Disproportionate minority contact in Canada: police and visible minority youth.

• Author: Fitzgerald, Robin T.

Introduction

The problem of disproportionate minority contact (DMC) with the criminal justice system has been well documented in Canada. Research has consistently demonstrated that Aboriginal peoples are over-represented as offenders at various points in the criminal justice system compared to their share of the general population (Beattie 2006; Brzozowski, Taylor-Butts, and Johnson 2006; Fitzgerald and Carrington 2008; La Prairie 1992; Roberts and Melchers 2003). Evidence also suggests that the overrepresentation extends to other groups including Black Canadians (Wortley and Tanner 2003; 2005) and Arab and West-Asian Canadians, particularly in the aftermath of 9-11 (Khalema and Wannas-Jones 2003). There is partial agreement about the extent to which this overrepresentation occurs at the earliest stages of the criminal justice process--police stops, searches, and arrests. However, there is also notable disagreement on this subject (see, e.g., the debate in the Canadian Journal of Criminology and Criminal Justice: Gabor, 2004; Gold, 2003; Melchers, 2003; Wortley and Tanner, 2003, 2005), precipitated by allegations of racial profiling by the Toronto Police Service, made in the Toronto Star (Rankin, Quinn, Shepard, Simmie, and Duncanson, 2002a; 2002b; 2002c).

In fact, DMC has been documented in all Western countries where such measurements are taken (Tonry 1997a: ix). For example, patterns of over-representation among visible-minority and recent immigrant populations have been reported in European countries (Albrecht 1997; Goodman and Ruggiero 2008; House of Commons 2007; Junger-Tas 1997; Killias 1997; Martens 1997; Miller 2010; Tournier 1997), among indigenous populations in Australia (Australian Bureau of Statistics 2008; Walker and McDonald 1995; Weatherburn, Snowball, and Hunter 2008) and New Zealand (Fergusson, Horwood, and Swain-Campbell 2003; New Zealand Department of Corrections 2007), and among visible minority groups in the United States (for recent reviews, see Bishop 2005; Kempf-Leonard 2007; Lauritsen 2005; Piquero 2008).

What is much more difficult to establish and more hotly contested is the explanation for any observed DMC with the police. In the American literature on the subject, there are two dominant explanations for DMC. The differential offending hypothesis suggests that over-representation results from racial differences in the incidence, seriousness, and persistence of offending (Lauritsen 2005). In contrast, the differential treatment hypothesis suggests that over-representation results from intended and/or unintended inequities in juvenile justice practice (Bishop 2005: 24). Alternatively, other commentators have argued that the reality likely incorporates both of these explanations (e.g., Piquero 2008). Our aim in this article is to examine systematically these possible explanations for disproportionate minority contact with the police among youth aged 12-17 in Canada. In doing so, we extend previous work on DMC with the police in three ways. First, the study is national in scope and is, to our knowledge, the first national-level study of DMC. Second, it distinguishes among different sub-versions of the differential treatment hypothesis and then systematically assesses the separate and simultaneous effects on DMC of differential involvement and of differential treatment. Finally, the analysis relies on youth self-report measures of police contact and offending behaviour and parent-report measures of race (see also Wortley and Tanner 2003; 2005) which overcome previously cited problems in relying on police-reported data to identify the offender's race (Doob 1991; c.f. Wortley 1999).

