THE MAIN THEORETICAL OR CAUSAL ANALYSIS PURPOSE OF THE ARTICLE IS TO PROVIDE A METHODOLOGY FOR DETERMINING AN OPTIMUM GEOGRAPHICAL ALLOCATION FOR VARIOUS GOVERNMENT PROGRAMS AS A STEP IN SEEKING TO EXPLAIN WHY THE OPTIMUM DIFFERS FROM THE ACTUAL. AS AN ILLUSTRATIVE EXAMPLE, IT USES THE ALLOCATION OF THE FUNDS OF THE LAW ENFORCEMENT ASSISTANCE ADMINISTRATION AMONG THE 50 STATES. THE METHODOLOGICAL PROBLEMS INCLUDE (1) ALLOCATING TO PLACES WHERE ONE MAKES AN ASSUMPTION OF LINEAR OR CONSTANT RELATIONS BETWEEN DOLLARS SPENT AND CRIMES REDUCED; (2) ALLOCATING WHERE ONE MAKES AN ASSUMPTION OF LOG-LINEAR OR DIMINISHING RETURNS RELATIONS BETWEEN DOLLARS SPENT AND CRIMES REDUCED; (3) USING DATA FOR ONE, TWO, OR MORE TIME POINTS FOR TWO OR MORE STATES; (4) TAKING INTO CONSIDERATION THE MINIMUM ALLOCATIONS PER PLACE AS DETERMINED BY THE POLICY MAKERS OR INTERNALLY-GENERATED FUNDS; (5) CONTROLLING FOR DEMOGRAPHIC AND OTHER VARIABLES THAT MIGHT INFLUENCE CRIME OCCURRENCE; (6) INTEGRATING ALLOCATION TO ACTIVITIES WITH ALLOCATION TO PLACES; (7) VARYING THE GOALS TO BE OPTIMIZED IN ORDER TO CONSIDER MAXIMIZING TOTAL RETURNS, MINIMIZING COSTS, DIFFERENTIAL WEIGHTING OF OUTPUTS, MINIMIZING OUTLIERS, OR MAXIMIZING THE NUMBER OF MODEL PLACES. (AUTHOR ABSTRACT MODIFIED)