NOT ONLY IS IT FEASIBLE TO APPLY PERFORMANCE MEASURES IN A JURISDICTION BUT IT IS FEASIBLE TO MAKE INTERJURISDICTIONAL COMPARISONS OF PERFORMANCE. ACCORDINGLY, CERTAIN KEY PERFORMANCE MEASURES CAN ASSIST OFFICIALS IN COURT, PROSECUTION, AND PUBLIC DEFENDER AGENCIES TO IMPROVE CRIMINAL PROCEEDINGS. IN THIS STUDY, STRUCTURED INTERVIEWS WITH 33 EXPERIENCED CRIMINAL JUSTICE PRACTITIONERS IN 13 JURISDICTIONS WERE CONDUCTED TO ELICIT THEIR VIEWS ON THE SELECTION OF ISSUE AREAS AND TWO DEMONSTRATION JURISDICTIONS AND ON THE VALUE AND SELECTION OF RELEVANT PERFORMANCE MEASURES. RESEARCHERS ALSO OBTAINED DATA ON 2,000 FELONY CASES IN MULTNOMAH COUNTY, ORE., AND DADE COUNTY, FL. A PILOT CASE AUDIT WAS MADE OF 20 BURGLARY-TYPE CASES. IN ADDITION, MAIL SURVEYS WERE ADMINISTERED TO 1,200 VICTIMS, WITNESSES, AND JURORS IN EACH JURISDICTION. ISSUE AREAS INCLUDED CHARGING STANDARDS, CHARGING ACCURACY, PLEA BARGAINING, SENTENCE VARIATION, EVENHANDEDNESS, DELAY, EFFICIENCY AND ATTITUDES OF LAY PARTICIPANTS. THE MEASUREMENT OF PERFORMANCE VARIED WITH THE AVAILABILITY OF DATA AND THE PRECISION OR AMBIGUITY OF THE PERFORMANCE MEASURES. AS A RESULT OF THIS STUDY RESEARCHERS SUGGEST AN INTEGRATED PERFORMANCE MEASUREMENT PROGRAM (IPMP), CONSISTING OF AN ENUMERATION OF REQUIRED DATA ELEMENTS AND PERFORMANCE MEASURES; STANDARDIZED DATA COLLECTION AND OUTPUT FORMS FOR EACH POLICY ISSUE ARE; FLEXIBLE MODULAR SOFTWARE PACKAGES FOR COMPUTING, DISPLAYING, AND ANALYZING PERFORMANCE MEASURES WITHIN EACH ISSUE AREA; AND GUIDELINES FOR CONDUCTING CASE AUDITS AT EACH MAJOR DECISION POINT. THE IPMP WOULD NECESSITATE THE ADMINISTRATION OF SAMPLING PLANS, STANDARD MAIL SURVEY QUESTIONNAIRES, AND STANDARD PERSONAL INTERVIEW QUESTIONNAIRES AND THE ANALYSIS OF RESPONSES OF DEFENDANTS. IT IS SUGGESTED THAT CASE AUDITS COMPLEMENT BUT NOT REPLACE STATISTICAL PERFORMANCE MEASURES AND THAT THE DATA COLLECTION BE CONSISTENT AND COMPLETE. THE DESIGN OF AN OPERATIONAL IPMP COULD BE EVEN MORE FULLY DEVELOPED WITH MORE ATTENTION TO THE AREAS OF EVENHANDEDNESS IN SCREENING, THE TRIAL, AND THE ALLOCATION OF PROSECUTORS' AND PUBLIC DEFENDERS' TIME. STATISTICAL MODELS SHOULD BE REFINED AND AN INFORMATION SYSTEM SELECTED. EACH JURISDICTION MUST DETERMINE WHETHER IMPLEMENTING A FULL IPMP IS COST BENEFICIAL; THIS CAN BE DONE ONLY AFTER SUCH AN APPROACH IS INSTALLED AND ITS BENEFITS CAN BE ASSESSED. NO SUPPORTING DATA ARE INCLUDED. FOR RELATED DOCUMENTS, SEE NCJ 36010-36011 AND 43817. (FCW)