The following are very small data sets of human birth weights (in kg) of either singleton births or individuals born with a twin. We are interested in the difference in mean weight between singleton babies and twin babies.

a. Construct a valid permuted sample from these data for this difference. (That these sample sizes are in reality too small for an effective permutation test.)

b. Assume that we wanted to test the difference in medians between these two groups. Would that change the way in which the permuted sample would be created?

Transcribed Image Text:

Singleton: 3.5,3.5, 2.7.2.7, 2.6.2.6, 4.44.4 Twin: 3.4.3.4, 4.2.4.2, 1.71.7