«WWW ’Ty'w" LE 1 ‘ if“: 9 :42 ES?" ‘3 t“f§:?s}h "fﬁ'zi'; In}? 2‘ .il' u'j: WI. 1“ .v i JIPIJﬂ J' . ...::'7. < a" ' “:4 . \ ' . -I.I:5:,7 ': IAIE ‘I o;_. 'l | . I main .; l I." 1.3. 1:... .H' 'Hf": ‘ I‘IIJ M v" “‘5' g": g; ' ‘5 IIIu' "H \ d ”"33”“ ﬁrm! '3 25‘. f'rtuv.”;-‘ f'I‘;II.5 555$." 5 5 5 115-13er 5 ~ 9' Eng! ‘ [’1'4' - i‘ W "in i5 :1 WI"? 42'. 1‘ 5-! ‘1 MM .. 'I~’-1 ‘ '5'?“ 5 l’lh-«ili’h't’ﬂ’ “£- “V" . ‘Hzﬂ. ‘13-? I?“ ~ «345553 54“” . - "5-) . . 4}." .. q ..‘..5.'_:—‘.=:x 25' " " 4’33 .’v I'r. 5 Ha- IT'S; "sffifpﬁ | 5.43345 ifkigsﬁ” - 3d! f’rs: ‘:"‘§2=3¢i #21:; ..‘ " 5 . Jvrf‘np ‘ H .3243. u. «.w ‘ .L . 1- _ u?"- ., W...“ .. a u '3' - ”-2! “- ’"t' .. 3‘. . . Au...— .— .‘.-}.-~;:- 3 .335” '."‘m.-l"‘7t' .. ._.. -2. 1. “33"? a .. . . w , - '3 - > .' .‘w , -* . . '2. 3 ' >~ u.- ‘- “-3..- am... -i , a. nc—n- - . .. '33:???- 2315254 k If}: ’I' ' , III! p.3.1"1"11.- ”E . 1' a2; 4"} A,.15'*I..;f1,'m’.5'rﬁrlz “3,; 'III: Wag. III M 34; fig! .3; akin I‘ v‘- I ”I,“ 'I‘” ‘ ‘ 5" Wm. WM. . “ M3,»...w‘ . 5 ‘ .3.. ~54; I'éI5r am @ﬂﬁw N If”. i .u‘ III" , t" 9" ‘ “D {Fir a??? : I I f . I‘ - 3‘ CI"; f4" E A».- Wilt“ nIM‘goI 'Ehzgﬁ 'M (W 5 . . ‘3 , - .w' _‘ "i" 'V --.. . .211“ '..~'-- .~r. I: ‘3 .u- I , .f! s.” " -.. . ....J. .. ~_;;“...C m~‘t¢——~ Jl, . .‘~.~'._. - ....._. r” 3" ‘ ,.. 7-0- uamw— - , . an I. r a L h. 1‘.--3;: ~...-" . .- JP ._.. .3 1 . r .33, w - .ﬁ,’“‘- ' ~— 31:." o- .. . ﬁr- u..— 7-.- __;..V haw-'5', ale-r n:~:‘;: --x-...- nu- 3:331: I 8 . 31:11.." 5. J35}?! 5555.] K W: "* ‘ )3" I»: '— 3 4,... u. w - 3—.-- o. THE-5‘3 r \‘J Ilﬂlﬂilil‘ljﬂﬂilﬂiiiiimIiilijﬁiiil 1570 66 6 This is to certify that the thesis entitled Departmental Controls of the Use of Excessive Force by the Police presented by Joseph Andrew Schafer has been‘accepted towards fulﬁllment of the requirements for M.S. degreein Criminal Justice Major professor Date 04/15/97 0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution LIBRARY Mlchlgan State Unlverslty ‘ PLACE IN RETURN BOX to remove We checkout from your record. TO AVOID FINES return on or before dete due. DATE DUE DATE DUE DATE DUE ~13U~ 95 19-98 MAGIC 2 933 IOEJZDUI Ulléu‘l L__J|__ M40020 ‘ DEC 3 02003 MAR O 6 2000 hr a. MSU I913! ““ - .- IA rr .. . u. A. . Ir'muuun M1 DEPARTMENTAL CONTROLS OF THE USE OF EXCESSIVE FORCE BY THE POLICE By Joseph Andrew Schafer A THESIS Submitted to Michigan State University in partial fulﬁllment of the requirements for the degree of MASTERS OF SCIENCE School of Criminal Justice 1997 ABSTRACT DEPARTMENTAL CONTROLS OF THE USE OF EXCESSIVE FORCE BY THE POLICE By Joseph Andrew Schafer There is a need for valid and reliable data about the use of force upon which law enforcement executives can base selection, training, and disciplinary decisions. This study attempts to meet this need through a secondary analysis of a nationwide survey of law enforcement agencies. Over 1100 agencies participated in the study, providing information about their policies on the use of force and their experiences with citizen complaints. The statistical analysis indicates that the use of Field Training Ofﬁcers, probationary periods, retraining, and certain types of in-service training has an effect on the rate at which citizen complaints of excessive force are sustained. In memory of Harjit Sidhu 07/22/74 to 12/02/86 I will not forget. iii ACKNOWLEDGMENTS I would like to thank my advisor, Dr. David L. Carter, for your encouragement, patience, advice, and guidance. Additional thanks also go to my committee members, Dr. Steve Mastrofski and Dr. Dennis Payne, who proofed early drafts and provided valuable input and guidance. Also, I must extend thank you’s to Tracy O’Connell, Barb Koons, and Dr. Christina DeJong who helped me make sense of this all. I owe much of where I am to my parents, John and Grace Schafer; your love, support and encouragement has meant more to me than you know. A special thank you to Dr. Clemens Bartollas, who introduced me to the Post- Newtonian Paradigm and encouraged me to seek balance in my life. I am still “keeping ever-lastingly at it”; I hope that I am doing you and RT. Lewis proud. I would be remiss if I failed to thank all of my esteemed colleagues from Poor Richard’s (Aaron, Canon, Chip, Geoff, Grady, Hud, Jody, Joel, Lou, and, of course, Earl)-- you never knew what you were starting. Here’s to life-long friendships, a cornucopia of nuggets, WWBT, and sweet potato pie. Long may you cheese. Finally, I must acknowledge the contributions made by Shelby J o Seegers. You have brought out what was always lurking inside of me. Without your love, patience, and encouragement, none of this would have been possible. Your sacriﬁces in this effort were by far the greatest. Words carmot express what I feel in my heart. Thank you. iv TABLE OF CONTENTS LIST OF TABLES ................................................................................... vii CHAPTER ONE INTRODUCTION ..................................................................................... 1 Legal Precedence ............................................................................. 3 The Problem to be Studied .................................................................. 7 The Purpose of the Study .................................................................... 8 Operational Deﬁnitions ..................................................................... 10- Limitations and Assumptions .............................................................. 12 Summary ..................................................................................... 16 CHAPTER TWO LITERATURE REVIEW ........................................................................... 18 Introduction .................................................................................. 18 The Nature of Police Force ................................................................ 20 Types of Force ............................................................................... 24 The Frequency of Force .................................................................... 26 Non-Deadly Force .................................................................. 27 Deadly Force ........................................................................ 29 Explaining Excessive Force ................................................................ 35 Departmental Controls on the Use of Force ............................................. 38 Minimum Education Standards .................................................. 39 Psychological Testing .............................................................. 41 Community Involvement .......................................................... 44 Investigating and Addressing Use of Force Incidents ......................... 46 The Role of Training.- ............................................................. 47 Probationary Periods and Field Training Programs ........................... 49 Summary ..................................................................................... 50 CHAPTER THREE METHODOLOGY ................................................................................... 52 Research Question ........................................................................... 52 Key Concepts ................................................................................ 52 Data and Sample ............................................................................ 54 Limitations ................................................................................... 60 V Variables to be Used ........................................................................ 62 Hypotheses ................................................................................... 64 Data Analysis Methods ..................................................................... 65 CHAPTER FOUR RESULTS ............................................................................................. 67 Citizen Reporting Options ................................................................. 67 Agency Reporting Requirements .......................................................... 72 Ofﬁcer Demographics ...................................................................... 74 Reported Citizen Complaints, Civil Suits, & Criminal Charges ..................... 77 The Rate of sustained Excessive Force Complaints .................................... 80 Summary of Results ......................................................................... 87 CHAPTER FIVE DISCUSSION ........................................................................................ 89 Discussion of Findings ..................................................................... 89 Policy Implications .......................................................................... 94 Future Research ............................................................................. 97 Summary .................................................................................... 101 BIBLIOGRAPHY .................................................................................. 102 vi Ta l Tal Tab l l { Tabl ) Tab). 1 Table I LIST OF TABLES Table 1 Survey Sampling and Response by Agency Type ................................. 56 Table 2 Responding Law Enforcement Agencies: Agency Type by Number of Sworn Personnel ........................................................................ 57 Table 3 Number and Percentage of Responding Law Enforcement Agencies by Agency Type and Jurisdiction Served ................................................ 58 Table 4 Outcome of Survey Question: “Does your department use any of the following methods to inform citizens about procedures for ﬁling complaints of police misconduct?” ................................................................. 68 Table 5 Outcome of Survey Question: “Which of the following types of assistance does your department provide when citizens ﬁle complaints of police misconduct?” ............................................................................. 68 Table 6 Outcome of Survey Question: “Does your department make complaint forms available for citizens at any of the following locations?” .................. 69 Table 7 Outcome of Survey Question: “Where can citizens ﬁle complaints of police misconduct?” ..................................................................... 70 Table 8 Outcome of Survey Question: “Can citizens ﬁle complaints of police misconduct against ofﬁcers in your department in the following ways?” ....... 71 Table 9 Agency Reporting Policies by Type of Force ...................................... 73 Table 10 Comparison of Gender Distribution Between All Sworn Ofﬁcers, Ofﬁcers with Citizen Complaints, and Ofﬁcers with Sustained Complaints .............. 74 Table 11 Comparison of Race/Ethnicity Distribution Between All Sworn Ofﬁcers, Ofﬁcers with Citizen Complaints, and Ofﬁcers with Sustained Complaints ............................................................................... 75 vii T; T2 Ta Ta Tat Tab Table 12 Comparison of Level of Education Distribution Between All Sworn Ofﬁcers, Ofﬁcers with Citizen Complaints, and Ofﬁcers with Sustained Complaints ................................................................................... 76 Table 13 Civil Suits and Criminal Charges Brought Against the Responding Agencies for Excessive Force Incidents (1991) ......................................... 78 Table 14 Number of Civil Suits and Criminal Charges Filed by Agency Type ............. 79 Table 15 Number of Civil Suits and Criminal Charges Filed by Jurisdiction Population .................................................................................... 79 Table 16 Results of T-Tests: Dependent Variable (rate of sustained use of force complaints per 100 ofﬁcers) by Various Independent Variables ..................... 82 Table 17 Results of ANOVA’s: Dependent Variable (rate of sustained use of force complaints per 100 ofﬁcers) by Various Independent Variables ..................... 84 Table 18 Results of AN OVA’s: Dependent Variable (rate of sustained use of force complaints per 100 ofﬁcers) by Various Independent Variables (types of in- service training) ............................................................................. 85 viii CHAPTER ONE INTRODUCTION Every year American law enforcement ofﬁcers, law enforcement agencies, and various units of government are the subjects of hundreds of lawsuits. Some of these claims are legitimate and arise from a tragic incident between the police and citizens. Other claims are comical and frivolous, such as the examples del Carmen (1991) cited: * A convict whose sentence was increased for escaping sued the county and its sheriff, accusing them of negligence in failing to prevent the . escape. * A man sued the state of Pennsylvania for $5 quadrillion, saying that suspension of his driver's license by the Transportation Department ruined his life and reputation. * A man sought revocation of the city charter, $15 million, and the right to combat police ofﬁcers to death. However comical these examples may be, litigation brought against the police is no laughing matter. Disposing of a frivolous claim still costs an agency money and disposing of a legitimate claims may require the agency to pay huge damage awards or out-of-court settlements. Furthermore, there are no disincentives to discourage citizens from ﬁling frivolous claims against government agencies. Plaintiffs who bring cases that clearly lack merit are subjected to judicial sanctions in less than one half of one percent of all law 2 enforcement-related civil claims decided by federal district judges (Kappeler, 1993). The real cost of liability lawsuits brought against the police can not be easily determined and has changed signiﬁcantly in the past three decades. It has been reported that in 1992 the average jury award in liability lawsuits ﬁled against municipalities was $230,000; by 1985, this average had risen to $2 million. Prior to 1966, only one lawsuit brought against a municipality had ever resulted in a jury award in excess of $1 million; by 1983, more the 350 such judgements had been made (Cisneros, 1986). More recently, Kappeler, Kappeler, & del Carmen (1993) found the average award in federal district court cases resulting in judgments against the police was over $120,000. However, in this study only about 8 percent (112 cases) of 1,359 cases examined resulted in a monetary award against the police. Meadows and Trostle (1988) reported similar ﬁndings for litigation against the Los Angeles Police Department. The average award in their study was $158,500, excluding court costs and attorney fees. Conﬂicting results have been reported in other studies, which have found that most liability judgements result in much smaller awards, with an average award of between $2,500 and $6,000 (del Carmen, 1991; Lant, Small & Wohlberg, 1979; Schmidt, 1976). The vast differences between the above studies may be a result of differing methodologies employed by their researchers. The true cost of liability cases is difﬁcult to assess accurately and basing ﬁndings on court rulings is inherently limiting because a large proportion of cases are settled privately. Law enforcement agencies must be aware of the risks they face in terms of the civil liabilities their employees may create and they must seek out ways to protect themselves from such claims. LEGAL PRECEDENCE The Civil Rights Act of 1871 was designed to provide a legal remedy for citizens whose constitutional rights had been violated. This Act has since been codiﬁed as Title 42 of the United States Code. Suits brought under this legislation are generally referred to as Section 1983 actions because Section I of the Civil Rights Act of 1871 provided a remedy for the deprivation of civil rights. This section was later codiﬁed as Section 1983 (Kappeler, 1993). Title 42 U.S.C. 1983 provides that: Every person who, under color of any statute, ordinance, regulation, custom, or usage, or any State... subjects any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights secured by the Constitution shall be liable to the party injured in an action at law... According to Davis (1993), three requirements must be met before a claim can be brought under Section 1983. First, the plaintiff must allege the deprivation of a right secured either by the Constitution or another law of the United States. Second, the defendant must be a "person" within the meaning of Section 1983. Finally, the defendant's action must have been under the color of law. After being ignored for nearly a century, Section 1983 has become one of the most common routes used to bring cases into the federal court system (second only to habeas corpus claims). In the early days of the civil rights era, Section 1983 was brought back to life by the United States Supreme Court (Barrineau, 1994). The cases that resurrected this statute was Monroe v. Pope (1961). The plaintiffs in Monroe (1961) sued 13 Chicago police ofﬁcers who allegedly broke into their residence without a warrant, forced them out of bed at gun point, 4 ransacked their house, subjected the family to verbal and physical abuse, detained the father at the police station for 10 hours, and then released him without ﬁling any charges. In the suit, Monroe alleged that the actions taken by the police constituted a deprivation "under color of law, ” of his constitutional rights. The Supreme Court agreed with this claim and held that Section 1983 was the appropriate vehicle for bringing similar claims to light. However, the Court also held that governmental units (e.g. the City of Chicago) were immune from liability under Section 1983. Monroe (1961) opened the door for 1983 suits, but awards were typically very small because police ofﬁcers are generally paid modest salaries. For the next 16 years, the court system continued to grant this immunity to municipalities (Barrineau, 1994). In 1978, the Supreme Court overruled the portion of the Monroe decision that provided municipal immunity. In 1971 , a group of New York City school teachers had brought a Section 1983 claim against the Department of Social Services, its Commissioner, the Board of Education, its Chancellor, the City of New York, and the Mayor of New York. The teachers sought injunctive relief and back pay for unpaid leaves of absence which they had been forced to take during their pregnancies. The Supreme Court handed down their decision in this case, Monet! v. Department of Social Services, in 1978. In Monell, the Court reversed its prior decision, ﬁnding that governmental units could be held liable for constitutional deprivations inﬂicted by their employees. This liability was contingent upon the plaintiff 's ability to establish that the deprivation had resulted from the ofﬁcial policy or custom of the unit of government. This ruling, in conjunction with the Civil Rights Attorney's Fees Awards 5 Act of 1976 (Title 42 U.S.C. Section 1988), dramatically altered the landscape of Section 1983 claims. Over night, suing law enforcement ofﬁcers, jail/detention personnel, and other criminal justice system employees became a potentially lucrative enterprise. Three cases stand out as landmarks which have structured the use of force by the police. Until the substantive due process standard was articulated in 1973 (Brown, 1991), most ofﬁcers had to rely on inadequate and inconsistent departmental policies and state guidelines as they exercised their right to use force (del Carmen, 1991). The substantive due process standard was articulated by the Second Circuit in Johnson v. Glick (1973). The plaintiff in Johnson claimed that he had been struck by a prison guard. In reviewing this claim, the Court of Appeals determined that, although the plaintiff had stated a claim under Section 1983, neither the Fourth Amendment’s prohibition against unreasonable searches and seizures, nor the Eight Amendment’s prohibition against cruel and unusual punishment, appeared to be applicable to the case. The majority opinion in Johnson (1973) concluded that “application of undue force by law enforcement ofﬁcers deprives a suspect of liberty without due process of law.” Four factors were articulated to guide substantive due process inquiry in excessive force claims. These factors were: 1. The need for the application of force 2. The relationship between the need and the amount of force that was used 3. The extent of injury inﬂicted 4. Whether force was applied in a good faith effort to maintain or restore discipline or maliciously and sadistically for the very purpose of causing harm. These factors were adopted by many of the Circuits and remained in full effect for twelve years. The Supreme Court’s 1985 decision in Tennessee v. Garner prompted many Circuits to re-examine their use of the substantive due process standard (Brown, 1991). In Garner (1985), two police ofﬁcers shot and killed a 15 year old, unarmed burglar. Garner had been observed ﬂeeing from the scene of a crime and had failed to stop when ordered to do so by a police ofﬁcer. The ofﬁcer who ﬁred the fatal shot stated he did not believe that Garner was armed. Under the existing Tennessee statute, the ofﬁcer's actions were legally justiﬁable. The Supreme Court concluded that the use of force was not justiﬁed under these circumstances because Garner posed no threat to the ofﬁcer or to others. In Garner (1985), the Court also reasoned that the use of force is governed by the Fourth Amendment and that this use of force must be reasonable (Brown, 1991). The majority stated that "there can be no question that apprehension by the use of deadly force is a seizure." Claims involving deadly force that fail to establish an underlying Fourth Amendment violation supersedes issues of qualiﬁed immunity and municipal liability. Despite the fact that it narrowed the range of cases in which the police might use deadly force, Garner left many unresolved issues that are still being determined in the lower courts. According to Hall (1988), three elements can be deduced from the opinion of the Court in Garner (1985). The use of deadly force may be justiﬁed if either: a.) the suspect must threaten the ofﬁcer with a weapon, or, b.) the ofﬁcer must have probable cause to believe the suspect has committed a crime involving the inﬂiction or threatened inﬂiction 7 of serious harm; when feasible, the ofﬁcer should also give warning. Some of the vagueness that shrouded the Gamer (1985) ruling was clariﬁed by the Supreme Court in Graham v. Connor (1989). The Graham ruling articulated which test shall be used in determining civil liability under Section 1983 in use of force cases. The Supreme Court established that such claims must be analyzed and judged under the Forth Amendment's "objective reasonableness" standard. That is, the reasonableness of a particular use of force should be judged from the perspective of a reasonable ofﬁcer and not with the clarity of hindsight. In other words, an ofﬁcer's actions are to be judged in light of the information the ofﬁcer possessed and in the context of the situation. THE PROBLEM TO BE STUDIED Police executives need valid and reliable data for selecting and training ofﬁcers, and making disciplinary decisions in order to protect their agency from the potential liabilities their employees may create. This study will examine the use of force by the police. The issue for this study is what effect certain hiring practices and other departmental factors have on the rate of sustained police misconduct complaints that departments experience. The effect these practices may have on the rate of sustained use of force complaints will be considered in terms of variables such as the use of pre-service psychological screening test, the use of ﬁeld training ofﬁcers, and variables relating to the procedures agencies use to investigate complaints of police misconduct. Conclusions will be made by measuring the relationship between these multiple factors and the rate of complaints reported by departments in a nation-wide study (Pate & Fridell, 1993). 