The importance of cross-validation, accuracy, and precision for measuring plumage color: A comment on Vaquero-Alba et al. (2016)

Abstract

Vaquero-Alba and colleagues published a study in The Auk: Ornithological Advances comparing objective color measurements of plumage taken in the field directly on a bird's body to those taken in the lab on collected feathers arranged to emulate the appearance of a bird's natural plumage. Although the field measures of plumage color were less repeatable than lab measures, the authors concluded that measurements taken in the field were more representative of a bird's "true color." Accordingly, they recommend that researchers should bring spectrophotometers into the field to measure color on live birds. We question the assumption that their field measurements represent true color and highlight concerns regarding their experimental design and methodology. Because they did not measure color of live birds in the lab or the color of plucked feathers in the field, they cannot directly test whether the assessment of color in the field on a live bird is superior. Also, rather than assume field measures are the most accurate or precise way to assess plumage color, we suggest cross-validation with other methodologies, such as digital photography, pigment biochemistry, or measures of a known color standard in both environments. Importantly, researchers should be aware of the limitations and advantages of various methods for measuring plumage color so they can use the method most appropriate for their study. Q2017 American Ornithological Society. ISSN 0004-8038, electronic ISSN 1938-4254.