Detecting, estimating, and correcting for biases in harvest data

Abstract

Hunting is important to many people because it provides food, recreation, and cultural identity, so proper management of wildlife is necessary. Wildlife agencies and researchers often rely on harvest data supplied by hunters, but interpretation of these data can be misleading when biases are not acknowledged, assessed, and corrected. We use harvest information collected by the Alaska Department of Fish and Game (ADF&G) from moose (Alces alces gigas) hunters to examine and correct 3 common biases in harvest data: heaping in responses of estimated effort (i.e., rounding), changes in report design, and nonreporting. We found that bias due to heaping was limited (2.8%). A large increase in special permits in 2004 (6.1% in 2003 and 40.1% in 2004) corresponded with increases in individuals with multiple permits (8.6% and 17.3%), which biased estimates of hunt participation calculated from permit data. Failure to correct for multiple reports per hunter also resulted in an artificial decline in success over time. Road access influenced reporting rates; rural Alaska residents without a road had the lowest reporting rate (67%) and rural with a road the greatest (82%). A statewide trend of 663 additional hunters per year calculated from raw permit data was eliminated once data were corrected for both multiple permits and nonreporting. Late reporters were also less likely to hunt (11.8%) than all reporters. Our research shows that survey data bias can significantly influence data interpretation, and wildlife managers must balance information needs, time constraints, and financial resources when determining which biases to correct. © 2015 The Wildlife Society.

Attenuating the nonresponse bias in hunting bag surveys: The multiphase sampling strategy

“…In practice, it is very unlikely to be able to gather relevant information about nonrespondents without contacting them. Accordingly, methods relying on auxiliary variables are generally not in use in our context (but see [ 51 ] for an example of imputation). Moreover, these methods may need assumptions which are difficult or impossible to verify.…”

Estimating bobcat and Canada lynx distributions in British Columbia

“…Trapper surveys likely reflected distribution more directly than harvest records and were the only data source that reported absence. The greatest problem with voluntary surveys is their high number of non‐respondents (Schmidt et al ); our response rate was low (23%), which hindered the spatial coverage of the province but was in line with other efforts (McDonald and Harris , Dorendorf et al ). The number of bobcats and lynx reported in the provincial vehicle‐kill records was also low.…”

Can hunting data be used to estimate unbiased population parameters? A case study on brown bears

“…However, our results showed that hunting records should be used cautiously when quantifying fluctuations in individual condition and population recruitment. To ensure that observed trends reflect true population processes, bias should be estimated [ 20 ] whenever possible. This could be achieved through a parallel longitudinal monitoring of a subsample of the population.…”