Co-offending and police notification: the differential reporting of young groups to the police

ABSTRACT

Despite a wealth of research on the decision to notify police following victimization, there has been relative inattention to the role of a basic feature of crime: the tendency for individuals to offend in groups. Following this, the current research examines the influence of co-offending on the decision to notify police using information from the National Crime Victimization Survey. Results indicate that the co-offending is significantly and positively related to the likelihood of notification. Findings further indicate that this relationship is moderated by co-offender characteristics, such that young co-offenders and, to a lesser extent, male co-offenders are more likely than their counterparts to be reported to police. These results have important implications for understanding the invocation of both formal social control and disparities in these processes, indicating that co-offending not only has a direct effect on notification, but that it also changes the effect of offender characteristics on this decision.

KEYWORDS:

• Victim reporting
• police notification
• co-offending
• NCVS
• victimization

Data availability statement

The data that support the findings of this study are openly available on the Inter-University Consortium for Political and Social Research data repository at https://doi.org/10.3886/ICPSR36834.v3.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We follow prior research in this area and use ‘co-offending’ and ‘group offending’ interchangeably, to refer to any instance in which two or more persons engaged in criminal activity together (see for example Lantz, 2021; McGloin & Piquero, 2009; Warr, 2002).

2 NCVS estimates may be biased due selection and nonresponse bias as well, but there is disagreement about the appropriateness of using weights to account for these biases in multivariate regression. Therefore, we follow Lohr and Liu (1994) and present results not adjusted using the incident weights provided by the NCVS.

3 Co-offender sex was coded as any male, rather than mostly male, given the theoretical focus on the effect of males as threatening. We used a similar coding scheme when coding race. We think this operationalization to be more effective than alternatives – like all male or all black – given that such coding schemes would essentially treat incidents involving, for example, three black suspects and one white suspect as part of the reference group (i.e. non-black). Such an operationalization would necessitate unwarranted assumptions about the function of race or sex that are neither informed by prior research nor consistent with our theoretical expectations.

4 Unfortunately, the NCVS does not collect individualized suspect-specific demographic information for incidents involving co-offenders (e.g. number of Black suspects), thus precluding the operationalization of these measures as proportional. That said, it is important to note that the majority of co-offending groups are homogenous by race, sex, and age (Lantz & Ruback, 2017).

5 Because the NCVS questionnaire asks respondents to indicate the age of offender(s) within a range (e.g. 21–29), we are unable to construct more precise measures of group age (e.g. mean age).

6 Series crimes are those incidents which are (a) the same or very similar to previous victimization incidents; (b) have occurred at least six times; and (c) in which the respondent is unable to recall individual dates and other details well enough to report incident details separately.

7 Supplementary analyses in which we switched the reference category for age revealed that incidents involving juvenile suspects are also significantly less likely to be reported to the police than those involving young adults.

8 As the reasons for reporting are not part of our primary analyses and only apply to a subset of our sample, they are not imputed.

9 Results similarly indicated that incidents involving male (O.R. = 1.55, p < .001) or Black (O.R. = 1.16, p < .001) offenders were also significantly more likely than their reference groups to be involved in co-offending incidents, net of controls.

10 While we recognize that scholars have employed different measures to assess severity, victim injury and weapon use are the two most frequently used indicators of this construct (see Lantz et al., 2022).