PROBLEMS WITH MEASURING DIET SELECTION OF TERRESTRIAL, MAMMALIAN HERBIVORES

The validity of methods used to measure diet selection has often been overlooked. This paper examines problems associated with the scale of food availability, perception of food by animals, measures of dietary composition, measures of food availability and diet selection indices. Most problems can be reduced by the choice of techniques consistent with the objective and degree of accuracy required, both of which should be clearly stated. In particular, the measured sample of food utilized by the animal should be spatially and temporally matched to the scale of food availability under consideration. In the past, researchers have tended to overstate the accuracy of their results. We support the view that the most appropriate analysis of diet selection is ranking food items in order of selection according to the value of the selection index.