An Exploration of Recidivism and the Officer Shuffle in Police Sexual Violence

Abstract

By examining a decade of newspapers accounts in the Midwest, the current study explores the nature of repeat or recidivist police sexual violence (PSV) in comparison to first-time offending in the characteristics of the offenders, nature of the PSV acts, and the departments and criminal justice system response. Results show that more than 41% of PSV cases are committed by recidivist officers who averaged 4 victims each over a 3-year span of offending. The idea of the officer shuffle is explored in the context of sexual offending by police, indicating that a minority of officers move across jurisdictions maintaining police certification despite allegations of violence and misconduct. Implications explore how policy changes, such as a national reporting system for police sexual offending, would impact problem-prone PSV offending.

Keywords

police misconduct sexual offending problem prone officers

Introduction

The conclusions of the Christopher Commission (1991) that a small group of identifiable officers accounted for a disproportionate number of citizen complaints of brutality and excessive force gave rise to a body of literature on the problem-prone officer. While there is compelling research that suggests much of police brutality and excessive force is committed by a minority of police officers (Geller & Toch, 1995; Lersch & Mieczkowski, 1996; Walker, Alpert, & Kenney, 2001), past research has not fully explored police sexual violence (PSV) as a pattern-prone offense.

There is good reason to suggest that PSV may have a strong recidivist population. First, research by Goldman (2003) and Goldman and Puro (1987, 2001) has confirmed the existence of the officer shuffle by which disreputable officers move across police jurisdictions, resigning at the onset of allegations to maintain their police certification and police employment. Most recently, Shockey-Eckles (2011) explored the hiring practices and existing laws that foster the practice of the office shuffle and the influence that these officers have on the image of law enforcement, suggesting that much of the misconduct that occurs, may be due to the behaviors of a few, “rogue or gypsy cops” (p. 292).

Second, several studies have mentioned the existence of the repeat PSV offender. McGurrin and Kappeler (2002) and Jarriel (1997) reported numerous officers in their sample had previously been convicted of sex offenses. Thomas Jarriel, reporter from ABC’s 20/20 (1997), exposed a small group of recidivist PSV offenders who had been accused of misconduct in several different police agencies. In addition, when interviewed about their first and secondhand knowledge of police sexual misconduct, police officers reported the existence of recidivist offenders, noting that certain officers engaged in more sexual misconduct than others.

This study explores the nature of recidivist or repeat PSV in comparison to nonrepeat offending in the characteristics of the offenders, nature of the sexual acts, and the criminal justice system and department response. Specifically, the present research seeks to address the following questions:

Research Question 1: Is there a pattern of a few recidivist officers with multiple allegations of sexual offending by citizens? Are these perpetrated by shuffling officers, who perpetrate violence across jurisdictions due to the practice of resigning at the onset of previous allegations?

Research Question 2: Is there a significant difference between the age and rank of those officers with repeated allegations of PSM versus nonrepeat offenders?

Research Question 3: Is there a significant difference in the nature of the PSV acts committed by repeat versus nonrepeat officer offenders, suggesting different forms of pattern offending?

Research Question 4: How does the response by the officers’ departments and local criminal justice systems differ in repeat offenders versus nonrepeat offenders?

The Nature of Sexual Offending

While police occupational deviance has been the subject of great interest and concern within the past four decades, much of the research has focused on corruption and graft (Geller & Toch, 1995; Sherman, 1980). Little research has focused on other means of misconduct (Ivkovic, 2005). There is, however, a small body of literature on the nature of police sexual offending, typically defined as police sexual misconduct or violence (Barker, 1978; Kraska & Kappeler, 1995; Maher, 2003, 2008; McGurrin & Kappeler, 2002; Sapp, 1997). From these studies the prevalence and definitions of the behaviors as well as the common characteristics of the victims and offenders can be determined.

Prevalence

Due to the hidden nature of sexual offending, it is hard to ascertain the prevalence of police sexual offending. Several researchers have tackled the blue wall of silence, which had previously restricted the access of citizens and researchers about police sexual offending, through surveying police officers regarding their first and secondhand knowledge of the frequency and nature of police sexual misconduct (Barker, 1978; Maher, 2003, 2008; Sapp, 1997; Shockey-Eckles, 2011). Barker (1978) called the police car a traveling bedroom, due to the high rates of sex he discovered officers engaged in on duty. In his study, 44% of officers reported between 10% to 100% of their fellow officers have or have had sex on duty. Maher (2003) found that more than 35% of all officers engage in some form of sexual misconduct (Maher, 2003). It is important to note that while less serious forms of police sexual misconduct (i.e., flirting on duty, consensual sex on duty, and pulling over a driver to get a closer look) were common, the most serious forms of sexual offending (i.e., rape, offenses with juveniles, and sexual shakedowns) accounted for only 3% of the offenses examined (Maher, 2003). From these studies, it is clear that less serious forms of police sexual offending are more common than criminal forms of offending.

