Sampling Simulation 1.3 2003/05/28 2003/07/14 15:23:05.917 GMT-5 David Lane lane@rice.edu David Lane lane@rice.edu Kyle Clarkson kclarks@rice.edu Elizabeth Gregory lizzardg@rice.edu Kevin Duh kevinduh@rice.edu Mariyah Poonawala mariyah@rice.edu Matthew Jeanes mjeanes@rice.edu Jeffrey Silverman jsilv@rice.edu

General Instructions This demonstration illustrates the difference between simple random sampling and stratified sampling. The population consists of 100 animals: 60 lions, 30 turtles, and 10 rabbits. The population is shown on the left. The default is for random sampling. If you push the "sample 10 animals" button, 10 animals will be sampled and the number of lions, rabbits, and turtles will be tabulated and shown. The animals sampled are shown in blue. If you push the button again, another sample will be drawn and the new tally will be shown to the right of the first tally. To start over, click the "reset" button. If you choose "stratified sampling" then the composition of the sample will match that of the population.

Step by Step Instructions Show Questions With random sampling, how often will the numer of lions sampled be exactly 6? Click "Sample 10 animals" repeatedly and count the number of times there are exactly 6 lions. Is it about half the time? Or is it more, or is it less? Click "Reset" and then choose "Stratified sampling." Click the "Sample 10 animals" button a few times and compare results to random sampling.

Summary With random sampling and a small sample size (such as 10), the composition of the sample is often very different from that of the population. This does not invalidate the use of random sampling because the mathematical models underlying statistical procedures take the randomness into account. Stratified sampling, by definition, assures that the composition of the sample matches the population on the variables for which it has been stratified.