8 Utilizing the results of this research, a law enforcement agency may be able to improve their methods for the selection and training of employees in order to limit their liability for employee misconduct and better serve their community. Three central objectives will be of paramount interest. First, the study will examine whether or not there are signiﬁcant differences in this rate of sustained force complaints based upon the selection and training procedures departments utilize (e. g. the use of pre-service psychological screening protocols or the use of Field Training Ofﬁcers). Second, the study will examine whether or not there are signiﬁcant differences in the rate of sustained force complaints based upon the ways in which agencies receive and process citizen complaints of police misconduct (e.g. the number of ways in which citizens can ﬁle complaints or requiring medical examinations for citizens complaining of excessive force). Finally, the study will examine whether various types of in-service training have any impact upon the rate of sustained use of force complaints that agencies experience (e.g. ﬁrearms requaliﬁcation or ofﬁcer survival tactics). THE PURPOSE OF THE STUDY The purpose of this study is to identify factors related to excessive force incidents which may be translated into organizational policy or practice to reduce the frequency rate with which police ofﬁcers use such force. The rate of sustained use of force complaints is one of many measures which could be used to examine the overall use of force by the police. It is composed of those incidents in which police ofﬁcers were deemed to have used an excessive amount of physical force against a citizen. The rate establishes a ' 9 baseline measure of how often police ofﬁcers use excessive force. The rate of sustained complaints does not capture all of the incidents in which police ofﬁcers use excessive force. The rate is only composed of those force incidents which are reported to the police, supported by sufﬁcient evidence to establish some degree of “guilt”, and ultimately deemed to be excessive by an investigatory body. Using the rate of sustained complaints as a measure of the actual use of excessive force requires the assumption that incidents are reported, investigated, and adjudicated with honesty and integrity. Despite these limitations, the rate of sustained force complaints may still be viewed as a strong measure of the actual use of excessive force. Other measures fail to account for such a large mrmber of force incidents or require observers to make subjective determinations of the reasonableness of a force incident. When examining data from a large number of agencies the rate of sustained force complaints is the best measure for approximating the use of excessive force (which is still amorphous to social science researchers and law enforcement executives). This study will consider the rate of sustained complaints in the context of independent variables relating to potential departmental controls on the use of force, including the use of pre-service psychological screening, the use of ﬁeld training ofﬁcers, and the procedures used in the investigation of use of force complaints. The results of this research could hold important implications for law enforcement administrators and human resource personnel. Claims ﬁled against police ofﬁcers and their employers are becoming increasingly costly. Even if an ofﬁcer and/ or agency is 10 absolved of any wrong—doing, the legal costs of simply being exonerated can be very high. In order to avoid these expenses it is important that agencies carefully select and train their employees. The three factors being considered are possible mechanisms which might assist agencies in controlling the use of excessive force by their employees. This, in turn, may lower the likelihood that agencies will suffer some form of legal repercussion for the conduct of their employees. In addition to these key devices, other potential departmental controls on the use of force will also be examined in less depth. OPERATIONAL DEFINITIONS The following are the operational deﬁnitions for some terms used in this study. Select terms as deﬁned by Pate and Fridell (1993) are: W: A police ofﬁcer’s commission of criminal offense; neglect of duty; violation of Departmental policies, rules, or procedures; conduct which may tend to reﬂect unfavorably upon the employee or the Department. The six types of misconduct that this study examines are: 1. Excessive/undue/unnecessary use of force; brutality (including use of weapons, cuffs, etc.) 2. False/unlawful arrest! imprisonment, improper detention/ interrogation (false charges ﬁled) Illegal/unlawful search or seizure Harassment, intimidation, threats, verbal abuse Abuse/ misuse of authority (conduct unbecoming an ofﬁcer) Improper language (demeaning, ethnic/ racial slur) 9‘99?” WM: An allegation of misconduct that is received from any source. W: The allegation of misconduct was ﬁled by either sworn or non-swom personnel of the law enforcement agency. 1.1m: The complaint was not based on facts as shown by the investigation, or the reported incident did not occur. 11 Exam: The incident occurred, but the action taken by the ofﬁcer(s) was deemed lawful and proper. W: The allegation was not supported by sufﬁcient evidence to justify disciplinary action against the ofﬁcer(s). Sustained: The allegation was supported by sufﬁcient evidence to justify disciplinary action against the ofﬁcer(s). Other terms of importance to this study include: Wm: The act of using a mental health professional and/or one of several conventionally accepted testing instruments in an attempt to assess the psychological well-being and/or the personality attributes of an applicant. Wm: A governmental unit charged with the routine enforcement of criminal laws, responding to calls for service, making arrests, serving search warrants, and maintaining civil order. W: The legitimate or illegitimate utilization of a law enforcement ofﬁcers right to seize a person or property through the employment of persuasion, negotiation, coercion, or physical methods (e.g. restraint holds, impact weapons, chemical weapons, ﬁrearms, etc.). W: The level of psychological or physical control required to seize a person or property for legitimate law enforcement purposes; the level will vary with the level of resistance encountered. W: Occurs when an ofﬁcer makes a good faith effort to seize control of a person or property and exceeds the level of control required to overcome resistance being encountered because the ofﬁcer lacks the training or tools to use a lower level of force. W: Any force that does, or reasonably could, result in death or serious bodily injury. W: The malicious and intentional use of more physical force than is reasonably necessary in seizing or detaining an individual. 12 LIMITATIONS AND ASSUMPTIONS Like so many other aspects of the criminal justice system, the police use of force is amorphous. There is little reliable data on the frequency with which police use non- deadly force. Departmental reporting requirements vary across jurisdictions. In one jurisdiction an ofﬁcer may have to report any level of physical contact (such as "come- along" and restraint holds). An ofﬁcer in a neighboring jurisdiction may not have to report using any force, short of shooting a citizen (Pate & Fridell, 1993). Although reporting practices have improved with time it is still difﬁcult to obtain a clear picture of the rate with which police use various levels of force. At least one study has found that the number of incidents in which police ofﬁcers caused the death of citizens has increased with time (Kobler, 1975b). However, this increase may be attributed to changes in reporting practices, increases in the population, or the unknown methodologies used to classify deaths in the study. Researchers have had mixed results in exploring the use of force (both deadly and non-deadly) by the police. There is no nationwide mechanism that requires police agencies to report the use of force by their employees (Blumberg, 1994). Because of this researchers must base conclusions on data that are incomplete, inaccurate, and, usually, derived from an unrepresentative sample of agencies. Studies relying on a small sample of agencies often have conﬂicting results because of the considerable variation in the rate of police use of force between various jurisdictions (Sherman & Cohn, 1986). It is easy to be mislead by statistics reporting the rates of police shootings. For example, during a ﬁve-year period New York City reported almost 3,000 incidents of 13 police-involved shootings; in a similar time ﬁame, Memphis reported 225 such incidents. When the Mean Annual Police Shooting Rate (per 1,000 ofﬁcers) were calculated, Memphis had a signiﬁcantly higher rate (33.5) than did New York (19.6). The Police Shooting Rate (per 1,000 violent felony arrests) established an even greater disparity between Memphis (56.98) and New York (16.71) (Fyfe, 1982). This illustrates how statistics related to police-involved shootings can be easily misinterpreted if not viewed in the proper context. Great care must be exercised in the analysis and interpretation of the data set used in this study, and other data sets relating to the use of force by the police. It is important to point out that the unit of analysis for this data set is the individual department, not the individual police misconduct complaint or the individual ofﬁcer. Demographic variables were aggregated for each department so accurate conclusions cannot be made about individual ofﬁcers or individual complaints. This study can not (and will not) attempt to examine the role an ofﬁcers age, gender, race/ethnicity, or level of education may play in their use of force, nor can it (or will it) examine the role a citizen’s race/ethnicity may have upon the likelihood of being the target of excessive force. In addition to problems with the data collection instrument, there are a number of potential errors that could be made on the part of the responding party. First, it is necessary to assume that the statistics reported by the responding departments were accurate. Unfortunately, the survey instrument may not have been as clear to the responding ofﬁcers as it was to the researchers responsible for its development. There are several parts of the survey that ask for demographic information 14 (including race, gender, average age, and level of education) for various groups of departmental employees. The original study solicited information about the aggregate characteristics of ofﬁcers. Because of this aggregation, it is not possible to irnpute a particular complaint to a particular ofﬁcer in order to make determinations about how ofﬁcer characteristics may affect the use of force. This limits the use of demographic information to descriptive purposes. In terms of potential errors it should be noted that respondents were asked to provide different aggregations of demographic information. Employees were grouped together as “all sworn personnel”, “ofﬁcers with citizen complaints”, and “ofﬁcers with sustained complaints.” With each layer of separation a new possibility of an error being made by the respondent surfaces. Either due to error, carelessness, or indifference, the subject ﬁlling out the survey may have given inaccurate statistics or made mathematical errors. Another potential problem with this data set is that it is necessary to assume that departments are utilizing a complaint investigation system that is both uniform and honest. It is doubtful that the investigation process is consistent between different agencies; some sources of variation might include agency size, agency type, the use of a citizen review board, and the presence of a formalized Internal Affairs Division. Different agencies are also likely to have varying deﬁnitions of what constitutes reasonable and acceptable behavior by their ofﬁcers in detaining citizens and effecting arrests. Furthermore, the review process in some agencies may simply perpetuate the existence of departmental bias and the “good old boy’s” system. 15 For example, Ofﬁcer X and Ofﬁcer Y (both white males) may be accused of using similar degrees of excessive force in effecting an arrest on a black male (for the sake of this example, assume they both intentionally kicked the subject after he had been handcuffed). Ofﬁcer X is employed by Any town P.D. , which takes citizen complaints seriously and has a citizen review board as a component of its investigative process. Ofﬁcer Y is employed by the Someville County Sheriff’s Department, which has a “conﬁdential" review process for citizen complaints. Ofﬁcer X might be suspended for a few days, while Ofﬁcer Y might only be told to “be more careful next time.” If these two incidents were reported to the authors of this study, they would have been coded in two distinctly different ways, even though the circumstances were approximately similar. Another possible source of error is that smaller agencies may not have the resources to devote to internal affairs in contrast to the budgets of larger agencies. This could work both for or against an ofﬁcer. On the one hand, lack of funds for internal affairs investigations in smaller agencies (which composed a sizeable segment of this sample) may result in a department failing to ﬁnd evidence of an ofﬁcer’s misconduct. On the other hand, this same lack of funding might fail to shed light upon evidence that might have exonerated an otherwise seemingly guilty ofﬁcer. The quality of investigations may also vary signiﬁcantly based both on the general mood of a department (either taking complaints seriously or backing ofﬁcers no matter what) and based on the available resources agencies have to investigate complaints. A ﬁnal area of concern is the failure of the original research team to provide clear instructions for the classiﬁcation of police misconduct complaints. The original authors 16 failed to deﬁne adequately what might constitute the various forms of misconduct (i.e. excessive force, verbal abuse, unlawful search, etc.). They also failed to state how a single incident with multiple forms of misconduct (for example, a use of excessive force accompanied by threatening or intimidating statements) should be classiﬁed. Some agencies may have only counted this as a single complaint, while others may have counted it as two complaints. The original researchers failed to provide instruction to the agencies on which method was preferred for the purposes of this study. In utilizing any agency-reported use of force data, there are many assumptions that must be made which may taint the resulting analysis. Unfortunately, there is only so much that researchers may do to control for these disparate investigatory practices. It must be accepted that any data obtained from different agencies is subject to this potential bias and this fact must be recognized in interpreting said data. SUMMARY The use of improper force by police ofﬁcers can create a costly liability for their employing agency. Because of this agencies must actively seek out ways in which they can protect themselves (and the communities they serve) from this liability. This study will analyze data from a sample of law enforcement agencies to identify what factors may provide such protection. The validity and reliability of this study’s data may be threatened by a number of potential errors. In accepting the ﬁndings of this analysis it is necessary to assume that respondents provided data that was free of intentional and/ or accidental errors, that 17 agencies had approximately uniform disciplinary procedures, and that respondents were able to comprehend the instructions provided by the research staff. Despite these limitations it is still possible to utilize this data set for the purposes of analysis. It can be assumed that the impact of this threats has been somewhat “diluted” by the large sample size. These threats do not nullify this data set; instead, they require that one proceed with caution in subsequent analysis and interpretation. CHAPTER TWO LITERATURE REVIEW INTRODUCTION Few issues ignite as much public passion as the use of force and charges of "police brutality" (Brave & Peters, 1993). The videotaped beatings of Rodney King and a group of illegal aliens on a California freeway have helped cement this passion in the public’s perception. It would seem that many citizens think that the use of physical force by the police is an integral part of their daily duties. But what do we really know about the use of force? How often do police ofﬁcer actually use force in the performance of their duties and how do they decide to use force? Have social scientists been able to determine accurately the frequency with which police ofﬁcers use force, and how have they done so? What can police executives do to establish some degree of control over the use of force by their employees? Before conducting an analysis of data relating to the use of force it is important to examine what is meant by the term “use of force." This chapter will explore the use of force (both deadly and non-deadly) by the police. First, this chapter will examine what the term “force” means in the police context. It is important to understand how the police ofﬁcer’s right to use physical force may signiﬁcantly inﬂuences the nature of the police- citizen relationship. A second focus of this chapter will be an examination of the different 18 19 types of force that the police may use in various situations. Police ofﬁcers have a number of different lethal and less-than-lethal force mechanisms that they may choose to use to gain citizen compliance. These mechanisms go beyond weapons sanctioned by law enforcement agencies to include the ofﬁcers person and seemingly harmless equipment that may be used as an ad hoc weapon by an ofﬁcer in an emergency situation. Third, this chapter will review literature which has attempted to assess the rate with which force, both excessive and acceptable, is used by the police. Before presenting the ﬁndings of this research study, it is critical that the reader appreciate how difﬁcult it may be to determine the true rate with which the police use force in the course of performing their duties. Different studies have determined such rates using various methodologies. These methods all have their strengths and weaknesses, but many of them yield an end product that can be highly misleading because of the context in which force is presented. A fourth consideration of this chapter will be the various explanations for why force is used (especially excessive force). Researchers have proposed that ofﬁcers possessing certain personality traits, demographic characteristics, or personal habits are more likely to use excessive and/or deadly force. Finally, literature discussing administrative concerns and controls will be reviewed in an attempt to assess what agencies may do to help ensure that their ofﬁcers use force within the context of the law and departmental policies. These controls will be of key importance in the analysis portion of this study. 20 THE NATURE OF POLICE FORCE CPolice ofﬁcers in modern society are entrusted with a unique level of power. The police are the only representatives of the government who are legally permitted to use some degree of physical force against citizens in the ordinary course of performing their duties (Kobler, 19751an0 other government agency is entrusted with this power; others must rely upon requests, persuasion, coercion, and legal means to compel citizens to comply with the law. Building inspectors, social workers, and tax auditors must resort to psychological tactics to compel an un-cooperative citizen to obey the law or their directives. A police ofﬁcer has the right to use physical force (within limitations) to secure citizen compliance. Because police ofﬁcers have the authority to use force the possibility of abuse will always exist (Harper, 1983; Papst, 1993). In a classic monograph on the police, Bittner (1970) describes that "the role of the police is best understood as a mechanism for the distribution of non-negotiably coercive force. " Citizens recognize that one of the underlying characteristics of a police ofﬁcer is the ofﬁcer’s implied right to exercise the use of force to obtain compliance. Bittner suggests that it is this implied right to use force that compels citizens to comply with police directives. This is not to say that the police always obtain compliance or solve problems through the use of force. Rather, police work consists of obtaining compliance and solving problems in the context of situations or encounters in which force may need to be used (Bittner, 1974). The power relationship between police ofﬁcers and the public is predicated upon the unquestioned authority of the ofﬁcer. The public expects the police to enter a 21 situation, seize control, and infuse order where there was confusion and chaos. When an ofﬁcer is confronted with an individual who is challenging his or her authority, the ofﬁcer may feel the need to re—establish the fact that he or she is in control of the situation (Reiss, 197 3). One way in which an ofﬁcer can re-establish his/her authority is to exercise the use of force (either physical or non-physical). The use of force to re-establish authority is not always a random or capricious behavior on the part of the police. Van Maanen (1978) examined this dynamic in depth and identiﬁed a process by which the police decide when, where, how, and against whom they might abuse their authority. Van Maanen found that police ofﬁcers use two questions in determining how to classify citizens they encounter. Ofﬁcers decide whether a citizen know what he/ she is doing and whether the person could act differently under the circumstances of the situation. If the ofﬁcer decides that the answer to both of these questions is “yes”, the he/she is most likely to use force or other extralegal tactics to castigate the citizen. Hunt (1985) identiﬁed two ways in which police ofﬁcers rationalize their use of force: excuses and justiﬁcations. One of the most important steps in the socialization of a rookie police ofﬁcer is learning to rationalize the use of force. By making these rationalizations, ofﬁcers "neutralize" (Sykes & Matza, 1957) their behavior to eliminate any sense of immorality or guilt that they might otherwise experience. Just as some observers have noted that ofﬁcers use humor to distance themselves emotionally from their jobs (Pogrebin & Poole, 1988), Hunt believes that ofﬁcers blame outside forces for "making" them use force so that they can establish this same psychological distance. 