Most of the past research on problem prone offending suggests that a small number of officers account for a disproportionate amount of citizen complaints. What is unclear from the research is if whether the complaints are in response to officer proactivity or due to egregious officer behaviors (Terrill & McCluskey, 2002). If sexual offending in officers follows this same pattern, it would suggest that there may be different kinds of problem-prone officers. Several theoretically relevant questions may contribute to this understanding.

Exposed by Goldman (2003), one potential factor may be the phenomena of officer shuffle, which refers to the practice by which discredited officers are allowed to shuffle between agencies amidst allegations of misconduct. Officers in the St. Louis area interviewed by Shocky-Eckles (2011) confirmed the existence of the officer shuffle whereby discredited officers are shuffled to other police agencies amidst allegations of misconduct and violence. Officers interviewed about their first and secondhand knowledge of PSM behavior by Maher (2008), also reported the existence of recidivist sexual offenders, noting that certain officers engaged in more sexual misconduct than others. This research has not been confirmed by sources other than firsthand experiences of police, primarily due to the difficulty in accessing reliable data sources.

Another potential factor that may determine the types of problem prone offenders is if police sexual offenders have a tendency to also commit general crimes or exclusively commit sexual offenses. There is growing support for a general theory of offending for sexual offenders (Hazelwood & Warren, 1990; Kruttschnitt, Uggen, & Shelton, 2000; Scully, 1990; Scully & Marolla, 1984) due to the common finding that sexual crimes are often accompanied by a variety of types of criminal activity, namely, property crimes. Problem-prone officers may commit a variety of types of police misconduct. In studies of general police misconduct, pattern offending is associated with more serious forms of misconduct although it is unclear if this finding remains true in sexual offending (Lersch & Mieczkowski, 1996). If the sexual offenses committed by repeat offenders are more serious over time, there are important implications for departments and criminal justice system response to earlier sexual offenses by police.

Defining Sexual Offending in Policing

Much of the difficulty in exploring sexual offenses committed by police is a lack in convention of what police sexual offending is. Researchers have differentiated police sexual misconduct from police sexual violence, noting that police sexual violence is “officially recognized and involves violence or the use of police force” (Maher, 2003, p. 357). By comparison, police sexual misconduct was defined as

any behavior by a police officer, whereby an officer takes advantage of his or her unique position in law enforcement to commit a sexually violent act, or to initiate or respond to some sexually motivated cue for the purpose of sexual gratification. (p. 357)

By defining sexual offenses committed by police along a continuum of behaviors that range from unobtrusive (i.e., “eying” victims, harboring sexually explicit material, and invasion of privacy), to obtrusive (i.e., custodial strip searches, deception to gain sexual favor, provision of services for sexual favors, and sexual harassment), and to criminal (i.e., sexual contact and/or sexual assault) the totality of PSV behaviors is revealed, rather than focusing exclusively on the most egregious and extreme behaviors.

Due to the high level of similarity between the two definitions, and the recognition of the importance of studying observed sexual behaviors, the study will maintain Kraska and Kappler’s (1995) original conceptualization. In this study, police sexual violence includes those situations that a female experienced a “sexually degrading, humiliating, violating, damaging or threatening act committed by a police officer through the use of force, fear or intimidation, or police authority” (Kraska & Kappeler, 1995, p. 93).

Officer Characteristics

A number of officer characteristics such as name,¹ age, tenure, rank, and prior police experience have been used as predictor variables for officer involvement with police sexual misconduct. McGurrin and Kappeler (2002) and Maher (2003) found that the majority of PSM offenders were line officers. The unique access officers have to victims combined with the unsupervised nature of patrol work were reasons that officers in Maher’s (2003) study reported misconduct as most likely to occur among line officers. McGurrin and Kappeler (2002) reported that the typical PSV offender had 10 years experience and was between the ages of 26 and 45.

Victims

PSV researchers have suggested that opportunity, power, authority, and isolation increases the likelihood of police sexual offending (Sapp, 1997). For police this would include citizens engaged in criminal behavior and unaccompanied victims (Kappeler, Sluder, & Alpert, 1998; Stevens, 1994). McGurrin and Kappeler (2002) found that sex workers are especially vulnerable targets for PSV offenses. Later work by Williamson, Baker, Jenkins, and Cluse-Tolar (2007) identified a category of officer offenders who performed sexual shakedowns, in which the officer demanded sex in exchange for freedom, based solely on the officer’s position of power. One victim in this study recalled being subjected to an invasive vaginal search by a male officer, after which she requested to be taken to jail. The officer had replied, “Well you can just give me some head and I won’t take you to jail” (p. 31).

