Abstract
The precautionary principle suggests that more conservative actions are prudent when levels of uncertainty are high. Decision makers often rely on intuition rather than specific methods to implement the precautionary principle. I demonstrate how Bayesian data analysis produces uncertainty metrics that can be easily blended with a mathematically explicit rendition of the precautionary principle. This results in decisions that are transparent, replicable, and exact. Societal values will often determine appropriate levels of precaution, so methods that elicit input from public stakeholders are needed.