The article “Controlled Clinical Trial of Canine Therapy Versus Usual Care to Reduce Patient Anxiety in the Emergency Department” (PLoS ONE 2019) reported on an experiment in which 80 adult hospital patients were randomly assigned to either 15 min with a certified therapy dog (m = 40) or usual care (n = 40). Each patient’s change in self-reported pain, depression, and anxiety (pre-treatment minus post treatment) was recorded. The researchers employed a rank-sum test to compare the two treatment groups on each of these three outcomes; the resulting test statistic values appear below.

Test the hypotheses H₀: μ₁ − μ₂  = 0 against H_a: μ₁ − μ₂ < 0 (1 = dog therapy treatment, 2 = control) for each of the three response variables at the .01 significance level. What can be said about the chance of committing at least one type I error among the three tests?

Transcribed Image Text:

Change in: W = Pain 1475 Depression 1316 Anxiety 1171