Ecologists need robust survey designs, sampling and analytical methods

Abstract

1. Research that yields conflicting results rightly causes controversy. Where methodological weaknesses are apparent, there is ready opportunity for discord within the scientific community, which may undermine the entire study. 2. We use the debate about the role of dingoes Canis dingo in conservation in Australia as a case study for a phenomenon that is relevant to all applied ecologists, where conflicting results have been published in high-quality journals and yet the problems with the methods used in these studies have led to significant controversy. 3. To alleviate such controversies, scientists need to use robust methods to ensure that their results are repeatable and defendable. To date, this has not occurred in Australia’s dingo debate due to the use of unvalidated indices that rely on unsupported assumptions. 4. We highlight the problems that poor methods have caused in this debate. We also reiterate our recommendations for practitioners, statisticians and researchers to work together to develop long-term, multi-site experimental research programmes using robust methods to understand the impacts of dingoes on mesopredators. 5. Synthesis and applications. Incorporating robust methods and appropriate experimental designs is needed to ensure that conservation actions are appropriately focused and are supported with robust results. Such actions will go a long way towards resolving the debate about the role of dingoes in conservation in Australia, and other, ecological debates.