THE COMPARISON OF PROPORTIONS IN THE PRESENCE OF LITTER EFFECTS

Considerable attention has been paid in recent years to statistical issues which arise when the effect of a new agent is assessed by comparing the proportion of affected animals in different treatment groups. The sampling unit for such experiments is often an intact group of animals, such as a litter, herd or flock. In this case, the application of standard statistical methods to the sample data may lead to spurious statistical significance, because the underlying assumption of statistical independence required for their validity is violated. Methods for dealing with this problem are discussed, using data from a published example to illustrate the various approaches that could be taken.