CARLSON'S STUDY SHOWS THAT THERE IS NO DIFFERENCE IN THE RATE OF RECIDIVISM AMONG CONVICTS WHO ARE PAROLED COMPARED WITH CONVICTS WHO SERVE OUT THEIR ENTIRE SENTENCE WITHOUT PAROLE - 64.4 PER CENT RECIDIVISM AMONG THOSE WHO HAD BEEN RELEASED ON PAROLE AND 64.5 PER CENT RECIDIVISM AMONG THOSE WHO HAD SERVED OUT THEIR ENTIRE SENTENCE WITHOUT PAROLE. THE MATHEMATICAL ANALYSIS OF CARLSON'S STUDY INDICATES ONLY ONE WAY TO LOWER THE RECIDIVISM RATE BELOW 50 PER CENT - THROUGH EDUCATION. IT IS SUGGESTED THAT INCENTIVES SHOULD BE GIVEN FOR CONVICTS TO INCREASE THEIR EDUCATION PARTICULARLY DURING THE LENGTHY SENTENCES SERVED IN PENITENCIARIES. THESE INCENTIVES COULD BE IN EITHER ONE OR A COMBINATION OF BOTH OF THE FOLLOWING FORMS: 'TIME-OFF' A CONVICT'S SENTENCE IN ACCORDANCE WITH HIS EDUCATION; AND INCREASED REMUNERATION WHILE INCARCERATED FOR CONVICTS WITH HIGHER EDUCATION. IN A BRIEF REPLY TO THIS ARTICLE, KENNETH A. CARLSON DISAGREES WITH THE AUTHOR'S INTERPRETATION OF THE RELATIONSHIP BETWEEN RECIDIVISM AND EDUCATIONAL LEVEL AND SUGGESTS THAT MORE EVIDENCE IS NEEDED THAT INCREASING ONE'S EDUCATION CAUSES LESS RECIDIVISM BEFORE THE PROPOSAL CAN BE SERIOUSLY CONSIDERED....ELW