<h2>METHODOLOGICAL EXPLANATION IN NONTECHNICAL LANGUAGE</h2> If any measurement is taken multiple times, say, for example, if the activity of a particular enzyme is measured in the blood of 6 rats, the measurement will not be the same for every rat; multiple factors will contribute to some variability. Results of this kind of experiment are often presented as a mean ( i.e ., an average) and a measure of variability, often the sd . The coefficient of variation (CV), when expressed as a proportion, is defined as the ratio of the sd to the mean; this normalizes the sd to the mean, thus making the CV a powerful tool to compare standardized variability among different experimental systems. In the rat experiment, the CV would be the ratio of the sd to the mean of the 6 enzyme activity measurements. If multiple sets, each n = 6, of these means and sd s are sampled from a single population of rats, the resulting collection of CVs will have its own distribution. From such a distribution, using assumptions we discuss below, we computed a range of values (limits) around the mean CV within which sample CVs would be expected to fall
/lp/fed-of-american-socs-for-experimental-biology/part-2-unusual-clustering-of-coefficients-of-variation-in-published-UVBr78fOfa