Notes
This essay presents the views of the authors on the general problem of selecting analytic measures. It does not necessarily reflect the views of the Northrop Corporation or any agency of the US Government. While accepting full responsibility for the arguments and conclusions expressed herein, we want to acknowledge Andrew W. Marshall's constant encouragement, wise counsel, and numerous suggestions. We also want to thank Eliot A. Cohen and Thomas A. Fabyanic for their invaluable comments on an earlier draft.