22 When ofﬁcers use an excuse to rationalize an act of force they are denying responsibility. For example, an ofﬁcer might blame the use of inappropriate force on the adrenaline rush of a high speed pursuit; the ﬂeeing subject brought the force upon himself because he chose to run from the police. In using such a rationalization the ofﬁcer is excusing his or her behavior by placing the onus upon the offender for compelling them to lose control of their emotions (essentially arguing that another person has more control of their emotional response then they themselves possess). On the other hand, ofﬁcers might attempt to justify their use of force because they perceived that their authority was being physically or symbolically threatened or violated. For example, situational justiﬁcation might occur when an ofﬁcer is making an arrest at a protest and is spat on by the handcuffed subject he/she is taking into custody. The ofﬁcer might perceive this as an affront to his/her authority and retaliate by striking the offender. This act would be seen by most ofﬁcers as being situationally justiﬁed because the ofﬁcer's authority was violated (Hunt, 1985). Social scientists commonly refer to the “police subculture.” This term refers to the set of values, beliefs and behaviors that are speciﬁc to the police and their occupational environment. The police are seen as alienating themselves from the non-police world. This alienation results in the formation of values and beliefs that are unique to police ofﬁcers; members of the non-police world are not perceived as being capable of understanding the world of the police (Meredith, 1984). The police subculture may embrace the use of force in situations where there is a perceived threat to the ofﬁcer, or the ofﬁcer’s authority (Pollock-Byrne, 1989). 23 An important question that this discussion raises is whether or not police ofﬁcers support and/or cover-up the use of force by their fellow ofﬁcers, even if they believe that the force used was excessive. Although Hunt (1985) does not address this directly, it is implied that while ofﬁcers may privately condemn such actions, they would never make such an admission to those outside of the police world. Similar conclusions were reached based on the results of ofﬁcer surveys (Brink & Lester, 1984; Lester & Brink, 1985). The survey responses in these studies indicated that ofﬁcers felt a sense of peer loyalty and the presence of a willingness to cover up for the misconduct of fellow ofﬁcers. The authors concluded that morality is a general attitude. Ofﬁcers who indicated a willingness to report the use of excessive force by a co—worker appeared to have high moral standards by also reporting a willingness to report the use of other forms of misconduct. In a comparison of the attitudes of police ofﬁcers and ﬁre ﬁghters towards the use of force by the police (Brodsky & Williamson, 1985), ﬁre ﬁghters were found to be signiﬁcantly more accepting of the use of excessive force. In light of other studies that have indicated that ofﬁcers a willing to cover-up the misdeeds of their co-workers, these results are intriguing. Fire ﬁghters (who the authors proposed may be viewed as representing the attitudes and outlooks of the general public) seemed to be more willing to condone the use of force than were police ofﬁcers, who are actually charged with using such force. However, the authors were quick to note that attitudes (the reported view of excessive force) are not necessarily the same as behaviors (the actual use of excessive force). When asked their opinion of the Rodney King incident (before the acquittal verdict 24 of 1992), 21% of a sample police ofﬁcers felt that King " got what he deserved, " while 59% believed that the ofﬁcers involved should be imprisoned (Lester and Arcuri, 1994). It is interesting to note that those ofﬁcer who supported the beating of King endorsed other additional items that indicated that they had a greater propensity towards abusing suspects. Support for the ofﬁcers in the King incident was also associated with the endorsement of items about the excitement, good pay, and power associated with their job. TYPES OF FORCE One of the clearest symbols of the police is the presence of a sidearm. Both the “old west” and the modern day sheriff share the “tin star" and ﬁrearm as symbols of their authority and power. In addition to ﬁrearms, there are many “non-lethal” weapons used by modern police ofﬁcers for both defensive and offensive reasons. These weapons are non-lethal if used properly, but many of them have the potential to cause death if misused. Perhaps the most common non-lethal weapon law enforcement agencies make available to their employees is the impact weapon (such as batons). A survey of police and sheriff agencies revealed that 93% of the responding organizations made at least one type of impact weapon available to their ofﬁcers (McEwen & Leahy, 1994). If properly used, an impact weapon can be a very effective tool against an armed and/or aggressive subject. Unfortunately, impact weapons can easily be misused or accidentally miss-aimed, resulting in severe damage or death (Cox, Buchholz, & Wolf, 1987). It is not uncommon for police ofﬁcers to carry a large metal ﬂashlight to provide 25 illumination.(Cox, Buchholz & Wolf, 1987; Cox, Faughn & Nixon, 1985). Some ofﬁcers carry a ﬂashlight because it may also be used as an alternative impact weapon. However, blows from a ﬂashlight are likely to fracture a bone, a particularly dangerous possibility if directed toward the head or shoulder (Cox, Faughn, & Nixon, 1985). The danger of using metal ﬂashlights as offensive or defensive weapons is compounded by the fact that few departments provide formal training to ensure that they are used in a manner designed to minimize the possibility of injury. These researchers have recommended that police managers should be aware of the heightemd potential for injury that may result from blows delivered with metal ﬂashlights. Chemical weapons are a common form of non-lethal weapon that are issued by over 70 percent of law enforcement agencies (McEwen & Leahy, 1994). In recent years many agencies have adopted “pepper spray" as their chemical weapon of choice (Hunter, 1994). These sprays are organically based inﬂammatory agents that mix small amounts of cayenne pepper derivatives with inert agents and propellents. Pepper spray results in tearing of the eyes, restricted breathing, and a strong burning sensation on exposed skin; the effects generally last 20 to 45 minutes (Hunter, 1994). Chemical agents allow ofﬁcers to temporarily incapacitate subjects without the risks of long-tenn injury or A less common type of non-lethal is the stun or “taser” gun. These electrical weapons are designed to deliver an electrical shock that results in a suspect being temporarily incapacitated (Vaughn & Ramirez, 1992). Many agencies have been reluctant to introduce electrical weapons into their arsenals and only 16 percent of the law enforcement agencies report the use of these devices (McEwen & Leahy, 1994). One 26 reason for this reluctance may be the high cost of training and heightened liability electrical weapons create. In addition, the electric shock these weapons deliver may be sufﬁcient to result in the death of a subject with a pre-existing susceptibility to heart attack (Vaughn & Ramirez, 1992). Some recent research has suggested that a substantial number of incidents involving police use of excessive force involve ofﬁcers who are untrained, under trained, or given insufficient policies. Some restraint holds, if improperly applied, may result in death due to positional asphyxiation; handcuffs that are too tight can cause permanent injury (Brave & Peters, 1993). In the early 1980's, sixteen black offenders died at the hands of the Los Angeles Police Department in separate incidents where ofﬁcers improperly applied carotid restraints (choke holds) (Anderson, 1995; Chemerinsky, 1991; Williams, 1993). While carotid restraints can be highly effective, many ofﬁcers do not have the training to apply them properly to avoid the risk of causing death (Creighton, 1983; Reay and Mathers, 1983). THE FREQUENCY OF FORCE After the Garner ( 1985) decision law enforcement agencies and their ofﬁcers had a clariﬁed understanding of when the use of deadly force was justiﬁed. The Supreme Court's actions in the Graham (1989) case were less decisive in deﬁning the appropriate application of non-lethal force; the " objective reasonableness" standard is not as concrete (Alpert & Smith, 1994). Non-deadly force also covers a much broader range of activities, from restraint holds, to the use of impact weapons, to the use of chemical 27 irritants. In the course of any given shift, it is much more probable that a police ofﬁcer will use non-lethal force than lethal force (Williams, 1993). Because of these two factors, when viewed collectively the liability potential of non-lethal force situations may be greater than the liability potential of lethal force situations. With over 17,000 independent law enforcement agencies enforcing various federal, state, and local laws on a daily basis it can be challenging to gather accurate information about the activities of these organizations and their employees. The Uniform Crime Report attempts to tabulate the number of crimes that occur within various jurisdictions. Obtaining accurate crime data through this system is simply not realistic and the resulting ﬁgures are little more than “best guesses. " Attempting to obtain national data on the rate with which ofﬁcers utilize different degrees would be equally, if not more, challenging. Due to the extremely sensitive nature of the subject it is not realistic to expect honest self- reporting from ofﬁcers about their use of force (McEwen, 1996). MW Various means can be used to determine the rate with which the police use physical force against citizens. In a 1966 survey of Denver residents, Bayley and Mendelsohn (1969) found that 4 percent of white respondents, 9 percent of black respondents, and 15 percent of Hispanic respondents indicated that they had personally experienced police brutality. More recently, a Gallup poll1 asked a sample of citizens “Have you ever been physically mistreated or abused by the police?” Approximately 5 1 The Gallup Poll, March 20, 1991. 28 percent of the respondents (9 percent of the non—white respondents) claimed to have been mistreated or abused by the police. It is crucial to note that some members of the sample may have misinterpreted what was meant by “mistreated or abused." There is no way of knowing how their perceptions of abuse compare with speciﬁc legal standards. Another way to estimate the rate with which police use force is by conducting systematic observations of ofﬁcers at work. Using this methodology, it has been estimated that the rate with which police use excessive force in citizen contacts varies from 0.30 % of the time (Worden, 1995), to 0.97 % of the time (Reiss, 1973), to 5.1% of the time (Friedrich, 1980). Observation studies using slightly different methodologies have found that police ofﬁcers used some degree of physical force in 7.9 percent (Bayley & Garofalo, 1989) to 12 percent (Fyfe, 1989) of all police-citizen encounters seen as having the potential for violence. The accuracy of the latter studies is vague because of the author's failure to adequately deﬁne that makes a situation "potentially violent. " Observation studies are limited in several respects. Most importantly, police ofﬁcers may not behave in a “normal" manner when they know they are being watched. Depending on the motivations of the individual ofﬁcer they may be less likely to use force because they know they are being watched or they may be more likely to use force in an attempt to impress the observer. Also, observers may not fully understand legal and policy constraints that might deﬁne what level of force is “excessive" and what level of force is “necessary” to effect an arrest. While some uses of force may appear particularly brutal to a civilian observer, they may be necessary and justiﬁable from the perspective of the police. 29 A third way in which researchers have estimated the rate with which police use force is through self-report. In self-report studies ofﬁcers are asked to complete a brief form every time they make an arrest or take a person into custody. These studies have indicated that ofﬁcers use some form of physical force in anywhere from 14.6 percent (Lundstrom & Mullan, 1987) to 22 percent (Garner, Buchanan, Schade & Hepburn, 1996) of all arrests. Self-report studies have potential weaknesses because they assume ofﬁcers honestly report the level of force they use in making an arrest. They are further limited by the fact that they look only at arrest situations (where the use of some level of physical force is a common necessity) and ignore the vast number of police-citizen encounters where no force is used (Alpert & Smith, 1994). A fourth way that researchers can obtain data about the use of force by the police is to ask departments to disclose the frequency with which citizens complained about the use of force by their officers during a given period of time. This method will be the focus of this particular study. The methodological weaknesses and limitations will be discussed elsewhere. W Over the course of a 10-month period, four New York City police ofﬁcers lost their lives in the line of duty. Shortly after this, an ofﬁcer fatally shot a ﬂeeing suspect in an unrelated case. An editorial appearing in the New York Times criticized the ofﬁcers actions, arguing that "if a policeman needs to defend his life, the use of force is permissible, but if he is chasing a suspect, he has no right to shoot the man. " Although 30 these events occurred in 1857 and 1858, they could easily represent the controversies that Americans are debating in the waning days of the 1990's (Geller, 1987). The use of deadly force by the police has been a topic of extreme debate in the United States for the past three decades. Numerous incidents of civil disorder in cities such as Los Angeles, Chicago, St. Petersburg, Detroit, Newark, and Miami (Anderson, 1995; Davis, 1993; Geller & Scott, 1992; More, 1992) have been largely attributed to fatal police—citizen contacts. It is very easy for the average citizen to lose perspective of the use of deadly force by the police when he or she focuses on a few isolated incidents. Because of the emotional nature of “police brutality" and “police shootings," it is understandable that citizens may not always place deadly force in the proper framework of the police ofﬁcer’s daily reality. In reality, the majority of police ofﬁcers do not use any form of physical force on a daily basis and may never use deadly force during their entire police career. It is regrettable that the police would ever have to use any level of force, particularly deadly force, in the commission of their duties. However, as long as the mandate of the police remains the enforcement of the law upon a public that is not always compliant, the possible use of force will remain a component of their occupation (Kobler, 1975b; Papst, 1993). In reality, any level of violence between the police and the public is exceedingly rare. In 1990, the Kansas City (MO.) Police Department received more than 500,000 calls for service; that same year the New York City Police Department received over four million 911 calls (Gest, 1991). Within this context, it may be easier to properly frame the magnitude of the use of deadly force by the police. The police do 31 use force and this force may, at times, be excessive. However, such occasions are relatively rare compared to the frequency with which the police come into contact with the public. Geller and Scott (1992) conducted an extensive analysis of various aspects of the use of deadly force against and by the police. In attempting to assess statistics such as the rate of deadly force incidents, huge discrepancies were found between the agencies which were questiomd and the studies which were analyzed. These ﬁndings lend support to the notion that interpreting data on the use of deadly force is a tricky enterprise. By manipulating how data are presented, it is possible to make use of force data support just about any hypothesis. Because of this, it is critical to understand the methodology that studies employ to make conclusions from deadly force data. Some have argued that the only accurate source of information on justiﬁable police homicides is the Public Health Service, which publishes these statistics as a part of their anmral Vital Statistics of the United States (Kobler 1975a, 1975b). According to the Vital Statistics, there has been a downward trend in the total number of civilians death since the early 1970's. In addition, the proportion of those civilians who were non-white has also been declining. However, several authors (Geller & Scott, 1992; Sherman & Langworthy, 1979) have proposed that these data are deﬁcient for several reasons and that they may under report the actual number of deaths by ﬁfty (More, 1992) to seventy percent (Kobler, 1975a). Law enforcement agencies that participate with the Uniform Crime Report are requested (but not required) to submit a Supplemental Homicide Report (SHR) for each 32 month. The form outlines the basic details of any homicides that took place within an agency's jurisdiction; justiﬁable homicides by police ofﬁcers would be included and duly noted as such. An examination of SHR’s for the years 1976 through 1988 revealed a sixteen percent decline in the number of homicides committed by police ofﬁcers during this time frame (Tennenbaum, 1994). It is important to note that because participation in the SHR program is voluntary, these conclusions are tentative, at best. The unit of analysis which might provide the most accurate context within which deadly force might be examined is the police-citizen encounter in which deadly force could be justiﬁably exercised (Binder, Scharf, & Galvin, 1982). This would provide a very clear picture of how often deadly force is applied in the context of those incidents where its use would be justiﬁed. While this would indeed be an informative context for studying deadly force, obtaining accurate data upon which to make such calculations would be difﬁcult. Using data reported by other researchers (Blumberg, 1989; Sherman & Cohn, 1986), Geller and Scott (1992) found that on average an ofﬁcer working in Jacksonville, Florida, would have to work 10 years before he or she statistically could expect to discharge a ﬁrearm at a civilian, 42 years before he or she statistically could expect to shoot a civilian non—fatally, and 139 years before he or she statistically could be expected to shoot and kill a civilian. Jacksonville reports a comparatively high rate of police-citizen violence. Using this same methodology an average ofﬁcer statistically could be expected to kill a suspect in the line of duty: every 193 years in Portland, Oregon; every 198 years in Dallas; every 594 years in Chicago; every 694 years in New York City; every 1299 33 years in Milwaukee; and every 7692 years in Honolulu. A study of the Chicago Police Department's shooting record from 1975-1977 revealed that, during this time period, there were 1,145 reported incidents where police ofﬁcers ﬁred upon civilians (Geller & Karales, 1982). The ofﬁcers struck the civilian in 26.9 percent of these incidents; 26 percent of those who were struck were fatally injured. The most common reasons cited for ﬁring upon civilians were: the use, or threatened use, of a ﬁrearm by the citizen (51%). The rate of civilians shot (per 100,000 residents) indicated that African Americans were shot at a rate over 3 time higher than caucasians (5.3 and 1.4, respectively), while hispanics were shot at a rate almost 2 times higher than caucasians (2.6 and 1.4, respectively). However, the rate of citizens shot (per 1,000 forcible felony arrests) was roughly equal for African Americans (22.4) and caucasians (21.0), while hispanics had a lower rate (17.6) (Geller & Karales, 1981). Horvath (1987) reported the results of a state-wide survey on deadly force incidents in Michigan from 1976-1981. A response rate of 69 percent was obtained (a "very good" rate according to Maxﬁeld & Babbie (1995)) yielding a data set of 551 incidents (fatal and non-fatal) in which deadly force was exercisedz. The predominant reason given for the use of deadly force was defense of life (58.1%). Most incidents occurred in a "metropolitan" area (80.4%) as opposed to an "outstate" area. Other studies have also found that city patrol ofﬁcers are involved in the majority of deadly force incidents (Kobler, 1975a). Interestingly, Horvath also found that 79 percent of the 2 Because the sample was not randomly selected, Horvath cautions that the resulting ﬁndings are not necessarily indicative of the actual incidents in the entire state. 34 civilians ﬁred upon in "metropolitan" regions were black, while in the outstate regions, black civilians were only ﬁred upon in a third of the incidents. The likely explanation for this phenomenon was that high concentration of the non-white population in the area Horvath deﬁned as "metropolitan. " It has been suggested that the implementation of new deadly force standards brought about by the Garner (1985) decision has resulted in a decline in the reported use of deadly force by the police. Prior to the Garner decision, Brown (1983) found that police ofﬁcers supported shooting policies that placed more restrictions upon their use of deadly force. According to Matulia (1982), agencies with more restrictive deadly force policies will have a lower rate of police-involved shooting incidents. In the late 1970's and early 1980's, the New York City Police Department implemented a new ﬁrearms policy that was strict enough to pass the Garner standard. This may have contributed to a 56 percent decline in the number of shooting incidents reported by the agency's ofﬁcers from 1973 to 1983 (Grennan, 1988). However, a similar restriction implemented by the Philadelphia Police had no appreciable effect on the rate with which ofﬁcers used deadly force (Waegel, 1984). Even before Gamer (1985) articulated the new standard for the use of deadly force many larger agencies had already enacted tougher policies that narrowed the number of situations in which deadly force could be used. Walker and Fridell (1992) conducted a survey of the police departments in the 100 most populous cities in America. Of those agencies responding3 over two-thirds (69.8%) indicated that they did not have to revise 3 Ninty-six departments responded. 