(No) Spillovers in reporting domestic abuse to police


Spillover effects in crime are typically studied as a result of offender behaviour. This study investigates whether spillover effects can occur in the reporting of domestic abuse by victims. Domestic abuse is a particularly interesting context because of its high prevalence but low reporting rate. Extending existing spatio-temporal Hawkes process specifications, we test for the presence of two spillover channels in all domestic abuse calls in a major English city. We find no evidence to support such effects in the reporting of domestic abuse.

Journal of the Royal Statistical Society: Series A (Statistics in Society)
Jan Povala
Jan Povala
Postoctoral Researcher

Postdoctoral researcher at the Alan Turing Institute.