THE EFFECT OF SAMPLING BIAS ON ESTIMATES OF ANGLER CONSUMPTION RATES IN CREEL SURVEYS

EPA guidance recommends that 30 grams per day be used to represent the consumption rate of fish caught from large bodies of water by a typical angler (EPA, 1989a). This estimate is based on the combined results of the Pierce et al. (1981) and Puffer et al. (1981) surveys of marine and estuarine anglers. An examination of these surveys demonstrates that the method used in both studies - creel survey - oversamples frequent anglers and produces a distribution of consumption rates that overestimates intake rates of the total angler population using the surveyed waterbodies. Weighting the individual survey responses by the inverse of the angler self-reported fishing frequency corrects this bias and produces a more accurate characterization of the total population of anglers using the surveyed waterbodies. This approach is an extension of the methodology used by both Puffer et al. (1981) and Pierce et al. (1981) to estimate the size of the total angler populations. The results of the reanalysis of the Pierce et al. (1981) survey indicate that the median consumption rate for the total angler population is 1.0 g/d. The results of the Puffer et al. (1981) reanalysis indicate a median consumption rate for total angler population of 2.9 g/d. The recalculated distributions of consumption rates were found to be consistent with the results of other angler surveys that use survey methods that do not oversample frequent anglers. The angler intake rate of 30 g/d corresponds to roughly the 90th and 95th percentiles of the total angler populations in the Pierce et al. (1981) and Puffer et al. (1981) surveys, respectively. The results of this paper indicate that the current estimate of 30 g/d significantly overestimates consumption for typical marine and estuarine anglers.