RESEARCHER VARIATION IN COLLECTION OF MORPHOMETRICS ON BLACK BEARS

Morphometrics are routinely taken on vertebrates for a variety of reas ons. One purpose is to predict weight (or other parameter) on the basi s of a single measurement or set of measurements. Variation in measure ments taken in the field, however, may negatively affect the predictiv e power of such equations. To assess this error, we estimated the vari ation in morphometrics of black bears (Ursus americanus) taken by 6 fi eld researchers. Eight body measurements were recorded by 3 teams of 2 researchers on 70 black bears in Great Smoky Mountains National Park in Tennessee and North Carolina. Measurements were summarized to deter mine intra- and inter-researcher variation present in the morphometric s. We ran simulations on predictive equations to determine how the obs erved variation affected the results. The intra-researcher simulations differed from one another by 1.5-2.6% of the average of the predicted parameters and the inter-researcher simulations differed by 3.4-8.1% of the mean of the predicted parameters. These differences may explain some of the error associated with predictive equations developed from morphometrics. We believe that it is imperative for researchers to mi nimize error in taking measurements by developing stringent standards for data collection and by reducing the number of personnel involved i n data collection.