In theory, obtaining a representative norm group is straightforward and simple. Consider a scholastic achievement test designed for sixth graders in the United states. The relevant population is all sixth graders coast to coast and in Alaska and Hawaii. A representative cross-section of these potential subjects could be obtained by computerized random sampling of 10,000 or so of the millions of eligible children. Each child would have an equal chance of being chosen to take the test; that is, the selection strategy would be simple random sampling