APPLET Use the simple random sample applet.

The Simple Random Sample applet can demonstrate how randomization works to create similar groups for comparative experiments.

Suppose that (unknown to the experimenters)

the 20 even-numbered students among the 40 subjects for the cell phone study in Example 3.11

(page 185) have fast reactions, and that the oddnumbered students have slow reactions. We would like the experimental and control groups to contain similar numbers of the fast reactors. Use the applet to choose 10 samples of size 20 from the 40 students.

(Be sure to click “Reset” after each sample.) Record the counts of even-numbered students in each of your 10 samples. You see that there is considerable chance variation but no systematic bias in favor of one or the other group in assigning the fast-reacting students. Larger samples from larger populations will on the average do a better job of making the two groups equivalent.