35 their deadly force policies in response to the Garner ruling. To some extent, at least for larger departments, Garner was merely symbolic of a change in police policies that had already taken place. Many of the more comprehensive studies of deadly force are also not as relevant as they may have once been. Since the Garner (1985) ruling was handed down, there have been changes in the way police use deadly force. Studies based on data from the pre-Garner period (Fyfe, 1982; Geller & Karales, 1981, 1982; Horvath, 1987) may not be accurate descriptions of the use of deadly force in the past twelve years. There is an absence of research that has attempted to make a comprehensive examination of deadly force in the post-Gamer area, particularly in rural areas serviced by smaller agencies. EXPLAINING EXCESSIVE FORCE Theories attempting to explain why some ofﬁcers use excessive force cover the full range of sociological, occupational, physiological, and psychological thinking. One theory links occupational stress to levels of violence. Another hypothesizes that ofﬁcers feel alienated and detached from the "non-cop" world and are, thus, more inclined to use force. A third theory suggests that the police become socialized to place a high value on being physically aggressive, while a fourth theory says that physically aggressive and power-hungry persons are drawn to policing as a career (Meredith, 1984). There is clearly little consensus among the experts. Several authors have observed that job-related stress can contribute to violent behavior by police ofﬁcers (Maraniss, 1991; Reese, 1988). Based on survey responses 36 of 65 police psychologists, Scrivner (1994) identiﬁed ﬁve "at-risk" proﬁles of police ofﬁcers that are most likely to exercise excessive force. These proﬁles are grounded in the personality of the individual ofﬁcer and can be just one of many factors, including organizational characteristics, that may intensify risk. The ﬁve proﬁles Scrivner developed were: Wofﬁcers characterized as having antisocial, narcissistic, or abusive tendencies that seriously impaired routine judgement and interactions with people (especially when their authority was challenged). MW ofﬁcers suffering from "burnout" or posttraumatic stress disorder (PTSD) resulting from untreated traumatic incidents. WW young/ inexperienced ofﬁcers who are immature, impulsive, and easily frustrated; these individuals experience disciplinary problems early in their career and the problems may persist if ignored. W "traditional" ofﬁcers who use heavy-handed patrol tactics; the use of force is a common means to establish their power and control as these individuals attempt to "get the job done" at any cost. W Not all ofﬁcers with personal problems use excessive force, however some do not cope well with personal loses (such as divorce or separation); personal problems can erode self-conﬁdence, making it difﬁcult to deal appropriately with emotionally charged patrol situations. Scrivner advocates the use of critical-incident debrieﬁng and improved ofﬁcer training by departments in order to prevent ofﬁcers from developing into any of these proﬁles. Similar recommendations were made by Reese (1988) in an analysis of psychological factors that can contribute to burnout and misconduct by police ofﬁcers. Researchers have also attempted to use a myriad of theories in attempting to explain when and why ofﬁcers decide to use deadly force. Because police ofﬁcers are stereotyped 37 as drinking too much coffee, it has been argued that this might somehow be linked to the use of deadly force. However, in controlled settings, coffee drinkers and non-coffee drinkers scored approximately the same in a test of their shooting judgement (Tirnm & Urban, 1993). Some have observed that black ofﬁcers are disproportionately represented among those ofﬁcers who use deadly force (Geller & Karales, 1981). However, black ofﬁcers are assigned more often to high crime precincts and judgement test have found no differences in the decision to use deadly force based on race, gender, or level of experience (Doerner & Ho, 1994). It has been suggested that the rate of fatal police-citizen encounters may be correlated to the rate of violent crime (Kania & Mackey, 1977; Lester, 1982; Matulia, 1982). This increase could be a result of a common acceptance of violence on the parts of the police and citizens in these communities. Or, as Waegel (1984) proposed, police agencies in high crime areas may respond to this crime with special tactics such as stakeouts and target crime units. These special tactics may have a higher likelihood of resulting in violence than would other enforcement practices. The attitudes of police ofﬁcers may under-go an important and signiﬁcant transformation after the ofﬁcer has been assaulted on the job. McMurray (1990) surveyed ofﬁcers employed by the Washington, DC, and Newark, New Jersey, police departments and found important attitude changes among those ofﬁcers who had been assaulted on the job within the preceding six months. Respondents indicated that they were now less casual with citizens, that they took their job more seriously, and that they were more cautious. Respondents also indicated that they were more likely to use force in order to ensure their 38 safety. This increased likelihood to use force does not necessarily mean that ofﬁcers would be abusive, but it could be interpreted as a possible indicator that officers might over-react in potential assault situations. A follow-up analysis of the judgement police ofﬁcers exercised in using deadly force found that ofﬁcers behaved approximately the same both before and after a deadly force incident (Holzworth & Brown, 1990). It should be noted that this analysis involved a small sample of ofﬁcers from a single agency. They had all been involved in the same deadly force incident. The shooting was quickly found to be justiﬁed and there had been little, if any, second-guessing of the actions the ofﬁcers had taken. Had the incident been more controversial, the follow-up outcome may have been very different. Also, while the deadly force judgement of the ofﬁcers did not change, it cannot be assumed that the ofﬁcers were not affected in some other way by the incident. The public often views excessive force (both deadly and non-deadly) as a highly racialized issue (Anderson, 1995). It is a common public perception that excessive force is generally a white (ofﬁcer) on black (citizen) problem. Reiss (1973) found that black citizens were less likely to be the victims of excessive force than were white citizens. In addition, Reiss also found that ofﬁcers were more likely to use excessive force against members of their own racial group. DEPARTMENTAL CONTROLS ON THE USE OF FORCE One of the most fundamental obligations of a law enforcement executive is to take steps to ensure that employees are performing their duties in the proper fashion (Blumberg, 39 1994). This obligation is grounded in both legal necessities (del Carmen & Kappeler, 1991) and ethical responsibilities (Hyams, 1991), and it extends from the executive down into the supervisory ranks (Callahan, 1989; Hyams, 1991). There will always be an element of an individual ofﬁcer’s behavior that will remain beyond an agency’s control. However, steps can be taken to limit and direct the way in which ofﬁcers use force in their contacts with citizens. The following discussion is intended to provide background information on departmental controls that will be of concern to this study. M" E! |° SI I I Evaluations of the effectiveness of minimum education standards on the subsequent job performance of police ofﬁcers have been somewhat mixed (Langworthy, Hughes, & Sanders, 1995). Part of the reasons for these mixed results might be the use of inconsistent outcome measures. For example, some studies have based their conclusions on the frequency of departmental awards, citizen complaints, on-duty injury, or the use of sick leave, while others have relied upon performance evaluations given by supervisors. Because various research efforts have based their conclusions on different standards for “successful" performance, it is understandable that they have achieved mixed results. Hoover (1975) outlined three rationales for upgrading the mean level of education of police ofﬁcers. First, education may improve the effectiveness of police performance because ofﬁcers will be able to “apply systems and technology" in controlling crime. Second, education may improve the effectiveness of police performance because ofﬁcers will be better equipped to maintain order through a balanced use of “social counseling" and 40 law enforcing. Third, education ensures that ofﬁcers properly exercise their discretionary powers. Carter, Sapp, and Stephens (1988b) hypothesized that college education for police ofﬁcers improves decision making, inculcate responsibility, engenders ﬂexibility and creativity, and helps employees develop empathy for minorities. In addition, post-high school education helps employees learn to understand and tolerate persons with differing lifestyles and background. Educated ofﬁcers develop critical thinking skills and mental ﬂexibility that may better prepare them for the often-times complex task of policing; skills which are of particular importance in the era of community-oriented and problem-oriented policing. College also improves communication skills and enhances the overall “professionalism" of police ofﬁcers. These arguments aside, the use of minimum education standards as a Bona Fide Occupational Qualiﬁcation (BFOQ) is legally justiﬁable (Carter, Sapp, & Stephens, 1988a). In Davis v. City of Dallas (1985), the 5th Circuit found that the job-relatedness of higher education could be quantitatively established. A suit was brought against the City of Dallas, which was employing a post—high school education requirement for police officer applicants. In its opinion the court acknowledged that such education standards had the potential for discriminating against minorities, but that the damage this potential discrimination might cause was out-weighed by the greater need to provide for public safety. Based on the Davis (1985) decision many departments have begun to adopt some level of minimum education standard for new employees. While requiring a completed 41 baccalaureate degree is still quite uncommon, the number of agencies requiring up to 60 credit hours of college course work (the equivalent of an associate degree) has slowly risen during the past two decades. 2 I l . I I I' There can be no arguing that the stressors placed upon police ofﬁcers are tremendous. While there is not a unanimous agreement it seems reasonable to conclude that stress may have a negative impact on an ofﬁcer's health, family life, job performance, and overall well-being. A logical question that arises is what can be done to ensure that persons entering law enforcement are psychologically strong and able to deal with these stressors in a positive manner? The current perspective is that agencies can successfully conduct psychological screening of their applicants in order to determine who is best suited to work as a police ofﬁcer (Bamhill, 1992). A recent study of the police departments in 59 major American cities found that 91.5 % of these agencies used some form of psychological interview when they were hiring new employees (Sanders, Hughs, & Langworthy, 1995). Standard & Associates, a Chicago-based psychological consulting ﬁrm recommends that all agencies, regardless of size, use a standardized psychological test in their hiring process“. Travis (1994) outlined some of the most common psychological tests used in the process of selecting new police ofﬁcers. The Minnesota Multiphasic Personality Inventory (MMPI) is currently the most popular objective psychological test used; in 1990, 59.7% 4 PR Newswire, April 15, 1991. 42 of all state and municipal agencies used the MMPI. The MMPI is the test of choice for Standard & Associates. They consider it to be the "world's most accurate. . .psychological assessment test”. By scoring a test-taker’s responses to 556 true-false questions, a personality proﬁle of that individual is developed. Police agencies use these proﬁles, sometimes in conjunction with psychiatric interviews, to "screen-in" candidates who appear to have a proﬁle consistent with "good" police ofﬁcers (Meier, Farmer, & Maxwell, 1987). Travis points out that there is considerable question regarding both the reliability and validity of using the MMPI as an indicator of future job performance. The results of an individual’s MMPI are translated to create a proﬁle of that individual. By using “scientiﬁc stereotypes”, certain predictions about future behavior can be made by comparing an individual’s proﬁle with the proﬁles of individuals whose personal and psychological histories are known. However, the predictive success rate of the MMPI may be only slightly better than chance and two trained psychologists operating independently may only agree on a candidate’s rating in approximately two out of three cases (Schoenfeld, Kobos, & Phinney, 1980). The second test Travis examined, the California Psychological Inventory (CPI), is much like the MMPI. Based on responses to true-false questions, a proﬁle of the candidate is developed. While its validity is also questioned, the CPI is considered to be more reliable than the MMPI. Travis also considered the Inwald Personality Inventory (IPI), which is the third most widely used test (generally used by state agencies and in 5 Ibid. 43 large municipalities). Unlike the MMPI and CPI, the IPI was designed speciﬁcally for testing and measuring law enforcement candidates. Despite claims that these three tests are excellent predictors of the future performance of police ofﬁcer candidates (many made by those who market these tests), most of the research evidence indicates that their strength is moderate, at best (Mills & Stratton, 1982). Even before examining the reliability and validity of these tests, it is necessary to consider if social science allows for accurate predictions in any future behavior. Among other problems, inventory-type tests are based on subjective criteria of what makes a good ofﬁcer. Agencies often do not have enough understanding of the tests they use to know whether they are really identifying the kind of people they want in their agency (Moriarty, 1989). Despite this body of evidence, agencies continue to reject candidates on the basis of invalidated testing strategies (Mills & Stratton, 1982). Psychological testing is not without its merits. From a legal-rational perspective, the use of testing is well advised and protects agencies from some liability. Tests can be used to detect certain psychological conditions or disorders that would make the candidate a problem employee in any job (Inwald, 1987). The problem arises when these tests are used to screen-out candidates whose condition is being deﬁned as occupationally inadequate (Meier, Farmer, and Maxwell, 1987). Most of the conventional tests used by law enforcement agencies are designed to detect these general mental inadequacies, not whether or not the test taker meets some ill-defmed occupational proﬁle. The role of psychological testing in police ofﬁcer selection might be best 44 understood as one of many mechanisms that may help agencies to ﬁnd the best candidates. It is crucial that testing procedures not remain static; agencies must continually re-evaluate the testing tool to ensure that it is appropriate for their selection standards (Hibler & Kurke, 1995). Also, agencies must use be careful in attempting to interpret test results that they are not adequately trained to understand (Moriarty & Field, 1994). W According to the United States Department of Justice ( 1993), good policing is based upon two equally important factors. The ﬁrst factor is the values upon which a police department operates; the organizational ethics, beliefs, and practices will deﬁne the nature of its relationship with the community it serves. Second, the practices a police department follows determine how citizens are treated in individual encounters. Police agencies can take steps to improve relations with their community by strengthening departmental values (from the top—down) and adjusting organizational practices (to ensure the proper use of discretion). Police managers must remember that, although citizens have deferred police power to the government, this deferral does not mean that citizens forfeit their right to fair treatment within the criminal justice system (Terrill, 1982). Citizen review boards are seen as one way in which the public can maintain some control over, and provide input to, law enforcement agencies (Anderson, 1995; Braunstein & Tyre, 1994; Terrill, 1982; Tyre & Braunstein, 1994). Civilian review organizations help ensure accountability by investigating and/or reviewing citizen complaints, investigating alleged misconduct, and 45 reviewing policy (Rudovsky, 1992). Many ofﬁcers oppose the notion of civilian involvement in police management based on one or more misconceptions about review boards (Anderson, 1995; Terrill, 1982). Some of these misconceptions include: misunderstanding the power and authority review boards possess, seeing review boards as "anti-cop", and believing that review boards introduce too many political variables into departments. In reality, the power of review boards is often limited to investigating complaints and recommending discipline; the ultimate decision usually lies with an agency's leader. Review boards often include police ofﬁcers, members of the academic community, and criminal law attorneys, all of whom are capable of making educated judgements about an ofﬁcer’s actions. There is no statistical data to support claims that police efﬁciency is hampered by the presence of a review board (Terrill, 1982). One key determinant of the role a civilian review board plays in a speciﬁc agency is the political climate in which the board was created. If a board is created in response to a public perception of police misconduct or abuse of authority it is likely that the board will have more control over the agency. However, if the creation of a board is initiated by a proactive police executive, the executive will appear conﬁdent in their agency and open to community involvement. In the end this will result in the agency maximizing the amount of control it has in relation to the board (Tyre & Braunstein, 1994). 46 I I'I' ”11'1le 11' There are certain steps that law enforcement agencies may take to help protect themselves in the event that they are the subject of legal action due to an ofﬁcer’s alleged use of force. In deadly force situations it is critical that agencies conduct a thorough investigation. A shooting that involves a police ofﬁcer is not just another homicide and it should be investigated accordingly. The ultimate objective of the investigation, ﬁnding out the truth, does not change. Agencies must be prepared to defend the actions of their ofﬁcers in a court of law. Special attention should be give to the collection of physical evidence (McCarthy, 1988). This evidence will allow the agency to determine whether or not their ofﬁcer(s) acted justly and whether or not witnesses are providing honest testimony. At times it may be possible to identify those ofﬁcers who appear to have a propensity toward violent encounters with citizens. If such an identiﬁcation is made, it is critical that agencies take steps to correct the situation. In the wake of the Rodney King beating, an independent commission examined the Los Angeles Police Department (LAPD) and the alleged corruption and misconduct of its ofﬁcers (Christopher, et al. 1991). The Commission found a small percentage of ofﬁcers who accounted for a large proportion of all the complaints ﬁled against the LAPD. Most disturbing was the fact that little, if any, intervention was done to address the question of why these ofﬁcers were the subjects of so many complaints. Agencies need to be aware of certain “red ﬂags” that might presage misconduct by their employees. During a police ofﬁcer’s career, cynicism, alienation, the lack of ethics, 47 and other factors can accumulate to cause certain dysfunctional tendencies (Janke, 1993). By identifying these factors at the earliest possible date, agencies can take certain measures to prevent a dysfunction from becoming a serious behavioral problem. Some have even suggested that psychological testing and performance assessments be conducted on a biennial basis (Janke, 1993). W Police work, like many other occupations, may lead to high rates of stress. The relationship between police ofﬁcers and the stresses they face are different from other occupations in two key ways. First, although other occupations, such as coal mining, have been classiﬁed as being more dangerous and stressful than policing, the dangers police ofﬁcers face are often not accidents. Second, although police ofﬁcers face stressors that are similar to those confronting other emergency service providers (ﬁre ﬁghters, paramedics, correctional ofﬁcers, etc.) the way in which they deal (or fail to deal) with this stress is dramatically different (Anson & Bloom, 1988). Occupational stress can result in ofﬁcers who alienate themselves from the non-police world (Meredith, 1984). Over time, stressors that are left uncontrolled can erode an ofﬁcer's self-control to the point where behavior that is based in rationality is replaced by behavior that is based in emotions (Carter, 1984). When not properly handled, stress can result in alcoholism, burn out, family trouble, misconduct, lower productivity, suicide, depression, and violence (Ellison& Genz, 1983; Gross-Farina, 1993; Kroes, 1976; Meredith; 1984; Travis, 1994). Law enforcement agencies have an obligation to make a good-faith effort to ensure that 48 their employees maintain an adequate level of mental health (Janke, 1993) in order to minimize the abuse of authority (Carter, 1984). Supervisors can play a role in this by reminding ofﬁcers to keep things in perspective while on duty (Gabor, 1994). Another way that law enforcement agencies can combat stress in their employees is through training. It has been shown that including stress management training as a part of academy curriculum can reduce the level of stress recruit ofﬁcers experience (McNulty, Jefferys, Singer, & Singer, 1984). It has also been suggested that a short course in anger management can increase an ofﬁcer's awareness of anger, but on-going training may be necessary if ofﬁcers are also expected to deal effectively with this anger (Abernethy & Cox, 1994). Integrating stress management training into pre-service and in-service training programs might assist ofﬁcers in deal with occupational stress, which is one way that police violence might be avoided. Police violence can also be contained through other pre-service and in-service training course. Pre-service training dealing with issues such as ethics (Harper, 1983), honesty (Webber, 1992), human relations (Harper, 1983; Matulia, 1982), and weapons proﬁciency (Hall, 1992). In—service should focus on the proper procedure for handling high-risk encounters (Fyfe, 1989; Smith, 1988), weapons proﬁciency (Hall, 1992), recent changes in legislation (Smith, 1988; Webber, 1992), and changes in departmental policies (Harper, 1983; Papst, 1993; Webber, 1992). If ofﬁcers are to operate within the structure of the law and organizational policy, they must be kept abreast of changes in these areas. Ofﬁcers must also maintain a high level of proﬁciency with the weapons they might use (e. g. defensive tactics, restraint holds, hand cufﬁng, baton usage, chemical irritants, 49 and ﬁrearms) and they must be reminded about when the use of different levels of force may be justiﬁed (Desmedt, 1984; Graves & Connor, 1992). Training needs to be as realistic and dynamic as possible to teach ofﬁcers how to apply technical knowledge and departmental policy in a situation that requires split-second decision making (Brown, 1983; O'Linn, 1992). This proﬁciency helps ensure that when ofﬁcers are confronted with a conﬂict or high-risk encounter they respond with the proper application of the proper level of force (Buchanan, 1993). Ell' 2.! IEHIHB Although there has been little research into the use