VIOLENCE BEGETS VIOLENCE … BUT HOW? A DECISION‐MAKING PERSPECTIVE ON THE VICTIM–OFFENDER OVERLAP*

• Date: 01 May 2016
• DOI: http://doi.org/10.1111/1745-9125.12102
• Published date: 01 May 2016

VIOLENCE BEGETS VIOLENCE . . . BUT HOW?

A DECISION-MAKING PERSPECTIVE ON THE

VICTIM–OFFENDER OVERLAP∗

MARGIT AVERDIJK,1JEAN-LOUIS VAN GELDER,2

MANUEL EISNER,3and DENIS RIBEAUD1

1Chair of Sociology/Criminological Research Unit, Swiss Federal Institute

of Technology Zurich (ETH)

2Netherlands Institute for the Study of Crime and Law Enforcement (NSCR)

3Institute of Criminology/Violence Research Centre, University of Cambridge

KEYWORDS: victimization, violence, victim–offender overlap, decision making, longi-

tudinal study

This study applied a decision-making perspective to examine the causal mechanisms

underlying the relation between violent victimization and offending. We theorized that

having been victimized affects an individual’s appraisal of subsequent potentially con-

flictive situations in such a way that victims become more attuned toward the benefits

of violence perpetration than toward its costs. Furthermore, we argued that this altered

appraisal mediates the relation between violent victimization and violent offending.

We tested these hypotheses by using data from the Zurich Project on the Social De-

velopment of Children and Youths, a longitudinal study of Swiss youth (N =1,013;

11–15 years of age). In line with expectations, path analysis results showed that prior

victimization influenced the appraisal of decision-making situations that, in turn, pre-

dicted subsequent self-reported violent offending. Importantly, these mediation effects

held when controlling for a variety of time-stable factors, such as self-control and risky

activities, as well as prior victimization and delinquency. Implications for research and

theorizing on the victim–offender overlap are elaborated in the discussion.

Prior research has demonstrated a strong association between violent offending and

victimization. Victims of violence are likely to commit violent acts themselves, and con-

versely, offenders have a relatively high probability of being victimized (see Jennings,

∗Additional supporting information can be found in the listing for this article in the Wiley Online

Library at http://onlinelibrary.wiley.com/doi/10.1111/crim.2016.54.issue-2/issuetoc.

Margit Averdijk and Jean-Louis van Gelder are both to be regarded as co-first authors of this

article. The research reported in this article was financially supported by the Swiss National Sci-

ence Foundation, the Jacobs Foundation, the Swiss Federal Office of Public Health, the Canton of

Zurich Ministry of Education, and the Julius Baer Foundation. The authors would like to express

their sincere thanks to the youths, parents, and teachers participating in the study. Moreover, the

authors are grateful to the interviewers and undergraduate students for their help in data collection

and coding.

Direct correspondence to Margit Averdijk, Chair of Sociology/Criminological Research

Unit, ETH Zurich, Weinbergstrasse 109, 8092 Zurich, Switzerland (e-mail: margit.averdijk@

soz.gess.ethz.ch; maverdijk@hotmail.com).

C2016 American Society of Criminology doi: 10.1111/1745-9125.12102

CRIMINOLOGY Volume 54 Number 2 282–306 2016 282

VICTIM-OFFENDER OVERLAP 283

Piquero, and Reingle, 2012, for a review). Whereas the victim–offender overlap is a highly

robust empirical research finding, theoretical explanations for it have been less forth-

coming (Lauritsen and Laub, 2007). Several theorists have argued that the relation be-

tween victimization and offending is spurious because a common third factor, such as

low self-control (Gottfredson and Hirschi, 1990), underlies both. Others, however, have

maintained that the relation is causal (Lauritsen, Sampson, and Laub, 1991). Although

there is empirical evidence for both positions, important questions remain. Specifically, as

Lauritsen and Laub (2007) remarked, to move the debate forward, research must go

beyond the commonly examined factors, such as demographic characteristics, risky

lifestyles, deviant peers, subcultural norms, and neighborhood characteristics, and exam-

ine aspects of decision making. This was the goal of the present study.

We addressed the causal mechanisms underlying the victim–offender overlap by exam-

ining how victimization alters people’s appraisal of (subsequent) potentially conflictive

situations. Our approach drew from both choice theories of criminal decision making and

appraisal theories of emotion, and it was premised on the idea that a full understanding of

why violent offending follows victimization requires an appreciation of the actual choice

process and its antecedents, an assumption that has thus far received little empirical and

theoretical attention. We posited that prior victimization influences how the reward pa-

rameters, i.e., the perceived costs and benefits of acting violently, are evaluated in such

a way that people become more attuned toward the benefits than toward the costs. This

altered situational appraisal, in turn, influences the likelihood of subsequent offending.

We tested this mediation hypothesis by using an encompassing longitudinal data set

that allows for disentangling the temporal sequence of victimization and offending. In ad-

dition, we employed a method that allows for testing whether changes in the choice pro-

cess are actually a result of the victimization event itself instead of a result of preexisting

differences between victims and nonvictims, such as a latent proneness toward victimiza-

tion. Specifically, we controlled for an extensive set of potentially confounding variables,

including a latent disposition for involvement in crime (i.e., prior offending, prior vic-

timization, and prior decision making), as well as self-control, anxiety and depression,

parenting, and risky lifestyles, thus, ruling these factors out as alternative explanations.

DECISION-MAKING PERSPECTIVE ON THE

VICTIM–OFFENDER OVERLAP

Whereas most research on the victim–offender overlap has focused on the effect of

delinquency on subsequent victimization, the present study adds to a growing body of

research examining victimization as a cause of delinquency.1Several studies have shown

that prior victimization increases the likelihood that an individual will resort to violent be-

havior at a future point in time (Berg et al., 2012; Manasse and Ganem, 2009; Turanovic

and Pratt, 2013). Although this relation seems to be well established, it is unclear why this

is exactly the case as existing theoretical perspectives fall short of explaining the mecha-

nisms underlying these findings. As Turanovic and Pratt (2013: 322) observed, “departing

1. Note that we did not consider a decision-making perspective to explain the effect of offending on

later victimization because victimization does not imply a choice process. Analyses not reported

here confirmed our intuition that decision-making characteristics do not mediate the effect of of-

fending on victimization.