SPECIFICALLY, IT ANALYZED THE RELATION BETWEEN COURT DELAY, COURT PRODUCTIVITY, AND THE DEMAND FOR COURT SERVICES AMONG ALL DISTRICT COURTS, AND THE DETERMINANTS OF DIFFERENCES IN COURT PRODUCTIVITY ITSELF. THE STATISTICAL METHOD USED WAS MULIPLE REGRESSION ANALYSIS APPLIED ACROSS ALL OF THE COURTS FOR EACH YEAR FROM 1968 TO 1974. THE TREND IN AVERAGE OUTPUT PER JUDGE OVER ALL COURTS WAS FOUND TO BE SIGNIFICANTLY HIGHER USING THE OUTPUT MEASURE THAN USING TOTAL CASES DISPOSED. COURT DELAY WAS FOUND TO BE CONSISTENTLY RELATED TO PENDING WORKLOAD PER JUDGE BUT ONLY WEAKLY RELATED TO OUTPUT PER JUDGE. OUTPUT PER JUDGE (PRODUCTIVITY) WAS FOUND TO BE STRONGLY RELATED TO DEMAND PRESSURE (TOTAL AVAILABLE WORKLOAD) AND SIZE OF COURT, BUT NOT SIGNIFICANTLY RELATED TO THE DIFFERENTIAL USE OF TRIALS. THESE FINDINGS SUGGEST THAT THE COURTS, ON THE AVERAGE, HAVE RESERVE CAPACITY, AND THE USE OF TRIALS IN PRACTICE IS NOT AS SIGNIFICANT A FACTOR IN LIMITING COURT OUTPUT AS IS GENERALLY BELIEVED. THE FINDINGS ALSO POINT TO SOME WEAKNESSES IN THE ANALYTIC FRAMEWORK ITSELF; REFINEMENTS TO ACHIEVE A MORE CONSISTENT AND ACCURATE ANALYSIS WITHIN THE GENERAL FRAMEWORK USED ARE SUGGESTED. APPENDED MATERIALS EXPLAIN THE METHODS USED TO CONSTRUCT THE OUTPUT AND DEMAND VARIABLES USED IN THE ANALYSIS AND PRESENT THE COURT DATA AND CORRELATION MATRICES. (AUTHOR ABSTRACT) (SNI ABSTRACT)...ELW