Skip to navigation – Site map

HomeSectionsSpace, Society,Territory2014Racially biased policing and neig...

2014
665

Racially biased policing and neighborhood characteristics: A Case Study in Toronto, Canada

Préjugés policiers liés au biais ethnico-résidentiel : une étude de cas à Toronto, Canada
Yunliang Meng

Abstracts

This study investigated race-and-place profiling in Toronto within a neighborhood context. It explored the spatial association between race-specific drug-related stops and neighborhood racial and socio-economic characteristics. The findings of this study suggest that Blacks are subject to disproportionately more stops for drug-related reasons in neighborhoods where more Whites reside and are less socio-economically disadvantaged, therefore confirming race-and-place profiling of Blacks in Toronto. However, race concentration and socio-economic disadvantage arguments fail to explain the spatial variations in drug-related stops of Whites. This result could be caused by the diverse ethnic origins and socio-economic backgrounds of White Torontonians. This article argues for the importance of a contextualized examination of racial profiling within the spatial context of neighborhoods and calls for democratic policing in Toronto. It also discusses the negative impacts of race-and-place profiling on Blacks in Toronto.

Top of page

Full text

This research was supported by a grant from the Ontario Ministry of Research and Innovation and Social Sciences & Humanities Research Council of Canada (SSHRC). The author would like to thank Dr. Uzo Anucha for her help and support during this research study.

Introduction

1In Canada, the police are empowered by common law to stop and search members of the public on the street. However, the exercise of this police power must be based on “a reasonable suspicion that the particular individual is implicated in the criminal activity under investigation” for it to be considered lawful (Mann, 2004, para. 34). Despite this legal limitation, police stop and search actions frequently sparked public debate and controversy. At the center of the controversy is the allegation of racial profiling, i.e. that the police often use race, skin colour, ethnicity, religion or a combination of these, rather than a reasonable suspicion, to single out an individual for greater scrutiny.

2For decades, Blacks in Toronto have complained that they are frequently stopped, questioned and searched by the police for “DWBBs – Driving While Being Black Violations” (Foster, 1996, p. 5). This controversy reached a climax in 2002, when Toronto Star (hereinafter referred to as the Star) published a series of news articles on race and crimes in Toronto. The articles revealed that Blacks in Toronto were highly over-represented in certain offence categories, including drug possession and “out-of-sight” traffic violations (e.g.: driving without a license or insurance). Based on these findings, the Star concluded that Toronto police were engaged in racial profiling or using their words stop people for little reason other than their skin colour” (Rankin et al., 2002, A8). In response to the articles appearing in the Star series, the Toronto Police Service commissioned an independent review to re-examine the Stars analysis results (see Gold and Harvey, 2003) and later denied all allegations of racial profiling. The Police Chief Julian Fantino declared, “we do not do racial profiling...There is no racism” (Toronto Star, 2002, p. A14). Meanwhile, a number of scholars participated in the ongoing discussion of racial profiling in Toronto (see Melchers, 2003; Wortley and Tanner, 2003). Although some researchers (i.e. Gold and Harvey, 2003; Melchers, 2003; Wortley and Tanner, 2004) pointed out data mistakes, methodological weaknesses, and inference errors associated with the analysis done by the Star, they have yet to provide concrete evidence to support the “we do not do racial profiling” argument. In 2012, the Star’s latest analysis showed that Black youth were more likely than White youth to be stopped and documented in each of the city’s 73 police patrol zones (Rankin and Winsa, 2012). The Toronto police defended the practice as good police work in high crime areas (Winsa and Rankin, 2012). Unlike his predecessor, Julian Fantino, Chief Blair has never directly defended racial profiling, calling it abhorrent (Winsa and Rankin, 2012). Ironically, he supported stopping citizens on the street, particularly minority youth, and recording the stoppages, stressing that it worked to reduce crime without giving any evidence to back up his claim (Winsa and Rankin, 2012). Given the tension and unchanged biased police attitude, more empirical studies are needed to examine the issue of racially biased policing in Toronto.

3Notwithstanding a few studies done by scholars, the media, and police, little research has focused on the spatial dimension of policing in Toronto. Police officers often develop and use an intricate knowledge of place in the field (Bittner, 1970; Brown, 1981; Sherman et al., 1989). Conceptions of public place and the people who occupy the space are linked to police assessment of perceived risk, which can influence police response and actions (Meehan and Ponder, 2002). As such, police-citizen interactions vary by neighbourhood context (Klinger, 1997) and police actions, such as racial profiling, arrests and searches, are “inextricably tied not only to race, but to officers’ conceptions of place, of what should typically occur in an area and who belongs, as well as where they belong” (Meehan and Ponder, 2002, p. 402). Some ecological studies of policing suggest that minorities are more likely to be stopped in areas where minorities look “out of place” (e.g. Meehan and Ponder, 2002; Stults et al., 2010). Others advance that disproportionate stops of minorities are likely to occur in crime “hot spot” areas where often coincide with disadvantaged neighbourhoods housing a high proportion of minorities (e.g. Roh and Robinson, 2009; Stults et al., 2010). Similar to the stoppage patterns of Blacks, police stops of Whites are also impacted by neighbourhood characteristics and tend to be spatially clustered (Gelman et al., 2007; Stults et al., 2010). Therefore, no matter what point of view the researchers hold, neighbourhood characteristics (i.e., racial and socio-economic dimensions) are important in differentiating police stops for different racial groups spatially. To fully understand racially biased policing in the context of place in Toronto, this study will examine whether disproportionate stops of Blacks and Whites tend to occur in areas where certain neighbourhood racial and socio-economic characteristics are also present. In this study, drug-related stops are chosen because they offer the more commonly suspected, if not also more likely, opportunities to observe racial bias in policing (Melchers, 2003).

4The paper is organized as follows. Section 2 provides a brief review of police knowledge of place and its association with racially biased policing, followed by the connection between police stops and neighbourhood characteristics. In addition, it offers a discussion of the key problem in testing for racial profiling in police stops: estimating the risk set, or “benchmark,” against which to compare the race distribution of stopped people. Section 3 presents a racially biased policing case study conducted in the city of Toronto, Canada, outlining a description of the study area, data source, variables used in this study, and analysis results in which the spatial relationship between the race-specific stops and neighbourhood racial and socio-economic characteristics is determined, followed by a discussion of the findings and conclusions in Sections 4 and 5.