and effect of probationary periods for new employees, much has been written about Field Training Ofﬁcer (FT 0) programs. The ﬁrst formalized FT 0 Program was introduced in San Jose, California, in 1972. These programs originated from recognized personnel problems and the need to improve training new ofﬁcers receive (McCampbell, 1987). New ofﬁcers are assigned to one or more experienced ofﬁcers over a series of time. The experienced ofﬁcer helps the new ofﬁcer “learn the ropes" of being a police ofﬁcer. FT O’s are generally selected from a pool of volunteers (McCampbell, 1987) and receive specialized training and incentive pay (Pagan, 1985). In addition to being a means by which agencies can teach new employees about the technical aspects of policing, FTO Programs can be used instill a sense of ethical values. The FTC may have a great deal of inﬂuence over a new employee. They help new ofﬁcers develop their personal values and their style of policing, and they may shape the 50 kind of police ofﬁcer that a “rookie” will become (Hyams, 1991). The FTC program can be an invaluable tool in an agency’s effort to obtain quality employees. Ofﬁcers who have completed an FTO Program indicate that it was instrumental in helping them to learn technical aspects of law enforcement, gain self-conﬁdence, understand departmental administrative processes, and work with people. FT 0’s indicate that they teach their “students” all of these skills, plus teaching them characteristics, such as discipline and industriousness (Fagan, 1985). Although limit research has explored how ofﬁcers react to an FTO Program, little attention has been paid to whether or not FTO’s actually improve ofﬁcer performance. SUMMARY The right to use force to obtain citizen compliance is a power that is unique to the police. The range of force that the police might use varies from the threat or implication that force might be used, to actually employing deadly force upon a citizen. Countless studies have attempted to assess the rate with which police ofﬁcers utilize various levels of force. Despite their best efforts, social science researchers have had little success in making accurate determinations of the rate of force, particularly the rate of deadly force. The use of excessive, unnecessary, or brutal force by the police has no single cause or cure (Hoffman & Crew, 1991). However, there may be certain steps that law enforcement agencies can take to reduce the rate with which their employees engage in this undesirable behavior. Agencies need to develop sound policies, provide strong leadership and supervision, and establish 51 an organizational climate that is based on integrity, honesty, and trust. In addition, there are factors within an agency’s control which may have an impact on the rate of sustained use of force complaints ofﬁcers accumulate. Some of these controls include: pre~service psychological testing, Field Training programs, probationary periods for new employees, community involvement, the investigation of citizen complaints, and in-service training. This study will use a pre—existing data to determine if these controls have an actual effect on the rate of sustained use of force complaints agencies experience. CHAPTER THREE METHODOLOGY RESEARCH QUESTION Does the use of departmental policies and practices that are commonly considered to control the use of excessive force by police ofﬁcers have a signiﬁcant impact on the rate of sustained citizen complaints experienced by law enforcement agencies? KEY CONCEPTS The dependent variable in this study is the rate of sustained use of force complaints (per 100 ofﬁcers) reported by the agencies that participated in the Pate and Fridell (1993) study. The accuracy of this rate is dependent upon: a.) a citizen’s willingness to come forward and ﬁle a complaint against an ofﬁcer he/she feels used excessive force against them b.) the presence of sufﬁcient evidence to establish that an act of excessive force did occur c.) a complaint processing structure within the ofﬁcer’s department that has the integrity to conduct an unbiased investigation and to punish offenders These three factors may not be present in many instances in which excessive force is used by a police ofﬁcer. Relying on the rate of sustained complaints is severely limited because there are so many excessive force instances that the rate does not capture. A law enforcement agency can not be categorized as being “good" or “bad" based 52 53 upon its rate of sustained complaints; quite the opposite may actually be true. Consider two agencies that have an equal number of actual excessive force incidents; based solely upon this criteria the two agencies could be seen as being equal. Agency A might have a complaint processing system that favors the ofﬁcers in all encounters (yielding a low rate of sustained complaints), while Agency B might have an unbiased processing system (yielding a high rate of sustained complaints). Based upon this criteria, Agency A appears to be the better organization because it has a lower rate of sustained complaints. In reality, the opposite would be true. Conceptually, having a “high” or a “low” rate of sustained complaints does not establish the actual frequency with which an agency’s employees use excessive force. Instead, this statistic provides a baseline— an accurate and valid indicator of the minimum rate with which excessive force is being used. Viewed alone, the rate for an individual agency means very little. But when the rates for a sample of agencies are compared, certain differences may become apparent. Unfortunately, this difference may be a result of ofﬁcers abusing the right to use force, or variance in the level of tolerance agencies have for such abuse, or a combination of these two factors. In other words, agencies with higher rates may either have ofﬁcers that are actually more abusive, or they may have a lower tolerance for such abuses. The independent variables used in this study can be conceptualized as representing ways in which law enforcement agencies may be able to control the use of force by their employees. For example, it has been proposed that Field Training Ofﬁcer (FTO) programs provide agencies with an opportunity to teach new recruits the right way to do 54 their job. If the actual outcome of an FTO program is an ofﬁcer who has been properly educated and trained, these ofﬁcers should know how to use force in a proper and reasonable manner. Thus, the rate of sustained excessive force complaints should be signiﬁcantly different for those agencies that have FTO programs. Independent variables have been selected based upon the existence of prior research that indicates their effectiveness at improving ofﬁcer performance. These variables can be seen as mechanisms which are within the control of a law enforcement agency and which can be hypothesized as having an impact upon the rate of sustained force complaints agencies experience. It is expected that there will be signiﬁcant differences between agencies based upon these independent variables. DATA AND SAMPLE Study ﬁndings are based on a secondary analysis of a recent nationwide study which was conducted by Pate and Fridell (1993) and sponsored by the Police Foundation. The project was intended to be an exploratory study of the experiences that law enforcement agencies reported to have had with citizen complaints of excessive force and the consequences these complaints had for the agencies and their employees. Pate and Fridell obtained a complete list of all the law enforcement agencies known to the Census Bureau based on the 1990 census. This list (the Law Enforcement Sector list) was "the most complete and exhaustive enumeration of such agencies available." From this list, it was possible to determine the individual agency's type, size, and jurisdiction population (in 1990). The Law 55 Enforcement Sector list is considered to be more comprehensive than the list of agencies maintained by the FBI’s Uniform Crime Report program, which only includes those agencies which chose to participate in this voluntary program. Of the 17,708 agencies reported on the Law Enforcement Sector list1 , 15,801 were found to be municipal police agencies, county sheriff/police agencies, and state police agencies. Other types of agencies were not of concern in this study. While the majority (66.4 %) of law enforcement agencies serve jurisdictions with populations below 10,000, the majority (61 %) of police ofﬁcers work in an agency which serves a jurisdictions with a population above 50,000 (Reaves, 1992a, 1992b). Because of this, Pate & Fridell employed a stratiﬁed sampling protocol to select agencies within bands of jurisdiction size (below 10,000; 10,000 to 24,999; 25,000 to 49,999; and 50,000 and above). The resulting sample size was 1,697 agencies. The selected agencies were mailed survey packages (containing a cover letter, the survey, and a return envelope) in August of 1992. The enclosed survey solicited information regarding reported use of force of various types, agency size, agency workload, general agency practices and policies, employee demographic characteristics, and facts pertaining to use of force complaints received by the agency for the 1991 calendar year. Agencies were assured that their responses would remain conﬁdential. After two rounds of follow-ups, the surveying had culled a responding sample size of ‘ The actual universe of American law enforcement agencies varies slightly depending on time and the deﬁnition of “law enforcement agency." The 17, 708 agencies Pate & Fridell (1993) found on the 1990 Law Enforcement Sector list was a higher ﬁgure than the 17,120 agencies indicated in the 1995 Sourceme (see the WW WSW-.1295, p. 39. 56 1,111 law enforcement agencies (65.5 % response rate). Table 1 presents the survey sampling and responses by agency type. At least one of the original researchers has indicated that the agencies that completed the survey can be considered to be representative of the universe of American law enforcement agencies (LA. Fridell, personal communication, March 5, 1997). In light of the sample size this consideration is reasonable. While the response rate did vary by agency type, the sample size should prevent this from posing a signiﬁcant threat to the reliability and validity of the resulting data. Table 1. Survey Sampling and Response by Agency Type Type of Number of Agencies Number of Agencies Response Agency Receiving Surveys Completing Surveys Rate Municipal Police 1,016 (59.9) 731 (65.8) 71.9 County Police 43 (2.5) 32 (2.9) 74.4 County Sheriff 588 (34.6) 303 (27.3) 51.5 State Police 50 (2.9) 45 (4.1) 90.0 Total 1,697 (99.9) 1,111 (100.1) 65.5 NOTE: Numbers in parentheses indicate column percents. Column totals may not equal 100 due to rounding. 57 Table 2 indicates the distribution of the responding sample agencies by agency type and number of sworn personnel. Agencies with higher numbers of sworn personnel were over-sampled based on the proportion of the universe of American law enforcement agencies which they represent, while agencies with smaller numbers of sworn personnel were under-sampled. Table 2. Responding Law Enforcement Agencies: Agency Type by Number of Sworn Personnel Number of ‘ County County Municipal State Sworn Sheriff’s Police Police Police TOTAL Personnel Dept. Dept. Dept. Agency 1000 and over 9 2 29 16 56 (3.0) (6.2) (4.0) (35.6) (5.0) 500 to 999 15 4 32 13 64 (5.0) (12.5) (4.4) (28.9) (5.8) 250 to 499 31 10 68 10 119 (10.2) (31.2) (9.3) (22.2) (10.7) 100 to 249 69 5 207 6 287 (22.8) (15.6) (28.3) (13.3) (25.8) 50 to 99 78 6 155 0 239 (25.7) (18.8) (21.2) (0.0) (21.5) 25 to 49 49 3 109 0 161 (16.2) (9.4) (14.9) (0.0) (14.5) 1 to 24 52 2 131 0 185 (17.2) (6.3) (17.9) (0.0) (16.7) TOTAL 303 32 731 45 1,111 (100.0) (100.0) (100.0) (100.0) (100.0) NOTE: Numbers in parentheses indicate column percents. 58 Table 3 indicates the distribution of the responding sample agencies by agency type and the size of the jurisdiction served. Again, in light of the large sample size, low response rates in certain cells pose a minimal threat to the reliability and validity of this data. Table 3. Number and Percentage of Responding Law Enforcement Agencies by Agency Type and Jurisdiction Served Population County County Municipal State Served Sheriff’s Police Police Police TOTAL Dept. Dept. Dept. Agency 50,000 and over 251 32 423 45 751 (52.4) (74.4) (80.3) (90.0) (68.3) 25,000 to 49,999 34 0 99 0 132 (50.7) (0.0) (71.2) (0.0) (64.1) 10,000 to 24,999 10 0 107 0 117 (43.5) (0.0) (61.5) (0.0) (56.8) Below 10,000 8 0 102 0 111 (42.1) (0.0) (58.0) (0.0) (56.9) TOTAL 303 32 731 45 1,111 (51.5) (74.4) (71.9) (90.0) (65.5) NOTE: Numbers in parentheses represent the percent of the surveyed agencies which responded. County sheriffs had the lowest response rate of all agency types, with just over half of the surveyed agencies participating in the study; state police agencies had the best response rate. The response rate tends to decline among agencies with smaller jurisdictional populations. However, the response rate of the largest jurisdictions may be slightly inﬂated by the high response rates of state and county police agencies (which all served jurisdictions with populations in excess of 50,000). 59 The unit of analysis for this data set is the individual law enforcement agency. Agencies provided information about their policies, practices, personnel, and experiences with citizen complaints during the 1991 calendar year. This unit of analysis limits the way in which this data set can be analyzed. Focusing on the agency allows one to examine how administrative practices (e.g. hiring and training practices, policies for investigating citizen complaints, etc.) may affect the rate of sustained complaints an agency experiences. However, focusing on the agency makes it more difﬁcult to examine how the traits of the individual ofﬁcer or the circumstances of the individual incident may have an affect on the outcome of complaints. The survey instrument requested that agencies indicate how many citizen complaints were received for various types of police misconduct (including excessive force) and the disposition of these cases. Using these data a rate of sustained complaints (per 100 ofﬁcers) can be calculated. Much of the analysis of this data set will be concerned with exploring whether or not these various administrative practices have any impact upon the rate of sustained use of force complaints experienced by agencies. Some of these practices will include conducting pre-service psychological screening of recruit officers, presence of a civilian review board, use of Field Training Ofﬁcers (FTO’s), selection of FTO’s, length of FTO programs, use of probationary periods, length of probationary periods, and the amount and type of in-service training provided to employees. Because the unit of analysis is “law enforcement agency," analysis of individual ofﬁcer characteristics cannot be used for more than descriptive purposes. Attempting to 60 correlate the demographic traits of an agency’s employees (age, race, gender, length of service, or level or education) with that agency’s rate of sustained complaints is not methodologically sound. The demographic information this data set provides is expressed in terms of the aggregate characteristics of groups (e. g. all sworn personnel, all ofﬁcers with citizen complaints, and all ofﬁcers with sustained complaints). Characteristics about individuals within these categories cannot be isolated for analysis beyond the descriptive level. LIMITATIONS Like all studies of the police, Pate and Fridell's ﬁnal product is not without its limitations. First, the list from which they derived their sample was dated, creating a possible error because certain eligible agencies may have been unintentionally omitted while agencies that were no longer eligible may have unintentionally been included in the sample that received surveys. Second, mail surveys have traditionally suffered from response rates that are lower than other survey techniques. Pate and Fridell employed a number of strategies to combat this problem and a response rate that is approaching 70 percent (65.5 percent) can be considered to be between “good” and "very good " in light of these problems (Maxﬁeld & Babbie, 1995). A third limitation of this study is the length and complexity of the survey instrument. Self-administered surveys are subject to certain critical limitations because they assume the respondent is accurate, unbiased, and cautious, and that they can understand the survey instrument. This study is no exception. The survey instrument 61 used by Pate and Fridell was quite lengthy (over 20 pages, including instructions and glossary) and required very speciﬁc information. Many smaller departments may not have had sufﬁcient records to provide accurate responses. At the same time, larger agencies may not have wanted to take the time to calculate accurately the answers for certain items. Respondents may have also failed to answer certain questions (intentionally or accidentally) or they may have approximated some answers in order to complete the survey in as little time as possible. Several critical questions were only completed by 20- 30% of the responding agencies. A ﬁnal area of concern is that respondents may have provided inaccurate information due to human error or the desire to show their agency in a favorable light. Even in reporting statistics as important as the use of deadly force, law enforcement agencies have frequently provided disparate rates to "ofﬁcial " researchers (i.e. the FBI) and "unofﬁcial" researchers (i.e. academicians and the media) (Geller, 1988; Geller & Scott, 1992). The results have left some researchers wondering which statistics to believe. Even though agencies were given the assurance of anonymity, they may have been hesitant to disclose the number of use of force incidents their ofﬁcers engaged in (if this data were even available). Despite all of these limitations, Pate and Fridell conducted one of the most exhaustive surveys of the police use of force available in contemporary criminal justice literature. As with any study of the use of force, certain assumptions must be made about the accuracy of the information provided by respondents. One cannot overlook the fact that it is very difﬁcult to obtain accurate information on the use of force practices in 62 which the police engage. The use of force complaints examined in this study were investigated, resolved, and reported upon by actors with questionable motivations. There are differences in the quality and integrity of the investigative efforts law enforcement agencies make in response to a citizen complaint. Some agencies will tend to side with their own ofﬁcers, no matter how gross the misconduct. Other agencies conduct fair and unbiased investigations in an attempt to discover the truth and to discipline ofﬁcers (as needed). Also, not all agencies would have sufﬁcient records to answer all of the questions on this survey. Furthermore, some respondents may have approximated certain items rather than taking the time to calculate the answer. Finally, the deﬁnitions of what constitutes various forms of misconduct (abuse of authority versus illegal searches or seizures) were not speciﬁed by the original researchers and they will not be uniform across the sample. After considering all the methodological limitations of this data set, one is left with a decidedly uncomfortable feeling. The only solace available is that there are probably few, if any, studies which have been conducted on a nationwide level. The best one can do is to take great care in interpreting the results of any analysis using this data set. VARIABLES TO BE USED The primary dependent variable being considered will be the rate of sustained use of force complaints (per 100 ofﬁcers). This rate was calculated by summing the number of citizen-generated and internally-generated reports of excessive force which were sustained through a departmental inquiry, as reported by each agency. The total number 63 of such sustained complaints was divided by the reported number of sworn personnel. Basing the rate on 100 ofﬁcers makes the rate more manageable and puts the rate into a more realistic perspective. This dependent variable will be examined using multiple independent variables, including the use of pre-service psychological screening, the presence of a civilian review board, the use of Field Training Ofﬁcers (FT O’s), the selection of FT O’s, the length of Fl‘O programs, the use of probationary periods, the length of probationary periods, and the amount and type of in—service training provided to employees. All of these independent variables could be seen as instruments through which an agency could attempt to control the use of force by its employees. These independent variables have been selected for two key reasons. First, the way in which they were reported by the participants in the original study makes them, presumably, accurate and valid. Second, and more importantly, the original study failed to report any ﬁndings (signiﬁcant or otherwise) relating to the possible correlation between these variables and the rate of sustained use of force complaints. Using the rate of sustained complaints (as opposed to the rate of received complaints) makes the potentially hazardous assumption that agencies conduct fair and unbiased investigations of the complaints they receive. It must be accepted that the rate of sustained complaints is probably well below the real rate with which excessive force is used. Not every incident of excessive force results in a citizen (or internal) complaint and not every citizen complaint (valid or false) is sustained. Every sustained complaint presumably indicates that an incident of excessive force actually took place. On the other 64 hand, not every citizen complaint that is received is actually based on an incident where excessive force was truly used. Of these two methods of measuring the use of excessive force, the rate of sustained complaints is the better measure. While the rate of sustained complaints may not be reliable (because not every real incident is captured), it is valid (because it reﬂects actual incidents). The rate of received citizen complaints is neither reliable (because not every real incident is captured), nor valid (because it reﬂects many false incidents that can not be considered to constitute excessive force). HYPOTHESES This study will assess whether certain mechanisms within a law enforcement agency’s control will affect the rate of sustained force complaints they experience. It shall be hypothesized that there will be signiﬁcant differences between law enforcement agencies based upon: 1. the presence of a Civilian Review Board and an Internal Affairs unit. 2. the speciﬁc requirements agencies have for receiving and investigating citizen complaints of excessive force. 3. the practice of requiring new employees to complete a Field Training Ofﬁcer (FTO) program and a probationary period, and the way in which these practices are administered (e. g. their length and the selection/ training of FTO’s). the agency’s response to ofﬁcers identiﬁed as using excessive force. 5. the type of law enforcement agency and the population of the jurisdiction it serves. 