A Report On Police Ethics from THE JOURNAL OF CRIMINAL JUSTICE(2001)

Abstract
This article identifies the characteristics of police officers (officers’ background attributes, arrest activity, and assignment) who most frequently receive complaints from citizens regarding the use of excessive force. The data for the study were obtained from a large mid-western municipal police department. The results show that arrest activity, officer age, and officer gender are most strongly related to the receipt of citizen’s complaints about excessive force and differentiate high-complaint officers from low-complaint officers. Implications of the findings are discussed. D 2001 Elsevier Science Ltd. All rights reserved.
Introduction
The ability to use ‘‘essentially unrestricted’’ coer- cive force lies at the core of the police function; it is this dimension of the police occupation that differ- entiates it from all others (Bittner, 1970). It is because of this core aspect of policing that the ‘‘...inappro- priate use of force is the central problem of contem- porary police misconduct’’ (Kerstetter, 1985, p. 149; Pate & Fridell, 1993). The inappropriate use of force has potentially devastating consequences for the police organization, the public, and the relationship between the police and the community. The goodwill created by community-oriented policing initiatives can be quickly destroyed given a well-publicized incident of police abuse of force. As demonstrated
on several recent occasions, the legitimacy of entire police organizations can be questioned as a result of such incidents (see Skolnick & Fyfe, 1993). Given the deleterious consequences of such inci- dents, it is not surprising that numerous studies have examined the causes, correlates, and control of police use of force; however, most of these analyses focus on police use of ‘‘deadly force,’’ which may or may not involve the actual abuse of force (e.g., Alpert, 1989; Binder & Scharf, 1980; Fyfe, 1988; Jacobs & Britt, 1979). Relatively little research has examined specifically what is perhaps the most significant form of force — that which is deemed excessive or brutal (Adams, 1995; Worden, 1995). Despite the relative rarity of research that exam- ines excessive force, one conclusion that can be drawn with some confidence is that a relatively small number of officers are responsible for a relatively large number of complaints about excessive force (Adams, 1999; Christopher Commission Report,
0047-2352/01/$ – see front matter D 2001 Elsevier Science Ltd. All rights reserved. PII: S0047-2352(01)00114-3
* Corresponding author. Tel.: +1-414-229-5443; fax: +1-414-229-4851. E-mail address: sgb@uwm.edu (S.G. Brandl).
Journal of Criminal Justice 29 (2001) 521–529
1991; Dugan & Breda, 1991; Lersch & Mieczowski, 1996). In an attempt to reduce the number of such complaints, it is first necessary to identify who these high-rate officers are, what characteristics these offi- cers have in common, and, ultimately, why these officers are subject to complaints with such fre- quency. Indeed, a few studies have attempted to identify who these ‘‘complaint-prone’’ officers are (Croft, 1985; Lersch & Mieczowski, 1996; Scrivner, 1994). It is rare, however, for such studies to look beyond individual characteristics in predicting the frequency with which officers receive complaints of excessive force. Such a narrow focus is problematic in at least two ways: (1) obviously, one cannot assess the impact of other possibly important (but not included) factors, and (2) it is difficult to determine with confidence that the relationships observed are genuine rather than spurious. This article represents a step toward ameliorating this gap in knowledge by examining the role of officers’ background character- istics, job assignments, and arrest activity in the receipt of citizens’ complaints about excessive force.
Theoretical and policy issues
Empirical studies and other discussions that have attempted to explain police use of excessive force have adopted several approaches1: excessive force (and the corresponding complaints that result from such actions) has been viewed broadly as a function of the characteristics of officers, situations, and organizations. Relevant to the primary interest in identifying the characteristics of ‘‘high-complaint’’ officers in this study, provided here is a summary of the literature regarding officers’ background char- acteristics, along with the role of officers’ job assign- ments and arrest activity in predicting use of excessive force complaints.
Officers’ characteristics
Studies that have adopted the individual-level approach rest on the belief that individualistic char- acteristics predispose officers to behave in a partic- ular way. To the extent that characteristics of officers are related to behavioral dispositions, a focus on the officer is important in understanding police decision making and behavior; however, such studies have been, at best, only moderately successful in explain- ing police behavior (Worden, 1989, 1995). Despite the inadequacy of the individual-level approach as an overall explanation of police behav- ior, particular relationships between officers’ charac- teristics and behavior may be significant and may have important policy implications (Worden, 1990).
