The Essentials of Biostatistics for Physicians, Nurses, and Clinicians

22 CHAPTER 2 Sampling from Populations

4 is E

5 is F

Now we must divide [0, 1) into fi ve equal parts.
So we get:

If 0.0000 ≤ U < 0.2000, then the index is 1.

If 0.2000 ≤ U < 0.4000, then the index is 2.

If 0.4000 ≤ U < 0.6000, then the index is 3.

If 0.6000 ≤ U < 0.8000, then the index is 4.

If 0.8000 ≤ U < 1.0000, then the index is 5.

The second uniform random number from the table is 29676, corre-
sponding to 0.29676. Now since 0.2000 ≤ U < 0.4000, the index is 2
corresponding to C. So now our sample includes A and C. Again, in
order to sample without replacement from the remaining four patients
B, D, E, and F, we divide [0, 1) into four equal parts and redefi ne the
indices as

1 is B

2 is D

3 is E

4 is F

For the intervals, we get:

If 0.0000 ≤ U < 0.2500, then the index is 1.

If 0.2500 ≤ U < 0.5000, then the index is 2.

If 0.5000 ≤ U < 0.7500, then the index is 3.

If 0.7500 ≤ U < 1.0000, then the index is 4.

The third uniform random number in the table is 69386. So U = 0.69386.
We see that 0.5000 ≤ U < 0.7500. So the index is 3, and we choose
patient E. Now we have three of the four required patients in our
sample. They are A, C, and E. So for the fi nal patient in the sample,