Morphology often relates to ecology in a well-defined manner, enabling prediction of ecological roles for taxa that lack direct observations, such a fossils. Diet is a particularly important component of a species' ecology. However, in order to predict diet it must first be codified, and establishing metrics that effectively summarize dietary variability without excessive information loss remains challenging. I will show how a dietary-item relative importance coding scheme can be used to derive multivariate dietary classifications for a sample of extant carnivoran mammals, and then used Bayesian multilevel modeling to assess whether that these scores can be predicted from a set of dental metrics, using body size as a covariate. There is no ``one size fits all'' model for predicting dietary item importance but model-averaged estimates perform especially well. I will show how models derived from living taxa can be used to provide novel insights into the dietary ecology of a few extinct carnivoran species. Significantly, this approach need not be limited to diet as an ecological trait of interest, to these phenotypic traits, or to carnivorans. Rather, this framework serves as a general approach to predicting multivariate ecology from phenotypic traits.