Criminal Justice System and Department Response

Past research has concluded that there is a lack of response to police sexual offending by state officials Kappeler & Kraska, 1995; McGurrin & Kappeler, 2002; Walker & Irlbeck, 2002). Many offenders receive very lenient, if any, organizational punishment or criminal sanctions for their indiscretions (Collins, 2003; Puro & Goldman, 1997). Hodgson (2001) notes, “significant reluctance within police institutions to develop or initiate new practices or policies that embrace non-violent conflict resolution or non-lethal responses to violence” (p. 540). There are many suggested reasons for these findings including the hegemonic masculinity inherent in police agencies that views sexual offending by police as consensual sex (Barker, 1978; Kraska & Kappeler, 1995; McGurrin & Kappeler, 2002; Sapp, 1997), and the difficulty in prosecuting sexual crimes in the criminal justice system (DuMont & Myhr, 2000; Spohn, Beichner, & Davis-Frenzel, 2001). Adding to the difficulty of punishing PSV offenders, previous research has found that the more serious allegations of misconduct have the lowest likelihood of being sustained in internal police review due to the higher evidentiary standards and the lack of credible witnesses (Griswold, 1994; Pate & Fridell, 1993).

For these reasons, victims of police sexual violence have turned to civil remedies. Egregious behaviors of police sexual misconduct have led to successful lawsuits by victims for violation under a variety of state and federal remedies (Escholz & Vaughn, 2001; Kappeler & Vaughn, 1997; Sapp, 1997; Seager, 1994; Shay, 1999; Vaughn & Coome, 1995) leaving departments to pay out millions of dollars in punitive damages. Due to union negotiations and limited certification revocation powers in many states, the departments are required to maintain the employment of the officers responsible for the damages, unless adjudicated for a felony offense in criminal court (Franklin, 2005). Officers accused of misconduct, unless convicted of felony charges, are often allowed to continue police employment placing victims in continued risk of harm (Goldman & Puro, 2001).

Theoretical Framework

Policing has long been confirmed as a site of hegemonic masculinity, which reinforces male power at the collective and cultural levels (Connell, 1995). Martin (1998, p. 324) acknowledges practices of “framing men as agents who actively create gender hierarchy at work” through the display of male power (Collinson & Hearn, 1994). Prokos and Padavic (2002) illustrate how hegemonic masculinity and the denigration of women as a symbolic group starts in police academy training. Research suggests that the sexist culture within many police departments does not stop there, manifested by sexual harassment and employment discrimination reported by female police officers (Collins, 2003; Martin, 1990; Rabe-Hemp, 2009; Walker & Irlbeck, 2002). In many ways these cultural values have masked PSV behaviors, providing an ideal setting for offenders to continue their predatory sexual behaviors within and outside the police station.

While past research on PSV has focused on the masculine culture inherent in policing, which provides justification for sexual offending, the research does not explain the impact of opportunity and the process of victim selection inherent in recidivist offending. PSV researchers have long held that opportunity, power, authority, and isolation increases the likelihood of police sexual offending (Sapp, 1997). Victims are selected based on weakness. For police this would include citizens engaged in criminal behavior (i.e., sex workers) and young, attractive victims (Kappeler et al., 1998; Stevens, 1994).

The opportunity of police sexual offending is increased by the likelihood that police work occurs in isolation although the presence of other officers has not shown to decrease other forms of police misconduct (Brooks, 1989; Parks, 1982). Interestingly, this is also confirmed by the lack of physical violence needed to perpetrate acts of sexual harassment, abuse, and violence on victims reported in other studies (Kraska & Kappeler, 1995; Walker & Irlbeck, 2002), instead relying on threats and quid pro quo inherent in police authority and power.

Method

Data Collection

The following research represents a content analysis of newspaper accounts of PSV offenses from 1996 to 2006. This study was designed to learn more about the nature of recidivism in PSV. To explore the nature of repeat sexual offending in newspaper accounts, a relatively exclusive sample of offenders captured over a period of time was necessary. In his longitudinal study of police misconduct, Harris (2009) found that offending peaked at the 3rd year and declined after the 6th year. A 10-year time frame was chosen for this reason. The Midwest was chosen due to the small group of states with comprehensive reporting of misconduct (see the Appendix for a complete listing of all newspapers searched). These particular newspapers were selected for their high circulation rates within their respective states and their representation of a relatively homogenous sample of cases. Cases of police sexual violence were identified by keyword searches of “police,” “sexual,” “police violence,” “sexual violence,” and “sexual harassment” in the Lexis-Nexis newspaper database.² Following the convention of Chermak, McGarrell, and Gruenewald (2006), researchers used a manifest approach to coding the data. Given the sensational nature of some news stories, manifest content (i.e., visible, surface content) is often blurred in an attempt to “sell the story.” Therefore, researchers focused solely on news stories, and excluding potentially “slanted material,” such as feature stories, editorials, and special issues.