For example, although officers’ traits (and the pre- sumed associated outlooks) may not be directly manipulated within the organizational setting, the traits identified as being conducive to the preferred policing style or orientation can serve as a basis for selection decisions. For instance, police departments across the country are encouraging officers (through internal incentives or selection standards) to become college-educated — the assumption being that college-educated officers ‘‘perform better’’ (Carter, Sapp, & Stevens, 1989; Geller & Scott, 1992). Research on the relationship between education and performance, and on education and use of excessive force in particular, has produced inconsistent and conflicting findings (see Worden, 1990). Cohen and Chaiken (1972) found that more educated officers were less likely to be the subjects of citizen complaints in general, and Cascio (1977) found that officers with higher levels of education were the subjects of fewer allegations of excessive force. In contrast, Croft (1985) found no differences by educa- tional level in her comparison of high- and low- excessive force complaint officers. Worden (1995) found that officers with bachelor’s degrees were actually more likely to use reasonable force and just as likely to use improper force as officers without bachelor’s degrees. In a related vein, many police departments are operating under a re-energized mandate of dramati- cally increasing the representation of female officers — with at least the partial intent of fostering a more ‘‘kind and gentle’’ policing orientation (see Skolnick & Fyfe, 1993, p. 137). Most researchers have found that gender does exert an influence on police behavior (cf. Alpert & Dunham, 1999; Worden, 1995). Bloch and Anderson (1974) found that female officers initiated fewer citizen encounters and made fewer felony and misdemeanor arrests. Morash and Greene (1986) also found that female officers made fewer arrests than male officers. Since most instances of excessive force occur during the course of arrest (Adams, 1999), it is not surprising that females are less likely to be the subjects of complaints of exces- sive force (Christopher Commission Report, 1991). Not only may females be less likely to be the subjects of excessive force complaints as a result of their arrest activity, but researchers have also suggested that female officers are more adept at avoiding violence and de-escalating potentially violent situations (Alpert & Dunham, 1999; Grennan, 1987; Milton, 1980). In any case, while the relationship between gender and complaints about excessive force is rather clear, the reasons for the relationship are subject to debate (Worden, 1993). Findings with regard to the relationship between race and excessive force are not as clear-cut. Some
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529522
theorists have argued that officers of color may take a more understanding approach in their interactions with members of racial minority groups (e.g., Mann, 1993). For example, as explained by Fyfe (1988):
Champions of representativeness in policing ...have long argued that one of the most promising routes to reducing police–citizen violence is to increase the percentage of minority officers. Presumably these officers are attuned more closely to the problems and folkways of the minority citizens who are dispropor- tionally the subjects of police deadly force and police attention generally (p. 195).
The empirical literature, however, lends little support to this claim. Garner, Schade, Hepburn, and Buchanan (1995) and Worden (1995) found that race was not predictive of use of force. Other studies have found that minority officers are more likely to be the subjects of complaints about excessive force (Gray, 1998), and that officers are more likely to use force against suspects of their own race (Alpert & Dunham, 1999). Some researchers have argued that these findings are an artifact of job assignment. Minority officers are commonly assigned to high-crime neigh- borhoods, which places them with greater frequency in situations that may require the use of force (e.g., Fyfe, 1988; Geller & Scott, 1992). Attention has also been given to officers’ age and length of service. Several studies have demonstrated that younger officers are more ‘‘active’’ than older officers. In particular, younger officers initiate more contacts, do more preventive patrolling, and record more crime reports (Crank, 1993; Friedrich, 1980; Sherman, 1980). One side effect of increased activity is that officers may find themselves more frequently in situations where they need to use force (Adams, 1999). Other researchers, however, have concluded that age does not matter in explaining use of force (Bayley & Garafalo, 1989; Worden, 1995). Besides the numerous incongruent and conflict- ing findings evident in the literature regarding offi- cers’ background characteristics, another problem (and perhaps a cause of at least some of the incon- gruities) was that most studies of police use of excessive force have examined officers’ attributes without considering important mediating variables, such as the characteristics of jobs to which officers were assigned or officers’ arrest activity. These variables are discussed below.
Job characteristics
Some literature suggest that officers’ use of exces- sive force is a function of the nature of the job, specifically the degree of specialization in the job and the crime level of the patrol beat to which officers are
assigned. With regard to job specialization, it has been argued that specialization encourages competi- tion among units and this, in turn, leads to abuse (Skolnick & Fyfe, 1993, p. 189). Further, the goal of specialized units, such as drug squads and gang units, is often to increase arrests. As the argument goes, when arrests by ‘‘any means necessary’’ are encour- aged, a ‘‘siege mentality’’ may develop, and this may encourage further the abuse of force (Skolnick & Fyfe, 1993). Unfortunately, there were no studies that empirically examined this proposition. With regard to the characteristics of the patrol beat, one might expect that high-crime areas foster conditions where the use of force is more frequently necessary (Toch, 1995). Similarly, it may be that officers are more likely to use (excessive) force in areas where residents are seen as most deserving of such police action and are least powerful in protecting themselves from it (Geller & Scott, 1992).
