STUDY SAMPLES WERE DRAWN FROM THE STATE SUPERIOR COURT'S SENTENCING DOCKET FOR FEBRUARY AND MARCH 1975, AND FROM THE BOARD'S HEARING DOCKET FOR THE SAME TIME PERIOD. IN THE CASE OF PAROLE ANALYSIS, THE ENTIRE RELEASE HEARING PROCESS, INCLUDING THE EXECUTIVE SESSION DURING WHICH THE BOARD DELIBERATES AND DECIDES THE CASE IN THE INMATE'S ABSENCE, WAS OBSERVED. ALTHOUGH SENTENCING HEARINGS WERE OBSERVED TO ASSESS PROCEDURES AND PARTICIPANTS IN THE SENTENCING PROCESS WERE INTERVIEWED OR SURVEYED, THE CRUCIAL PART OF SENTENCING -- THE NEGOTIATIONS BETWEEN THE PROSECUTOR AND THE DEFENSE COUNSEL -- WAS CLOSED TO OBSERVERS. MULTIVARIATE ANALYSES OF PROBATION AND PAROLE DECISION CRITERIA SUGGEST THAT PAROLE BOARD DECISIONMAKING IS MORE STRUCTURED THAN COURT DECISIONMAKING. IN LIGHT OF THE SPECIFIC, LEGAL CRITERIA AVAILABLE TO THE COURT AND THE COMPARATIVE LACK OF PAROLE CRITERIA, THIS FINDING IS THE OPPOSITE OF WHAT WOULD HAVE BEEN ASSUMED. THE DATA REVEAL NO PARTICULAR GOAL PATTERNS IN EITHER DECISION FUNCTION WERE FOUND, LARGELY REFUTING THE HYPOTHESIS THAT BOTH PROCESSES WERE DOMINATED BY CONCERNS WITH RETRIBUTION AND INCAPACITATION AND THAT LEGAL CRITERIA CARRIED THE GREATEST WEIGHT IN EXPLAINING DECISIONS. LARGE CRITERIA HAD MARGINAL IMPACT IN BOTH ANALYSES; TREATMENT VARIABLES CARRIED SOME WEIGHT, BUT IN A RELATIVELY INCONSISTENT FASHION. ORGANIZATIONAL, CONSTITUTIONAL, POLICY, AND OTHER IMPLICATIONS OF THE FINDINGS ARE DISCUSSED. SUPPORTING DATA, DETAILS OF THE ANALYSIS, AND A BIBLIOGRAPHY ARE PROVIDED. (AUTHOR ABSTRACT MODIFIED--LKM)