more positively evaluated. Naturally this risk isn’t completely avoided (not only for cultural inclinations, think about the extroversion question in a questionnaire for a job position characterized by frequent contact with the public).
Also the vocational test can be rather useful because the examined subjects could not having found the job to which they are best suited, particularly in this case since, as we will see, the chosen individuals where rather young.
So tests interpreted by professional psychologist could represent a good integration to the open questions devised by the astrologers.

The astrologers’ requests have been taken into consideration also for the choice of the person to identify: the astrologers asked the subjects of an age equal or greater than 30 years with very different professional situations.

The academics gathered 23 volunteers 30-31 years old (a big difference of age would have eased the identification of the correct birth date based only on the photo).

The chosen individuals had rather different careers; in the article they do eight examples: a prostitute, a never-do-well son of a politician, a bum, a lawyer, a showman, a seamen, a journalist and a firefighter.

The astrologers had then the task of linking every birth chart to the right dossier, giving also their confidence level on the choice, and even with the possibility of choosing alternatives.

In the end the astrologers made a very meagre figure, with correct associations varying from zero to three.


Critical points of the experiment


To begin I have to say that interpreting a psychological test is not like measuring a person’s height or weight. The error margin is much greater and, if the psychologists made an error, this could have potentially influenced the judgement of the astrologers.
I do not know in detail the test given to the volunteers, I can only say that often, the closed questions tests to which I answered, had also questions that I did not believe could be applied to me, so I had in fact to nearly invent the answer, but with the fear of altering the test. Anyway this objection could only derive from my ignorance on the subject.

Using 30 years old subject reduces the usefulness of their job, this is so because, at a so young age, it is not certain that one has already found his main job, even in the United States, “the land of opportunities”. Anyway this problem could be reduced by the vocational test that, if the psychologists made a good job, can correct possible discrepancies and allow correct identifications anyway.

I end with what I believe is the main reason that afflicted the experiment: the illusion that everything is in the birth chart.
It is not so and cannot be so, in the same hour and place can be born both the son of a rich catholic entrepreneurial family and
the daughter of poor atheist fishermen. Among the life of the two we could certainly note parallelisms and similarities, but taken as a whole thei
r lives would be rather different.
he birth chart shows which influxes the native has to express, but gives only very limited informations about the context where they are expressed.
According to the context the same influences can express themselves in a very different, even twisted way, that could even make the difference between an honest life and a criminal one.

A birth is like the introduction of a variable in a system, the outcome depends from the characteristics of the system and their compatibility with the inserted variable.
For example pouring some water can give very different results: if it is poured out on a plant the result would be positive; if on the floor the result would be indifferent or moderately negative and, to conclude, if poured on an active electronic apparel we would have made a good deal of damage. The birth chart allows to know the variable, but gives rather limited informations on the system where it is introduced.

For example the birth chart of the never-do-well politician’s son could be not much different from the bum’s one.
It doesn’t end here, the firefighter theme could show a strong Mars influence (danger, fire) that could be shown also in that of the lawyer (Mars as contentions and conflicts), while the nativity of a firefighter’s co-worker could show rather different influences (such as the influence of fire signs or of the Sun; maybe or even completely different elements could be predominant, suggesting more the urge of helping others).

I do not exclude that, an astrologer that worked on many nativities of firefighters and lawyers brought to their profession by a strong Mars influence, can find hints that allow to distinguish among the two chances with ease, but without a similar experience it could be more difficult than one could initially think.

So I believe that it is this the main reason of the failure of the astrologers involved in the test: on one side they have overvalued the potentialities of the birth chart and on the other side they adopted a too light-hearted approach in their research for hints allowing the right identification.

We have to add that 23 birth charts are probably too much, with a strong risk of finding more charts compatible with the informations they were looking for.
I understand that finding the correct chart among two or maybe even three can bring the fear of flukes, but 23 are really too many.
It is true that the two researchers affirm that, when they brought the results at the Indiana Federation of Astrologers, the attendants reacted with incredulity and split themselves into groups, each working on only five datasets, failing even with this reduced set.
But probably they worked too quickly, they should have brought the informations at home and worked on them for some days.


The conclusions of the two sides


The astrologer’s failure is evident and, understandably, the researchers believed of having disproved astrology.

They made also some strange considerations, for example showing surprise for the fact that the six astrologers, that affirmed of using the same approach (six astrologers working in the same way? Is that possible?) have made very different associations.
They thought that the association between the birth charts and the dossiers would have been identical using the same method, but in astrology the human factor is very relevant and, even applying the same method (seriously?), there is a great margin of interpretation; we are examining persons, not things.

If everybody converged on the same wrong profiles maybe astrology (or that astrological school at least) would have been disproved, but if the results diverge simply the astrologers made different errors; there was only a real possibility of converging results… choosing the right ones.

Anyway, reflecting on the matter, McGrew and McFall had also an intuition maybe excessive, but partially correct: they affirms that the complexity of astrology can bring to find confirmations to anything you are looking for in a birth chart.
Human beings are complex and contradictory creatures, inside everyone there is a bit of everything, even Gandhi will have desired the death of somebody on some occasion; so it is probable that if, instead of carefully looking what there is in the chart, we look for what we want to find, we do great mistakes.


Notwithstanding the fact that the two researchers had occasion for speaking with the astrologers, understanding them and the discipline, they seem to have maintained a rather approximative idea of the art.
They affirm, in fact, that selecting a range of correct data among 23 is much easier than obtaining these data form a birth chart, considering the infinity of possible combinations.
t could even be easier, but in truth the astrologers receive from the customers some support informations, many of them similar to those gathered for the experiment, in order to understand the context where the influences shown by the birth chart are expressing themselves.
A customer doesn’t go to an astrologer in order to test his ability, but to obtain informations or resolve a problem, so he cooperates with the astrologer giving as much informations as possible.

The Indiana Federation of Astrologers remained much surprised by the test results. So it tried to botch up a defence publishing an article on the “Journal of Research of the American Federation of Astrologers” affirming: “ many cases the correct answer contained the attributes we had chosen, but in a different [astrological] position… one big mistake was in agreeing to use young subjects. This was the Saturn/Neptune conjunction, of course, which produced may “lost souls”… like medicine, the law and theology astrology may not always give quantifiable results – but it works, none the less.” (quote taken form the article of McGrew and McFall).

Personally I do not believe that the Saturn/Neptune conjunction can create a series of nativities impermeable to a scientific test; instead I think that the affirmation about the subject age is at least partially correct, because, as already hinted, it is not sure at all that they already found their vocation, in the majority of cases, anyway, their personality should be completely developed or nearly so.
The affirmation about the chosen attributes, instead, is an admission of guilt, of negligence on the astrologer’s part, that probably should have examined with more attention the birth charts, seeking all the “rival” charts and then identify the right one. Still we have to understand the difficulty of the test, also for the great number of birth charts.

With a similar reaction it is no surprise if, the two psychologists, conclude the article showing stupor for the resistance of belief systems in face of disconfirming evidences.

But, what the two researchers are missing, is that the astrologers have already seen their discipline working in their daily practice. So, a test that does not demonstrate it, at their eyes can be a test with something wrong or a test badly executed by the astrologers, but surely can’t represent the disproving of astrology.

- The Cat