Inverse prediction is important in a wide variety of scientific and engineering contexts. One might use inverse prediction to predict fundamental properties of a material by using multiple measurements (responses) obtained from it. This can be accomplished by inverting parameterized forward models that relate the measurements to the properties of interest. Recognition of the uncertainty in the estimated forward models leads to an errors-in-variables approach for inverse prediction. The forward models (and their associated uncertainty) can also be used to analyze how well the various responses complement one another for inverse prediction over the range of the factor-space of interest. One may find that some of the responses are redundant or non-informative. Simple analysis and examples are used to illustrate how one may select an informative and discriminating subset of response variables among candidates in cases where the number of response variables that can be acquired during inverse prediction is limited by difficulty, expense, and/or availability of material.