For inclusion, the acts of PSV³ had to occur between the years of 1996 and 2006, must be against private citizens rather than coworkers,⁴ and also must have had clear resolutions (i.e., adjudicated against the officer, department, or police board decision, or lawsuit decided/settled) by the end of 2009.⁵ Only after a civil lawsuit was decided or settled against an officer, an accused offender was found guilty, or a police review board made a decision against the accused officer,⁶ was the case included for study. Previous allegations of PSV were only included if an offender was found guilty of sexual offenses at a later time. After the data for each case was collected, cases that indicated lawsuits had been filed were further researched through the Lexis-Nexis legal database to determine the details and status of civil suits. Researchers engaged in two procedures to ensure reliability of the measurement of variables: interrater method and test-retest method. Due to the pragmatic language of some news stories, it can be difficult to ascertain the nature of a variable. For example, a news story may not explicitly state an officer is a “recidivist,” but may instead contain written language that indicates such a status. Therefore, three independent raters participated in the coding of news stories. Finally, the principal investigator engaged in a test-retest method with approximately 80% of the final sample of cases.

Measures

Recidivist measures

The independent variable in this analysis was repeat offender and was measured as s dichotomous variable (1 = repeat offender, 0 = nonrepeat offender). Due to the exploratory nature of the research, repeat officers were defined as officers who had been found guilty of engaging in PSV and had been accused of PSV in the past. Three additional measures were created to explore recidivist police behaviors including the years between offenses (calculated by taking the most recent offense and subtracting the first reported allegation), the total number of victims reported, and if the officer had a previous police posts (previous background = 1, none = 0), which they left amid investigation (see Table 1).

Table 1.

Descriptive Statistics for Variables (n = 106).

Variable	M	SD	Min	Max
Recidivist measures
Repeat offender	0.42	0.50	0	1
Time between offenses	3.29	4.59	0.01	20
Number of victims	2.25	2.81	1	21
Left previous post amid allegations	0.14	0.35	0	1
Offender characteristics
Officer age	36.25	8.22	24	60
Officer tenure	8.73	6.70	0.33	31
Rank	1.22	0.57	1	3
Officer on duty	0.78	0.42	0	1
Nature of the PSV act
Criminal form of violence	0.69	0.46	0	1
Victim age	21.71	8.73	7	48
Victim juvenile	0.49	0.50	0	1
Victim prostitute	0.13	0.33	0	1
Physical injury to victim	0.26	0.44	0	1
Accomplices	0.17	0.38	0	1
Other alleged crimes	0.21	0.41	0	1
CJ and department response
Criminal charges filed	0.80	0.40	0	1
Offender fired	0.45	0.50	0	1
Prison or jail sentenced	0.35	0.48	0	1
Investigation by Police only	0.46	0.50	0	1
DA involved	0.60	0.49	0	1
Lawsuit filed	0.26	0.44	0	1

Officer characteristics

The research question regarding PSV officer characteristics was measured by officer age, officer tenure, officer rank (Administrator = 3, Command = 2, Line = 1), and officer on duty (on duty = 1, on personal time = 0).

Nature of the PSV act

To answer the research question, “Is there a significant difference in the nature of the PSV acts committed by repeat versus nonrepeat officer offenders?” PSV acts were recoded as criminal behavior (1), as opposed to unobtrusive or obtrusive behavior (0). This allowed for a comparison of the severity of the PSV acts between repeat and nonrepeat PSV offenders. The following additional variables were also created: Victim age, victim juvenile (under 18 = 1, 18 and above = 0), victim reported as a prostitute (prostitute = 1, not a prostitute = 0), physical injury to the victim (injury = 1, no injury = 0), accomplices were involved in the act (accomplices = 1, alone = 0) and offender was accused of other crimes as well (other crimes too = 1, sexual offending exclusively = 0)

Criminal justice and department response

Several dichotomous case disposition variables were created to measure the CJS and department response to PSV acts. They were all dummy coded: criminal charges filed (filed = 1, no charges filed = 0), offender was fired (fired = 1, retained = 0), offender was sentenced to jail or prison (jail or prison = 1, community sanctions or fines = 0), there was an investigation by the department (1 = police investigation, 0 = no investigation), the investigation involved the District/State’s Attorney’s office (1 = DA involvement, 0 = police investigation), and lawsuit filed (suit filed = 1, no suit filed = 0).

