Research Question
Can criminal recruiters be identified and ranked by crime harm levels in a London borough, and if so, how long is the predictive window of opportunity for targeting them with crime prevention efforts?
Data
This study deploys 5 years of Metropolitan Police Service crime data, relating to one of the 32 London boroughs in that time period. The data structure allowed identification of all suspects linked to the same crime report and all crime reports linked to the same suspects. Identification of linked suspects and their associated crime harm was undertaken using Structured Query Language (SQL) and ColdFusion Markup Language (CFML) via a web-based application.
Methods
All offenders were ranked by the number of co-offenders they acquired, as well as the total Cambridge crime harm index weight of the detected offences associated with them.
Findings
The highest harm recruiters are shown to be up to 137 times as harmful as the average offender, with one recruiter committing the same number of crimes as another but having 97 times more crime harm. Recruiter populations are highly dynamic, with few potential targets persisting from year to year over multiple years.
Conclusions
This study suggests that criminal recruiters are readily identifiable from police data, but police would only have a short window of opportunity to use deterrent or other preventive strategies with them once they are identified.