Testing Therapeutic Touch
Summary. Therapeutic Touch (TT) practitioners claim they can treat medical problems by manipulating what they call the “human energy field.” In this study the investigators tested the ability of TT practitioners to detect a human energy field. Each practitioner attempted to detect the experimenter's hand on 10 trials. This video segment also stimulates discussion of topics such as: science vs. pseudoscience, the placebo effect, observer expectancy effects, and double-blind vs. single-blind control.
Video. Scientific American Frontiers: Beyond Science (1997). The segment begins approximately 47 minutes into the program and runs for about 7 minutes. This segment is also part of the 2002 SAF program A Different Way to Heal. The segment, Healing Touch, can be viewed online (as of August 2016) -- it is episode 10 of season 12. Pause the video after Alan Alda says “The hands belong to a trained therapeutic touch practitioner. A coin toss determines where Emily will position her hand. Then the practitioners have to say where the hand is.” After the students have analyzed the data, show the remainder of the segment.
Original Research. Rosa, L., Rosa. E., Samer, L., & Barrett, S . (1998). A close look at therapeutic touch. Journal of the American Medical Association, 279, 1005-1010.
Statistical Analysis. For the initial and follow-up tests, perform a directional (one-tailed) one-sample t-test . The null hypothesis is that the population mean = 5. The data were taken from Figure 2 in Rosa et al. showing for each of the 28 TT practitioners the number of trials (out of 10) on which they were correct. If practitioners cannot detect an energy field then the expected performance is 50% correct. Although Rosa et al. do not report the results of a one-tailed t-test for all 28 practitioners, your students could do the test (the results are: M = 4.39, SD = 1.663, t(27) = -1.93, p > .05).
Published Results. Results of a one-tailed one-sample t-test are reported separately for the initial test of 15 practitioners (M = 4.67, SD = 1.74, t = .72) and the follow-up test of 13 practitioners (M = 4.08, SD = 1.44, t = -2.22). Note that because the observed means are less than 5, and the test is one-tailed, it wasn't really necessary to perform a t-test at all.
Published conclusions. "To our knowledge, no other objective quantitative study involving more than a few TT practitioners has been published, and no well-designed study demonstrates any health benefit from TT. These facts, together with our experimental findings, suggest that TT claims are groundless and that further use of TT by health professionals is unjustified." (p . 1009).
Table 1. Realistic data showing the number of correct responses (out of 10) on the initial test of 15 practitioners and the follow-up test of 13 practitioners.
Initial Test, Follow-up Test
The Excel file for this activity contains the same data as shown here in Table 1.