In addition, sentencing guidelines and parole release guidelines should be implemented together, and both forms of guidelines should rest on a shared philosophical orientation in which sentencing and incarceration goals are explicit. Study data came from a sample of 454 armed robbers released from a Massachusetts State prison from 1977 to 1982. The study aimed to determine whether parole, by virtue of its control over time served, acted as an equalizer of disparate sentences. Legal and extralegal factors accounted for only 12 percent of the sentence variability but 62 percent of the variance in time served. Most sentencing variability therefore remained unexplained. Prison behavior and minimum sentence accounted for much of the variability in the length of incarceration. Different factors were related to each decision stage. A path model of the relationships accounted for 59 percent of the variance in time served. Although parole reduced sentencing disparity partly through the influence of unwarranted extralegal characteristics, these decisions were not necessarily more biased. Including prison behavior also accomplished sentence equalization. Data tables, figures, appendixes presenting additional data, and an 88-item bibliography are supplied.