Many diagnostic imaging experiments are characterized by the presence of several observations for each patient studied. Evaluation of metastases with different imaging modalities in patients with cancer or examination of multiple artery segments in patients with heart abnormalities are some examples of such studies. Data obtained from multiple observations per patient are cluster correlated and should not be analyzed by using standard statistical methods because of correlations within a subject. In this article, positron emission tomographic studies are used as a framework to review statistical methods for the analysis of clustered data. Some simple statistical methods that account for correlation within a subject and that can be applied to conventional and well-known statistical methods, such as the chi(2) and t tests, are introduced. One of these methods is illustrated by using a brief analysis of data from a positron emission tomographic study, which demonstrates how resulting conclusions may be incorrect if appropriate techniques are not applied. Alternative methods that can handle multiple observations and dependency within a subject for diagnostic imaging studies are discussed.