Arrest activity
Yet another argument is that the use of excessive force is a function of officers’ arrest activity. From this perspective, physical arrests put officers at risk of having to use force and having this force perceived as ‘‘excessive’’ by suspects and others. According to this reasoning, citizens’ complaints of excessive force are a byproduct of arrests. As Toch (1995) proposes, ‘‘an officer may be highly productive and may initiate a larger-than-usual amount of enforcement activity. He may disrupt the felonious plans of many disgrun- tled (and complaint-prone) offenders’’ (p. 100). Indeed, the existing research would lead one to believe that this is true; when force is used by the police, it is usually in arrest situations (Adams, 1999). In summary, the literature contained many as- sumptions about factors that contribute to officers’ use of excessive force. When these assumptions were subject to empirical tests, the results often were unsupportive, inconsistent, or even contradictory. This article brings additional empirical evidence to the understanding of the relationship between police officers’ background attributes, arrest activity, assign- ment, and citizens’ complaints about excessive force. With the accumulation of such evidence, one may develop a more complete understanding of the char- acteristics of complaint-prone officers.
Methodological issues
There are a number of methods that can be used to collect data on police behavior (and excessive force in particular), but each has limitations and biases (Adams, 1995; Garner et al., 1995). For example,
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529 523
the observational approach will invariably lead to problems associated with trying to observe uncom- mon events (Bayley & Garafalo, 1989). This is particularly true given the common research conclu- sion that police use of excessive force (and physical force generally) is a rare event (Adams, 1999). Another significant concern with data collected through observation is reactivity (Mastrofski & Parks, 1990). As for self-report data, they will likely reflect problems of distortion and compliance (see Alpert, Kenney, & Dunham, 1997). Little question, depart- mental records, such as use of force reports, will also have problems of intentional distortion and compli- ance (Adams, 1995). Citizen complaints as a source of data, as used in this study, raised at least two other important issues. The first issue related to the complaint process and the relative difficulty of filing a complaint against an officer. It was important to realize that the complaint receipt and investigation processes, which were largely controlled by the police themselves, may have a substantial impact on the resulting data. No question, it was more difficult to file a complaint in some departments than in others. Further, given the possible hurdles to be navigated in filing a com- plaint, citizens (or certain groups of citizens) may typically be less willing or interested in filing complaints against officers. While this issue may have had the most impact on data in a cross- departmental comparison, which was not the case here, the issue needs to be considered in evaluating the resulting data of this study. The second issue related to definitional matters. In essence, the question is ‘‘What constitutes excessive force?’’ The literature offers definitions and discus- sions of various forms of force (improper force, unnecessary force, brutality, force used excessively, excessive force, among others) (see Adams, 1995; Garner & Maxwell, 1999; National Institute of Jus- tice, 1999; Skolnick & Fyfe, 1993). Definitions aside, what constituted various forms of force — and what differentiated one from another — was a matter of perspective. Police officers may have a perspective on what is appropriate force that differs from researchers as well as from citizens (especially those citizens upon which force was used). On one hand, the distinctions between different forms of force may be useful, as different forms of force may be influ- enced by different factors and remedied with different interventions (Worden, 1995). On the other hand, one may argue that these distinctions were not important because, from a citizen’s perspective, it was immate- rial if the police used ‘‘excessive’’ or ‘‘unnecessary’’ force (as defined empirically or legally) — what matters was that the citizen perceived unfair, unjust, or otherwise unequal treatment from the police, and it
was this perception that may have carried dysfunc- tional consequences for the police and the relation- ship between the police and the community. So what is excessive force? As discussed below, in this study, excessive force was measured as it was perceived (and reported) by citizens, and as categorized by the police department from which the data were obtained.