Limitations

Newspapers report the most sensational stories. Although newspapers are a good source of information, especially in dealing with topics that would otherwise be difficult to analyze, newspaper articles are not always complete accounts of the actual events. This leads to the problem of missing data. For this reason a sample number is listed for each analysis. Cases that did not have a clear resolution (n = 20) were excluded from the analysis. Furthermore, for some of the states in this study (i.e., Illinois, Iowa, and Michigan) there are media information restrictions that prohibit a police department from releasing the name of an officer under investigation unless there has been a criminal conviction or decision rendered by the police board. Kappeler et al. (1998) argue, “police unions say that these restrictions protect innocent officers from bogus claims, but they also provide police with protections not available to other citizens” (p. 99). By state law, police internal investigations are kept from the eyes of the public. In other words, the media accounts most likely best represent those acts that are most egregious. For this reason, the severity of the misconduct, according to the Kraska and Kappeler (1995) continuum, and the injury to the victim are coded for each offense.

In future research, the study can be expanded and improved in several ways. First, future research should replicate this research on a national scale. Variability by region may be explored to determine if PSV reporting varies by city or state. Second, there may be additional forms of police sexual offending to be explored, including sexual offending against coworkers. Finally, decisions by state administrative law judges or state Peace Officer Standards and Training (POST) board findings may be used to confirm newspaper and lawsuit information currently available. This would also enable researchers to gauge the validity and reliability of newspaper and lawsuit account information in exploring PSV instances.

Lastly, the number of cases under review is small in number. To follow cases over time, the sample was limited to the Midwest. These factors limit the generalizability of the findings. However, due to the hidden nature of PSV and the exploratory nature of PSV recidivism, the results while limited, are compelling. Despite these limitations, the utility of newspaper articles as a primary data source has proven an effective measure of studying police misconduct when we examine how such news stories assign legal culpability in such cases (Hirschfield & Simon, 2010). Lersch and Feagin (1996) utilized 15 major newspapers containing accounts of violent encounters between police officers and citizens, revealing that only 13% of cases resulted in disciplinary action. This crucial finding demonstrated the usefulness of newspaper accounts of police activities, provided the lack of nationwide, systematic data on police brutality. The same convention holds for lack of national data on police use of force (Lawrence, 2000) and police-involved litigation (Archbold, Lytle, Weatherall, Romero, & Baumann, 2006). With regard to PSV, it is impossible to determine the level of accuracy of newspaper reporting of PSV against citizens as there is no baseline to compare against. Thus it is imperative for researchers to utilize more innovative means to uncover more about this important issue.

Findings

Repeat Offenders

Of the 106 cases under examination, 44 (41.5%) involved a repeat offender who had a history of PSV, leaving 62 police officers who were nonrepeat sexual offenders (Table 1). Repeat offenders in this sample averaged 4.09 reported victims (SD = 3.73). Of the total 220 victimizations reported in this analysis, problem-prone sexual offenders accounted for 158 or 72% of the victims (Table 2). Recidivist offenders ranged from 2 to 21 previous allegations of police sexual violence in addition to the allegations for which they were found guilty. The average number of years from the first offense to the most recent offense was 3.29 years (SD = 4.59), with a median of 1.33 years.

Table 2.

Number and Percentage of Offending Officers and Victimizations.

No. of allegations	No. of offending officers	% of offending officers	No. of victimizations	% of victimizations
1	62	58.49	62	28.18
2	16	15.09	32	14.54
3	14	13.20	42	19.09
4	4	3.77	16	7.27
5	3	2.83	15	6.81
6	2	1.88	12	5.45
7	2	1.88	14	6.36
8	1	0.94	8	3.63
9	1	0.94	9	4.09
10	1	0.94	10	4.54
Total	106	100	220	100

Offender Characteristics

Bivariate comparisons between recidivist sexual offenders and nonrepeat offenders’ characteristics were measured to answer the second research question, “Is there a significant difference between the age or rank of those officers with repeated allegations of PSV versus nonrepeat offenders?” (See Table 3). In keeping with past research, PSV offenders are likely to be line officers between 28 and 44 (Maher, 2003; McGurrin & Kappeler, 2002) There are few statistically significant differences between repeat and nonrepeat offenders, except officer tenure and prior police experience. Recidivist officers had about 2.5 fewer years of experience than nonrepeat offenders, but had left a prior police post, lending some evidence for the officer shuffle in police sexual violence cases, and at the very least suggests a potential difference in the offending trajectories of repeat versus nonrepeat PSV offenders.

Table 3.

Officer Characteristics.

