The tests carried by the English mathematician Dr Soal are often cited as powerful evidence for the existence of ESP phenomena. However there is strong evidence that Soal engaged in fraud in order to obtain his results and thus such claims should be viewed with caution.
Soal originally became interested in psychical research when he conducted a lengthy series of ESP tests, hoping to provide independent corroboration of Rhine's work. He tested 160 people over a five-year period and analyzed the total of 128 350 guesses against the targets (symbols on the cards) they were attempting to 'see'. He found nothing but chance results and promptly stopped his ESP research, criticizing Rhine for what he considered must be errors in the methods he used to produce positive results.
That might have been the end of the story, had it not been for the influence of yet another English researcher, Whately Carington. In his own ESP tests, using drawings as targets, Carington had discovered strange displacement effect. Sometimes a subject would miss the target he was trying to guess and instead reproduce the previous day's targets or even the one to be selected at random the next day. Carington urged Soal to re-examine his statistics and look for such a 'psychic displacement'. The mathematician did this, and sure enough the effect was found in the results produced by two subjects, Basil Shackleton and Gloria Stewart. Both showed positive and negative displacement at times, and Soal continued his ESP work using Shackleton and Steward as his subjects.
The results of experiments conducted with Shackleton between 1941 and 1943 were extremely impressive and were taken up by parapsychologists as evidence of the existence of ESP. But 20 years later Mrs. Gretl Albert, who had been involved in the tests as an am agent, claimed she had several times seen Soal altering the figures. A recent re-examination of the Soal statistics suggests that this is exactly what he did. In order to ensure that the cards used in the experiments were picked at random, Soal used the standard laboratory technique of referring to Chamber's logarithmic tables and Tippett's random number tables (although he did not indicate exactly how he used them). What has been discovered is that the random lists Soal used in his experiments do not match the standard ones. A study by Betty Markwick, published in 1978, has revealed that certain long sequences of numbers are repeated many times. This need only mean that Soal was using a small pool of random numbers and would not be necessarily affect the validity of the experiment. However, Miss Markwick has discovered that the long repeated sequences are in fact not identical; they are sometimes interrupted by extra numbers, and that these, where they occur, show a remarkable correspondence with the ESP 'hits' recorded by Soal. Remove them and the scores fall to chance levels. Summing up this evidence, Miss Markwick states that 'all the experimental series in card-guessing carried out by Dr. Soal must, as the evidence stands, be discredited.'
See: Betty Marwick, "The Soal-Goldney Experiments with Basil Shackleton: New Evidence of Manipulation," Proceedings of the Society for Psychical Research, 56, 211.