A Collection of Strange Beliefs, Amusing Deceptions, and Dangerous Delusions

Critical Thinker's Dictionary

From Abracadabra to Zombies | View All

occult statistics

Occult statistics are statistics used as the handmaiden of occult theorizing, in much the same way that philosophy was used by theology during medieval times, viz., to justify beliefs in supernatural beings and occult forces.

Parapsychologists, astrologers, theologians, and others who seek anomalies to guide them to transpersonal wisdom and insight into the true nature of the universe, are now able to use computers to do extremely complex statistical analyses of monumental masses of data. When they find a statistically significant correlation between or among variables, they are extremely impressed and consider the discovery to be proof of the occult or the supernatural. To the occult statistician there is no such thing as a spurious correlation.

For example, William Dembski's The Design Inference: Eliminating Chance through Small Probabilities is said to "provide a mathematical foundation for the types of statistical inferences parapsychologists use to identify paranormal phenomena. In particular, the book shows how to deal with statistical experiments whose p-values are extremely small (like those that regularly come up in parapsychology experiments). This work is clearly relevant to Carl Jung's idea of synchronicity. [It] promises to put synchronicity on a solid scientific footing" (Rabi Gupta, personal correspondence).

Likewise, The Princeton Engineering Anomalies Research program (closed in 2007) led for nearly thirty years by Robert Jahn, Dean of the School of Engineering and Applied Science, claimed that in their experiments where human operators tried to use their minds to influence a variety of mechanical, optical, acoustical, and fluid devices, they got results that were not likely due to chance and could "only be attributed to the influence of the human operators."

Legions of parapsychologists, led by such generals as Gary Schwartz and Dean Radin (and his love of meta-analysis), have also appealed to statistical anomalies as proof of ESP. Statistician Jessica Utts of the University of California at Davis  gave her imprimatur to U.S. government studies of ESP and remote viewing. Many occultists have claimed that certain dreams must be clairvoyant and cannot be explained by coincidence because they defy the laws of probability.

It was not long ago that astrologers were claiming that Gauquelin had found the Holy Grail with his statistics showing the so-called "Mars effect." More recently, millionaire playboy Gunter Sachs published Die Akte Astrologie, which uses insurance company data analyzed by professors of statistics at the University of Munich to prove astrology is true.

Obviously, this list could go on forever; it could include the Bible Code and various proofs of the existence of a god on the grounds of improbability that chance could explain the nature of the universe or some complex aspect of it, such as the genetic code.

skeptics unimpressed by occult stats

Skeptics are unimpressed with arguments that assert improbabilities for what has already happened. Whatever has already happened is obviously not an impossible event. Accurately calculating the odds of either the genetic code or the universe occurring by "chance," i.e., by natural laws alone without the design of a divine being, is impossible. Analogies to a monkey typing up Hamlet by chance, a Boeing 747 assembling itself, or to a Mona Lisa being "created" by nature, are irrelevant and notably without impact on skeptics.

Skeptics are not very impressed by statistical anomalies generated by those in quest of occult forces. Sometimes parapsychological colleagues have discovered that statistics were generated by fraudulent means, e.g., the work of Walter J. Levy at Rhine's Institute of Parapsychology (Williams 191, 319). The history of ESP research is a paradigm of dishonesty and incompetence (Rawcliffe, Randi), though it should be mentioned that the two major incidents of fraud (Levy and that of S. G. Soal), though suspected by skeptics, were uncovered and reported by true believers. Skeptics have noted many times while investigating the statistical claims of paranormal researchers that there are often significant problems with subjective validation, confirmation bias, optional starting and stopping, the clustering illusion, the regressive fallacy, etc.

Sometimes the variables being correlated are ambiguous or vaguely defined, if defined at all, so that practically anything can count in support of the occult hypothesis. What is a "great" athlete or a "rebel"? Sometimes the methods of finding patterns are deceptive and inappropriate, e.g., finding hidden messages in texts. As John Ruscio notes, "If you look in a fantastic number of places, and count anything that you stumble upon as supportive evidence, you are guaranteed to discover meaning where none exists" (45).

