A new analysis of 92 studies from 27 countries conducted by ecologists at the University of Massachusetts Amherst suggests that many recent multi-species studies of wildlife communities often incorrectly use the analytical tools and methods available.