Method
The data for this study were collected from a large mid-western municipal police department. The pop- ulation that the police department served was approx- imately 45 percent White and 40 percent African– American. The largest segment of the workforce was engaged in the service industry with government being the second largest employer. The unemploy- ment rate was usually slightly higher than the national average. In 1993 (the year of this study), there were 2,529 violent crimes reported per 100,000 persons (not including rape as those statistics were not avail- able) and 5,926 property crimes per 100,000 persons (not including larceny theft) (Federal Bureau of Investigation, 1994). In the study department, citizens could file complaints alleging police misconduct at any of the district stations. These complaints should be written and filed in person by the person alleging the misconduct or his/her representative. Citizens’ complaints about police use of excessive force were investigated by civilian personnel employed by the department and assigned to a designated unit of the police department. The Internal Affairs Unit of the police department was responsible for investigating other allegations of police misconduct, not complaints about excessive force. For the most serious sustained cases, a citizen review board could become involved in determining the proper sanctions for officers and could make sanction recommendations to the chief. As part of the routine data collection and storage capabilities of the agency, the department maintained all sworn officers’ personnel information on the departmental computer system. These records included data on background characteristics of offi- cers (sex, race, education, age, length of service)2, officers’ unit of assignment (patrol, narcotics, gang crimes), patrol area assignment (precinct), number of The data for this study were collected from a large mid-western municipal police department. The pop- ulation that the police department served was approx- imately 45 percent White and 40 percent African– American. The largest segment of the workforce was engaged in the service industry with government being the second largest employer. The unemploy- ment rate was usually slightly higher than the national average. In 1993 (the year of this study), there were 2,529 violent crimes reported per 100,000 persons (not including rape as those statistics were not avail- able) and 5,926 property crimes per 100,000 persons (not including larceny theft) (Federal Bureau of Investigation, 1994). In the study department, citizens could file complaints alleging police misconduct at any of the district stations. These complaints should be written and filed in person by the person alleging the misconduct or his/her representative. Citizens’ complaints about police use of excessive force were investigated by civilian personnel employed by the department and assigned to a designated unit of the police department. The Internal Affairs Unit of the police department was responsible for investigating other allegations of police misconduct, not complaints about excessive force. For the most serious sustained cases, a citizen review board could become involved in determining the proper sanctions for officers and could make sanction recommendations to the chief. As part of the routine data collection and storage capabilities of the agency, the department maintained all sworn officers’ personnel information on the departmental computer system. These records included data on background characteristics of offi- cers (sex, race, education, age, length of service)2, officers’ unit of assignment (patrol, narcotics, gang crimes), patrol area assignment (arrests made (index arrests, total arrests; as deter- mined by arrest reports filed), and the aggregate number of citizen complaints received. These data were recorded and maintained on an annual basis. All data analyzed in this study were for the year of 1993. In this department, citizens’ complaints about excessive force were categorized into ten groups:
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529524
(1) menaced verbally,3 (2) menaced physically, (3) menaced with weapon, (4) handled roughly, (5) handcuffs too tight, (6) battery with fist, (7) battery with object, (8) battery with pistol, (9) battery: shot, and (10) battery: other. Unfortunately, only the aggregate number of complaints received by each officer was available in the personnel file and the investigative dispositions of the complaints were not specified. In 1993, there were 2,868 complaints received by the police department that were classi- fied into these ten categories.4 Three groups of officers were selected for analysis in this study. First, a random sample (n=800) of all police officers assigned to patrol and specialized units (gang crimes and narcotics) was selected (i.e., ‘‘all officer sample’’). The second group consisted of a random sample (n=600) of officers who received two or less citizen complaints that alleged an act of excessive force in 1993 (‘‘low-complaint officer sample’’). The third group consisted of all officers (n=200) who received three or more use of excessive force complaints in 1993.5 These 200 officers received a total of 717 complaints,6 or 25 percent of the total number of citizen complaints of excessive force filed in 1993. At the same time, this subset of officers constituted less than 10 percent of all officers employed in the department. A comparison of the high-complaint officer group and the low-complaint officer group allowed one to identify how high-complaint officers differed from low-complaint officers. Further, separate analyses of the ‘‘all officer sample’’ allowed one to identify the impact of the individual, assignment, and arrest variables on the number of excessive force com- plaints received by officers. As such, the data were analyzed first by comparing distributions of variables across the high- and low-complaint officer groups (using t tests), then examining the correlations between variables, and finally by estimating an OLS regression equation with the number of citizen complaints received as the dependent variable.
