THE USE OF A SYSTEMS PERSPECTIVE IN EVALUATING OFFICIAL DATA REQUIRES THAT ANY DETAILED ANALYSIS OF A PARTICULAR SUBSYSTEM (POLICE, COURTS, OR CORRECTIONS) MUST BE RELATED TO THE REST OF THE CRIMINAL JUSTICE SYSTEM. THE MAJOR THRUST OF SYSTEMS ANALYSIS IS TO RELATE HOW ONE SUBSYSTEM'S OPERATION IS AFFECTED BY AND AFFECTS THE REMAINING SUBSYSTEMS. THUS, IN ADDITION TO OBTAINING STATISTICS FOR EACH SYSTEM PART, EVALUATORS MUST USE THE STATISTICS ON ALL STAGES IN CONJUNCTION IN ORDER TO RELATE OUTCOMES IN ONE STAGE TO THOSE IN OTHERS. THE FIRST TASK, ESTABLISHING CASE FLOW, DETERMINES THE AMOUNT OF CASE LOSS THROUGHOUT THE SYSTEM AND THE POINTS AT WHICH IT IS TAKING PLACE. THE FOLLOWING POINTS ARE THE ONLY ONES WHICH CAN BE CONSIDERED WHEN USING OFFICIAL STATISTICS: (1) NUMBER OF OFFENSES REPORTED TO POLICE TO OFFENSES KNOWN TO POLICE (NUMBER REPORTED--UNFOUNDED), (2) OFFENSES KNOWN TO POLICE TO OFFENSES CLEARED OTHERWISE AND CLEARED BY CHARGE, (3) OFFENSES CLEARED BY CHARGE TO OFFENSES BROUGHT TO COURT, (4) OFFENSES BROUGHT TO COURT TO OFFENSES RESULTING IN CONVICTION, AND (5) OFFENSES RESULTING IN CONVICTION TO SENTENCING. THE FEASIBILITY OF USING POLICE AND COURT DATA PUBLISHED BY STATISTICS CANADA IS EXAMINED, AND GENERAL CRITICISMS OF OFFICIAL STATISTICS AS DATA SOURCES ARE BRIEFLY REVIEWED. THE ANALYSIS OF STATISTICS CANADA DATA SUGGESTS THAT THE TYPES OF DATA COLLECTED AS WELL AS THE METHODS OF TABULATING AND RECORDING THEM MUST UNDERGO SIGNIFICANT REVISION IN ORDER TO BE SUITABLE FOR CANADIAN RESEARCH ON THE CRIMINAL JUSTICE SYSTEM. EVEN WHEN USED FOR THE PURPOSE OF ANALYZING OFFICIAL RESPONSE TO CRIME, OFFICIAL STATISTICS FAIL TO MEET THE OPTIMISTIC EXPECTATIONS OF MANY RESEARCHERS. CONSEQUENTLY, RESEARCHERS STUDYING OFFICIAL RESPONSE TO CRIME CANNOT RELY ON OFFICIAL STATISTICS AS A DATA SOURCE BUT MUST COLLECT DATA SPECIFIC TO THEIR PARTICULAR PURPOSES. DATA MUST BE COLLECTED WHICH, MINIMALLY, PERMIT THE DETERMINATION OF CASE FLOW, THE REASONS FOR CASE LOSS, AND COMPARABILITY OF POLICE AND COURT STATISTICS TO ENSURE COMMONALITY OF CASES OVER TIME. TABULAR DATA, FOOTNOTES, AND REFERENCES ARE INCLUDED. (PRG)