In its ecological sense, biomonitoring refers to the recording of observations of animal or plant populations, often over space and time. Owing to the possible presence of spatially and temporally correlated observations, the “what, where, when, and how” of biomonitoring studies must be carefully considered, beginning with the choice of appropriate response variables to measure. There are many available methods to analyze data from biomonitoring field studies or experiments. The standard toolbox of statistical techniques, particularly in the area of regression analysis, now includes generalized linear models and generalized additive models. These allow the incorporation of non-Gaussian (non-normal) error terms, and can describe nonlinear relationships using nonparametric smoothing functions. Analysis of a single dataset may include liberal use of graphics, handling of data with heterogeneous variances and below detection limits, and dealing with irregular coverage through space and time. A multipronged approach to analysis of complex datasets that often arise from biomonitoring studies is not unusual.