6. the in-service training practices agencies employ. The review of the prior research would suggest that all of these variables may have some 65 importance in determining the outcome of citizen complaints. As a result, agency practices in these categories should result in differentiation within the study sample. In addition, it shall be hypothesized that their will be no signiﬁcant differences between law enforcement agencies based upon: 7. the use of pre-service psychological screening mechanisms to select new ofﬁcers. As the literature review indicated, prior research has generally indicated that psychological screening instruments are ineffective in predicting an individual’s potentiality for future violence. DATA ANALYSIS METHODS For the purpose of this study, the hypotheses will be tested using two statistical methods. First, dichotomous independent variables will be tested using a t-test. T-tests compare the values of a dependent variable for the two categories of a dichotomous independent variable. This comparison will determine whether there is signiﬁcant variance in the two categories based upon the values of the dependent variable. In other words, a t-test will determine whether the rate of sustained complaints is signiﬁcantly different for two groups of agencies (indicating that the rate of sustained complaints was affected by the independent variable). The second statistical method to be used will be AN OVA’s. An AN OVA conducts a function that is similar to a t-test, except it looks for variance within an independent 66 variable with more than two categories. A ﬁnding of signiﬁcant variance based upon the results of an AN OVA means that the rate of sustained complaints differs between agencies based upon the independent variable. The ﬁndings of these two statistical methods will be augmented with relevant frequencies, contingency tables, and descriptive statistics (means, medians, and modes). CHAPTER FOUR RESULTS CITIZEN REPORTING OPTIONS Tables 4 through 8 reﬂect departmental practices relating to when, where, and how citizens were able ﬁle complaints against agency employees in 1991. Within the sample of responding agencies, these items all had very high completion rates, with fewer than one percent of the agencies failing to provide answers. Table 4 presents some of the possible means through which agencies may have informed the public about their procedures for ﬁling complaints. The majority of the responding agencies did not report using any of the ﬁve speciﬁc categories of procedures to inform the public about their complaint process. Unfortunately, the data set provides no elaboration on the category most widely used by agencies (“other"). Table 5 indicates the types of assistance agencies may have provided to citizens who wished to ﬁle complaints of police misconduct. Some of the most common forms of assistance included informing citizens of the case's ﬁnal status (91.2%), providing assistance to non-English speaking citizens (72.9%), and providing a complaint form (71.1%). Only slightly more than one-half of the agencies provided citizens with a copy of their complaint report (53.6%) or informed citizens of any disciplinary action taken as a result of the complaint (52.7%). The form of assistance that police agencies used most 67 68 Table 4. Outcome of Survey Question: “Does your department use any of the following methods to inform citizens about procedures for ﬁling complaints of police misconduct?” ll Yes No Posters 28 (2.5) 1077 (96.9) Flyers 152 (13.7) 953 (85.8) Newsletters 64 (5.8) 1041 (93.7) Public Service 96 (8.6) 1008 (90.7) Announcements Citizen Complaint 87 (7.8) 1018 (91.6) Information Hotline Other 241 (21 .7) 863 (77.7) NOTE: Figures in parentheses indicate row percents. Table 5. Outcome of Survey Question: “Which of the following types of assistance does your department provide when citizens ﬁle complaints of police misconduct?" I Yes =No___J| Provides complaint form 790 (71.1) 318 (28.6) Provides bilingual complaint form 109 (9.8) 998 (89.4) Ofﬁcer complete complaint form 670 (60.3) 436 (39.2) Civilian employee completes complaint form 218 (19.6) 888 (79.9) Assist non-English speaking citizens 810 (72.9) 295 (26.6) Provides copy of complaint report to citizen 595 (53.6) 508 (45.7) Inform citizen of case ﬁnal status/disposition 1013 (91.2) 94 (8.5) Inform citizen of disciplinary action taken 586 (52.7) 521 (46.9) NOTE: Figures in parentheses indicate row percents. irrﬁequently was providing bilingual complaint forms to citizens. This was practiced by only 9.8% of the responding agencies. Table 6 indicates the various locations where agencies may have made complaint forms available to citizens wishing to ﬁle a complaint of police misconduct. The most common location was the agency headquarters facility (75.9 percent). Assuming such forms were provided by the individual agency, there were no other common area where a citizen could obtain a such a document. The far right column of Table 6 indicates that either the responding agency did not provide complaint forms or their jurisdiction did not 69 contain the listed facility for form distribution. Table 6. Outcome of Survey Question: “Does your department make complaint forms available for citizens at any of the following locations?” || Yes No N/A Agency headquarters 843 (75.9) 160 (14.4) 104 (9.4) Score from or mini/mobile stations 91 (8.2) 457 (41.1) 557 (50.1) Fire stations 10 (0.9) 826 (74.3) 269 (24.2) Public libraries 11 (1.0) 848 (76.3) 246 (22.1) Public housing management ofﬁces 13 (1.2) 746 (67.1) 345 (31.1) Ofﬁce of City Clerk 85 (7.7) 749 (67.4) 271 (24.4) City Council ﬁeld ofﬁces 26 (2.3) 654 (58.9) 424 (38.2) Grassroots 25 (2.3) 758 (68.2) 319 (28.7) Other 118 (10.6) 551 (49.6) 433 (39.0) NOTE: Figures in parentheses indicate row percents. 70 Table 7 indicates the various locations where a citizen could ﬁle a complaint against a police ofﬁcer. The distribution presented in this table indicates that most agencies did not allow citizens to ﬁle complaints anywhere other than at the agency headquarters (and/or district or precinct stations) or the city hall. In many cases, other facilities do not appear to have been available for such purposes (for example, 71.1 percent of the agencies indicated that they did not have a civilian complaint review agency where complaints could be ﬁled). Table 7. Outcome of Survey Question: “Where can citizens ﬁle complaints of police misconduct?" Ii Yes No N/A Agency headquarters 1102 (99.2) --- 7 (0.6) District/precinct stations 305 (27.5) 134 (12.1) 668 (60.1) Store front or mini/mobile station 111 (10.0) 306 (27.5) 689 (62.0) Civil Service Commissions 144 (13.0) 395 (35.6) 566 (50.9) Board of Commissioners 257 (23.1) 321 (28.9) 529 (47.6) Civilian complaint review agency 57 (5.1) 258 (23.2) 790 (71.1) City Hall 470 (42.3) 349 (31.4) 287 (25.8) Other 123 (11.1) 480 (43.2) 504 (45.4) NOTE: Figures in parentheses indicate row percents. 71 The various ways in which civilians could ﬁle their complaints are presented in Table 8. Generally, it would appear that agencies had many convenient ways in which complaints could have been ﬁled. The vast majority of agencies reported that they would accept citizen complaints in person (98.9 percent), through the mail (88.7 percent), and over the department’s main telephone line (83.4 percent). Over the past few years, many agencies have developed web pages sites on the World Wide Web (WWW). Using WWW technology, it is possible for computers to instantaneously exchange information. Were this study to be replicated in 1997 , it might be interesting to note how many agencies would allow citizens to ﬁle complaints electronically using the Internet or similar technologies. Table 8. Outcome of Survey Question: “Can citizens ﬁle complaints of police misconduct against ofﬁcers in your department in the following ways?" ll Yes No Anonymous 784 (70.6) 323 (29. 1) Through the mail 985 (88.7) 123 (11.1) By telegram 835 (75.2) 270 (24.3) Over the telephone 927 (83.4) 181 (16.3) (main department line) Over the telephone 566 (50.9) 541 (48.7) (other department line) In person 1099 (98.9) 9 (0.8) NOTE: Figures in parentheses indicate row percents. 72 AGENCY REPORTING REQUIREMENTS The survey used by Pate and Fridell (1993) also asked the responding agencies to disclose the reporting policies they had for different types of force. Agencies could indicate that their reporting policy for a particular type of force (e.g. the use of handcuffs) was either mandatory, optional, or none (meaning the agency had no such policy). Agencies also had the option of indicating that a reporting policy was “not applicable" (N/A). This would be used when an agency did not have a reporting policy for a particular type of force because they did not allow their ofﬁcers to use this method of force, such as an electrical device or a choke hold. Agencies indicated a wide variety of reporting policies for the various types of force that their ofﬁcers might have employed. The distribution of these different reporting policies is presented in Table 9. It must be noted that the presence of a reporting policy on paper does not mean that ofﬁcers will consistently and/or truthfully report their use of force. The percentage of agencies with mandatory reporting policies appears to decrease proportionally with the relative severity of force. For example, virtually all agencies require ofﬁcers to report incidents in which they ﬁre their weapon, while fewer agencies mandate the reporting of the use of batons, and most agencies do not mandate the reporting of the use of a ﬁrm grip or handcuffs. 73 Table 9. Agency Reporting Policies by Type of Force Type of Force Mandatory Optional None N/A No Answer Civilian killed 1083 (97.5) 1 (0.1) 14 (1.3) -- 13 (1.2) Civilian shot & wounded 1082 (97-4) 1 (0.1) 15 (1.4) -- 13 (1.2) Civilian shot, but not hit 1075 (96-8) 3 (0.3) 17 (1.5) -- 16 (1.4) Electrical devices 424 (38.2) 12 (1.1) 428 (38.5) 184 (16.6) 63 (5.7) Chemical agents 799 (71.9) 47 (4.2) 166 (14.9) 63 (5.7) 36 (5.7) Batons 918 (82.6) 79 (7.1) 79 (7.1) 11 (1.0) 24 (2.2) Other impact devices 620 (55.8) 36 (3.2) 285 (25.7) 118 (10.6) 52 (4.7) Flashlight 870 (78.3) 84 (7.6) 118 (10.6) 15 (1.4) 24 (2.2) Twist lock 423 (38.1) 282 (25.4) 321 (28.9) 44 (4.0) 41 (3.7) Bodily force 689 (62.0) 240 (21.6) 145 (13.1) 9 (0.8) 28 (2.5) Unholster weapon 257 (23.1) 333 (30.0) 488 (43.9) 10 (0.9) 23 (2.1) Swarm 275 (24.8) 233 (21.0) 491 (44.2) 53 (4.8) 59 (5.3) Finn grip 197 (17.7) 318 (28.6) 535 (48.2) 21 (1.9) 40 (3.6) Neck restraint 687 (61.8) 96 (8.6) 213 (19.2) 72 (6.5) 43 (3.9) Handcuff 289 (26.0) 323 (29.1) 458 (41.2) 14 (1.3) 27 (2.4) “Come alongs” 255 (23.0) 317 (28.5) 466 (41.9) 38 (3.4) 35 (3.2) Dog attacks 802 (72.2) 26 (2.3) 196 (17.6) 52 (4.7) 35 (3.2) Vehicle ramming 954 (85.9) 12 (1.1) 81 (7.3) 40 (3.6) 24 (2.2) NOTE: “N/A” indicates that the use of this type of force or device is not allowed. Figures in parentheses indicate row percents. 74 OFFICER DEMOGRAPHICS Tables 10, 11, and 12 compare the distribution of gender, race, and level of education between all sworn ofﬁcers, those ofﬁcers with citizen complains, and those ofﬁcers with sustained citizen complaints. The only complaints reﬂected in these tables are for (alleged) excessive force incidents ﬁled in 1991. Given the sporadic and variable response rate these items obtained, no concrete interpretations or conclusions can be drawn from their data. Table 10 indicates that 88.1 percent of the ofﬁcers employed by agencies participating in this study were male. It is interesting to note that male ofﬁcers seemed to compose a disproportionately high portion of ofﬁcers with complaints, but don't seem to be any more or less likely to have their complaints sustained. Males accounted for 95.5 percent of the officers with citizen complaints and 95.7 percent of the ofﬁcers with sustained citizen complaints. Complaints against both male and female ofﬁcers were sustained approximately 5 percent of the time. Table 10. Comparison of Gender Distribution Between All Sworn Ofﬁcers, Ofﬁcers with Citizen Complaints, and Ofﬁcers with Sustained Complaints All Sworn Ofﬁcers with Ofﬁcers with % of Complaints Ofﬁcers Citizen Sustained That Were _ Complaints lComplaints Sustained Male 144,521 (88.1) 9752 (95.5) 536 (95.7) 5.50 Female 19,553 (11.9) 458 (4.5) 24 (4.3) 5.24 TOTAL 164,074 10,210 560 5.49 NOTE: Figures in parentheses indicate column percents. The percent of all respondents completing the “all sworn ofﬁcers", the “citizen complaints", and the “sustained complaints” portions of this item were approximately 97%, 45%, and 22%, respectively. 75 The distribution of ofﬁcers based on race/ ethnicity is presented in Table 11. The most signiﬁcant ﬁnding from this table is the complaint experiences of those ofﬁcers who were classiﬁed as "other" (which would include ofﬁcers of Asian or Native American descent). While only one percent of all ofﬁcers are classiﬁed as "other" these ofﬁcers received 7.3 percent of citizen complaints. Almost half of the ofﬁcers classiﬁed as "other" had an excessive force complaint ﬁled against them in 1991. However, less than one percent of these complaints were eventually sustained. It must be noted that the number of agencies providing race/ethnicity information on ofﬁcers with citizen complaints and sustained complaints was low (approximately 44 percent and 22 percent, respectively). Table 11. Comparison of Race/Ethnicity Distribution Between All Sworn Ofﬁcers, Ofﬁcers with Citizen Complaints, and Ofﬁcers with Sustained Complaints All Sworn Ofﬁcers with Ofﬁcers with % of Ofﬁcers Citizen Sustained Complaints Complaints Complaints That Were I .. Sustainedﬁl White 126,092 (79.6) 7,162 (70.5) 452 (75.2) 6.31 Black 18,391 (11.6) 1,247 (12.3) 89 (14.8) 7.14 Hispanic 12,387 (7.8) 1,017 (10.0) 53 (8.8) 5.21 Other 1,592 (1.0) 737 (7.3) 7 (1.2) 0.95 TOTAL 158,462 10,163 601 5.91 NOTE: Figures in parentheses indicate column percents. The percent of all respondents completing the “all sworn ofﬁcers”, the “citizen complaints", and the “sustained complaints" portions of this item were approximately 97%, 44%, and 22%, respectively. The distribution of ofﬁcers based on level of education is presented in Table 12. information). 76 Table 12. It appears that ofﬁcers with some college (an Associates degree or less) were slightly more likely to be the subject of a citizen complaint and to have that citizen complaint sustained. Ofﬁcers with advanced degrees had more of their complaints sustained, but due to the low frequency of this category, no clear conclusions can be made. It is critical to point out that conclusions about this demographic variable are confounded by the low response rate to this item (refer to the note under Table 12 for speciﬁc response Comparison of Level of Education Between All Sworn Ofﬁcers, Ofﬁcers with Citizen Complaints, and Ofﬁcers with Sustained Complaints All Ofﬁcers Ofﬁcers with % of Complaints Sworn with Citizen Sustained That Were Ofﬁcers Complaints Complaints Sustained Advanced 858 23 6 26.09 Degree (2.5) (1.2) (1.6) BA/BS 7,204 337 53 15.73 Degree (20.6) (18.1) (14.4) Associated 4,987 341 67 19.65 Degree (14.3) (18.3) (18.3) Less than 2 7,906 474 94 19.83 yrs. College (22.6) (25.4) (25.6) High School 13,979 689 147 21.34 or GED (40.0) (37.0) (40.1) . TOTAL 34,934 1,864 367 19.69 NOTE: Level of education refers to the highest level completed. Numbers in parentheses are column percents. The percem of respondents completing the “all sworn ofﬁcers", the “citizen complaints", and the “sustained complaims" portions of this item are approximately, 70%, 33% , and 17%, respectively. 77 CITIZEN COMPLAINTS, CIVIL SUITS, AND CRIMINAL CHARGES Of the 1,111 agencies in this sample, approximately 75 % provided information about the number of citizen complaints they received for alleged police misconduct (there was some variation based on the type of misconduct and its source of origin). The agencies responding to this portion of the survey reported receiving 41,179 complaints of ofﬁcer misconduct in 1991. The most common type of misconduct reported was excessive force, with 16,249 complaints (39.5 % of all complaints). The vast majority (96.1 %) of these complaints were received from citizens, as opposed to being reported by other ofﬁcers (3.9 %). Only 891 (5.5 %) of the excessive force complaints were eventually found to be sustained. There is a signiﬁcant disparity between the percentage of citizen-generated complaints that were sustained (4.7 %) and internally-generated complaints that were sustained (24.6%). In terms of all the misconduct complaints reported, 8.7 percent of the citizen-generated complaints were sustained, in contrast to the 34.7 percent of the ofﬁcer- generated complaints which were sustained. These ﬁndings might indicate that misconduct complaints ﬁled by police ofﬁcers are given more weight and viewed as more legitimate than are complaints ﬁled by members of the public. Agencies were also asked to disclose information about the number of civil suits and criminal charges that had been brought against their ofﬁcers during 1991 (suits and charges were limited to those relating to the alleged or actual use of excessive force). Most agencies refused or failed to provide information about these matters. Table 13 presents the data provided by those agencies which disclosed this information. 78 Table 13. Civil Suits and Criminal Charges Brought Against the Responding Agencies For Excessive Force Incidents (1991) CIVIL CRIMINAL SUITS CHARGES Number of agencies refusing 782 763 to answer this item (70.4) (68.7) Number of agencies reporting 48 296 zero cases (4.3) (26.6) Number of agencies reporting 281 52 one or more cases (25.3) (4.7) Number of Cases Reported 2558 122 Mean number of cases 7.8 0.35 SET agency responding) NOTE: Figures in parentheses indicate column percents. As Table 13 indicates, most agencies that reported data on their experiences with civil suits had been the subject of 1 or more suits during 1991. However, only a small proportion of those agencies that reported relevant data indicated that their ofﬁcers had been brought up on such charges. Table 14 provides a breakdown of civil suits and criminal charges by agency type, while Table 15 provides this information by the size of the jurisdiction served by the responding agency. Most civil suits and criminal charges were brought against municipal police departments (64.4 % and 71.2 % , respectively) and county sheriff’s departments (22.4% and 17.3%, respectively). In terms of jurisdiction size, larger agencies were the subject of the vast majority of all civil and criminal matters (85.1% and 80.8%, respectively). 79 Table 14. Number of Civil Suits and Criminal Charges Filed by Agency Type Agency Type Civil Suits Criminal Charges County Sheriff 63 (22.4) 9 (17.3) County Police 16 (5.7) 3 (5.8) Municipal Police 181 (64.4) 37 (71.2) State Police 21 (7.5) 3 (5.8) TOTAL 281 (100.0) 52 (100.1) NOTE: Figures in parentheses indicate column percents. Column totals may not equal 100.0 due to rounding. Table 15. Number of Civil Suits and Criminal Charges Filed by Jurisdiction Population Jurisdiction Population Civil Suits Criminal Charges 50,000 and over 239 (85.1) 42 (80.8) 25,000 - 49,999 28 (10.0) 4 (7.7) 10,000 - 24,99 7 (2.5) 4 (7.7) Under 10,000 7 (2.5) 2 (3.8) TOTAL 281 (100.1) 52 (100.0) NOTE: Figures in parentheses indicate column percents. Column totals may not equal 100.0 due to rounding. Agencies responding to the Pate and Fridell (1993) survey were also questioned about the total amount they had paid in civil litigation during 1991. As with the other questions relating to legal proceedings, most agencies refused or failed to answer this 80 item. However, 99 agencies indicated that they had paid some amount of civil damages during 1991. The reported total amounts of all civil settlements and awards paid by each individual department ranged from $306 to $11,005,937. The average total amount paid (by those agencies reporting that they had paid some amount) was $496,014.27. The median total amount paid was $25,000; the mode total amount paid was $5,000. THE RATE OF SUSTAINED EXCESSIVE FORCE COMPLAINTS Based upon these reported sustained use of force complaints, it was possible to calculate the rate of complaints per unit of ofﬁcers. One hundred ofﬁcers was selected as the proper unit to place this rate in a manageable context. This rate was highly variable, ranging from 0 (a value reported by 541 agencies) to 33.3 (reported by a 3 ofﬁcer department which had one sustained force complaint). Using this rate as a dependent variable, statistical tests were performed using multiple independent variables. Independent variables were selected if they were within the control of a police department to determine if they had any impact on the dependent variable. For example, departments can control the amount and type of in-service training they provide to their ofﬁcers. Statistical tests were conducted using the rate of sustained complaints and information about departmental training practices to determine if a signiﬁcant relationship existed. By knowing what types of training might reduce the use of force by their ofﬁcers, departments may take steps to control such force. T-tests were conducted using the rate of sustained force complaints and fourteen different independent variables. These independent variables were survey items that 81 related to departmental practices; all the items yielded dichotomous “yes" or “no" responses. T-tests are used to determine “whether two population means are equal based on the results observed in two independent samples” (Norusis, 1995). In this analysis t—tests were used to determine whether population means for those departments answering various questions with a “yes" (meaning their agency used this practice) were equal to the population means for those departments answering “no" (meaning their agency did not use this practice). For example, a t-test was used in this study to determine if the mean rates of sustained force complaints for agencies that used pre-service psychological screening was equal to the mean rate for agencies that did not use such pre-service screening. Table 16 presents the t-test results for the dependent variable when it was grouped using fourteen different independent variables. According to the results of these tests, only four of the independent variables had results which were signiﬁcant at the .05 level. These results demonstrate that there were signiﬁcant differences between the rate of sustained complaints for those agencies answering the item with a “yes" and those agencies answering the item with a “no.” The four items that achieved signiﬁcance were: the presence of a policy requiring ofﬁcers who become aware of citizens with complaints to report this to a supervisor; the use of probationary periods for recruit ofﬁcers after completing academy training; the use of a formalized Field Training Ofﬁcer program; and the use of a formalized training curriculum for Field Training Ofﬁcers. 