Are there patterns in how coefficients of variation (CV's)
vary across taxonomic group, age class, or survey method?
Here's a breakdown of mean CV's according to such variables from
the annual surveys contained in the Amphibian Count CV Tool. More specific
cross-variable analyses can be made at that site, where the mean CV is given for
A CV is calculated by dividing the standard
deviation by the mean. To remove the effects of trend over time, we used the
standard deviation of the residuals from a regression of the
counts against time. Associated with this method are inherent assumptions
which you may find disagreeable due to sticky statistical issues like model
misspecification, autocorrelation, and effect of survey length on variance.
For this reason, we include raw counts in the database
for you to analyze with your preferred method.