Why are we not evaluating multiple competing hypotheses in ecology and evolution?

• Main Authors: Gustavo S. Betini, Tal Avgar, John M. Fryxell
• Format: Article
• Language: English
• Published: The Royal Society 2017-01-01
• Series: Royal Society Open Science
• Subjects: platt; chamberlin; strong inference; model selection; scientific method
• Online Access: https://royalsocietypublishing.org/doi/pdf/10.1098/rsos.160756

The use of multiple working hypotheses to gain strong inference is widely promoted as a means to enhance the effectiveness of scientific investigation. Only 21 of 100 randomly selected studies from the ecological and evolutionary literature tested more than one hypothesis and only eight tested more than two hypotheses. The surprising rarity of application of multiple working hypotheses suggests that this gap between theory and practice might reflect some fundamental issues. Here, we identify several intellectual and practical barriers that discourage us from using multiple hypotheses in our scientific investigation. While scientists have developed a number of ways to avoid biases, such as the use of double-blind controls, we suspect that few scientists are fully aware of the potential influence of cognitive bias on their decisions and they have not yet adopted many techniques available to overcome intellectual and practical barriers in order to improve scientific investigation.