Literature Review

Police Knowledge of Place and Racially Biased Policing

5Place at once refers to the buildings, streets, monuments and open spaces etc. located at a specific geographic location. Now, it is a word often used as the conceptual fusion of physical space and human activity that gives areas an identity (Abrahamson, 1996). When it comes to a patrol officer’s sector assignment, a place is not simply a geographic designation; but the territory over which the officer exerts jurisdictional claims about acceptable and unacceptable behaviour (Manning, 1997). To exercise the jurisdictional powers, police officers often develop and use an intricate knowledge of place such as area knowledge (Bittner, 1970), territorial knowledge (Brown, 1981), and knowledge of hot spots (Sherman, 1989). Police knowledge of place comes from not only the physical environment of the place, but the people who occupy the place (Sacks, 1972). For the police, race and socio-economic conditions are strongly tied to their knowledge of place. They know which neighbourhoods are Whiter and wealthier, Blacker and poorer, or some combination of the two, and where in their own neighbourhood, the racial and class composition differs (Brown, 1981). Such knowledge is a resource for constructing the meaning of place. However, the police’s knowledge of place is shared and impacted by the community, media, and their occupational and professional demands and experiences (Brown, 1981; Alpert and Dunham, 1988). As a result, their knowledge of place often contains bias and prejudice.

6Police knowledge of place impacts how police officers response towards physical and social surroundings. For example, when police officers practice stop-and-searches on streets, police decisions are often influenced by subjective perception or biased personal knowledge of suspects, crimes and patrol areas. Race together with place may be used as a proxy for threat or, at a minimum, for suspiciousness in police stops. Thus, police’s reactions to similar cases may significantly differ depending on neighbourhood characteristics and racial background of suspects. If this is the case, a racial group may experience excessive police stops in some neighbourhoods, not because of formal social control efforts at the agency level, but because of abuse of discretion at the individual officer level. Accordingly, disproportionate stops of a racial group in some neighbourhoods can be considered as empirical evidence that supports race-and-place profiling allegations. However, researchers have not reached consensus regarding the spatial distribution of race-and-place profiling phenomenon.

Police Stops and Neighbourhood Characteristics

7The police usually apply diverse crime control strategies in different neighbourhoods. Problem-oriented policing strategies and crime “hot spot” analyses are considered as appropriate methods to allocate limited police resources. More police patrols are likely to be assigned to neighbourhoods that have more reported crimes (Doerner, 1997) and police patrols have a significant positive relationship with police stops (Roh and Robinson, 2009). Thus, disproportionately more police stops practiced in crime hot spots can be perceived as police efforts to reduce crime rates in those areas, and police stops and crimes are expected to be spatially correlated. Meanwhile, different races tend to spatially cluster in different areas of a large metropolitan city such as Toronto (e.g. Fong, 2006; Hulchanski, 2010). In a highly ethnically diverse metropolitan city, crime “hot spots” are often found in impoverished neighbourhoods that may house a high proportion of minorities. Roh and Robinson (2009) argue that a significantly higher police presence and vigorous law enforcement are a common police practice in disadvantaged and racialized neighbourhoods. Therefore, some researchers have advanced that living in or often travelling to such neighbourhoods could lead to high likelihood of people’s contact with police, thus increasing the likelihood of minorities being stopped by police due to the overrepresentation of minority groups in the areas (Roh and Robinson, 2009; Stults et al., 2010). Other research has proposed an “out of place” theory which suggests that the likelihood of racial minorities being stopped is greater in areas where the racial minorities look “out of place” than in areas where their skin colour is predominant (Meehan and Ponder, 2002; Stults et al., 2010). This is because geographic concentration of disadvantage is very pronounced in minority neighbourhoods, so police officers tend to suspect that racial minorities might be engaged in criminal activities when they appear in White and wealthy neighbourhoods. As a result, disproportionately more stops of minorities are likely to happen in areas less racialized and disadvantaged. Similar to the stoppage patterns of Black people, police stops against Whites are also impacted by neighbourhood characteristics and tend to be spatially clustered. Ridgeway (2006) notices that the Whites are subject to disproportionate more stops in White and wealthy neighbourhoods. Gelmen et al. (2007) advance “racial incongruity theory” to explain high stop rates for Whites detected in Black dominant neighbourhoods. Stults et al. (2010) find that the presence of a large percentage of White residents in a neighbourhood decreases the stop rate of Whites and that the clusters of high stop rates of Whites tend to be overlapped with neighbourhoods dominated by either White or non-White residents. Certainly, as some studies suggest (e.g. Risse and Zeckhauser, 2004; Gelmen et al., 2007), if minorities are subject to more police surveillance because they are walking or driving in predominantly White neighbourhoods, Whites are subject to such scrutiny as well when they walk or drive in areas where minorities predominate. However, such an argument fails to acknowledge that although minority areas are often lively, historically interesting, and culturally rich, police do not generally confuse Whites’ interest in such areas with a predilection for crime (Lever, 2005). Therefore, disproportionate police stops of Whites in minority dominant areas may be used as preliminary rather than conclusive evidence to support accusations of race-and-profiling of Whites.

