The explanatory variable denoting whether the criminal entered a prisoner work programme seems to significantly decrease survival time by 6.3%. However, this coefficient is statistically insignificant, and thereby exerts no explanatory power on our model. Following the number of prior convictions …show more content…
On the other hand, the variable which looks at marital status at the time of incarceration finds that an individual who was married prior to being jailed has an increase on the survival time by 34.1%. This could be given by the effect of partners and family support. In addition, years of schooling seems to have a positive effect on the survival time by 2.3% for every year of education. This may indicate that prisoners who have a degree or a type of higher education may be more likely find a job or have had a previous job according to their skills. The age variable looks at the effect on being arrested again on our survival time, and it finds that for every month that our individual ages by, the survival time increases by 0.4%. This however is a very small effect and contradicts the variable regarding prior convictions given every time a prisoner goes into jail he should consequently get older but not necessarily reducing their probability of being arrested …show more content…
Results confirmed the relevance of gender, age, nationality, access to employment and prior convictions on recidivism. This author also found significant differences in hazards of recidivism by type of initial offense, while controlling for prison fixed effects. Although, this study found that neither marital status, nor education, nor homelessness, are associated with a significant change in the instantaneous probability of recidivism. Releases who declared they had a job at entry face a significantly lower hazard after release (12% lower), which is consistent with rational choice theory (trade-off between legal and illegal activities) (Monnery,