Mars effect - dead or alive? Dissenting from J.W. Nienhuys' 'retrospect.' - response to J.W. Nienhuys, Skeptical Inquirer, vol. 6, p. 24, 1997

Is the Mars effect dead? Dr. Nienhuys' retrospect on the Mars effect (Nienhuys 1997) might be premature. Why? He makes his case by referring to Gauquelin's athletes database, which he deems biased. Yes, Gauquelin's published database of N = 2,888 is biased, as was shown with Gauquelin's cooperation several years ago (Ertel 1988). Ertel visited Gauquelin in Paris and copied N = 1,496 unpublished cases that he found in Gauquelin's files, combined them with the published sample, and thus obtained N = 4,384. This total is bias-corrected.

Ertel provided this database to the Dutch skeptics, and in his article Nienhuys makes use of it. So Nienhuys or others who have this data should be able to make the simple comparison shown in Figure 1 between the distribution of all of Gauquelin's data and that of the CFEPP. The birth distribution of Gauquelin's corrected athletes' sample across twelve Mars sectors can be seen in the dashed curve of Figure 1. Even with bias removed, a remarkable peak of births with Mars rising and a slight peak with Mars culminating is apparent (birth excess in the two key sector zones).

The figure also shows the distribution of the French skeptics' athletes (CFEPP) across the twelve Mars sectors. Strangely, this distribution hardly deviates from Gauquelin's, and the correlation CFEPPC-Gauquelin is r = 0.85 which is highly significant (p = 0.003) (CFEPP and Gauquelin deviations from chance expectancy must be used here as variables, not original frequencies). Nienhuys, placing full confidence in the CFEPP's data, says that the Mars effect "most likely has been an illusion after all" (p. 24). But the correlation shown in Figure 1 is difficult to class as an illusion. For both samples, the athletes' births peak considerably at the point where Mars rises. The hump in Gauquelin's data when Mars culminates is not replicated in the CFEPP's data, but this hump is small anyway. Moreover, an astounding parallel drop is apparent in the two samples down to sector number 6 and a rise is seen again up to sectors 7-10 when Mars sets and reaches its lower culmination. Nienhuys already noticed, in Gauquelin's data and in his suggestions for changes to the CFEPP sample, these secondary deviations from expectancy, though he felt they were due to some "secondary" selection bias on Gauquelin's part. But how can such bias on Gauquelin's side slip into the CFEPP's data as well, since these were collected independently, "from scratch," as Nienhuys says.

It is true that the CFEPP's primary birth deviations only (Mars percentage for rise plus culmination) do not reach the conventional significance level of p = .05. Very significant only is the Mars percentage in Gauquelin's bias-corrected total (p = 0.00007). Why is the CFEPP's critical percentage not significant while the correlation of its distribution with Gauquelin's distribution is? Shouldn't we try to find an answer to Nienhuys' question ". . . Where are those famous champions that the CFEPP deviously excluded to suppress the Mars rate?" (p. 29)? This is a question that we need to take seriously rather than rhetorically?

Notes

1. We make no comment on the CFEPP's intentions, but the answer to Nienhuys's question can be found in sample adjustments suggested or endorsed, but not followed, by the CFEPP in their own report, and also in one important omission from that report. Readers will find details of these and other points in our full article, "The Mars Effect in Prospect. Dissenting from J.W. Nienhuys' 'Retrospect'" at Ken Irving's 'Planetos' web site http://members.aol.com/kirving.

References

Ertel, S., 1988. Raising the hurdle for the athletes' Mars effect. Association covaries with eminence. Journal of Scientific Exploration 2: 53-82

Nienhuys, J.W., 1997. The Mats effect in retrospect. SKEPTICAL INQUIRER 6:24-29

Suitbert Ertel is a professor at the Institute for Psychologie, George-August-Universitat, 37073 Gottingen, Germany. Kenneth Irving writes from Staten Island, New York.

COPYRIGHT 1998 Committee for the Scientific Investigation of Claims of the Paranormal
COPYRIGHT 2000 Gale Group