The effect of observer experience on English Breeding Bird Survey population trends

Capsule Including observer effects in population models is unlikely to improve population trend estimates. Aims We test the hypothesis that the observer's experience has a significant effect on the detection of birds, and subsequently impacts on estimates of abundance and population trends. Methods Two models were used to test the effect of observer experience: (1) assuming that effects of observer naïvety operate only in the first year, and express experience as a binomial variable; (2) assuming effects of experience improve year‐on‐year, and express as a continuous variable. For each model, experience was included in annual site‐by‐year log‐linear models with Poisson error terms. Results Significant observer‐experience effects were found for up to half of the species analysed depending on the model. However, there were no patterns in the direction of these negative effects. Importantly, including observer experience in analyses of population trends significantly affected estimates of change in only one of the 76 species tested. Conclusions For Breeding Bird Survey data in England, there is no consistent first‐time observer‐experience effect across species and including observer experience in the population models is unlikely to improve population estimates.