Benchmarking Methods

8From a methodological perspective, the key problem in testing for racial bias in police stops is estimating the risk set or benchmark, against which to compare the racial distribution of stopped people. To date, the two most common approaches have been to use census data (i.e. population or driver census) and conduct surveys (i.e. traffic survey) in places in which observers tally the race distribution of stopped people (Grogger and Ridgeway, 2006). However, the use of either method has two vital problems (see Melchers, 2003; Grogger and Ridgeway, 2006). First, the use of census and survey data can create two interrelated errors: base error and aggregation error. Both data are an appropriate base or denominator for statistics that measure prevalence. However, counts of police stops – the numerator in racial profiling studies – measure the number of stoppage incidents. Using counts of police stops divided by population numbers from census or survey data creates base error. A single individual can be stopped multiple times by police over any defined period, so a small group has a significant impact on how a larger group encompassing them is perceived. That leads to aggregation error. Two errors together can further skew interpretations of findings. Second, the unreasonable assumption of randomness of reported stops is associated with the use of census or survey data. Melchers (2003) argues that it wastes the limited resources of police if patrol deployment and stoppage of people or vehicles are random over space and time. Police patrols and stops are most effectively deployed when they focus on when and where problems are most expected. Other than the common weaknesses, census data do not take into account the racial composition of pass-by people or drivers. People stopped by police in a neighbourhood are not necessarily the residents of the neighbourhood. Survey data are very expensive to collect and people’s racial backgrounds are hard to be identified in a multi-ethnic and night environment.

9Data of race-specific arrestees is considered as an improved benchmark in the studies of racially biased policing (see Gelman et al., 2007; Ridgeway and Macdonald, 2010). The new benchmark has relatively less weaknesses. For example, the index is a ratio, so it would not introduce base error and aggregation error. It does not assume that reported stops are randomly conducted, since it is created under the logic that police stops are correlated with crimes. Therefore, data of race-specific arrestees will be used as the benchmark against which to compare the racial distribution of stopped people in this study.

10Given the possible relationships between race-specific stops and the neighbourhood characteristics, the current study hypothesizes that drug-related stop/arrest ratios for Blacks and Whites are spatially associated with neighbourhood racial and socio-economic characteristics in Toronto. Specifically,

  • H1a: Will the drug-related stop/arrest ratios for Blacks increase, as the level of concentrated disadvantage and racialization increases?

  • H1b: Will the drug-related stop/arrest ratios for Blacks increase, as the level of concentrated disadvantage and racialization decreases?

  • H2a: Will the drug-related stop/arrest ratios for Whites increase, as the level of concentrated disadvantage and racialization increases?

  • H2b: Will the drug-related stop/arrest ratios for Whites increase, as the level of concentrated disadvantage and racialization decreases?

Case Study

Study Area

11The city of Toronto lies on the northwest shore of Lake Ontario, the easternmost of the Great Lakes. Downtown Toronto is the primary central business district in Toronto, Ontario, Canada (See Figure 1). Located entirely within the former municipality of old Toronto, the area is made up of the city’s largest concentration of skyscrapers and businesses. The busiest highway in Canada – Highway 401 passes through North of Toronto. In the south of Toronto, Gardiner Expressway moves across the downtown area and is connected with Highway 401 by Don Valley Parkway (See Figure 1). Home to more than 2.6 million people, it is the most populous and diverse city in Canada and it is the center of one of Canadian most dynamics regions – the Greater Toronto Area (GTA).

12According to the 2006 population census, 47% of Toronto’s population reported themselves as being part of a visible minority group. The City of Toronto’s visible minority population has increased by 11% since 2001, and by 32% since 1996. Given the racial diversity (see the descriptive statistics for visible minority in Table 1), Toronto is a particular important geographic location in which to conduct a study of race-and-place profiling.

Figure 1. Study Area: The City of Toronto

Figure 1. Study Area: The City of Toronto

Table 1. Visible Minorities as % of Population in Toronto (City of Toronto, 2013)

Visible Minority

2006 Census

South Asian

12.0 %

Chinese

11.4 %

Black

8.4 %

Filipino

4.1 %

Arab/West Asian

2.6 %

Latin American

2.6 %

Southeast Asian

1.5 %

multiple

1.3 %

Korean

1.4 %

Not included elsewhere

1.0 %

Japanese

0.5 %

Total

46.9 %

13Toronto is a multicultural city and has a number of ethnic neighbourhoods that flavour the city’s character. For example, neighbourhoods with a large cluster of Chinese (i.e. Steeles L’Amoreaux) are located in the Northeast of Toronto and Chinatown in the downtown area. South Asians are clustered in neighbourhoods such as Malven and Jamestown located in the Northeast and Northwest corners of Toronto. Concentrations of Blacks are mostly at the scale of an apartment building (Qadeer and Kumar 2006). The Jane-finch neighbourhood has a relatively large proportion of Black residents (20.2%).

Data and Variables

14Police stop data were gathered from Field Information Report or well known as 208 cards, filled out by Toronto police in mostly non-criminal encounters with the public. The data include details on contact ID, Person ID, age, gender, place of stoppage, contact time, birth place, skin colour (i.e.: Black, White, Brown, and Other), and stop reason (i.e. drug, general investigation, loitering, and traffic etc.) for each stop incident. The 2008 data capture details from 7,062 drug-related 208 cards filled out by 6,595 individuals. In other words, a single person can be stopped a number of times during the year of 2008 and recorded in multiple 208 cards. It should be noted that the skin colour data were reported by individual police officers based on their own judgment, so it is possible that for example, some south Asians or some Latin Americans were recorded as Blacks in 208 cards. Arrest data were from Toronto police arrest and charge data collected in the field between 2004 and 2008. This dataset contains information regarding age, gender, charge category, arrest time, skin colour, case ID, and place of arrest for each arrestee. During 2004 and 2008, 54,870 arrests were made against 32,816 individuals for drug-related reasons in Toronto. In other words, a single person can also be arrested multiple times during 2004 and 2008 and be recorded multiple times in the police arrest and charge database. Police stop and search data and data of race-specific arrestees were obtained at police patrol zone level. There are 73 police patrol zones in Toronto (see Figure 1).

