Requiring conventional sample size calculations might be an impediment to completing translational or early clinical studies, write Peter Bacchetti and his colleagues at the University of California, San Francisco, in a Perspectives article in Science Translational Medicine. "There is little theoretical or empirical support for the conventional requirement that every study must have at least 80% power," the authors write. Instead, they prose a few alternatives: the "value of information" approach, a method based on estimating the effect and confidence intervals, and one that evaluates the cost and return of increasing the sample size. "Practical considerations, including the costs of larger versus smaller sample sizes, inevitably drive sample size choices, and the reality of diminishing marginal returns implies that this is scientifically valid," the authors add.
Sample Size and the Need for Power
Jun 17, 2011