	Repeat offender		Nonrepeat offender
	%	N	%	N
Officer on duty
Chi-square = 2.85, 1 df, p = .06	86.0	37	71.90	40
Left previous post amidst allegations. Chi-square = 4.41, 1 df, p = .03*	22.70	10	8.10	5
Officer age
Mean	37.62		35.41
(SD)	(8.05)		(8.3)
t = 1.20, 89 df, p = .21
Officer tenure
Mean	7.14		9.85
(SD)	(4.28)		(7.98)
t = −1.75, 72 df, p = .05*
Officer rank
Mean	1.25		1.20
(SD)	(0.59)		(0.55)
t = 0.42, 100 df, p = .68

p ≤ .05.

Nature of the Offense

To explore the nature of the sexual offenses committed by repeat PSV offenders and nonrepeat offenders, the criminal behaviors, number of victims reported, and details of the offenses were examined (see Table 4). Only two distinctions were found between repeat and nonrepeat offenders. First, repeat offenders were statistically more likely to victimize a juvenile than nonrecidivist offenders. More than 13% of recidivists and 11% of nonrepeat offenders offended against prostitutes (n = 7). Eighteen percent (n = 8) of recidivist officers committed their offenses with accomplices compared to 16.1% of the nonrepeat offenders. Second, that repeat PSV offenders were more likely to have had previous police experience, which they left amid allegations of PSV (22.7% vs. 8.1%, χ² = 4.41, p ≤ .03), lends some evidence to the practice of officer shuffle in PSV allegations. PSV offenders tended to commit the same types of behaviors over time, rather than behaviors that increased in severity over time as indicated by similar rates of injury to victims and criminal forms of misconduct for both nonrepeat and recidivist offenders. In keeping with a general theory of offending, 27.3% of the recidivist sample was also accused of other crimes unrelated to the sexual allegations against them (i.e., drug crimes, burglary, domestic violence) compared to 15.9% of the nonrepeat offenders. These differences were not statistically significant.

Table 4.

Nature of PSV Act.

	Repeat offender		Nonrepeat offender
	%	N	%	N
Victim juvenile	30.3	10	62.5	30
Chi-square = 8.11, 1 df, p = .00***
Victim prostitute	13.6	6	11.7	7
Chi-square = 0.09, 1 df, p = .51
Physical injury to victim	32.5	13	20.7	12
Chi-square = 1.74, 1 df, p = .15
Criminal form of violence	61.4	27	74.2	46
Chi-square = 1.98, 1 df, p = .09
Accomplices	18.2	8	16.1	10
Chi-square = 0.08, 1 df, p = .51
Other alleged crimes	27.3	12	15.9	10
Chi-square = 1.94, 1 df, p = .13
Victim age	24.7 (9.2)		20.1 (8.4)
Mean
(SD)
t = 2.04,63 df, p = .05*

p ≤ .05.

Response

For case disposition characteristics, there appear few differences in how departments and the criminal justice system handled pattern-prone offenders in comparison to other one-time offenders (Table 5). Most significant is how cases are investigated. Of the cases investigated, they may be handled by the state’s attorney or prosecutor, sometimes in conjunction with the police; by an independent third party or by a police board or commission; or handled solely by a police supervisor/internal affairs. Incidents by nonrepeat offenders were more likely to be disposed of by internal police investigations (χ² = 5.35, p ≤ .05). Following the investigations, both recidivating and nonrepeat offenders have criminal charges filed in over three fourths of the incidents and are jailed or imprisoned for the acts in about one third of the cases.

Table 5.

Criminal Justice System and Department Response.

	Repeat offender		Nonrepeat offender
	%	N	%	N
Criminal charges filed	75.0	33	83.6	51
Chi-square = 1.18, 1 df, p = .20
Offender fired	47.7	21	42.4	25
Chi-square = 0.29, 1 df, p = .39
Prison or jail sentenced	47.7	13	37.0	20
Chi-square = 0.14, 1 df, p = .41
Investigation by police only	32.5	13	56.9	29
Chi-square = 5.35, 1 df, p = .02*
DA involved	68.3	28	55.0	33
Chi-square = 1.80, 1 df, p = .15
Lawsuit filed	46.5	20	12.9	8
Chi-square = 15.03, 1 df, p = .00***

p ≤ .05. ***p ≤ .00.

One clear distinction between the disposition of repeat and nonrepeat offenders, is the percentage of incidents in which lawsuits were filed (χ² = 15.03, p ≤ .000). To test the robustness of the relationship between cases of PSV involving lawsuit(s) and pattern prone behavior, a multivariate model was estimated using logistic regression (results presented in Table 6). The results suggest that after controlling for a myriad of case disposition characteristics, recidivist PSV acts were significantly more likely to prompt lawsuits than nonrepeat offenders (b = 2.46, SE = 0.73, p = .01). The odds of a lawsuit based on recidivist PSV are 11.70 times greater than for nonrecidivist PSV, which is a robust indicator of the liability of departments that hire and maintain officers with previous PSV allegations (Vaughn, 1999).