Results
Table 1 allows for a comparison of the character- istics of low- and high-complaint officers. Table 1 shows that high-complaint officers are significantly more likely to be younger, less experienced, and assigned to the highest crime areas7 compared to the low-complaint officers. The most striking differ- ence between high- and low-complaint officers is with respect to arrest activity: high-complaint officers made twice as many index arrests and nearly three times as many total arrests as their low-complaint counterparts. Finally, while there are no meaningful
differences across educational level or assignment, there is a substantive (but still not significant) differ-
Table 1 Background characteristics, job assignment, arrest activity, and complaints received for low-complaint officers and high-complaint officersa
Variable
Low-complaint officer sample (n=600)
High-complaint officer population (n=200)
Gender 1=Female 17.5 3.0 2=Male 82.5 97.0 t value=0.05
Race 1=White 65.2 46.5 2=Minority 34.8 53.5 t value=3.8
Education 1=Completed high school
43.2b 41.3c
2=Some college 39.6 43.9 3=College degree/more 17.1 14.8 t value=0.0
Mean age (range) 40.0 (22–61) 34.9 (24–53) t value=24.29**
Mean length of service (range)
12.3 (1–35) 7.3 (1–26)
t value=23.81**
Assignment 1=Specialized unit 9.8 8.5 2=Patrol unit 90.2 91.5 t value=0.01
Patrol area 1=Lower-crime areas 44.4d 37.2e 2=Highest-crime areas 55.6 62.8 t value=4.33**
Mean index arrests (range)
12.8 (0–77) 24.2 (0–94)
t value=36.77**
Mean total arrests (range)
62.9 (0–374) 177.8 (0–553)
t value=136.70** a Percentages are listed. b n =525. c n =196. d n=541. e N =183. * P<.05. ** P <.01.
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529 525
ence across gender; males are over-represented among high-complaint officers. Table 2 shows the zero-order correlations between the variables of interest for the ‘‘all officer sample.’’8 In Table 2, it is seen that officer gender (r =.08; P < .05), officer age (r = .23; P <.01), length of service (r= .20; P<.01), and number of index arrests made (r=.31; P<.01) are most closely related to the number of complaints received. Again, male officers, those who are younger and have less experience, and those who made more index arrests are the most likely to receive com- plaints about excessive force. The zero-order correlations also allow for an examination of the relationships between variables traditionally included in an individual-level approach (i.e., race, gender, age, education, length of service) and other variables (i.e., job characteristics and arrest activity) that may mediate their effect on citizen complaints. With regard to race, Table 2 shows that
minority officers are more likely to be assigned to higher-crime areas (r=.27; P<.01); however, minor- ity officers are not more likely to make more arrests (r=.05) nor are minority officers likely to receive more complaints (r=.01). With regard to age, these analyses show that younger, less experienced officers are more likely to make more arrests ( r = .36; P<.01) despite that they are not more likely to be assigned to high-crime areas (r= .02). Finally, edu- cation is not significantly correlated with assignment, patrol area, arrests made, or complaints received. To isolate the impact of each variable on the number of citizen complaints about excessive force, an OLS regression analysis was performed on the ‘‘all officer sample.’’ The results are presented in Table 3.9 Like the earlier analyses, these results show that the number of index arrests made, along with officer age and gender, are predictors of the number of complaints about excessive force; those officers who are younger, who are male, and who
Table 3 OLS regression analysis of complaints received for all officer sample (n=800)
Variable Coefficient Standard error t ratio b
Gender .23 0.08 3.01** .11 Race .04 0.06 0.60 .02 Education .02 0.04 0.50 .02 Age .02 0.00 4.36** .17 Assignment .08 0.10 0.80 .03 Patrol area .02 0.06 0.31 .01 Arrests made .02 0.00 6.78** .25 Constant .61 0.26 2.33 R2 .12 Adjusted R2 .12 n 800
Mean substituted for missing data. * P<.05. ** P <.01 (two-tailed test).
Table 2 Relationships between officers’ characteristics, arrest activity, assignment, and complaints received for all officer sample (n=800)
Gender Race Education Age
Length of service Assignment
Patrol area
Arrests made
Complaints received
Gender 1.0 Race 0.07 1.0 Education 0.02 0.10* 1.0 Age 0.23** 0.18* 0.19** 1.0 Length of service 0.32** 0.29* 0.19** 0.85** 1.0 Assignment 0.02 0.02 0.05 0.04 0.01 1.0 Patrol area 0.07 0.27** 0.02 0.02 0.06 n.a. 1.0 Arrests made 0.06 0.05 0.02 0.36** 0.36** 0.07* 0.02 1.0 Complaints received 0.08* 0.01 0.02 0.23** 0.20** 0.02 0.00 0.31** 1.0
* P <.05 (two-tailed test). ** P <.01 (two-tailed test).
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529526
make index arrests with greater frequency are the most likely to receive more complaints about the use of excessive force.
