Misidentification is a problem for many ecological data sets and dealing with this source of observational error is increasingly important as we rely more on publicly collected data sets and large species occurrence databases. We have been working to quantify rates of misidentification and to develop statistical methods that properly deal with these errors when making inferences.
