The study found that young inmates who harbored a fear of other inmates while incarcerated had a higher rate of recidivism, contrary to one study hypothesis. Living under restrictions on their freedom while incarcerated had no significant effect on recidivism. Inmates whose incarceration deprived them of strong, close ties with friends and loved ones that existed prior to imprisonment were deterred from reoffending, but inmates who had no such close ties prior to incarceration were not deterred from reoffending by not having outside social contacts while in prison. These findings, particularly the finding that fear of other inmates did not deter reoffending, shows the difficulty of applying rational choice theory to a cross-section of individuals exposed to adverse circumstances, many of whom may not act rationally in accordance with reasonable assessments of how to avoid pain and gain rewards. The dataset came from the Hanover Prison Study, which involved a study of 2,405 German male juvenile offenders who were younger than 21 years old at the time of the offense for which they received a first-time incarceration. All of the men entered a juvenile custodial institution between 1998 and 2001. During the custodial period, at least one standardized interview was conducted for each inmate immediately after his incarceration. A second interview was conducted with 1,874 inmates to determine their attitudes in the course of interacting with other inmates after a period of incarceration. In the study's focus on reoffending after release from prison, data were obtained on the reoffending of 623 offenders released after August 2002. The analysis of reoffending focused on first offenses within 6 months after release, as recorded in the German National Register of Offenses. 1 table, 3 figures, 6 notes, and 41 references