15The city of Toronto’s racial and socio-economic data were gathered from 2006 census data at dissemination area (DA) level – the smallest units used by Statistics Canada to aggregate census data. Racial groups considered in this study are Blacks and Whites. Since ‘race’ is not a category in the Canadian census of population, this study used the ‘visible minority’ variable (see Table 1) to identify Black population in each DA in Toronto. White population in each DA was quantified using the non-visible minority population minus the aboriginal population. The level of racialization in each DA was represented by the percentage of Whites in the residential population (see Table 2). The indices used to describe the level of disadvantage in each DA included the percentage of single-parent families, the unemployment rate, and government transfer payments as a percentage of total income (see Table 2). Racial and socio-economic data at DA level were carefully aggregated to the level of police patrol zones for neighbourhood level analysis using “Polygon in Polygon Analysis” in Hawth’s analysis tools designed for ArcGIS 9.3 (Beyer, 2004). Although a police patrol zone is not explicitly considered as a neighbourhood, police patrol zone boundaries have been widely used in extant research as a unit of analysis that corresponds to actual neighbourhood boundaries (Roh and Robinson, 2009). Hence, Toronto police patrol zone boundaries were used as actual neighbourhood boundaries in this study. To calculate drug-related stop/arrest ratios for Black or White people at neighbourhoold level, the number of drug-related stops of Blacks or Whites in 2008 was divided by the average number of drug-related arrests against Blacks or Whites between 2004 and 2008 in each police patrol zone. The average number of drug-related arrests between 2004 and 2008 was used as denominator to avoid zero divisor error, since there were no drug-related arrests made in some neighbourhoods during a specific year. A high race-specific stop/arrest ratio indicates excessive police stops or simply broadened suspicion of individuals based on race. A low race-specific stop/arrest ratio suggests under-stops of a racial group or a reluctance to stop more often persons of one race.

Table 2. Neighbourhood Racial and Socio-economic Characteristics

Neighbourhood Racial and Socio-economic Characteristics

Unit

Level of Racialization

Whites in the residential population

%

Level of Disadvantage

Single-parent families

%

Unemployment rate

%

Government Transfer Payments

%

Results

Descriptive Statistics for Police Stops at City Level

16Descriptive statistics regarding population, drug-related stops, and drug-related arrests of Blacks and Whites in Toronto are shown in Figure 2. At the city level, although only 8.4% of Toronto’s population was Blacks, they made up 26.0 % of the people who were stopped because the police suspected them of committing drug-related crimes. In addition, 24.0% of the drug-related arrests were made on Blacks. When it comes to Whites, they made up 53.1% of Toronto’s population, but they only accounted for 58.9% of the drug-related stops and 56.7% of the drug-related arrests. The drug-related stop/arrest ratio is 1.04 for Whites and 1.08 for Blacks at city level. By comparing the two ratios, it is clear that Blacks were slightly overrepresented in the drug-related stops conducted by Toronto police.

Figure 2. The Percentages of Blacks and Whites in the population of Toronto and the Percentages of Drug-related Stops and Arrests for Blacks and Whites

Figure 2. The Percentages of Blacks and Whites in the population of Toronto and the Percentages of Drug-related Stops and Arrests for Blacks and Whites

Spatial Association

17Neighbourhoods are not closed and isolated areas, but exhibit interdependency where events that occur in one neighbourhood affect events that occur in spatially proximate neighbourhoods (Stults et al., 2010). Previous crime studies in Toronto have provided empirical evidence for this claim, particularly revealing that various crimes, such as harassment, theft, break and entering, sexual assault, minor and major assault, robbery, shoplifting, and drug offences, are not randomly distributed across space (see Charron, 2009). However, the spatial proximity of police stops in Toronto has yet to be examined. As criminal behaviours, police stops in Toronto are not expected to be randomly distributed. Rather, as mentioned above, this study hypothesizes a spatial clustering of disproportionate police stops for Blacks and Whites in areas where certain neighbourhood racial and socio-economic characteristics are also present.

18To test the two hypotheses in this study, Bivariate global Moran’s I (Anselin et al., 2002) was used to gain information on the extent to which values for race-specific stop/arrest ratios observed at a given location showing a systematic association with neighbourhood racial and socio-economic characteristics observed at the “neighbouring” locations across the study area. A Bivariate global Moran’s I of zero indicates no spatial correlation between the two variables. A positive Bivariate global Moran’s I value suggests a spatially similar cluster in the two variables, while a negative value would imply a spatially dissimilar cluster in the two variables. The results of Bivariate global Moran’s statistics are listed in Table 3 as follows.

Table 3. Bivariate Global Moran’s I Values for Race-specific Drug-related Stop/arrest Ratios and Neighbourhood Racial and Socio-economic Characteristics

The stop/arrest ratios for Whites

The stops/arrest ratios for Blacks

White population (%)

0.044

0.365*

Single-parent families (%)

-0.030

-0.328*

Unemployment rate (%)

-0.006

-0.378**

Government Transfer Payments (%)

-0.005

-0.359*

** Spatial autocorrelation is significant at p < 0.01
* Spatial autocorrelation is significa
nt at p < 0.05

19The Bivariate global Moran’s I statistics show an overall positive spatial correlation between the stop/arrest ratios for Blacks and the percentages of Whites in the population and an overall negative spatial correlation between the ratios and the indices quantifying the level of disadvantage (see Table 3). The results suggest that disproportionately more drug-related stops against Blacks are likely to happen in neighbourhoods less racialized and disadvantaged. Bivariate global Moran’s I values close to zero are obtained between the stop/arrest ratios for Whites and the percentages of Whites in the population, and between the ratios and indices quantifying the level of disadvantage (see Table 2). The results indicate that no significant spatial correlation was found between the two sets of variables globally. However, the magnitude of spatial associations is not necessarily uniformed over the study area.

20In addition to testing the spatial association globally for the variables, LISA (Local Indicator of Spatial Association) cluster maps (Anselin, 2003) were used to illustrate the geographic clustering of the stop/arrest ratios and neighbourhood racial and socio-economic characteristics. For example, the similarity or dissimilarity of the percentages of Whites in the residential population near locales is determined by the LISA statistics at p < 0.05 (Anselin, 2003) and displayed by the LISA cluster maps, which show five different types of spatial association (see Figure 3a for example): (1) high-high, for neighbourhoods with high percentages of White residents that are also in close proximity to neighbourhoods with high percentages of White residents; (2) low-low, for areas with low percentages of Whites in the residential population that are also in close proximity to areas with low percentages; (3) high-low, for neighbourhoods with high percentages of Whites in the population but are proximate to neighbourhoods with low percentages; (4) low-high, for neighbourhoods that have low percentages of Whites in the population, yet are in close with proximity to neighbourhoods with high percentages; (5) not significant, for areas where there is no significant spatial clustering. In this study, neighbouring patrol zones are determined by queen contiguity, whereby any patrol zones sharing either boundaries or vertices are regarded as neighbours. Then, generated is the same spatial typology for each of the three disadvantage indices and the race-specific stop/arrest ratios.