Table 6.

Logistic Regression Estimates for Lawsuit Filed.

	b	SE	Odds ratio
Repeat offender	2.46**	0.73	11.70
Nature of PSV act
Criminal form of violence	–1.15	0.77	0.32
Physical injury to the victim	–0.33	0.75	0.72
CJS and department response
Prison or jail sentenced	1.21	0.76	3.37
Offender fired	0.09	0.65	1.09
Constant	–1.84*	.79	.16
Chi-square = 21.76, 5 df, p = .00

p ≤ .05. **p ≤ .01.

Discussion

In conclusion, this research provides support for existence of the problem-prone offender in PSV. A group of repeat officers are responsible for the majority of the sexual offending known to the public. Almost three quarters of victims in the study were victimized by a repeat or problem-prone offender. Each repeat offender in the sample averaged four reported victims over an average of 3 years. However, the findings do not suggest that PSV offenders are general offenders who commit a variety of criminal offenses. Instead, PSV offenders tended to commit the same types of PSV behaviors, as indicated by similar rates of injury to victims and criminal forms of misconduct. These findings confirm previous criminal career research and underscore the importance of identifying problem-prone PSV offenders earlier in their careers. Few differences were reported between nonrepeat and repeat offenders, with the exception of officer tenure and prior police experience.

It is unclear why problem-prone PSV offending was associated with less tenured police officers although that relationship is also commonly cited in the general police misconduct literature (Lersch & Mieczkowski, 1996). Future research should attempt to capture the career trajectories of police sexual offenders to analyze the onset and desistance of the behavior. In his longitudinal study of police misconduct, Harris (2009) found that police offending peaked at the 3rd year and declined after the 6th year. The means for both recidivist and nonrepeat officers in this study exceeded the 6th year mark, suggesting a potential difference between general misconduct and sexual offending trajectories. To replicate in sexual offending, citizen allegations of police sexual offending should be examined as well as adjudicated behaviors (see Williamson et al., 2007).

The finding that repeat PSV offenders were more likely to have had previous police experience that they left amid allegations of PSV, lends some evidence to the practice of officer shuffle in PSV allegations (Goldman, 2003; Goldman & Puro, 1987, 2001; Shockey-Eckles, 2011). Fourteen percent of cases (n = 15) illustrated this shuffle whereby officers accused of PSV were allowed to resign and resume police employment in other agencies. For example, a Lindenhurst, Illinois police officer, Troy Hetz, was allowed to resign from his post when a 19-year-old woman filed suit claiming he forcibly fondled and kissed her under threat of arrest. Shortly after securing a patrol position with nearby department Holiday Hills, Hetz was charged and later convicted for exchanging sex atop his patrol car for a verbal warning for a 20-year-old women driving on a suspended license (“Ex-police Officer Starts Serving Jail Sentence,” 2004). Kenneth Norton, Police Chief in Colon, Indiana was arrested following charges that he fondled a woman in her home less than a year after he had been suspended as a Deputy in the same county for similar behavior. After the Chief’s retirement, numerous others came forward to report other incidents of sexual misconduct (“Ex-Chief Admits to Sex Charge,” 2005). In both of these cases, officers were allowed to resign from their earlier posts following allegations of sexual impropriety with recommendations from superiors for future employment, allowing them to continue their sexual offending in a neighboring town. In Mount Vernon, Iowa students of a nearby college protested the hiring of an officer who had a reputation for PSV behavior in a neighboring town until the officer agreed to resign from duty (“Cornell Students Protest Hiring of Police Officer,” 1999). In all of these cases, the departments were aware of the offending officer’s previous behaviors when hired. These cases represent a pattern of “shuffling” problem-prone PSV offenders and elucidate many of the problems with the current official response to PSV.

There are several important remedies that can be taken to address problem-prone PSV offenders. First, the results of this analysis underscore the importance of strengthening the criminal justice system and department response to PSV behavior to deter future acts. Relying on local police departments may not be a comprehensive or realistic solution to PSV offending, exemplified by the low rates of investigation and organizational punishments observed in the study. Police abuse scholars warn, “When we try to use criminal law as a substitute for standards that should be applied within a profession or occupation, we almost invariably are disappointed with the results” (Skolnick & Fyfe, 1993, p. 198). Exacerbating the difficulty of charging police with PSV, sexual cases are typically one of the hardest types of cases to prove in criminal court due to the private nature of the events and the he said/she said testimony (Du Mont & Myhr, 2000; Spohn et al., 2001), adding to the hesitancy of State Attorneys to charge police in misconduct cases (Human Rights Watch, 1998). Furthermore, because cases against police officers are usually difficult to win, prosecutors contend that it is best to pursue only the cases with the greatest chances of conviction, which are typically the most egregious. Recidivist police offending was more likely to receive attention by the criminal justice system in the form of investigation, charging, and conviction. One reason for these findings was that previous victims testified to offenders’ later behavior, strengthening the evidence presented in pattern-prone PSV cases. At the trial of Milwaukee Patrol Officer Steven Lelinski, six previous victims testified to being sexually assaulted by him. Three of the women had previously reported allegations to authorities, but the district attorney declined prosecution. The testimony of seven victims made a compelling case against Lelisnki, who was sentenced to more than 20 years in prison in 2007 (“Third Charge Expected Against Police Officer,” 2006). Instead, official policies and training on PSV and reporting procedures are necessary to counteracting the “consensual sex” myth, followed by strengthening the supervision of line officers by police executives so they are alert to the potential of PSV and can immediately investigate rumors and allegations of potential offenses if necessary.

