Stone M. H. (1997) Statistics "Prove" Ether Exists! Rasch Measurement Transactions 11:1 p. 541.

Carver's (1993) reanalysis of data from the famed 1887 Michelson-Morley experiment demonstrates that the hypothesis, "the speed of light depends on the direction of travel through the ether", cannot be rejected even at a .001 level of statistical significance. Thus ether is confirmed and Einstein's Theory of Relativity invalidated. Relativity's consequences, such as the Atomic Bomb, are seen to be misunderstandings at best, hoaxes at worst.
The moral of Carver's story is that significance testing is never decisive.
More informative than statistical testing, however, is plotting the data points as Michelson & Morley (1887) did. Their Figure is reproduced here. The upper curve is for the noon observations and the lower for the evening ones. The dotted curves represent one-eighth of the amplitudes away from the X-axis predicted by the "ether" theory. The solid lines represent the empirical results. The solid lines (the effect size) do exhibit some waviness and are indeed statistically "significantly" different from the straight lines predicted by non-ether theory. But do they provide support for the ether theory? The slight empirical waviness indicates that less than 1% of the variance of the speed of light is associated with direction, a substantive rejection of the ether theory. Relativity can be reinstated!
Several methodological conclusions result from comparing the graph with the "significance" analysis:
1) Well designed graphs always convey the meaning of results better than summary statistics and significance test results, and with less opportunity for confusion.
2) Effect sizes must be reported along with significance tests for correct conclusions to be drawn.
3) Type I errors (failures to reject incorrect null hypotheses) can occur even at the very demanding .001 level.
Mark Stone
Adler School of Professional Psychology
Chicago, Illinois
Carver R (1993) The case against statistical significance testing, revisited. The Journal of Experimental Education 61:4287-292.
Michelson A, Morley E (1887) On the relative motion of the earth and the luminiferous ether. American Journal of Science, 34, 333-341.
Statistics "Prove" Ether Exists! Stone M. H. Rasch Measurement Transactions, 1997, 11:1 p. 541.
The URL of this page is www.rasch.org/rmt/rmt111c.htm
Website: www.rasch.org/rmt/contents.htm