Discussion
This study represented an attempt to better under- stand the relationship between police officers’ back- ground characteristics, arrest activity, assignment, and citizens’ complaints about excessive force. An exami- nation of the relationships between these factors allowed one to be more confident that observed relationships were genuine rather than spurious, and to assess the relative influence of a range of variables on the frequency with which officers were the subject of use of force complaints. That said, the findings of the study were enlightening as much for what was not found as what was found. For instance, of the officers’ characteristics examined, only age and, to a lesser extent, gender appear to have an impact on the like- lihood of receiving complaints and differentiate high- complaint from low-complaint officers. All else equal, younger officers were more likely to receive more complaints about excessive force than older officers, and female officers were less likely to be the subject of citizen complaints of excessive force than male offi- cers. One interpretation of these findings was that not only were younger (male) officers likely to make more arrests, they may at the same time be more likely to resort more quickly to physical force in arrest situa- tions. Moreover, perhaps both of these tendencies — to make more arrests and to be more physical in these situations — were an effort to create or enhance a sub- culturally valued ‘‘gung-ho’’ or ‘‘kick-ass’’ reputation (Toch, 1995). In the same vein, it appeared that females might be more adept at avoiding violence or de-escalating potentially violent arrest situations, as others have suggested. In addition, female officers, already often at the fringe of the police sub-culture (Worden, 1993), might not be as desirous of a ‘‘hard- nosed’’ reputation as their male counterparts. None of the analyses conducted here would lead one to believe that officers’ race or level of education played a role in the receipt of excessive use of force complaints. Taken as a whole, these findings lead one to question the wisdom of several of the traditional assumptions about how officers with various charac- teristics behave or perform. This study provided no evidence to support the claim that complaints about excessive force were a function of job specialization. It was clear that job specialization was not necessarily an ‘‘organizational evil’’ at least in terms of the propensity of officers assigned to such units to receive complaints about excessive force. Perhaps such tendencies could be
mitigated by proper leadership, supervision, and temporary assignments to such units. Further, complaints did not appear to be a function of the contexts in which officers work. In general, the results did not support the claim that patrol area was related to the likelihood of receiving complaints from citizens. It appeared that officers were at risk of receiving complaints not only from citizens in high- crime areas but from those who live (or work) in lower-crime areas as well. Finally, the strongest evidence provided in this study was that complaints were related to arrests made. The analyses clearly offer the most support for this argument; officers who made more arrests also received more complaints about the use of excessive force. The number of index arrests made seemed to most clearly differentiate officers who received more complaints from those who received few complaints. This conclusion, however, was not an unambiguous one. It was not clear from these analyses whether arrests led to complaints, or the use of force led to arrests being made to ‘‘cover’’ the illegitimate use of force. In any case, the data suggested that arrests came at a price: greater arrest activity was linked to citizen complaints about excessive force. The findings of this study highlighted at least three issues for consideration in future research. First, it would be worthwhile to question further the causal relationship between arrest activity and complaints of excessive force. Specifically, what is the common causal order between arrests and complaints? Are complaints necessarily tied to arrests? Perhaps a useful approach to studying this issue is an examina- tion of the dynamics of the interactions between police officers and citizens in actual encounters — encoun- ters when force is used as well as when it is diverted. The relationship between arrest activity and excessive force complaints may have implications for training, supervision, and performance appraisal systems. Research designed to disentangle and clarify the relationship may inform such policy development. Second, additional efforts that seek to explain the variation in the frequency in which officers’ receive complaints would be worthwhile. Given the purpose of the present study and the nature of these data (officers as the unit of analysis), the focus was limed to the background characteristics of officers, assign- ment, and arrest activity. Remaining pieces of the puzzle may include organizational factors (such as incentive systems and organizational philosophy), situational factors (see Worden, 1995), and additional individual level variables (e.g., officer personality; see Toch, 1995). Finally, it would be beneficial to specify how
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529 527
on which excessive force is measured. As such, it would be useful to replicate available studies, includ- ing the present one, with different sources of data for the measurement of excessive force. With such an approach, inconsistencies in findings across studies may be specifically attributed to the nature of the data. Carefully crafted research to address these questions could contribute significantly to an under- standing of police use of excessive force and aid police officials in reducing the frequency of it.