21To demonstrate the degree of overlap between the spatial clustering of race-specific stop/arrest ratios and the levels of racialization, and between the spatial clustering of race-specific stop/arrest ratios and the levels of disadvantage, the LISA clusters of the race-specific stop/arrest ratios are overlaid with the LISA clusters of the levels of racialization and disadvantage at neighbourhood level (see Figure 3a and 3b, Figure 4a and 4b, Figure 5a and 5b, Figure 6a and 6b). The Figure 3a, 4a, 5a, and 6a show that the clusters of high stop/arrest ratios for Blacks tend to be in areas where there is spatial clustering of high percentages of Whites in the population and low levels of socio-economic disadvantage. Although police officers stop individuals for different reasons, not only to arrest, but also, for example, to obtain information for other crimes, the findings here do reveal a significant amount of police and Black citizen interactions in White and wealthy neighbourhoods in Toronto. Also, this finding suggests an enormous amount of police discretion in terms of stopping Black suspects in the White dominated and advantaged neighbourhoods in Toronto. Conversely, the clusters of low stop/arrest ratios for Blacks tend to occur in areas where there are clusters of low percentages of Whites in the population and high levels of socio-economic disadvantage (see Figure 3a, 4a, 5a, and 6a). In other words, Blacks tend to be stopped less often in racialized and disadvantaged neigbhourhoods. This finding suggests a relatively less amount of police discretion in terms of stopping Black suspects in their own racialized and disadvantaged neighbourhoods in Toronto.

22However, different spatial patterns are found for the spatial clustering of stop/arrest ratios for Whites and the levels of racialization; and the spatial clustering of stop/arrest ratios for Whites and the levels of disadvantage. The clusters of high stop/arrest ratios for Whites are overlapped with areas where there is spatial clustering of either high or low percentages of Whites in the population (see Figure 3b) and with areas where there is spatial clustering of either high or low levels of disadvantage (see Figure 4b, 5b, and 6b). In other words, Whites are subject to disproportionately more drug-related stops in either White or non-White neigbhourhoods and they are more likely to be stopped in either advantaged or disadvantaged neigbhourhoods. This finding indicates that the amount of police and White citizen interactions and police discretion in terms of stopping White suspects can not be explained only by neighbourhood racial and socio-economic characteristics. Police stops for Whites in Toronto may be related to other social and spatial factors that are, as yet, not fully understood.

Figure 3. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Percentages of White Residents in the Neighbourhoods

Figure 3. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Percentages of White Residents in the Neighbourhoods

Figure 4. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Percentages of Single Parent Families at Neighbourhood Level

Figure 4. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Percentages of Single Parent Families at Neighbourhood Level

Figure 5. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Unemployment Rates at Neighbourhood Level

Figure 5. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Unemployment Rates at Neighbourhood Level

Figure 6. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and Government Transfer Payments as a Percentage of Total Income at Neighbourhood Level

Figure 6. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and Government Transfer Payments as a Percentage of Total Income at Neighbourhood Level

Discussion

Drug-related Stop/arrest Patterns for Blacks

23The global spatial analysis shows that neighbourhood racial and socio-economic characteristics are associated with drug-related stop/arrest ratios for Blacks. In addition, the LISA cluster maps suggest that areas with high percentages of White residents and better socio-economic conditions are overlapped with areas with the high stop/arrest ratios for Blacks. These findings corroborate the view that Toronto police are more likely to stop Black youth who looked “out of place” and provide empirical evidence to support the race-and-place profiling of Blacks in Toronto.

24Race-and-place profiling of Blacks in Toronto could produce hidden distortions in crime statistics, since this disproportionate number of stops may lead to more arrests in White-dominated and less disadvantaged neigbhourhoods. Guided by problem-oriented policing strategies, arrest statistics would dictate that relatively greater number of Blacks will be stopped in those neighbourhoods in the future. This will become a vicious cycle that even the strictest law enforcement advocates would admit is patently unfair. On the other hand, the race-and-place profiling phenomenon undermines the effectiveness of policing, because it increases the chance of missing real criminals hiding in other racial groups. For example, to disproportionately target Blacks in a neighbourhood, the police must allocate disproportionately more resources to targeting the innocent Blacks. As a result, there are disproportionately fewer resources assigned to search for other real offenders in the neighbourhoods since police only have limited resources at any given time. The race-and-place profiling of Blacks is also likely leading to the stigmatization of Blacks, undermining of their trust in the justice system, heightening of anxiety about their safety and freedom, and curtailing of their sense of belonging. While race-and-place profiling reflects police racist attitudes and contributes to public harm, it is given an “official seal” of approval, since race-and-place profiling is implemented by police authority in Toronto. Even in the event that statistically, Blacks are more inclined to commit certain crimes or at least to be arrested and convicted, there is no empirical evidence to support that race-and-place profiling of Blacks do, in fact, reduce Blacks crime rates in Toronto.

25In Toronto, Whites occupy quality neighbourhoods in disproportionate numbers (Darden and Kamel, 2002). Race-and-place profiling detected in this study suggests that there are large and desirable neighbourhoods in Toronto where Blacks are “out of place,” do not “belong,” and should not walk or that Blacks can only enjoy the benefits of such areas at considerable risk to their pride, security, convenience and anonymity. Consequently, race-and-place profiling in Toronto can reinforce residential segregation and racial antagonism which could result in a significant and negative impact on socio-economic integration of minority groups such as Blacks in this study. If Black crime hot spots have been identified in White and advantaged neigbhourhoods, Toronto police should target the specific geographic locations (e.g., addresses, blocks) rather than the neighbourhoods as a whole, to alleviate even further the negative impressions of singling out Blacks in White dominant advantaged neighbourhoods for stop and searches. Indeed, most crime hotspot literature also focuses on the clustering of crimes at the place level rather than at the neighbourhood level. However, such a policing strategy is not still free from the issue of adverse impact on a few selected neighbourhoods because innocent residents living in and around crime hot spots, as well as the targeted minority groups (i.e. Blacks) in and around hot spots, can still be unduly influenced by disparate law enforcement.