82 Table 16. Results of T-Tests: Dependent Variable (rate of sustained use of force complaints per 100 ofﬁcers) by Various Independent Variables Independent Variable “YeS' Sig mean / 11 Does your jurisdiction have a Civilian Review Board? .6222 / 60 .525 Does your department have a policy requiring that complaints receive .5655 / 282 .372 review OUTSIDE of the chain of command where the ofﬁcer is assigned? Does an Internal Affairs unit conduct this outside review? .5263 / 196 .720 Does your department publish summary information on complaint .5611 / 194 .507 investigations? Does your department take photos when an individual ﬁles an excessive .4934 / 792 .661 force complaint? Does your department require that citizens who ﬁle excessive force .4650 / 467 .516 complaints get immediate treatment by a physician? Does your department obtain a copy of the medical report when individuals .5019 / 785 .684 obtain medical treatment? Does your department have a policy requiring that ofﬁcers who become .5556 I 630 .004 * aware of a citizen with a complaint of excessive force to report this information to a supervisor? Does your department regularly review and investigate use of force reports .5293 / 638 .280 ﬁled by your ofﬁcers even if no citizen complaint was ﬁled? Is a psychological or psychiatric evaluation required for a pre-service .5427 / 602 .137 ofﬁcer? Does your department have a probationary period for recnrits after .5230 / 686 .000 * completing the academy training? Does your department have a formalized Field Training Ofﬁcer (FTO) .5832 / 555 .001 * program? Does your department have requirements for selecting Field Training .5411 / 511 .458 Ofﬁcers? .5701 / 431 .062 * Is there formal um ' required for the Field Training Ofﬁcers? * t-test signiﬁcance < .05 83 Other independent variables were tested using AN OVA’s. An AN OVA provides information similar to that obtain through a t-test, but an AN OVA is used to compare the means of more than two population. For example, an AN OVA was used to determine if the mean rates of sustained force complaints were equal for departments serving jurisdictions of various sizes. An AN OVA cannot determine which group(s) are “unequal"; it can only determine whether a signiﬁcant inequality exists between the groups being compared. Table 17 presents the results of AN OVA’s performed on eleven independent variables. These variables relate to the practices that departments may use for handling citizen complaints, training new personnel, or treating ofﬁcers identiﬁed as using excessive force. Departments had the option of indicating the counseling and retraining for ofﬁcers identiﬁed as using excessive force was either mandatory, optional, or not available. The categories for type of law enforcement agency and population of jurisdiction served are listed in Tables 14 and 15, respectively. The length of the probationary period was either less than 12 months, 12 months, or more than 12 months. The length of the Field Training program was either less than 12 weeks, 12 weeks, or more than 12 weeks. Three of the ANOVA’s presented in Table 17 achieved signiﬁcance at or below the .05 level. These three variables were: the number of types of assistance available to citizens ﬁling complaints; the use of retraining for ofﬁcers identiﬁed as using excessive force; and the type of law enforcement agency. The results of these AN OVA’s do not indicate which type(s) of law enforcement agency/agencies have higher or lower rates of 84 sustained use of force complaints. All that can be determined is that at least one type of agency has a mean rate which is signiﬁcantly different from the other rates. Table 17. Results of AN OVA’s: Dependent Variable (rate of sustained use of force complaints per 100 ofﬁcers) by Various Independent Variables Independent Variable F D.F. F Ratio Prob. Number of procedures to inform citizens of complaint process 1.2219 5 .2968 (see Table 41) Number of types of assistance available to citizens ﬁling 1.9801 9 .0388 * complaints (see Table 4-2) Number of locations where complaint forms are available 0.4835 6 .8209 (see Table 4-3) Number of locations where complaint forms can be ﬁled 0.4438 7 .8747 (see Table 4-4) Number of ways in which complaint forms can be ﬁled 1.3219 6 .2446 (see Table 4-5) Use of counseling for ofﬁcers identiﬁed as using 1.4652 2 .2317 unnecessary/excessive force Use of retraining for ofﬁcers identiﬁed as using 3.9487 2 .0197 * unnecessary/excessive force Type of law enforcement agency 3.8774 3 .0091 * Population of jurisdiction served 0.7271 3 .5360 Length of probationary period for new ofﬁcers 1.0720 2 .3429 Le of Field T ' ' ro for new ofﬁcers 0.4501 2 .6378 * ANOVA signiﬁcance < .05 85 Table 18 presents the results of AN OVA’s performed on the departmental participation requirements for thirteen different types of in-service training. Only two types of training were found to be signiﬁcantly different (the use of non-lethal weapons and community-oriented policing). It must be pointed out that the results of these ANOVA do not indicate which level of training (mandatory, optional or not provided) is signiﬁcantly different from the other two levels, nor can it be determined if the rate at this level is higher or lower than at the other levels. Table 18. Results of AN OVA’s: Dependent Variable (rate of sustained use of force complaints per 100 ofﬁcers) by Various Dependent Variables (types of in-service training) ll Dependent Variables F Ratio F Prob. (Type of In-Service Training) Use of non-lethal force 1.3892 .2500 Stop and approach 0.2640 .7681 Use of deadly force 1.1717 .3105 Crisis intervention 0.9435 .3898 Ofﬁcer survival 1.5597 .2109 Use of non-lethal weapons 3.6703 .0260 * Cultural sensitivity 0.8599 .4237 Code of ethics 2.9502 .0530 Human relations 1.9405 .1444 Stress management 1.2821 .2781 Physical combat 1.6421 .1943 Community-oriented policing 3.7883 .0231 * Firearms requaliﬁcation 0.4614 .6306 * ANOVA signiﬁcance < .05 NOTE: Agencies indicated whether different types of training were mandatory, optional, or not provided. In all instances, the number of degrees of freedom is 2. 86 Due to limitations in this data set, there is no way to assess the quality of training that ofﬁcers received. It would be logical to assume that the quality of training offered by police departments covers the entire spectrum of quality. Because of this diversity, the effect of high quality programs and the effect of low quality programs may off-set each other. For example, agencies that have a mandatory human relations training program that is of a high quality might notice an effect on their rate of sustained complaints. However, this effect may be masked by those agencies that have a lower quality mandatory program which yielded no effect. For the purposes of this study, it must be assumed that all training programs were of approximately equal quality. The author believes that such an assumption is safe in light of the size of the data sample. There will be variations in quality between agencies, however, these variations most likely “average out." Some agencies provide their ofﬁcers with inferior and inadequate training; this could be due to resource limitations or poor administration. Other agencies have the personnel/ fmancial/ equipment resources and the administrative integrity to provide high quality training to their employees. These types of agencies are probably in the minority and effectively cancel each other out. The majority of law enforcement agencies most likely make honest attempts at providing their ofﬁcers with quality, meaningful training. While improvements could be made in these training programs, the agencies are making good faith efforts at providing the best training possible with the resources that are available. If the quality of training in all of these agencies could be quantiﬁed and graphed, it is very likely that the result would be an approximation of a bell curve. The quality of other police practices (e.g 87 internal investigations or ﬁeld training) may be conceptualized using this same model. The author believes that the signiﬁcant results revealed in this analysis are not spurious, but are the product of a real effect that these variables have upon the frequency and manner with which the police use force. Existing research has already indicated that agencies which employ programs such as FTO’s have ofﬁcers who are more effective in the performance of their duties. While these research ﬁndings are mixed they generally support the hypotheses that departments have the capacity to take steps that may have an effect upon the use of excessive force by their employees. SUMMARY OF RESULTS Based upon the survey results, there is wide variation in where, when, and how citizens can ﬁle complaints against police ofﬁcers. Variation is also present in the reporting policies different departments have for incidents were various types of force are employed by their ofﬁcers. The majority of all patrol ofﬁcers covered by this survey were males. Male ofﬁcers also received most of the citizen complaints and were the subjects of most sustained complaints for excessive force. The average number of civil suits (among agencies disclosing such information) filed against each department was 7.8 in 1991 alone. Having criminal charges ﬁled against an ofﬁcer(s) was less common, affecting only one in three agencies that provided such information. Most civil and criminal cases were directed at municipal police departments and agencies serving jurisdictions with populations of over 50,000. When agencies lost a civil suit or settled a civil suit out of court, the average per annum cost for 88 all such cases was nearly half a million dollars. The results of the statistical analysis indicate that there are several variables that are within the control of most police departments and appear to affect the rate of sustained excessive force complaints agencies experience. These variables include: the presence of a policy requiring ofﬁcers who become aware of citizens with an excessive force complaint to report this knowledge to a supervisor, the use of probationary periods for new ofﬁcers, the use of formalized Field Training programs, requiring Field Training Ofﬁcers to undergo formalized training, the number of types of assistance available to citizens ﬁling complaints, the use of retraining for ofﬁcers that use excessive force, the type of law enforcement agency, and in-service training on the use of non-lethal weapons and community-oriented policing. CHAPTER FIVE DISCUSSION DISCUSSION OF FINDINGS The results of this study present some interesting and valuable information relating to the use of force by police ofﬁcers and how their employers may be able to place some limitations upon this use. For example, Table 9 presents the agency reporting requirements for different types of force ofﬁcers may use. It is difﬁcult to believe that in 1991, just over one percent of the responding agencies had no formal policy requiring that ofﬁcers to report that they killed a civilian in the line of duty. Given the litigious nature of our society, the lack of such a critical policy could be a huge liability for these agencies. Table 9 also indicates that a sizeable percent of agencies have no reporting requirements for the use of several forms of force that could create a great deal of liability for law enforcement agencies. For example, in examining the policies relating to the use of electrical devices, more agencies indicated that they had no reporting policy (428 agencies) than indicated that they had a mandatory reporting policy (424 agencies). Signiﬁcant portions of the agencies indicated that they had no reporting policy for several other forms of “high-risk” force (types of force that have the potential to cause serious 89 90 injury or death). Included types force would be: - Chemical agents (14.9% indicated no reporting policy) - Flashlights (10.6% indicated no reporting policy) - Neck restraints (19.2% indicated no reporting policy) - Dog attacks (17.6% indicated no reporting policy) - Vehicle ramming (7.3 % indicated no reporting policy) These results would demonstrate that in 1991 there was a need for agencies to improve aspects of their policies relating to the use of various types of force. It is interesting to note that only 5.5% of the complaints received by the agencies participating in this study were subsequently sustained. One factor that might contribute to this low rate is the fact that a sizeable number of citizen complaints are not valid (either an ofﬁcer did not use any force or the force used was deemed to be objectively reasonable upon being investigated). Another factor contributing to the low rate might be the presence of police departments which are reluctant to admit that their officers may have acted in an inappropriate manner. The former factor is one reason why the rate with which complaints were sustained is “naturally” low; the latter factor is one reason why the rate may be “artiﬁcially" reduced. It is possible that agencies which utilized more of the controls explored in this study may actually have higher rates of sustained complaints. This higher rate could be a product of the fact that these agencies have more integrity and responsibility throughout their structure. This high level of integrity may cause these agencies to more readily recognize and admit when their ofﬁcers make mistakes or act in an egregious manner. Agencies that care greatly about in-service training may also care enough about their reputation to conduct honest and thorough investigations upon receiving a citizen 91 complaint. Thus, these “high integrity” agencies might have a high rate of sustained complaints, not because ofﬁcers working in these agencies use more excessive force, but because they refuse to ignore such behavior. It is not surprising that the distribution of civil suits and criminal charges brought against law enforcement agencies and their employees were heavily skewed towards municipal police departments and those agencies serving larger populations. State police agencies accounted for a very small proportion of these legal actions. Pate and F ridell (1993) suggest that this is probably the case because these agencies tend to focus more on trafﬁc enforcement; state police agencies also tend to have more rigorous personnel selection processes and training programs, which may eliminate a larger number of unsuitable police ofﬁcer candidates. Suits that are settled out of court or decided in favor of the citizen may be very costly to the defendant agency. Among those agencies that paid such settlements and awards in 1991, the mean total amount paid was nearly half a million dollars, while the median was $25,000. To a smaller jurisdiction this latter amount could be an ofﬁcer’s salary for an entire year or a large portion of a small town’s operating budget. When talking about these kinds of dollar amounts, controlling excessive force takes on a new sense of urgency. T-tests found signiﬁcant differences between departments based on the following four factors: 1. A policy requiring ofﬁcers who become aware of a citizen with an excessive force complaint to report this information to a supervisor. 2. The use of probationary periods for recruits after the completion of academy training. 92 3. The use of a formalized Field Training Ofﬁcer (FTO) program. 4. Requiring formalized training for FTO’s While the signiﬁcance of these factors may not be surprising, the lack of signiﬁcance among other factors is even more startling. Neither the presence of a Civilian Review Board or an Internal Affairs unit had signiﬁcant effects on the rate of sustained complaints. The same was true of several variables relating to how complaints are processed. Through the use of AN OVA’s, signiﬁcant differences were found between departments based upon ﬁve independent variables: 1. The number of types of assistance available to citizens wishing to ﬁle complaints of police misconduct. 2. The use of retraining for ofﬁcers identiﬁed as using unnecessary or excessive force. 3. The type of law enforcement agency. 4. In-service training on the use of non-lethal weapons. 5 In-service training on community-oriented policing. Even though larger jurisdictions were the subject of more civil suits and criminal charges, the population of the jurisdiction served did not have an effect on the rate of sustained use of force complaints. Although the use of a probationary period or a Field Training Ofﬁcer program had a signiﬁcant impact on the rate, the actual length of such initiatives was not found to have a signiﬁcant impact. Of the seven hypotheses stated in Chapter Three, one was proven, one was not proven, and ﬁve yielded mixed results. The ﬁrst hypothesis was not supported through statistical analysis. This hypothesis dealt with the use of Civilian Review Boards and Internal Affairs units. Strictly in terms of the rate of sustained complaints agencies reported, these two mechanisms do not appear to have a signiﬁcant affect. They certainly 93 may aid in the administration of law enforcement agencies in ways that this analysis did not reveal. Hypotheses two through six all yielded mixed results, depending upon the variable being used in their assessment. For the most part, the rate of sustained force complaints does not appear to be affected by the ways in which organizations receive and investigate complaints from citizens. Field training and probationary periods have an effect, as does training FTO’s, but the length of such programs did not have a statistically signiﬁcant affect. When ofﬁcers were identiﬁed as having used excessive force, retraining has an effect on the rate, but the same is not true of the use of psychologists. The type of agency has an effect, but not the population of the jurisdiction it serves. Finally, the results of this study indicate that most in-service training does not effect the rate of sustained use of force complaints. The seventh hypothesis was the only hypothesis that was supported through statistical analysis. This hypothesis stated that the use of pre-service psychological screening tests would not achieve a level of signiﬁcance. This result is consistent with the ﬁndings of prior research, including the Independent Commission on the Los Angeles Police Department (Christopher, et a1. 1991), which noted that psychological screening devices were not effective in identifying subtle abnormalities, poor impulse control, or violent tendencies in patrol ofﬁcer candidates. While these ﬁndings do not indicate that psychological testing has no place in law enforcement agencies, the failure to achieve a level of signiﬁcance lends support to the argument that such testing does not routinely identify violence-prone ofﬁcers. 94 These study ﬁndings indicate that there are steps that law enforcement agencies and executives can take to affect the frequency with which their ofﬁcers use excessive force based on at least one outcome. It must be acknowledged that these results would not necessarily be the same if a different outcome were being used, such as the rate of ﬁled complaints, the rate of ofﬁcer reported use of force incidents in all citizen contacts, or the rate of ofﬁcer reported use of force incidents in arrests. Alternatively, there may be many other controls that would indicate signiﬁcant differences if they had been encompassed by this data set. POLICY IMPLICATIONS The ﬁndings of this research indicate that law enforcement agencies can take steps to control the use of excessive force by their employees. Training is at the core of this ability to control force. This analysis examined thirteen types of in-service training and found two that had an effect on the rate of sustained complaints (one measure of the use of force). Because the quality of training could not be assessed, other types of training may also have an affect if they are of quality. Additional forms of training not included in this analysis may also have an affect. A munber of other studies have suggested that training is an essential component in managerial efforts to ensure accountability and responsibility for the use of force. This type of training must go beyond simply showing ofﬁcers how to use force; it must also integrate teaching ofﬁcers when, and at what level, to use force. Ofﬁcers need realistic training that incorporates both practicing the act of applying a particular level of force and 95 practicing the decision making process that ofﬁcers should use in encounters requiring the use of force (Brown, 1984). Other components of training that are worthy of consideration might include: instruction on keeping tactical advantages that will allow ofﬁcers to avoid unnecessary escalation in potentially violent encounters (Alpert & Smith, 1994); problem anticipation, an emphasis on restraint techniques, and training to avoid obvious provocations in confrontational situations (Bayley & Garofalo, 1989); instructing ofﬁcers on how to diffuse potentially violent encounters (Blumberg, 1994); and, emphasizing perceptual processes and threat assessment, decision making under duress, and anger management (Scrivner, 1994). As a possible alternative to psychological selection methods, agencies might consider focusing more resources on the background investigation in the personnel selection process. The Christopher Comnrission (Christopher, et a1. 