Background

Explanations of disproportionate minority contact

Recent reviews of DMC research find that empirical studies typically work within a framework of hypotheses including differential involvement in crime, differential treatment, or some combination of the two (Bishop 2005; Engen, Steen, and Bridges 2002; Huizinga, Thornberry, Knight, Lovegrove, Loeber, Hill, and Farrington 2007; Piquero 2008). The differential involvement hypothesis asserts that higher rates of police contact among minority youth are brought about by their higher rates of involvement in crime and/or in the types of crime that have the highest priority for police; for example, serious violent crime. As Tonry (1997b: 11) puts it, "[I]n most countries [where the subject has been researched] ... the general conclusion, as in the United States, has been that most of the measured disparity ... appears to be attributable to offending differences." This position implies that the criminal justice system treats offenders equally regardless of their race or ethnicity; and as a consequence, controlling for self-reported offending behaviour should eliminate differences between minority and non-minority youth in their contact with the police (e.g., Bishop and Frazier 1992; Hindelang, Hirschi, and Weis 1981; Sampson and Lauritsen 1997; Terry 1967b). The extent to which minority youth are disproportionately involved in offending appears to be contingent on several factors. In her careful examination of the issue, Lauritsen (2005) found that race differentials in juvenile offending in the United States depended on the type of data used to measure offending; for example, police-generated arrest data, victim surveys, or self-reported offending data as well as type of offence and location of the offence. Nonetheless, according to Lauritsen the available data suggest "disproportionate black, and to a lesser extent, Latino involvement in violent crime, but fewer differences in other types of crime" (2005: 97).

The differential treatment hypothesis says that DMC results from decision-making processes that operate differently for members of different races/ethnicities at various stages of criminal justice processing. At the initial stage of police contact, the differential treatment hypothesis posits that members of certain minority groups are more likely to be selected by police for stops, searches, arrests and/or formal charges, regardless of their actual involvement in crime (Bishop 2005; Fagan and Davies 2000; Petrocelli, Piquero, and Smith 2003; Piliavin and Briar 1964). The potential for differential treatment is, in principle, quite large at the stage of police handling of juveniles because of the relatively large degree of discretionary decision making that can be exercised by the police despite legislative and policy constraints (Bishop 2005: 38; Werthman and Piliavin 1967).

A common theme underlying explanations of differential treatment is the criminal justice system's greater focus on social characteristics that may also be more highly associated with race/ethnicity (Bishop 2005; Engen et al. 2002: 196; Free 2001; Kakar 2006; Pope and Feyerherm 1981; Tracy 2005). Thus, the differential treatment hypothesis can be seen as having three sub-versions: (1) police, as organizations and as individuals, tend to pay more attention to certain types of people (e.g., young, male, unsupervised, low socio-economic status, etc.), and some visible minority groups tend to be disproportionately represented in some of these groups; (2) police tend to pay more attention to certain kinds of places (e.g., areas with higher socio-economic disadvantage, higher perceived social disorder, or crime hotspots) and members of some visible minority groups tend disproportionately to be found in these places; or (3) police are affected by race/ethnicity itself in their decision making. Sub-versions 1 and 2 do not concern police racial bias per se; rather, they concern disproportionate minority possession of personal, family, and neighbourhood risk factors for police contact; that is, systemic discrimination that may result from certain racial/ethnic minority groups being disproportionately represented in social locations that are at greater risk of police attention (Wilson 2009).

The differential involvement and differential treatment hypotheses cannot be assumed to be mutually exclusive, and in fact, may work in tandem to explain DMC (Bishop 2005; Piquero 2008). However, there are few systematic analyses of the simultaneous influence of these two phenomena on the chances of minority juvenile contact with the police in particular. A small number of empirical studies have looked specifically at whether the effects of juveniles' race on contact with the police remain significant after controlling for individual offending behaviour and risk factors such as age, gender, family socio-economic status and structure, parenting, and/or peer relationships (Farrington, Loeber, and Stouthamer-Loeber 2003; Fergusson et al. 2003; Huizinga et al. 2007; Wortley and Tanner 2005). The weight of evidence from these studies suggests that neither offending nor these various risk factors explain higher minority youth contact with the criminal justice system; the implication of this finding is differential treatment on the basis of race.

In many cases, the risk factors examined in these studies can be linked to both police attention and individual involvement in offending behaviour. For example, being young, unsupervised, or living in a socio-economically disadvantaged neighbourhood may increase the chances of delinquent behaviour among youth (differential involvement) as well as their chances of receiving police attention (differential treatment). As a result, accounting for these risk factors to explain disproportionate minority contact without specifying their possible role does not help to differentiate between the two hypotheses.

We propose that, when data are available on self-reported crime, then personal, family, and neighbourhood risk factors should be treated as risks for police attention rather than for actual crime. To the extent that these risk factors affect the propensity to commit crime, their influence on police contact should already be accounted for by including...