Whether a Criminal is Likely to Re-offend? A Statistical Analysis Using Crimes Data from Broward County Florida

Abstract

Security issues have always been a significant threat to the safety of citizens in every country, and many of these re-arrested criminals have negatively impacted social security. Therefore, predicting and studying the factors of a criminal's re-entry to prison will significantly help maintain social order and improve the civil society happiness index. This study, it will show what elements are predicted to influence a criminal's return to prison and what aspects will have a higher proportion and weight based on the collected data set. In the dataset, each re-admission inmate is categorized according to gender, age range, race, records, etc. Use the Logistics model and OLS model to build a model to predict what factors most directly lead to a criminal being arrested and imprisoned again. Data research has proved that the "number of priors" is the factor that most affects the recidivism rate of criminals.