1992) noted that the IAPD pre-service background investigation focused almost exclusively on the applicant’s drug use and sexual orientation, while overlooking potential indices of violent tendencies. The Christopher Commission also recommended that the LAPD and other agencies pay increased attention to the background of applicants and the use of in-service psychological testing. Field Training Ofﬁcer (FTO) programs can be an invaluable tool in molding new employees. Along with a carefully supervised probationary period, FTO programs give agencies an opportunity to seek out areas in new employees that need to be reinforced. Agencies also have a chance to assess whether or not new employee should be retain for 96 further employment. However, caution must be used to make sure the ﬁeld training program does not over-emphasize technical skills, is not too short and/or too demanding, and does not use trainers who may be too young and inexperienced (Eisenberg, 1981). The potential value of these two programs should not be underestimated; agencies should remember the old adage that “we reap what we sow." An equally important factor in monitoring the use of force is the level of supervision that ofﬁcers receive (Geller & Toch, 1995). All of the policies in the world do not matter if patrol ofﬁcers are sent to work every day (or night) without adequate supervision. Front line supervisors, such as sergeants and lieutenants, are the key to ensuring that an agency’s policy is also its reality. If an ofﬁcer knows that a policy means more than the paper it is written on (through enforcement by supervisors) it is logical to assume that they will pay more attention to these policies. In addition, supervisors who become aware of an excessive force violation will set a dangerous trend if they do not address this violation through the proper disciplinary channels. It is critical that agencies pay attention to the early warning signs that Scrivner (1994) and others have described. At times, executives may tend to look the other way when ofﬁcers use unnecessary force or “cut constitutional corners in pursuit of laudable... goals" (Geller & Scott, 1992). Instead, executives need to help such ofﬁcers ﬁnd other ways to achieve their goals within the structure of the law and organizational policy. Setting this tone of respect for the law and individual rights is one of the key obligations a law enforcement executive bares. In at least one city, Fyfe (1988) found only one factor that could be used to consistently predict the general frequency with which ofﬁcers 97 used deadly force: the identity of the city’s mayor. Above all else, law enforcement executives need to exercise care in the development and implementation of new policies, programs, and training strategies aimed at reducing police-citizen violence. It is very difﬁcult to fully understand how these strategies may actually impact the way in which ofﬁcers use force. Because of this, administrators must strive to maintain a balance between ensuring accountability and allowing officers to protect themselves (and others). Ofﬁcers need to know that there are limitations on their right to use force and that they will be held accountable for misusing this right. However, these limitations must not be so restrictive that an ofﬁcer risks injury in order to avoid the wrath of his/her employer (Malcolm, 1990). FUTURE RESEARCH For the most part, the use of force by the police is still a mystery to social science researchers. Because there is no nationalized reporting process, we simply do not know how often ofﬁcers employ various types of force. Even in departments which have rigorous reporting requirements, the statistics are likely nothing more than a best guess. While observational studies have made some progress towards estimating the frequency of force, they still have limitations. These studies are very time and cost intensive, so they have only been done on a limited basis in larger jurisdictions. In addition, given the relative infrequency with which the police use force, they have only captured a very small sample of force incidents (primarily low-level force incidents). It would be ideal if there was a nation-wide reporting system (similar to the 98 Uniform Crime Report) that could capture data on use of force incidents. Even if logistical problems (e. g. money) can be overcome, there is still one major impediment to the success of such a system: the willingness (or lack there of) of police ofﬁcers to honestly disclose information on the force they use. At the present time, our society is highly sensitive and litigious. Ofﬁcers are not oblivious to this fact and many may be afraid to disclose any force they used (even reasonable and justiﬁed force) for the fear of being sued and/or disciplined. Until such fears can be alleviated, this author is skeptical that any self-report statistics would be accurate. However, by fearing to disclose this “dark ﬁgure” of force (justiﬁable uses of force) ofﬁcers may be doing themselves more harm than good. By hiding the majority of incidents where force low-levels of force are used in a reasonable manner, public attention remains focused on those minority incidents in which the right to use force is grossly abused. The net result of this lack of information is that the frequency with which the police use of force remains largely unknown. Social science researchers do not know the frequency with which ofﬁcers exercise their right to use force. They do not understand what precipitates the use of force and how ofﬁcers decide to use a particular level on a particular individual. They do not know how often the right to use force is abused. They don’t fully understand what departments can do to direct and control the way in which their ofﬁcers use force. McEwen (1996) proposed that improved reporting of the use of force offers four 99 beneﬁts to local police departments. Better reporting would: 1.) Show the public that the police department wants to control its use of force 2.) Help the department establish better training programs 3.) Serve as an early warning system about ofﬁcers prone to excessive use of force or use of excessive force 4.) Serve as an early warning system about systemic problems that needlessly put ofﬁcers in untenable situations where they are criticized if they do and criticized if they do not use coercive force By showing agencies that they will beneﬁt by improving reporting practices, ofﬁcial records will be strengthened. Stronger ofﬁcial records may be the most promising approach for developing a national picture of police use of force. One way in which future research could still make valuable contributions would be to conduct a replication of Pate and Fridell’s (1993) original study. Replications could be done in two distinctly interesting ways. The ﬁrst approach would be to readminister a survey tool to the original sample; this method is currently being used by the original research staff( L.A. Fridell, personal communication, March 5, 1997). Agencies which participated in the ﬁrst survey have been sent a new survey that targets speciﬁc variables (in contrast to the “scatter-shot" approach used in the original instrument). The researchers hope that they will be able to develop a longitudinal picture of the use of force (and other variables, including ofﬁcers killed in the line of duty) in the participating agencies. A second possible method would be to administer a similar survey tool to a larger sample of agencies. One of the factors that may have hampered the response rate in the original study was the survey instrument’s length and lack of clarity. It would be possible to develop a more concise survey which might target speciﬁc variables relating to the 100 frequency and outcome of citizen complaints for excessive force. Key independent variables that would be of interest in the resulting analysis might also be included in the new format. An argument could be made that a shorter survey would yield a stronger response rate. Future research should also strive to address three assumptions that the present project was forced to make: quality, integrity, and climate. All three of these factors, if properly assessed, could dramatically alter the ﬁndings of this study. Self-report surveys do not lend themselves to measuring the quality of a training program, the integrity of a department’s complaint process, or the organizational climate. Although measuring these factors may not be impossible, this task is challenging when surveys of over a thousand agencies are being collected. Within those agencies which indicated the use of various training mechanisms (e. g. FTO’s) there will be signiﬁcant variance in the structure and quality of these training programs. This variance could alter the effect that a particular form of training may have on the rate of sustained force complaints. Another variable which might have a possible effect on this rate would be the integrity of a department’s investigative process. All else being equal, agencies with more integrity and honesty will report a higher rate of sustained complaints than will agencies with less integrity and honesty. A ﬁnal factor which could be seen as effecting the rate would be the organizational climate. Quality training and rigorous policies are only as good as the executives and supervisors who enforce them on a day-to-day basis. 101 SUMMARY During the past three decades researchers have learned much about the police use of force. However, there is still a great deal that has not been learned. Until such a time when researchers and police managers have the ability to easily identify violence-prone ofﬁcers, they must seek out strategies to reduce the rate with which excessive force is used. This study has attempted to further this cause by seeking ways in which law enforcement agencies can control and limit the use of excessive force by their employees. The key ﬁndings are that, all issues of quality and integrity being equal, agencies may be able to effect the rate of sustained complaints through the use of FTO programs, probationary periods for new ofﬁcers, retraining in the use of force, and in-service training on community-oriented policing and the use of non-lethal weapons. Future research needs to attempt to make improvements in what we know about the rate with which ofﬁcers use force, including reasonable force, excessive force, and deadly force. At the present time, there is no accurate record of such incidents. Until more dependable frequency statistics have been established, all conclusions based upon the existing data must be interpreted with caution. In the interim period, law enforcement agencies would be well advised to focus their attention on personnel selection, training, supervision, and monitoring. Until research establishes departmental controls that have greater reliability, these factors seem to offer the most promise for limiting the use of excessive force by police ofﬁcers. BIBLIOGRAPHY BIBLIOGRAPHY Abernethy, A. D., & Cox, C. (1994). Anger management training for law enforcement personnel. W12, 459-466 Alpert, G. P., & Smith, W. C. (1994). How reasonable is the reasonable man?: Police and excessive force IoumalnLCnminallanLandaninolongi. 481 -501 Anderson, KC. (1995). W San Diego, CA: Lucent Books. Anson, R. H., & Bloom, M. E. (1988). Police stress in an occupational context. IoumaLQLEerceScrenceandAdmmrstramm 229-235 Bamhill, W. (1992, August 11). Early warnings: Identifying violence- -prone police ofﬁcers. Iheﬂashmgtonﬂst. P B5 Barrineau, HE. (1994). W (2nd ed.). Cincinnati, OH: Anderson Publishing. Bayley, D.H., & Garofalo, J. (1989). The management of violence by police patrol ofﬁcers. W 1-25. Binder, A., Scharf, P., & Galvin, R. (1982). We; Qfﬁcers (National Institute of Justice No. 79-NI-AX-0134). Washington, DC: National Institute of Justice. Bittner. E. (1970). Ihefnnctiomfthenolmmodemsocietx. Chevy Chase. MD: National Institute of Mental Health. Bittner, E. (1974). Florence Nightingale 1n pursuit of Willie Sutton: A theory of the Police In H Jacob (Eda) IhemtentiaLforrefonnmnminalJustice (pp 1744) Beverly Hills, CA: Sage. Blumberg, M. (1994). Police use of excessive force: Exploring various control mechanisms InAR Roberts (Ed) Criticalissuesincrimeandjusticemp 110- 126) Thousand Oaks, CA: Sage. 102 103 Braunstein, S., & Tyre, M. (1994). Selling your community on a citizens' review board. WU). 12-14. Brave, M.A., & Peters, J.G., Jr. (1993). Liability constraints on human restraints. WC), 28-35. Brink, W. T., & Lester, D. (1984). Reporting fellow police ofﬁcers for brutality: A preliminary study W54. 36 Brodsky, S. L., & Williamson, G. D. (1985). Attitudes of police toward violence. 125W 1179-1180 Brown, J .I. (1991). Deﬁning “reasonable” police conduct: Graham v. Connor and excessive force during arrest. W18, 1257-1286. Brown, M.F. (1983). Shooting policies: What patrolrnen think. W 59(5), 35-37. Brown, M. F. (1984). Use of deadly force by patrol ofﬁcers. Training implications. IoumalntﬁolmﬁcrmandAdmmrsttatanz. 133-140 Buchanan, G.W. (1993). Managing police use of force. WW), 20. Callahan, M. (1989). Municipal liability for inadequate training and supervision: Divergent views. Wilma). 24-30. Carter, D.L. (1984). Theoretical dimensions in the abuse of authority by police ofﬁcers. Wheel. 224-236. Carter, D.L., SaPP. A.D., & Stephens, D.W. (1988a). Higher education as a Bona Fide Occupational Qualiﬁcation (BFOQ) for police: A blueprint. Amedcmlgnmal W0). 1-27. Carter, D. L., Sapp, A. 1)., & Stephens, D. W. (1988b). Wise W Washington D C: Police Executive Research Forum. Carter, D.L., & Sapp, AD. (1989). The effect of higher education on police liability: Implications for police personnel policy. WWO), 153- 166. Chemerinksy, E. (1991). Policing the police. W02), 32-36. 104 ChristOPher W st 31 (1991) W W Los Angeles. CA: City of Los Angeles. Cisneros, H. (1986, April 13). High time for cities to ﬁght back. W Chronicle. section 6, p. 4. Cox, T. C, Buchholz, D. J., & Wolf, D. J. (1987). Blunt force head trauma from police impact weapons: Some skeletal and neuropsychological considerations. LonmaLQf BolrceScreni‘eandAdnnmsttatrani. 56-62. Cox, T. C., Faughn, J. 8., & Nixon, W. M. (1985). Police use of metal ﬂashlights as weapons: An analysis of relevant problems. LQuma1_Qf_P_Qlic_e_Science_and AdministrationJZ. 244-250 Creighton, B.W. (1983). Carotid restraint: Useful tool or deadly weapon? Trial, 12(5), 102-106. Davis, R.L. (1993). W Washington, DC: Federation of Tax Administrators. Davis v. City of Dallas, 777 F.2d 205 (5th Cir.) (1985). del Carmen, R.V. & Kappeler, V.E. (1991). WWW AlchchlaaLenforsementmrsonneL Englcwood Cliffs, NJ: BradY- del Carmen, R.V. (1991). Municipalities and police agencies as defendants: Liability for ofﬁcial policy. WHO), 1-17. Desmedt, J. C. (1984). Use of force paradigm for law enforcement. mm W 170-176 Doerner, W. G., & Ho, T. (1994). Shoot-don' t shoot: Police use of deadly force under simulated ﬁeld conditions. W0), 49- 68. Eisenberg, T. (1981). Six potential hazards inherent in developing and implementing Field Training Ofﬁcer (FTO) programs. W0), 50-51. Ellison, K.W., & Genz, J .L. (1983). W Springﬁeld, IL: Charles C. Thomas. Fagan, M. M. (1985). How police ofﬁcers perceive their Field Training Ofﬁcer. loumaLoonhceScrermndAdmrmstraanl. 138-152. 105 Friedrich, R. J. (1980). Police use of force: Individuals, situations, and organizations. .11 1 v.11. 1. 1110 ’0 -_ H.111... _ ‘1‘ ‘ 82-97. Fyfe, J. J. (1982). Blind justice. Police shootings in Memphis. W W11. 707- 722 Fyfe, J .J . (1988). Police use of deadly force: Research and reform. 1115119; MILL 165-205. Fyfe, J.J. (1989). Police/citizen violence reduction project. Wm W6). 18-23. Gabor, T. (1994). The 1990s: The time for aggressive police ofﬁcers. EBLLaw WM“). 15-17. Garner, J., Buchanan, J., Schade, T., & Hepburn, J. (1996). Undemandingm W (Series: Research in Brief, NCJ 158614). Washington, DC: National Institute of Justice. Geller, W.A. (1987). MW (crime ﬁle study guide). Washington, DC: National Institute of Justice. Geller, W. A. & Karales, K. J. (1981). Shootings of and by Chicago police: Uncommon crises (Part I: Shootings by Chicago police). WM ansLCriminologx12. 1813- 1866 Geller, W. A. & Karales, K. J. (1982). Shootings of and by Chicago police: Uncommon crises (Part II: Shootings of police, shooting correlates and control strategies). WW1. 331- 378 Geller W...A &Scott. M (1992) Wuhan W; Washington D C.: Police Executive Research Forum. Geller, W. A. & Toch, H. (1995). Improving our understanding and control of police abuse of force Recommendations for research and action. In W. A. Geller & H. 277-337). Waishington, D. C.: Police Executive Research Forum. Gest, T. (1991). Why bmtality persists. In W. Dudley (Rd). W (PP. 49-51). San Diego, CA: Greenhaven Press 106 Graham v. Connor, 490 U.S. 366 (1989). Graves, F.R., & Connoer, G. (1992). The FLETC use—of—force model. M 911121.120). 56-57. Grennan, S. A. (1988). New York City shooting incidents by police ofﬁcers. W 992 Gross-Farina, S. (1993). Fit for duty? Cops, choirpractice, and another chance for healing. LlnixersinLQtMialeanLRexmAl. 1079- 1164 Hall, J.C. (1992). Firearms training and liability (part 1). Wm Bulletin._61(12). 18-23- Harper, RB. (1983). Accountability of law enforcement ofﬁcers in the use of deadly force. WW 119-151. Hibler, N. S., & Kurke, M.I. (1995). Ensuring personal reliability through selection and training. In M I Kurke & E. M. Scrivner (Eds ) We 21.1mm (pp. 57-91). Hillsdale, NJ: Lawrence Erlbaum Associates. Hoffman. P.. & Crew, JM. (1991). W New York: American Civil Liberties Union. Holzworth, J. R., & Brown, C. W. (1990). Follow-up analyses of police judgements WW1 95-104 Horvath, F. (1987). The police use of deadly force: A description of selected characteristics of intrastate incidents. WWW 226-238. Hoover. LT. (1975). W Washington, DC: National Institute of Law Enforcement and Criminal Justice. Hunt, J. (1985). Police accounts of normal force. M1; 315.341. Hunter, J .C. (1994). Focus on use of force: Pepper spray. W W6). 24-26- Hyams, M.T. (1991). Communicating the ethical standard. W 18(10), 127-133. 107 Inwald, R. E. (1987). Use of psychologists for selecting and training police. In HW More & PC Unsinger (Eds.) EnlichanageriaLuansxchnlngLand W Springﬁeld, IL: Charles C. Thomas. Janke. GA (1993). Dismnnunnalnnlmnnfﬁnenﬂanagnmenmmgmfnuhe fugue. Sacramento, CA: California Commission on Peace Ofﬁcer Standards and Training. Johnson v. Glick, 481 F .2d 1028 (2d Cir.) (1973). Kania, R., & Mackey, WC. (1977). Police violence as a function of community characteristics. W 27-48. Kappeler, V.E. (1993). WWW Prospect Heights, IL, Waveland Press. Kappeler, V. E., Kappeler, S. F. & del Carmen, R. V. (1993). A content analysis of police civil liability cases: Decisions of the federal district courts, 1978-1990. lenmal ntCnmmaLIustthJL 325- 337. Kobler, A. L. (1975a). Figures (and perhaps some facts) on police killing of civilians in the United States, 1965- 1969. W 185-191. Kobler, A.L. (1975b). Police homicide in a democracy. WM 31, 163-184. Kroes, W.H. (1976). W Springﬁeld, IL: Charles C. Thomas. Lant, D., Small, J ., & Wohlberg, D. (1979). Suing the police in federal court. W 781-824. Lester, D. (1982). Civilians who kill police ofﬁcers and police ofﬁcers who kill civilians: A comparison of American cities .InurnalnﬂinlmSniennnaniAdmimstratinn. 1Q. 384-387. Lester, D., & Arcuri, AF. (1994). How did police ofﬁcers view the Rodney King incident? PerenntinnandMntnLSkillﬂ 1382. Lester, D., & Brink, W. T. (1985). Police solidarity and tolerance for police misbehavior Emhnlnginalknnnsﬂ 326. Lundstrom, R., & Mullan, C. (1987). The use of force: One department’s exPerience. W0). 6-9. 108 Malcolm, AH. (1990, September 4). Change in enforcement is said to cut violence between police and the public. We, p.32. Maraniss, D. (1991). Job-related stress can cause police brutality. In W. Dudley (Ed.), Wig (pp.67-71). San Diego, CA: Greenhaven. Matulia, K]. (1982). Wm Washington, DC: National Institute of Justice. Maxﬁeld. M.G.. & Babbie. E. (1995). Researchmethndsfnncriminaljnstinnann 91111111191921. Belmont, CA: Wadsworth. McCarthy, R. (1988). Real issues and answers on police use of deadly force. Ihe W00). 33-37. McCaInPbell. M.S. (1987).Etnld_tta1mng_fnLnnhc§_nfﬁners._Ihe_state_nLthe_art. Washington, DC: National Institute of Justice. McEwen. T. (1996). NatinnaLdathnlInctinLannlinLusnannrne (NCJ- 160113). Washington, DC: National Institute of Justice. McEwen, T., & Leahy, F. (1994)EinaLRennrt:1&ss.than.lnthaLfnrce - . -- ' Washington, D. C. National Institute of Justice. McMurray, H. L. (1990). Attitudes of assaulted police ofﬁcers and their policy implications WW 4448 McNulty, S., Jefferys, D., Singer, 6., & Singer, L. (1984). Use of hormone analysis 1n the assessment of the efﬁcacy of stress management training in police recruits. WWW 130-132. Meadows, R. J., & Trostle, L. C. (1988). A study of police misconduct and litigation. Findings and implications. lnumalnLContrannranLCriminaLlustian 77- 92 Meier, R. D., Farmer, R. E., & Maxwell, D. (1987). Psychological screening of police candidates. Current perspectives. Wand Administratinn15. 210-215 Meredith, N. (1984). Attacking the roots of police violence. W 18(5), 20-26. Mills, M.C., & Stratton, J .G. (1982). The MMPI and the prediction of police job Performance. EBILanLEnfnrnnmnntBulletinLﬂQ). 10-15. 109 Monell v. Department of Social Services, 436 US. 658 (1978). Monroe v. Pape, 365 US. 167 (1961). More, H.W. (1992). W Cincinnati, OH: Anderson Publishing. Moriarty, AR. (1989). Police psychological screening: The third generation. Ihe W160). 36-40. Moriarty A R. & Field. M.W.P_nl1ce_nfﬂnnr_selent1nn._A_handb_onk_fnrJan WIS. Springﬁeld, IL: Charles C. Thomas. Norusis, MJ. (1995). Wham Englewood Cliffs. N.J.: Prentice-Hall. O'Linn, MK. (1992). The gaps in use-of-force policies and training. Ihe_P_Qliee ma). 52-54. Papst, J .W. (1993). Basic guidelines for administrative control. Ihe_P_oliee_Chief, ﬂ)(8), 20-26. Pate. AM.. &1Fridell. LE (1993) Pnhmsnntfnrnnnfﬁmlxennrtngnm WWW (National Institute of Justice 91 --IJ CX— 0028). Washington, D. C. The Police Foundation. Pogrebin, M.R., & Poole, ED. (1988). Humor in the brieﬁng room: A study of the strategic uses of humor among police. WWW 183-210. Pollock-Bymc, 1M. (1989). WWW Paciﬁc Grove, CA: Brooks/Cole Publishing. Reaves, B..A (1992a). W Bureau of Justice Statistics Bulletin. Washington, D. C.: U. S. Department of Justice. Reaves, B.A. (1992b). Bureau of JuStice Statistics Bulletin Washington, D. C.: U. S. Department of Justice. Reay, D.T., & Mathers, R.L. (1983). Physiological effects resulting from use of neck holds. W0). 12-15. 110 Reese, J. T. (1988). Psychological aspects of policing violence. In J. T. Reese & J M. Horn (Eds.) Enlimnsxrhnlngsaﬂperatinnalassistancn (pp 347- 361) Washington D. C. U. S. Department of Justice. Reiss, A. J., Jr. (1973). Police brutality— Answers to key questions. In A. Niederhoffer & A S Blumberg (Eds 1W2: (pp. 321-.331) San Francisco: Rinehart Press. Rudovsky, D. (1992). Police abuse: Can the violence be contained? W W 465- 501. Sanders, B., Hughs, T., & Iangworthy, R. (1995). Police ofﬁcer recruitment and selection: A survey of major police departments in the US. WM), 1-4. Schmidt, W. (1976). Recent developments in police civil liability. mm W 197- 202 Scrivner, E. (1994) Police brutality. In M. Costanzo & S. Oskamp (Eds.), MW (pp. 181-202). Thousand Oaks, CA: Sage. Schoenfeld, L.S., Kobos, J.C., & Phinney, LR. (1980). Screening police applicants: A study of reliability with the MMPI. W11. 419-425. Sherman, L. W, & Cohn, E. G. (with Gartin, P. R., Hamilton, E. E., & Rogan, D P) (1986) W Washington D C Crime Control Institute. Sherman, L. W., & Langworthy, R. H. (1979). Measuring homicide by police ofﬁcers. lnumalnLCriminaLIamndLriminnlngx1Q. 546-560 Smith, PM. (1988). Inservice training for law enforcement personnel. EBLLaw WW6), 20-22. Sykes, G. M. & Matza, D. (1957). Techniques of neutralization: A theory of delinquency WW 664-670 Tennenbaum, A. N. (1994). The inﬂuence of the Garner decision on police use of deadly force lnumalnLCriminaLIanLandcmninnlngLﬂi. 241 2-60 Tennessee v. Gamer, 471 US. 1 (1985). Terrill, R. J. (1982). Complaint procedures. Variations on the theme of civilian participation lmmlniﬂnlicnScieimniAdminimﬁnmJﬂ. 398407. lll Timm, H. W., & Urban, G. D. (1993). The effects of caffeine on shooting judgement and reaction time under simulated law enforcement conditions. W EnlmndsnmmaLBsxchnlngLﬂQ). 2- 8 Travis, MA. (1994). Psychological health tests for violence-prone police ofﬁcers: Objectives, shortcomings and alternatives. WM 1717-1770. Tyre, M., & Braunstein, S. (1994). Building better civilian review boards. E111 WU”. 1014. United States Department of Justice, Community Relations Service. (1993). 311.1‘ 1 1.1.1 1.11:. 11.1 1‘1..‘ 1;.11 111 ‘ .11 .‘1 Washington, DC: Department of Justice. Van Maanen, J. (1978). The Asshole. In P. K. Manning & J. Van Maanen (Eds. ), W (pp. 221 -.238) Santa Monica, CA. Goodyear. Vaughn, M. S., & Ramirez, C. R. (1992). Police civil liability for use of stun and taser guns WMWI) 1-3 Waegel, W. B. (1984). The use of lethal force by police: The effect of statutory change. WW 121-140. Walker, S., & Fridell, L. (1992). Forces of change in police policy: The impact of Tennessee v. Garner. Amedemlnumalnﬁﬁnﬂeeﬁa), 97-112. Webber, N.O., Jr. (1992). The role of internal affairs in police training. EBLLaw EnfnrcnmenLBnllntianZ). 6—8. Williams, G.H. (1993). Controlling the use of non-deadly force: Policy and practice. HamardBlanklnttnLlnumaLJQ. 79-104. Worden, R. (1995). The causes of police brutality: theory and evidence on police use of force In W Geller & H Toch (Eds). AndJnsnannLalmnderstandingand WW (pp. 31-60). Washington, D. C.: Police Executive Research Forum. IES H "iiiiiijlﬂﬁlijgijiuiu11111111811 066 Q