26The significant spatial correlations in Table 3 and the LISA maps suggest that race-and-place profiling of Blacks was a department-wide phenomenon rather than behaviour of few police officers. In other words, Toronto police had done a reasonable job of ensuring that those recruited to the police do not display an overt racial bias, but unintentional and intentional forms of prejudice and discrimination still exist in police stops and are responsive to place. The police sensitivity training failed because a focus on individual police attitudes and behaviour misses the underlying societal and occupational structural problems that produce race-and-place profiling. Because race-and-place profiling is a product of pervasive and institutionalized patterns of racial segregation in the society, it is not helpful to treat race-and-place profiling as an effect of individual officers, consciously or unconsciously taking advantages of their occupational positions to act on their individual prejudices (Meehan and Ponder, 2002). In other words, if race-and-place profiling reflects society wide patterns of a generalized attitude about who belongs where, focusing on police attitudes or cultural sensitivity will not resolve the problem. In addition, police officers are often socially conservative in personal attitude and lifestyle and express prejudices against minorities, but it lacks of empirical evidence to prove that their personal attitude and lifestyle can translate to discrimination on duty (Black and Reiss, 1967). Therefore, discussion of race-and-place profiling must move beyond a consideration of the intent or motivation of police officers in individual cases which dominates legal thinking, to an examination of its embeddeness as practice within the organizational context of the police and their relationship to larger societal context from which discrimination, whether intentional or unintentional, emanates (Feagin and Eckberg, 1980).

27Given the race-and-place profiling findings, the current study calls for democratic policing in Toronto which is often understood as the antonym of inequality, seeking an equitable distribution of police service or police control over the public (Roh and Robinson, 2009). Central to the concept of democratic policing, especially in a participatory or a deliberative democracy, is community control of the police through community empowerment and community participation (Sklansky, 2008). According to this aspect of democracy, a community should not be limited as a consumer of police service but should be a co-producer of police services, playing a key role in the production process. The pubic must participate in the policing decision-making process and determine the amount and the types of police services and police protection. In this sense, disproportionate stops of Blacks, no matter how “effective such stops seem for battling crime in Toronto, would be perceived as undemocratic, so the community members could veto the practice. In other words, the Toronto police must take into consideration the community’s needs and demands when developing a policing strategy. The police also should strive for community support before implementing the strategy, especially if there is a chance that the strategy may arouse resentment among the community members that they (or their communities) are being treated in an inequitable way.

Drug-related Stop/arrest Patterns for Whites

28This research finds that the global spatial association between neighbourhood racial and socio-economic characteristics and drug-related stop/arrest ratios for Whites is insignificant in Toronto. The LISA cluster maps show that areas with high drug-related stop/arrest ratios for White people are overlapped with areas where the percentage of White residents and the level of disadvantage are either high or low. The findings suggest that race concentration and socio-economic disadvantage arguments fail to explain the spatial variations in drug-related stops for Whites. This may be due to the complicated ethnic composition of White Torontonians. Toronto has very large Italian and Portuguese communities, about 8% and 4% respectively of the total city population in 2006. In addition, although Latin Americans are not considered as “White” in the Canadian Census, they are likely to be coded as “White” in 208 cards. Latin Americans constituted 2.6% of the total Toronto population in 2006. The socio-economic conditions of these three groups are relatively worse compared with those of Whites with British Isles origins and may also be subject to race-and-place profiling. Therefore, the spatial patterns of police stops against Whites may be skewed in Toronto, due to the inclusion of the three groups under the homogenizing label: White. It should be noted that the origins of Blacks in Toronto are also diverse. 29.8% of Blacks reported multiple ethnic origins. The top ancestral origins among Blacks were Caribbean and African such as Jamaican (22.8%), Haitian (13.9%), Somali (4.4%) and Trinidadian/Tobagonian (3.7%) (Statistic Canada, 2006). There were also Blacks who reported British Isles (10.9%) and Canadian (10.8%) origins (Statistic Canada, 2006). Although the origins of Blacks are not homogenous, race concentration and socio-economic disadvantage arguments help explain the spatial variations in drug-related stops for Blacks as a whole in Toronto. Based on the conflicting results obtained in this study, raised is the question of why race concentration and socio-economic disadvantage theories successfully explained the spatial variations in drug-related stops for Blacks as a whole, but not for Whites in Toronto.

Limitations

29It is acknowledged that this study is not without limitations. First, this study can not overcome the Modifiable Areal Unit Problem (MAUP), since the visual representations of the hot spots on the LISA cluster maps are affected by the size, shape, and orientation of the police patrol zones. Second, skin colour data recorded in 208 cards are not exactly equal to visible minority or ethnicity data in Canadian census. For example, Toronto police may have difficulties to distinguish dark skin South Asians from Blacks and differentiate Latin Americans from Whites by their appearance. Thus, chances are that some people who were coded as Blacks or Whites in 208 cards might be counted as belonging to other racial groups in the Canadian Census. If this is the case, the number of Blacks or Whites in the drug-related stop data is not accurate. Third, the arrestee benchmark may be a more plausible approach, but it also contains weaknesses. For example, arrests can be made some distance away from where crimes actually occurred. Racial bias could also exist in arrests made by police officers, so using the arrestee benchmark could cancel racial bias in police stops to some extent.

Conclusions

30In general, the issue of racially biased policing has not been examined as deeply in Canada as it has been in the United States. This research contributes methodologically and empirically to the study of racially biased policing in the context of neighbourhoods in Toronto. Previous racial profiling studies done in Toronto were carried out at city wide level ignoring neighbourhood-level processes (see Gold and Harvey, 2003; Melchers, 2003; Wortley and Tanner, 2003; 2004). This study demonstrates how neighbourhood racial and socio-economic characteristics can contribute to spatial variations of police stops involving Blacks and Whites. The analysis of this research accounts for the effects of spatial proximity of neighbourhoods to nearby neighbourhoods that exhibit similar patterns of races, socio-economic conditions, and police stops. The findings of this study provide empirical evidence to support that Toronto police are engaged in race-and-place profiling when they practice drug-related stops in White dominated and less disadvantaged neighbourhoods. However, spatial variations in drug-related stops of Whites can not be explained by race concentration and socio-economic disadvantage at neighbourhood level. The inconsistency could be caused by the diverse ethnic origins and socio-economic backgrounds of Whites in Toronto.

31In terms of social policy, this research is significant in that it emphasizes the importance of neighbourhood in the examination of racial profiling and calls for democratic policing in Toronto. For real change to occur, it is vital to identify the characteristics of the areas that place minority individuals at the high risk of being racially profiled and pass this information to the police and neighbourhood representatives so that the practice can be eventually abandoned in the future (Parker et al., 2004). It is also essential for police organization first to consider each neighbourhood’s needs and demands when developing a policing strategy and then to strive for neighbourhood support before implementing it.

32This research recognizes the difficulties and challenges for Toronto police to establish a balance between promoting public safety and protecting individual rights. While Toronto Police Services struggle to maintain the delicate balance between preserving individual human rights and maintaining public safety, the race-and-place profiling debate will continue regarding who are targeted, where are targeted and how to prevent police from race-and-place profiling. Meanwhile, researchers need to conduct more theoretically guided and methodologically sound empirical studies that generate reliable findings to better inform policy in the field. First, this study has verified the race-and-place profiling of Blacks, but more research is needed to verify whether other minority groups are subject to race-and-place profiling too in Toronto. Second, transportation modes (i.e., walking, driving, or biking) of the stopped people can not be differentiated in this study because such information is not available in 208 cards. If it is available, it would be interesting to compare stop/arrest ratios for people from the same race but using different modes of transportation, since racial bias is more likely to be introduced in the police interactions with pedestrians and bikers whose skin colour is known before they are stopped. Third, given that the Blacks and Whites themselves are not homogeneous, further research is needed to examine police stops for specific Black or White groups under the homogenizing label: “Black” and “White”. Forth, more research is needed to explore why race concentration and socio-economic disadvantage theories successfully explained the spatial variations in drug-related stops for Blacks as a whole, but not for Whites in Toronto.

Top of page

Bibliography

Abrahamson M., 1996, Urban Enclaves : Identity and Place in America, New York, St. Martins Press.

Alpert G. and R. Dunham, 1988, “Policing Multi-ethnic Neighborhoods: The Miami Study and Findings for Law Enforcement in the United States”, New York, Greenwood.

Anselin L., I. Syabri and O. Smirnov, 2002, “Visualizing Multivariate Spatial Correlation with Dynamically Linked Windows”. In Anselin L. and Rey S. (eds.), New Tools for Spatial Data Analysis: Proceedings of the Specialist Meeting, Santa Barbara, University of California.

Anselin L., 2003, “Spatial Externalities, Spatial Multipliers, and Spatial Econometrics”, International Regional Science Review, Vol. 26, No. 2, 153-166.

Beyer, H. L. 2004. Hawth’s Analysis Tools for ArcGIS. Retrieved from http://www.spatialecology.com/htools (last visited July 5, 2013).

Bittner E., 1970, The Functions of Police in Modern Society, Rockville, National Institute of Mental Health.

Black D. and A. J. Reiss, 1967, “Patterns of Behavior in Police and Citizen Transactions”, in Black D. and Reiss A. J. (eds.), Studies of Crime and Law Enforcement in Major Metropolitan Areas: Vol 2. Report to the President’s Commission on Law Enforcement and the Administration of Justice, Washington, DC, Government Printing Office.

Brown M. K., 1981, Working the Street: Police Discretion and the Dilemmas of Reform. New York, Russell Sage Foundation.

Charron M., 2009, “Neighbourhood Characteristics and the Distribution of Police-reported Crime in the City of Toronto”, Crime and Justice Research Paper Series, Catalogue no. 85-561-MIE, Ottawa, Statistics Canada.

City of Toronto, 2013, Toronto’s Racial Diversity, retrieved from http://www.toronto.ca/toronto_facts/diversity.htm (last visited July 5, 2013).

Darden J. T. and S. Kamel, 2002, “The Spatial and Socioeconomic Analysis of Aboriginals and Whites in the Toronto CMA”, The Canadian Journal of Native Studies, Vol. 22, 33-62.

Doerner W., 1997, An Introduction to Law Enforcement: An Insider’s View, New York, Butterworth-Heinemann.

Feagin, J. R. and D. Eckberg, 1980, “Discrimination: Motivation, Action, Effects, and Context”, Annual Review of Sociology, Vol. 6, 1-20.

Fong E., 2006, Inside the Mosaic, Toronto, University of Toronto Press.

Foster C., 1996, A Place Called Heaven: The Meaning of Being Black in Canada, Toronto, Harper-Collins.

Gelman A., J. Fagan and A. Kiss, 2007, “An Analysis of the New York City Police Department’s ‘Stop-and-Frisk’ Policy in the Context of Claims of Racial Bias”, Journal of the American Statistical Association, Vol. 102, No. 479, 813-823.

Gold A. and E. Harvey, 2003, Executive Summary of Presentation of behalf of the Toronto Police Service. Toronto, Toronto Police Service.

Grogger J. and G. Ridgeway, 2006, “Testing for Racial Profiling in Traffic Stops from behind A Veil of Darkness”, Journal of the American Statistical Association, Vol. 101, 878-887.

Hulchanski, J. D., 2010, The Three Cities Within Toronto: Income Polarization Among Toronto’s Neighbourhoods, 1970-2005. Toronto, Cities Centre Press, University of Toronto.

Klinger D. A., 1997, “Negotiating Order in Patrol Work: An Ecological Theory of Police Response to Deviance”, Criminology, Vol. 35, No. 2, 277-306.