Notes
1. For the most part, these were also the approaches used to analyze other forms of police behavior (see Riksheim & Chermak, 1993; Sherman, 1980; Worden, 1989). 2. Within the departmental computer system, the variables were defined as follows: sex (female/male), race (White/minority), education (competed high school/some college/college degree or more), age (in years), length of service (in years). By necessity, the same measurement schemes are used in these analyses. 3. Theoretically and conceptually, ‘‘menaced verbally’’ is a questionable form of excessive force; however, given the nature of these secondary data, one is forced to accept it. Fortunately, this type of complaint constituted a small proportion of all excessive force complaints filed in this department in 1993 (see footnote 4 below). 4. Based on an analysis of a separate data file on the 2,868 complaints filed in 1993, it was determined that approximately 78 percent of these complaints alleged either ‘‘battery with fist’’ (n = 1118; 39.0 percent of total), ‘‘handled roughly’’ (n=717; 25.0 percent), or ‘‘battery with object’’ (n=404; 14.1 percent). Only 5.9 percent (n=169) alleged ‘‘menaced verbally.’’ In addition, 57 percent of the complaints (n=1634) allegedly involved ‘‘bruises,’’ ‘‘abra- sions,’’ and or ‘‘pain only.’’ Only 2.4 percent of the complaints (n = 68) allegedly involved broken bones, disfigurement, internal injuries, or death. The nature and circumstances of citizens’ alleged injuries in this set of complaints are similar to those reported in other studies (see Alpert & Dunham, 1999). 5. No question, designation of those officers with three or more complaints in 1993 as ‘‘high-complaint’’ officers is an arbitrary one and is guided more by empirical and statistical necessity than theory. Given the requisite data, it would be advisable that officers be placed on a multi-point continuum that captured with more precision the frequency by which officers received complaints (see Toch, 1995). Nonetheless, the rather simple classification scheme used here is not unlike that used by others (e.g., Worden, 1995). 6. Of the 200 officers, 127 received three complaints, forty-seven received four complaints, sixteen received five complaints, five received six complaints, three officers received seven complaints, one officer received eight complaints, and one received nine complaints. 7. Based on crime statistics provided by the department and discussions with departmental personnel, the nine areas identified as the highest crime areas in this study clearly
experience a high level of violent and property crime in relation to the other precincts. Further, these areas were typically identified by officers as the most difficult and demanding to work. As a consequence of these beats being identified as unique in relation to the others, the present classification scheme was used. Different treatments of this variable in the analyses (beats divided in half, beats divided into quartiles) had no appreciable affect on any of the results. 8. The ‘‘all officer sample’’ had the following characteristics: approximately 83 percent of officers were male, 64 percent White, 43 percent completed some college, the mean age of officers was approximately forty years, and on average these officers had twelve years of service. Approximately 92 percent of the officers were assigned to patrol units; of the officers assigned to patrol units, 63 percent worked in the sixteen low-crime areas (precincts). The mean number of index arrests recorded by officers was twelve, while the mean number of total arrests for the year was sixty-six. 9. This study cannot be considered an attempt to explain police use of excessive force because the focus of the research is only on individual-level variables (and the individual as the unit of analysis). Accordingly, the sole purpose of the multivariate analysis is to assess the impact of the independent variables on the dependent variable. As such, the amount of variation explained should be of secondary concern (see Worden, 1989 ,1995 for similar arguments). In addition, given the high correlation between officer age and length of service, length of service was excluded from the OLS regression analysis. For the sake of comparison, the OLS regression analysis was run with length of service included instead of age and the results were virtually identical.
References
Adams, K. (1995). Measuring the prevalence of police abuse of force. In: W. Geller, & H. Toch (Eds.), And justice for all: understanding and controlling abuse of force. Wash- ington, DC: Police Executive Research Forum. Adams, K. (1999). What we know about police use of force. In: Use of force by police: overview of national and local data. Washington, DC: National Institute of Justice. Alpert, G. P. (1989). Police use of deadly force: the Mi- ami experience. In: R. G. Dunham, & G. P. Alpert (Eds.), Critical issues in policing. Prospect Heights: Waveland Press. Alpert, G. P., & Dunham, R. G. (1999). The force factor: measuring and assessing police use of force and suspect resistance. In: Use of force by the police: overview of national and local data. Washington, DC: National In- stitute of Justice. Alpert, G. P., Kenney, D. J., & Dunham, R. (1997). Police pursuits and the use of force: recognizing and managing the ‘‘pucker factor’’: a research note. Justice Quarterly, 14, 371–385. Bayley, D. H., & Garafalo, J. (1989). The management of violenceby policepatrol officers. Criminology, 27, 1–27.
