Twenty-five convicted burglars, drug abusers, and sex offenders who agreed to probation with polygraph testing in lieu of a prison sentence were selected as the experimental group. Polygraph tests were conducted on the subjects every 90 days for 2 years (July 1, 1983, to July 1, 1985). Subjects were informed that failure of two independent tests would result in a revocation hearing. The right against self-incrimination was waived as a condition of accepting probation. The control group consisted of 243 offenders convicted of the same crimes as the experimental group. They were monitored over the same period as the experimental group under the condition of probation supervision alone. A chi square test was used to determine any statistically significant difference in the reoffending of the two groups. The polygraph was apparently a valid means of measuring whether probationers were reoffending, as there was no external evidence that test findings were erroneous. The experimental group had a statistically significant lower rate of revocation than the control group, thus indicating that polygraph surveillance deterred recidivism. 1 table and 15 references.