THE TWO PAROLE SYSTEMS STUDIED WERE THE U.S. PAROLE COMMISSION, WHICH USES JUSTICE MODEL-BASED GUIDELINES TO STRUCTURE ITS DECISIONMAKING, AND THE PENNSYLVANIA BOARD OF PROBATION AND PAROLE, WHICH OPERATES WITHOUT GUIDELINES AND REPRESENTS THE REHABILITATIVE MODEL. INTERVIEWS WERE CONDUCTED WITH 112 FEDERAL INMATES AND 91 PENNSYLVANIA INMATES SHORTLY BEFORE THEIR FIRST PAROLE HEARINGS. THREE PERCEPTUAL DIMENSIONS WERE MEASURED; (1) CERTAINTY; I.E., WHETHER THE INMATES FELT THEY COULD PREDICT THE OUTCOME OF THEIR HEARING; (2) CONTROL; I.E., INMATES' BELIEF THAT THEIR ACCOMPLISHMENTS IN PRISON WOULD INFLUENCE THE PAROLE DECISION; AND (3) APPROVAL; I.E., INMATES' PERCEPTIONS OF THE GENERAL FAIRNESS OF THE PAROLE PROCESS. THE TWO GROUPS DID NOT DIFFER ON THE PERCEIVED CERTAINTY DIMENSION; IN FACT, 76 PERCENT OF STATE SUBJECTS AND 72 PERCENT OF FEDERAL SUBJECTS DID ACCURATELY PREDICT THE OUTCOME OF THEIR HEARINGS. HOWEVER, STATE PAROLEES SCORED SIGNIFICANTLY HIGHER THAN FEDERAL PAROLEES ON THE CONTROL AND APPROVAL DIMENSIONS, AND THERE WAS A POSITIVE CORRELATION BETWEEN THESE TWO DIMENSIONS. MANY FEDERAL INMATES COMPLAINED ABOUT THE PAROLE GUIDELINES, INDICATING THAT WHAT THEY ACCOMPLISHED IN PRISON MADE NO DIFFERENCE. THE FINDINGS SUGGEST THAT, DESPITE THE USE OF CONCRETE, STANDARDIZED DECISIONMAKING CRITERIA, THE JUSTICE APPROACH DOES NOT AFFORD INMATES GREATER PRECEIVED CERTAINTY REGARDING THEIR RELEASE ON PAROLE. THE FINDINGS ALSO INDICATE THAT THE REHABILITATIVE MODEL ALLOWS INMATES GREATER PERCEIVED INFLUENCE OVER THE PAROLE PROCESS, WHICH IN TURN LEADS THEM TO EXPRESS LESS HOSTILITY ABOUT THE PAROLE SYSTEM. IMPLICATIONS OF THE FINDINGS ARE DISCUSSED. SUPPORTING DATA AND REFERENCES ARE INCLUDED.