In my school each class of girls stays together and takes the same classes all day, so these became easy groupings to use as clusters. Each grade is composed of one or more subsections depending on the number of students in the grade. In all there are M = 185 students divided into N = 11 classes. I decided that given the time it would take to collect sibling data from each student that I would randomly select four clusters for the sample, then visit at the end of their math class and ask each student how many siblings she has. Thankfully we have a high attendance rate and I needed only one visit to each class to collect data from every student. There were few practical problems in establishing the frame. Since each class of students doesn't necessarily have math class every day it took two days to collect all the data. I purposely chose this population and parameter of interest to minimize difficulty in data collection. Both a listing of classes and of all students were readily available to me. b) The class groupings are: 6A, 6B, 7A, 7B, 7C, 8A, 8D, 9A, 10A, 11A, and 12A. Using