Integrating police-recorded crime data and DNA data to study serial co-offending behaviour
| Author | Sabine De Moor,Christophe Vandeviver,Tom Vander Beken |
| Published date | 01 September 2018 |
| Date | 01 September 2018 |
| DOI | http://doi.org/10.1177/1477370817749499 |
| Subject Matter | Articles |
https://doi.org/10.1177/1477370817749499
European Journal of Criminology
2018, Vol. 15(5) 632 –651
© The Author(s) 2018
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1477370817749499
journals.sagepub.com/home/euc
Integrating police-recorded
crime data and DNA data
to study serial co-offending
behaviour
Sabine De Moor
Ghent University, Belgium
Christophe Vandeviver
Ghent University, Belgium
Research Foundation—Flanders (FWO), Belgium
Tom Vander Beken
Ghent University, Belgium
Abstract
When studying offending behaviour, researchers primarily rely on police-recorded crime data,
even though such data contain only detected crimes and known offenders. Using DNA data,
which also contain information on unknown offenders, enables researchers to link offenders
by identifying their presence at shared crime scenes. In this paper we combine police-recorded
crime data with DNA data to study serial co-offending behaviour. We focus on the changes the
networks of crimes obtained from police-recorded crime data undergo when integrated with
data from unknown offenders in the DNA database. We demonstrate that an integrated dataset
reveals more and larger networks of crimes with a larger spatiotemporal spread compared with
the police-recorded crime data only.
Keywords
DNA data, police-recorded crime data, spatiotemporal analysis, unknown offenders
Corresponding author:
Sabine De Moor, Department of Criminology, Criminal Law and Social Law, Ghent University,
Universiteitstraat 4, 9000 Ghent, Belgium.
Email: Sabine.DeMoor@UGent.be
749499EUC0010.1177/1477370817749499European Journal of CriminologyDe Moor etal.
research-article2018
Article
De Moor etal. 633
Introduction
Two types of data sources are primarily used for studying offending behaviour: police-
recorded crime data and self-report crime surveys or victimization surveys (Kirk, 2006;
Maxfield et al., 2000; Thornberry and Krohn, 2000). Both data sources have particular
advantages, but they share one major disadvantage: they include only information on
detected or (self-)reported crimes (De Moor et al., 2017). As a result, police-recorded
crime data may underestimate the true magnitude of crime. This is compounded by the
low clearance rates of detected crimes in most Western countries, where only about 20
percent are cleared (Lammers and Bernasco, 2013; Vandeviver et al., 2015).
Furthermore, even if an offender is known to the police or reports their offences in a
survey, they may not mention accomplices or additional offences (Alarid et al., 2009).
Kocsis and Irwin (1998: 199) state that as ‘some serial criminals have committed more
offences than those for which they are charged, it would be inappropriate to assume that
a person found guilty of only a single offence could not be a serial offender’. As a result,
undetected crimes by known offenders, and unknown offenders and their crimes, remain
out of range when studying offending behaviour. This affects the generalizability and
applicability of research results. Simply extrapolating the findings for known offenders
to unknown offenders raises serious methodological questions. For example, Johnson
(2013: 144) acknowledges the bias that the under-reporting and under-recording of crime
may cause when using police-recorded crime data, but concludes that ‘there probably is
no better available data to work with’.
It is possible that the effectiveness of crime prevention initiatives and law enforce-
ment deployment strategies could be compromised if they are based solely on analysing
known offenders’ offending behaviour. Arguably, prioritizing law enforcement strategies
that target known offenders is lowering detection and clearance rates and making
unknown offenders invisible. It is therefore important that the offending behaviour of
unknown offenders is also understood, in order to develop new and inclusive prevention
and law enforcement strategies.
DNA databases could be used to overcome the constraints of traditional crime data
sources since they include information on both known and unknown offenders. Two
types of DNA profiles are generally stored in a DNA database: unknown forensic profiles
extracted from biological stains found at crime scenes (for example, a drop of blood),
and reference profiles obtained from stains taken directly from known individuals (for
example, a buccal swab from a suspect or victim) (Home Office, 2014). In forensic case-
work, forensic profiles are compared with reference profiles in order to identify a suspect
within a specific crime case. If the forensic profiles are stored in a database, it is then
possible to compare profiles between different crime cases. DNA traces involving the
same unknown offender, found at different crime scenes at different time points (that is,
a serial offender), can be linked. The presence of other unknown co-offenders can also
be revealed through their attendance at shared crime scenes.
DNA databases have been the subject of many different studies on a range of issues,
from the legal and ethical concerns surrounding the use and expansion of DNA data-
bases, to the opportunities they provide for operational policing. Their use in crimino-
logical research to learn more about criminal behaviour and the broader criminal context
To continue reading
Request your trial