Key points are not available for this paper at this time.
Shipley M J (Department of Epidemiology and Population Sciences, London School of Hygiene and Tropical Medicine, London WC1E 7HT, UK), Smith P G and Dramaix M. Calculation of power for matched pair studies when randomization is by group. International Journalof Epidemiology 1989, 18: 457–461. A formula is given for the calculation of the statistical power of paired intervention trials in which the units of randomization are groups rather than individuals. Factors affecting the power of such trials are also investigated and an example is given where the power of group randomization is compared to that of individual randomization.
Shipley et al. (Sun,) studied this question.