Skeptics have noted that many times something seems to be statistically improbable when, in fact, it is not improbable at all. Some spurious correlations are due to lack of clarity regarding the variables; others are due to incorrect calculation of the odds. Both errors are common occurrences regarding so-called clairvoyant dreams.

Finally, skeptics are unimpressed with artificially evoked statistical anomalies because such anomalies are expected to occur with some frequency given the vast number of trials that are made.

Correlating just a couple dozen variables with one another will produce a matrix containing nearly 300 correlation coefficients. By convention, results that occur at a level expected by chance just 5 percent of the time are called "statistically significant." We can therefore expect about fifteen spuriously significant correlations within every matrix of 300 (Ruscio, 45).

Each of those spurious correlations is a temptation to see causal connections where there are none and to engage in post hoc theorizing to explain non-existent mysterious forces.

Finally, many of the alleged statistical anomalies discovered by parapsychologists may be due to their equipment and methods. Some use machines that they call "random event generators." However, "It is not clear that any of these machines is truly random. Indeed, it is generally believed that there are no truly random machines. It may be that lack of randomness only begins to show up after many trials" (Park 2000: 199). The data from psi experiments should remind us that statistical significance does not imply importance. Science that claims to have identified barely detectable causal agents observed near the threshold of sensation, which are nevertheless asserted to have been detected with great accuracy and be of great significance, is one of the signs of what Irving Langmuir called pathological science and Bob Park calls voodoo science. As Ernest Rutherford once said: “If your result needs a statistician then you should design a better experiment.”*

See also Bible Code, clustering illusion, confirmation bias, ESP, Forer effect, law of really large numbers, Mars effect, numerology, optional starting and stopping, post hoc fallacy, regressive fallacy, remote viewing, and selective thinking.


further reading

books and articles

Gilovich, Thomas. How We Know What Isn't' So: The Fallibility of Human Reason in Everyday Life (New York: The Free Press, 1993).

Hansel, C.E.M. ESP and Parapsychology: A Critical Re-evaluation (Buffalo, N.Y.: Prometheus Books, 1980).

Hansel, C.E.M. The Search for Psychic Power: ESP and Parapsychology Revisited (Buffalo, N.Y.: Prometheus Books, 1989).

Huff, Daryl. How to Lie with Statistics (W.W. Norton & Company, 1954).

Hyman, Ray. 1996. "The Evidence for Psychic Functioning: Claims vs. Reality." Skeptical Inquirer.

Lee,  K. L,, J. F. McNeer, C. F. Starmer, P. J. Harris and R. A. Rosati. 1980. Clinical judgment and statistics. Lessons from a simulated randomized trial in coronary artery disease.  Circulation; 61;508-515. (When the stats show that nothing is going on, scientists sometimes pull out subgroups from the data to find correlations. Beware: correlations between subgroups may be spurious and due to chance.)

McDonald, John."200% Probability and Beyond: The Compelling Nature of Extraordinary Claims in the Absence of Alternative Explanations," Skeptical Inquirer, January/February, 1998.

Paulos, John Allen. A Mathematician Reads the Newspaper (Anchor Books, 1996).

Paulos, John Allen. Innumeracy: Mathematical Illiteracy and Its Consequences (Vintage Books, 1990).

Rawcliffe, Donovan Hilton. Occult and Supernatural Phenomena(New York: Dover Publications, 1988).

Randi, James. Flim-Flam! (Buffalo, New York: Prometheus Books,1982).

Seife, Charles. 2010. Proofiness: The Dark Arts of Mathematical Deception. Viking Adult.

Ruscio, John. "The Perils of Post-Hockery," Skeptical Inquirer, November/December 1998.

Williams, William F. Encyclopedia of Pseudoscience (Facts-on-File, 2000).

websites

STATS - Statistics and the Media

Coincidences: Remarkable or Random? by Bruce Martin

Bad Science

news

The Dark Art of Statistical Deception The tendency of academics, politicians and pundits to generate ... numerical falsehoods from data — and the tendency of the public to believe the results — is a phenomenon cleverly explored in the new book “Proofiness: The Dark Arts of Mathematical Deception,” by Charles Seife.

Last updated 22-Dec-2013

© Copyright 1994-2013 Robert T. Carroll * This page was designed by Cristian Popa.