Studies Of The I Ching: I. A Replication - Statistical Data Included

Journal of Parapsychology, The, June, 2001 by Lance Storm, Michael A. Thalbourne

MICHAEL A. THALBOURNE [1]

ABSTRACT: L. Storm and M. A. Thalbourne (1998-1999) carried out an experiment (N = 93) with the Chinese book of divination, the I Ching (J. Blofeld, 1968), which contains 64 hexagrams (6-line structures) and associated readings. Three coins are thrown 6 times to generate 1 of these hexagrams. Participants selected 16 of 64 hexagram-descriptor pairs, based on their emotional and/or cognitive states of mind. It was predicted that 1 of the 16 choices would come up as a "hit" ([P.sub.MCE] = .25). The proportion of hits was marginally significant. Transliminality and 6 factors on R. B. Cattell's 16 Personality Factor Questionnaire (16PF; R. B. Cattell, H. W. Eber, & M. M. Tatsuoka, 1970) correlated significantly with hitting success. Number of changing lines (given by throwing 3 heads or 3 tails) was not significantly different from chance, but number of changing lines did correlate significantly with transliminality. Number of changing lines also correlated with 5 factors on the 16PF. Finally, transliminality cor related with 5 factors on the 16PF. A replication experiment (N = 107) was conducted in 1999. Hexagram hitting was significant, but the transliminality--hitting correlation and the 6 hitting--16PF correlations were not significant. Also, the transliminality--changing lines correlation and the 5 changing--lines--16PF correlations were not significant. However, 4 transliminality--16PF correlations were significant. The successful parapsychological outcomes were interpreted according to the theory of psychopraxia (L. Storm & M. A. Thalbourne, 2000; M. A. Thalbourne, 1982, in press-a).

This is a report of a replication of an experiment we conducted in 1998 (Storm & Thalbourne, 1998-1999). We carried out a study with the I Ching (an ancient Chinese system of divination; Blofeld, 1968) to determine, among other things, whether transliminality [2] "might function as a connecting principle between paranormal effects and other personality variables" (Storm & Thalbourne, 1998-1999, p. 100). Participants were also required to complete the Transliminality Scale (Thalbourne, 1998) and Cattell's 16 Personality Factor Questionnaire (Cattell, Eber, & Tatsuoka, 1970).

According to traditional usage, to consult the "oracle" of the I Ching one must generate a hexagram, which involves, first, posing a meaningful question to the I Ching (yes-no questions are excluded), followed by the repeated casting of 64 yarrow sticks to generate all six lines of the hexagram. (The modern method involves throwing three coins six times, which was the method used in both I Ching experiments.) We provided further details about the "coin method," and background material concerning the origin, structure, and use of the I Ching, in our previous article (Storm & Thalbourne, 1998-1999, pp. 101-103).

One of the earliest experiments with the I Ching was conducted by Rubin and Honorton (1971, 1972). Participants were asked to generate a hexagram in response to a specific question. They were each given two readings: (a) the correct one and (b) a control reading. Both readings were rated on a scale of 1 to 10 by the participant according to how relevant or accurate he or she felt them to be (the difference between the two ratings being the dependent variable). There was no overall significant result for the whole sample, but a sheep-goat rating scale had also been administered, and participants who believed in ESP scored significantly higher than those who disbelieved.

A replication of Rubin and Honorton's (1971, 1972) experiment was conducted by Thalbourne, Delin, Barlow, and Steen (1992-1993). Participants followed the same procedure of generating a hexagram and rating two readings (the correct one and a systematically selected control). There were three planned "parapsychologically relevant" hypotheses, and the following results were found: (a) although the mean difference score was above chance, it was not significantly so, (b) there was no significant positive correlation between scores on the Australian Sheep--Goat Scale (Thalbourne & Delin, 1993) and difference scores, and (c) those who believed in the efficacy of the I Ching scored significantly higher on difference scores than those who disbelieved.

Regarding finding (b), Lawrence (1994) used Rosenthal's (1991, p. 63) Formula 4.2 for testing the significance of the difference between effect sizes to test the nonsignificant finding of Thalbourne et al. (1992-1993) against Rubin and Honorton's (1971, 1972) significant finding. When the original mean difference scores were converted to effect size estimates, Lawrence found no significant difference between the two. He concluded that Thalbourne et al. had effectively replicated Rubin and Honorton's result.

Thalbourne (1994) conducted another experiment using the I Ching but this time focused on changing lines exclusively. Using a personal (but ecologically sound) database, he first found that the number of changing lines in mid-1992 was significantly higher than mean chance expectation (1.66 changing lines, where MCE= 1.5 changing lines).