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529528
Binder, A., & Scharf, P. (1980). The violent police–citizen encounter. Annals of the American Academy of Political and Social Science, 452, 111–121. Bittner, E. (1970). The functions of police in modern society. Rockville: National Institute of Mental Health. Bloch, P., & Anderson, D. (1974). Policemen on patrol: final report. Washington, DC: Police Foundation. Carter, D. L., Sapp, A. D., & Stevens, D. W. (1989). The state of police education: policy direction for the 21st century. Washington, DC: Police Executive Research Forum. Cascio, W. F. (1977). Formal education and police officer performance. Journal of Police Science and Administra- tion, 5, 89–96. Christopher Commission Report. (1991). Report of the independent commission of the Los Angeles Police Department. Cohen, B., & Chaiken, J. M. (1972). Police background characteristics and performance: summary. New York: Rand Institute. Crank, J. P. (1993). Legalistic and order maintenance behav- ior among police patrol officers: a survey of eight mu- nicipal police agencies. American Journal of Police, 7, 103–126. Croft, E. B. (1985). Police use of force: an empirical anal- ysis. PhD Dissertation. State University of New York at Albany. Ann Arbor: University Microfilms. Dugan, J. R., & Breda, D. R. (1991). Complaints about police officers: a comparison among types and agencies. Journal of Criminal Justice, 19, 165–171. Federal Bureau of Investigation. (1994). Crime in the United States. Washington, DC: U.S. Department of Justice. Friedrich, R. J. (1980). Police use of force: individuals, sit- uations, and organizations. Annals of the American Academy of Political and Social Science, 452, 82–97. Fyfe, J. (1988). Police use of deadly force: research and reform. Justice Quarterly, 5, 165–205. Garner, J., & Maxwell, C. D. (1999). Measuring the amount of force used by and against the police in six jurisdic- tions. In: Use of force by the police: overview of na- tional and local data. Washington, DC: National Institute of Justice. Garner, J., Schade, T., Hepburn, J., & Buchanan, J. (1995). Measuring the continuum of force used by and against the police. Criminal Justice Review, 20, 146–168. Geller, W. A., & Scott, M. (1992). Deadly force: what we know. Washington, DC: Police Executive Re- search Forum. Gray, M. K. (1998). Police characteristics and internal af- fairs complaints. Paper presented at the annual meeting of the American Society of Criminology, Washington, DC. Grennan, S. (1987). Findings on the role of officer gender in violent encounters with citizens. Journal of Police Sci- ence and Administration, 15, 78–85. Jacobs, D., & Britt, D. (1979). Inequality and police use of
deadly force: an empirical assessment of conflict hypoth- esis. Social Problems, 26, 403–412. Kerstetter, W. (1985). Who disciplines the police? who should? In: W. A. Geller (Ed.), Police leadership in America: crisis and opportunity. Chicago: American Bar Foundation. Lersch, K. M., & Mieczowski, T. (1996). Who are the prob- lem-prone officers? An analysis of citizen complaints. American Journal of Police, 15, 23–44. Mann, C. R. (1993). Unequal justice: a question of color. Indianapolis: University Press. Mastrofski, S. M., & Parks, R. B. (1990). Improving obser- vational studies of police. Criminology, 28, 475–496. Milton, C. H. (1980). Women in policing. Washington, DC: Police Foundation. Morash, M., & Greene, J. (1986). Evaluating women on patrol: a critique of contemporary wisdom. Evaluation Review, 10, 230–255. National Institute of Justice. (1999). Executive summary. In: Use of force by police: overview of national and local data. Washington, DC: National Institute of Justice. Pate, A. M., & Fridell, L. A. (1993). Police use of force: official reports, citizens complaints, and legal conse- quences. Washington, DC: Police Foundation. Riksheim, E. C., & Chermak, S. M. (1993). Causes of police behavior revisited. Journal of Criminal Justice, 21, 353–382. Scrivner, E. M. (1994). The role of police psychology in controlling excessive force. Washington, DC: National Institute of Justice. Sherman, L. (1980). Causes of police behavior: the current state of quantitative research. Journal of Research in Crime and Delinquency, 17, 69–100. Skolnick, J. H., & Fyfe, J. J. (1993). Above the law: police and the excessive use of force. New York: Free Press. Toch, H. (1995). The ‘‘violence-prone’’ police officer. In: W. Geller, & H. Toch (Eds.), And justice for all: under- standing and controlling police abuse of force. Wash- ington, DC: Police Executive Research Forum. Worden, A. P. (1993). The attitudes of women and men in policing: testing conventional and contemporary wis- dom. Criminology, 31, 203–237. Worden, R. E. (1989). Situational and attitudinal explana- tions of police behavior: a theoretical reappraisal and empirical assessment. Law and Society Review, 23, 667–711. Worden, R. E. (1990). A badge and a baccalaureate: poli- cies, hypotheses, and further evidence. Justice Quarterly, 7, 565–592. Worden, R. E. (1995). The ‘‘causes’’ of police brutal- ity: theory and evidence on police use of force. In: W. Geller, & H. Toch (Eds.), And justice for all: understanding and controlling police abuse of force. Washington, DC: Police Executive Research Forum.
S.G. Brandl et al. / Journal of Criminal Justice 29 (2001) 521–529 529