From prison to the streets: Can importation work in reverse?

The purpose of this study was to determine whether prison misconduct is capable of predicting rearrest and whether criminal thinking plays a role in this relationship. Two hypotheses, a variable-centered hypothesis and a person-centered hypothesis, were tested in a group of 951 male federal inmates, with results showing that at the variable level, prison misconduct predicted arrests for more serious crimes (person offenses) but not for less serious crimes (nonperson offenses). In an effort to examine the prison misconduct-community recidivism relationship at the person level, latent class growth analysis (LCGA) was performed on offending indicators across 3 nonoverlapping time periods: before prison (prior convictions), in prison (annual rate of incident reports), and after prison (total subsequent arrests). The LCGA results revealed a large decelerating pattern and 3 smaller accelerating patterns. Criminal thinking successfully distinguished between the 2 patterns after controlling for age and criminal history. The present findings suggest that prison can have a deterrent effect (decelerating pattern) or a criminogenic effect (accelerating patterns) after a person's release from prison and that criminal thinking may put one at risk for entry into the accelerating pattern. (PsycINFO Database Record