Police Discretion with Young Offenders

V. Situational Factors Affecting Police Discretion

In this chapter, we assess the impact on police decision-making with young persons of factors specific to the individual incident and the apprehended youth. Circumstances of the incident which we examine include: the seriousness of the crime, as indicated by the type of offence, the presence or use of a weapon, and the harm done to a victim; victim-related circumstances, including the expressed preference of the victim for a particular course of action by police, the type of victim (person or business), and the relationship, if any, between the victim and the offender; accomplice-related aspects, including whether there were accomplices, whether any was an adult, and whether this was apparently a gang-related crime; whether the apprehended young person was intoxicated at the time of the incident; and the location and time of day of the incident. We examine the following characteristics of the apprehended youth: his or her prior record of criminal activity, age, gender, race, demeanour, any delinquent peer group or gang affiliation, home and school situations, and the involvement of the parents.

For each of these possible influences on police decision-making, we have tried to assess its impact in two ways. First, in our interviews with police officers, we asked all officers who were currently, or had recently been, involved in decision-making with apprehended youth, to what extent each factor had an impact on their decision whether to use informal action, refer to alternative measures, or lay a charge (or recommend the latter actions, if the decision was not theirs). At least one officer from each police service in the sample was asked these questions. In smaller police services and detachments, where we interviewed only one or two officers, the current assignments of the persons who answered these questions ranged from patrol to commanding officer, but all were currently, or had recently been, directly involved in decision-making with youth. In larger police services, where we interviewed between two and seven officers, these questions were not posed to senior management, since they had generally not been involved in this kind of decision-making for several years or more.

The answers for each factor were coded on a Likert scale, ranging from "major factor" to "not a factor". A major factor indicates that the officer takes this factor into consideration practically every time s/he decides whether to charge, use alternative measures, or deal with a young person informally. A factor indicates that this factor does play a role in decision-making, but does not carry as much weight as a major factor, and is not necessarily considered in every case. A minor (secondary) factor denotes an answer to the effect that the factor sometimes plays a role; however, its impact is case-specific. We also coded a factor as minor (secondary) when an officer said that it plays a role, but in a secondary manner, in conjunction with other factors; thus, it does not have a primary or independent impact on decision-making. Finally, the answer was coded as not a factor if the respondent clearly stated s/he never considered this factor in his or her decision-making. In some cases we have reported these categories combined into "not a factor/ minor factor" and "factor/major factor".

For each of these factors, we provide an overall assessment of its weight in the police decision concerning the disposition of the incident and offender, and any variations in the weight given to it by our respondents. We looked for variations across the regions of Canada, types of community, whether or not the police service's jurisdiction includes a First Nations reserve or a significant number of aboriginals living off-reserve, the level and types of youth crime in the community, the type of policing, whether the police or the Crown make the decision concerning charging, and officer characteristics such as the level of authority in the police organization, the location of service (patrol, GIS, youth squad, etc.), gender, years of service, specific training for youth crime, and previous youth squad experience.

Occasionally, respondents volunteered that a factor also affects their decision-making concerning the method of compelling appearance, and this is noted where appropriate. However, a detailed analysis of the decision-making around compelling appearance is in Chapter II, Section 7.

In contrast with Chapters III and IV, where the police service or detachment was the unit of analysis, our unit of analysis here is the individual police officer, and his or her views concerning the factors which affect the exercise of his or her discretion. [86]

The second method which we used to assess the impact of situational factors on police discretion was multivariate analysis of statistical data from the UCR2 Survey. The data which we analysed include 38,727 young persons apprehended in 2001 by 186 municipal police services and provincial police detachments which respond to the UCR2.[87] For each apprehended young person, the decision which we analysed (i.e. the dependent variable) was the police disposition: whether the young person was charged (or recommended to be charged in Crown screening jurisdictions) or processed otherwise (i.e. by informal action or referral to alternative measures, although these two actions are unfortunately not distinguished in the available data). The factors whose impact were analysed include: the type of offence (using grouped Criminal Code classifications), the number of prior apprehensions of the youth, the youth's age, sex, and race (aboriginal or not), whether the incident involved a lone offender or accomplices, any weapon present, any injury suffered by a victim, and any relationship between a victim and an apprehended person. Using multivariate analysis, the impact of each factor was assessed, while holding other related factors constant; also, the relative weight of the various factors was estimated.[88]

This statistical analysis is similar to that used in a previous study of police discretion with young offenders in Canada in 1992-1993 (Carrington, 1998a), but there are two innovations. One is the use of UCR2 data for 2001. Not only are these data more recent, but they include substantially more police services than were included in the UCR2 Survey during the period of the earlier study. However, the more important innovation is the inclusion of the young person's record of prior contacts with the police (apprehensions) as an independent variable. Although this information is not captured in UCR2 records, it was constructed by a record linkage project carried out by the Canadian Centre for Justice Statistics especially for this project. The method of construction of the prior record variables is described in the Methodological Appendix.

Each of these two sources of data - opinions of police officers provided in interviews, and statistical data from the UCR2 Survey - has its own strengths and weaknesses, which are discussed in the Methodological Appendix. One major drawback of the UCR2 data is that several factors which have been identified in the literature as having an impact on police discretion with young offenders are not captured in the UCR2 Survey. These are: the victim's expressed preference concerning the disposition of the incident, whether the incident was gang-related, the young person's demeanour, home and school situations, gang or delinquent peer group affiliations, and the level of involvement of the parents. Therefore, we must rely on officers' opinions concerning the impact of these factors. Furthermore, many of the variables which we used to look for variations in the impact of factors - such as the characteristics of the officer making the decision, and of the police service in which s/he works - are also not available in the UCR2 Survey; so, our assessments of impact based on the UCR2 data are not differentiated by these variables, as the results from the interview data are.


  • [86] Actually, the unit of analysis is the interview; where more than one officer participated in an interview, we coded the responses as through they represented the views of one individual.
  • [87] Details of the subset of police services are in the Methodological Appendix. For more information on the methodology of the UCR2 Survey, see e.g. Canadian Centre for Justice Statistics.
  • [88] Details of the statistical analysis are given in the Methodological Appendix.

Date modified: