Factors Influencing the Probability of Clearance and Time to Clearance of Canadian Homicide Cases, 1991-2011
| DOI | 10.1177/1088767916686148 |
| Author | Cristina Pastia,Garth Davies,Edith Wu |
| Published date | 01 August 2017 |
| Date | 01 August 2017 |
| Subject Matter | Articles |
research-article2017
Article
Homicide Studies
2017, Vol. 21(3) 199 –218
Factors Influencing the
© 2017 SAGE Publications
Reprints and permissions:
Probability of Clearance
sagepub.com/journalsPermissions.nav
https://doi.org/10.1177/1088767916686148
DOI: 10.1177/1088767916686148
journals.sagepub.com/home/hsx
and Time to Clearance of
Canadian Homicide Cases,
1991-2011
Cristina Pastia1, Garth Davies1, and Edith Wu1
Abstract
Both homicide and clearance rates in Canada have declined significantly over the
last few decades. The existing literature predominantly focuses on differentiating
between cleared and uncleared cases, with less attention given to circumstances
affecting the time taken to solve homicides. To address this gap, data were taken
from the Homicide Survey; 11,297 police-reported homicides in Canada between
1991 and 2011 were examined using both Kaplan–Meier survival analysis and Cox
regression. Although results showed some notable differences in clearance times
among demographic variables, incident characteristics—notably drug or gang
involvement—largely overshadowed the effects of victim characteristics.
Keywords
clearance, drugs, correlates, policing, gang, gender, elderly
Any discussion of homicide clearance is incomplete without the acknowledgment that,
over time, the percentage of cleared homicides in North America has declined notably,
despite the falling rate of homicides themselves. In Canada, although homicide rates
increased from 1.28 (per 100,000) in 1961 to 3.03 in 1975, they have since been on a
long decline, falling to 1.44 in 2013 (Statistics Canada, n.d.-b). Similarly, in the United
States, the homicide rate was 4.8 (per 100,000) in 1961; it then peaked twice—once in
1980 (10.2) and again in 1991 (9.8)—and finally dropped to a record low 4.6 in 2013
1Simon Fraser University, Burnaby, British Columbia, Canada
Corresponding Author:
Cristina Pastia, School of Criminology, Simon Fraser University, 8888 University Drive, Burnaby, British
Columbia, Canada V5A 1S6.
Email: cpa10@sfu.ca
200
Homicide Studies 21(3)
Figure 1. Homicide rates and clearance percentages, 1960-2010.
Note. Data for homicide rates taken from Statistics Canada (n.d.-b) and clearance from Mahony and
Turner (2012).
(Federal Bureau of Investigation, n.d.). Several explanations have been offered for
these fluctuating homicide rates, with varied success. The rise and fall of violent crack
cocaine markets have been found to impact the homicide rate in the United States
(Wallman & Blumstein, 2005), but this explanation was less salient in Canada, which
did not experience a similar epidemic. Stiffer penalties for firearm-related offenses
and increased screening for firearm purchases in particular were mostly related to a
reduction in U.S. homicide rates (Irvin, Rhodes, Cheney, & Wiebe, 2014; Wallman &
Blumstein, 2005; Webster, Crifasi, & Vernick, 2014; Wodarz & Komarova, 2013).
Similarly, an increase in gun ownership has been linked to an increase in non-stranger
homicides (Siegel et al., 2014). But in Canada, where gun-related homicides made up
a much smaller proportion of homicides than the United States, increased gun control
has not been shown to conclusively affect overall homicide rates (Langmann, 2012;
Leenars & Lester, 1994). Finally, a reduction in the proportion of the population
between the ages of 15 and 24 showed a very small, but significant, reduction in the
homicide rate in both countries (Leenars & Lester, 2004; Sprott & Cesaroni, 2002).
The reduction in North American homicide rates has been accompanied by an over-
all decrease in homicide clearance rates. The fluctuations in both homicide rates and
clearance rates in Canada are displayed in Figure 1. A homicide is considered cleared
when a perpetrator has been identified and arrested, or when a perpetrator has been
identified but cannot be arrested due to an extraordinary circumstance (e.g., suicide of
the perpetrator); the latter is known as an exceptional clearance and accounts for a
Pastia et al.
201
minor percentage of all clearances (Jarvis & Regoeczi, 2009). The fact that a case was
cleared does not guarantee that the accused will later be convicted of the offense in
court. In the early 1960s, homicide clearance percentages in both the United States and
Canada were more than 90% (Federal Bureau of Investigation, n.d.; Mahony & Turner,
2012; Regoeczi, Kennedy, & Silverman, 2000). Despite a return to near-1960s homi-
cide rates, however, both countries have experienced record low clearance rates within
the last decade: 69.6% in 2008 for Canada and 60.7% in 2006 for the United States.
Research on the topic of homicide clearance has been dominated by two perspectives:
victim devaluation and situational/incident characteristics. This study was informed
by both these approaches.
Victim Devaluation
The concept of victim devaluation stems from Black’s (1976, 1980) work on the
behavior of law, which begins with the premise that society is stratified. Broadly
speaking, stratification here refers to the unequal distribution of wealth, which was
affected by demographic factors (Black, 1976). The application of the law varies
according to an individual’s rank. A victims’ rank or social status was often measured
using sex, age, and race; those of a lower status were more likely to be female, young,
and non-White (Black, 1976). “Lower status” victims received “less law” than those
of higher ranks, particularly when the former were victimized by someone of a higher
status. The phrase “less law” referred to a lower likelihood of police attention, arrest,
or conviction, and a greater likelihood of a less severe sentence for the perpetrator. In
other words, this perspective argued that victims of lower ranks were valued by law
enforcement less than victims of higher ranks. In fact, homicides with victims of a
lower social status may not have been classified as homicides at all; alternatively, they
may have been given less attention and were less likely to be cleared.
Overall, empirical support for this perspective was mixed. On one hand, a recent
qualitative investigation into attitudes of homicide detectives revealed that detectives
saw some homicide victims as “true victims”—for instance, law-abiding citizens, par-
ticularly the very young or old, or police officers. Incidents with such victims com-
prised a small proportion of homicides, but they were prioritized over those with
victims involved in criminal lifestyles or who were members of an at-risk demo-
graphic, reinforcing the notion that officers valued some victims over others (Hawk &
Dabney, 2014). On the other hand, although women were considered to be of lower
social status than men, and therefore, cases with female victims should be less likely
to be cleared, this was often not the case. Several studies have found the exact oppo-
site; that is, homicides with female victims were more likely to be cleared (Lee, 2005;
Regoeczi, Jarvis, & Riedel, 2008; Regoeczi et al., 2000; Taylor, Holleran, & Topalli,
2009). Others found no significant difference in the likelihood of clearance according
to sex (Alderdan & Lavery, 2007; Banziger & Killias, 2014; Puckett & Lundman,
2003; Wellford & Cronin, 1999). Only a handful of studies found a significant sex dif-
ference in line with the victim devaluation perspective (Jiao, 2007; Roberts & Lyons,
2009, 2011). The findings pertaining to victim age and clearance rates have similarly
202
Homicide Studies 21(3)
been at odds with the victim devaluation thesis. Although, according to Black (1976),
the young were usually considered to be of a lower rank, evidence has suggested that
homicides with young victims were often more likely to be cleared (Jiao, 2007; Lee,
2005; Puckett & Lundman, 2003; Regoeczi et al., 2008; Regoeczi et al., 2000; Roberts
& Lyons, 2009, 2011; Taylor et al., 2009; Trussler, 2010).
Race appeared to be extremely inconsistent as a predictor of clearance rates, owing,
at least in part, to disparities in the manner in which race has been measured. Race has
often been operationalized as a dichotomy, with White and non-White as the catego-
ries. Some studies found no significant racial difference between those two groups
(Puckett & Lundman, 2003; Roberts, 2007), whereas other studies were consistent
with the victim devaluation perspective. Lee’s (2005) research suggested that homi-
cides with White victims were more likely to be cleared. Regoeczi et al. (2008) exam-
ined homicide clearance by looking at both binary outcomes (cleared vs. not cleared)
as well as time to clearance; homicides with White victims were more likely to be
cleared, but not once was accounted for. Jiao (2007) found a significant difference
between racial groups, but cases with White victims were no longer significantly more
likely to be cleared once incident characteristics were considered. However, Regoeczi
et al. (2000) delved into homicide clearance in the United States and Canada and
found that, at the national level, homicides with non-White victims were more likely
to be cleared. This does not hold true when the authors examined homicides at the state
and provincial level, as there were no significant racial differences in homicide clear-
ance in the state of New York or Ontario. When victim and offender race combinations
(with race dichotomized as White and non-White) were explored, Roberts and Lyons
(2009)...
To continue reading
Request your trial