Dominic Sprott (A-16) conducted an experiment with rabbits in which the outcome variable was the fatty infiltration in the shoulder mass (PFI, measured as a percent). At baseline, 15 rabbits had a randomly chosen shoulder muscle detached. The shoulder was then reattached. Six weeks later, five randomly chosen rabbits were sacrificed and the differences in the PFI between the reattached shoulder and the nondetached shoulder were recorded (group A). Six months later, the 10 remaining rabbits were sacrificed and again the differences in the PFI between the reattached shoulder and the nondetached shoulder were recorded (group B).

Percent Fatty Infiltration Difference (Nondetached–Reattached)
Group A Group B 2.55 1.04 1.38 0.9 3.29 0.75 0.2 0.99 0.36 0.29 1.79 0.74 1.11 0.85 0.3 Can we conclude, at the .05 level of significance, that the treatments have a differential effect on PFI between the two shoulder muscles? What is the p value for the test?
In each of the Exercises 15 through 29, do one or more of the following that you think are appropriate:

(a) Apply one or more of the techniques discussed in this chapter.

(b) Apply one or more of the techniques discussed in previous chapters.

(c) Formulate relevant hypotheses, perform the appropriate tests, and find p values.

(d) State the statistical decisions and clinical conclusions that the results of your hypothesis tests justify.

(e) Describe the population(s) to which you think your inferences are applicable.

(f) State the assumptions necessary for the validity of your analyses.
15. The purpose of a study by Damm et al. (A-17) was to investigate insulin sensitivity and insulin secretion in women with previous gestational diabetes (GDM). Subjects were 12 normal-weight glucosetolerant women (mean age, 36.6 years; standard deviation, 4.16) with previous gestational diabetes and 11 controls (mean age, 35 years; standard deviation, 3.3). Among the data collected were the following fasting plasma insulin values (mmol/L). Use the Mann–Whitney test to determine if you can conclude on the basis of these data that the two populations represented differ with respect to average fasting plasma insulin level.

Controls Previous GDM Controls Previous GDM 46.25 30.00 40.00 31.25 40.00 41.25 30.00 56.25 31.25 56.25 51.25 61.25 38.75 45.00 32.50 50.00 41.25 46.25 43.75 53.75 38.75 46.25 62.50