Lever A., 2005, “Why Racial Profiling is Hard to Justify: A Response to Risse and Zeckhauser”, Philosophy and Public Affairs, Vol. 33, No. 1, 94-110.

Mann R. V., 2004, 52 Supreme Court of Canada, retrieved from the Canadian Legal Information Institute website: http://canlii.ca/t/1hmpl (last visited December 5, 2011).

Manning P. K., 1995, “The Police Occupational Culture’’, in Bailey, W. (eds.), Encyclopedia of Police Science, New York, Garland.

Meehan A. and M. Ponder, 2002, “Race and Place: The Ecology of Racial Profiling African American Motorists”. Justice Quarterly, Vol. 19, No. 3, 399-430.

Melchers R., 2003, “Do Toronto Police Engage in Racial Profiling?” Canadian Journal of Criminology and Criminal Justice, Vol. 45, No. 3, 347-366.

Parker K. F., J. M. MacDonald, G. P. Alpert, M. R. Smith, and A. Piquero, 2004, “A Contextual Study of Racial Profiling: Assessing the Theoretical Rationale for the Study of Racial Profiling at the Local Level”, American Behavioral Scientist, Vol. 47, 943-962.

Qadeer and Kumar, 2006, Ethnic Enclaves and Social Cohesion. Canadian Journal of Urban Research, Vol. 15, No. 2, 1-17. 6.

Rankin J., J. Quinn, M. Shephard, S. Simmie and J. Duncanson, October 20, 2002, “Police Target Black Drivers”, Toronto Star, A8.

Rankin J. and B. Winsa, March 9, 2012, “Known to Police: Toronto Police Stop and Document Black and Brown People Far More than Whites” Toronto Star, retrieved from http://www.thestar.com/news/insight/article/1143536--known-to-police-toronto-police-stop-and-document-black-and-brown-people-far-more (last visited June 5, 2012).

Ridgeway G. and J. Macdonald, 2010, “Methods for Assessing Racially Biased Policing”, in Rice S. K. and White M. D. (eds.), Race, Ethnicity and Policing: New and Essential Readings, New York, NYU Press.

Risse M. and R. Zeckhauser, 2004, “Racial Profiling,” Philosophy and Public Affairs, Vol. 32, 131-170.

Roh S. and M. Robinson, 2009, “A Geographic Approach to Racial Profiling: The Microanalysis and Macroanalysis of Racial Disparity in Traffic Stops”, Police Quarterly, Vol. 12, No. 2, 137-169.

Sacks H., 1972, “Notes on Police Assessment of Moral Character”, in Sudnow D. N. (eds.), Studies in Social Interaction, New York, Free Press.

Sherman L. W., P. R. Gartin, and M. E. Buerger, 1989, “Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place”. Criminology, Vol. 27, 27-55.

Sklansky D. A., 2008, Democracy and the Police, Stanford, Stanford University Press.

Statistics Canada, 2006, Profile of Ethnic Origin and Visible Minorities for Canada, Catalogue No. 94-580-X2006002.

Stults B., K. Parker and E. Lane, 2010, “Space, Place, and Immigration: New Directions for Research on Police Stops”, in Rice S. K. and White M. D. (eds.), Race, Ethnicity and Policing: New and Essential Readings, New York, NYU Press.

Toronto Star, October 19, 2002, “There is no Racism, We do not Do Racial Profiling”, Toronto Star, A14.

Wortley S. and J. Tanner, 2003, “Data, Denials and Confusion: The Racial Profiling Debate in Toronto”. Canadian Journal of Criminology and Criminal Justice, Vol. 45, No. 3, 367-390.

Wortley S. and J. Tanner, 2004, “Discrimination or Good Policing? The Racial Profiling Debate in Canada”. Our Divers Cities, Vol. 1, 197-201.

Winsa P. and J. Rankin, April 14, 2012, “Police Service Board Decision on ‘Carding’ Stuns Activists”, Toronto Star, retrieved from http://www.thestar.com/news/gta/artic-le/1161490 --police-service-board-decision-on-carding-stuns-activists (last visited June 5, 2012).

Top of page

List of illustrations

Title Figure 1. Study Area: The City of Toronto
URL http://journals.openedition.org/cybergeo/docannexe/image/26165/img-1.jpg
File image/jpeg, 3.4M
Title Figure 2. The Percentages of Blacks and Whites in the population of Toronto and the Percentages of Drug-related Stops and Arrests for Blacks and Whites
URL http://journals.openedition.org/cybergeo/docannexe/image/26165/img-2.png
File image/png, 71k
Title Figure 3. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Percentages of White Residents in the Neighbourhoods
URL http://journals.openedition.org/cybergeo/docannexe/image/26165/img-3.jpg
File image/jpeg, 124k
Title Figure 4. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Percentages of Single Parent Families at Neighbourhood Level
URL http://journals.openedition.org/cybergeo/docannexe/image/26165/img-4.jpg
File image/jpeg, 156k
Title Figure 5. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and the Unemployment Rates at Neighbourhood Level
URL http://journals.openedition.org/cybergeo/docannexe/image/26165/img-5.jpg
File image/jpeg, 164k
Title Figure 6. LISA Cluster Maps of Race-specific Stop/Arrest Ratios and Government Transfer Payments as a Percentage of Total Income at Neighbourhood Level
URL http://journals.openedition.org/cybergeo/docannexe/image/26165/img-6.jpg
File image/jpeg, 152k
Top of page

References

Electronic reference

Yunliang Meng, « Racially biased policing and neighborhood characteristics: A Case Study in Toronto, Canada », Cybergeo: European Journal of Geography [Online], Space, Society,Territory, document 665, Online since 06 February 2014, connection on 28 March 2024. URL : http://journals.openedition.org/cybergeo/26165 ; DOI : https://doi.org/10.4000/cybergeo.26165

Top of page

About the author

Yunliang Meng

Assistant Professor, Department of Geography,
Central Connecticut State University, United States of America
mengy@ccsu.edu

By this author

Top of page

Copyright

CC-BY-4.0

The text only may be used under licence CC BY 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search