Shinichi Nakagawa, William K Cornwell, Corey T Callaghan 

In macroecology, a classic empirical observation has been positive relationships between local abundance and species' range, known as the abundance-occupancy relationships (AORs). The existence of this empirical relationship has informed both theory development and applied questions. Notably, the spatial neutral model of biodiversity predicts AORs. Yet, based on the largest known meta-analysis of 16,562,995 correlations from ~3 billion bird observations, this relationship was indistinguishable from zero. Further, in a phylogenetic comparative analysis, species range had no predictive power over the global mean abundance of 7,464 bird species. We suggest that publication and confirmation biases may have created AORs, an illusion of a ‘universal’ pattern. This nullification highlights the need for ecologists to instigate a credibility revolution like psychology, where many classic phenomena have been nullified.