Abstract: | One concept that has been widely supported in experiments on predicting the outcome of majority-rule voting committees is the Core, the set of alternatives that cannot be defeated by any other in a binary contest. Unfortunately, the applicability of these experiments to organizations is limited by a set of rigid controls intended to ensure that the experiments are internally valid. This paper tests the Core in a laboratory environment that relaxes these controls. The Core is found to be highly predictive under these relaxed conditions, thereby enhancing confidence in its external validity. |