Are we selecting appropriate metrics to assess human impacts on biodiversity?

Biased and subjective choices in the variable selection processes used in ecological studies commonly lead researchers to reach misleading conclusions regarding patterns of biodiversity response to disturbances. Nevertheless, little attention has been given to these processes in the majority of studies published to date. Here, we assess the extent to which variables commonly employed in ecological studies correspond to those deemed to be most important by researchers of the same studies. Specifically, we examined both biodiversity (response) and environmental (explanatory) metrics from a comprehensive literature review and compared their use with their relative importance, according to a survey with the studies’ authors. We used the literature concerning land use change effects on dung beetles as our study case. Our results highlight marked disparities between researchers opinion and their choice of variables in published papers. We suggest that these disparities are due to the high costs of sampling and processing some variables, as well as to logistical constraints and researchers own bias. If current practices and these discrepancies persist then our understanding of the biodiversity consequences of land-use change will remain compromised, while further undermining our confidence in the results of ecological studies