Second, increasing the power and presence of Peace Officer Standards and Training (POST) boards may be a solution to repeat PSV offenders. In many states, POSTs have limited authority to discipline offending officers (Goldman & Puro, 2001). For example, in Illinois, prior to 2004, an officer could only been decertified if the officer was adjudicated of a felony, Patrol Officer Ernest Marsalis was the subject of more than 20 citizen complaints about PSV behavior toward citizens. Following the settlement of US$625,000 dollars to one victim, her attorney commented on the practice of restricted decertification, “A woman had to be raped before they forced him to resign” (“$625,000 Settlement in Alleged Cop Rape,” 2000, p. 26). Illinois law has since then been updated to include convictions of specified misdemeanors, including sexual offenses, and misdemeanors involving bodily harm.⁷

The findings of this study suggest that victims are finding some redress in the civil system. Part of determining liability for acts of PSV entails the confirmation of whether the violence was a foreseeable consequence of agency activities (Vaughn, 1999). Courts have used the offender’s previous incidents of PSV and the assailant’s disciplinary record and/or previous offense history as a means of determining if the violence was foreseeable by the department (Vaughn, 1999). Unexpectedly, in reviewing the lawsuits filed in these cases, several were filed by the officers themselves seeking redress from their departments for releasing public information about their sexual offenses rather than victim litigation. In the future, consideration must be given to implementing qualified immunity laws protecting police administrators who report information concerning officer PSV behavior to future potential employers. Currently, the threat of lawsuit may be sufficient to induce police administrators not to report to potential employers⁸ (Puro & Goldman, 1997). In this study, many departments faced questions of liability in reporting officers’ PSV behaviors to other departments. If police agencies and their personnel have qualified immunity and are protected for their good faith effort, reporting of PSV may increase. Likely this will also reduce the practice of the officer shuffle.

Third, greater access to reporting mechanisms for police misconduct may be another solution. The creation of the International Association of Directors of Law Enforcement Standards and Training (IADLEST) National Decertification Index (NDI) as a national reporting mechanism for police decertification is a step in the right direction. The purpose of the NDI is to serve as a national registry of certificate or license revocation actions relating to officer misconduct. However, there are several obstacles to making NDI a national clearing house for PSV behaviors. First, NDI permits only the POSTs to report to them, not local agencies. Second, local agencies cannot access the decertification records held by NDI without POST permission. Finally, NDI only reports decertification, not other forms of police discipline.

In conclusion, the harm caused by repeat offenders of police sexual violence is great, not only to their victims but also to the reputation of their coworkers and the liability of the departments that employ them. While this exploratory study is an important first step, more research is needed. Despite the estimates that more than 35% of all police engage in some form of sexual misconduct (Maher, 2003), compared to 1% of officers engaged in police brutality (Bureau of Justice Statistics, 2011), only a handful of studies are dedicated to PSV. More research on this important form of victimization is needed to determine mechanisms to deter future offending and further victimization.

Footnotes

Appendix

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Cara E. Rabe-Hemp is an Associate Professor in the Department of Criminal Justice Sciences at Illinois State University. She received her Ph.D. in Criminal Justice from the University of Illinois at Chicago in 2005. Her research interests include: gender and policing, small and rural policing, citizen perceptions of the police, and police deviance. Her recent publications have appeared in the Women and Criminal Justice, Police Quarterly, Feminist Criminology, Journal of Criminal Justice, and Policing: An International Journal of Police Strategies and Management. In 2010, Cara was awarded the University Research Initiative Award.

Jeremy Braithwaite is currently a graduate student at the University of California, Irvine. His research interests include: special populations of sexual offenders, juvenile delinquency, and evaluative research.

References

Archbold

C. A.

Lytle

Weatherall

Romero

Baumann

(2006). Lawsuits involving the police: A content analysis of newspaper accounts. Policing: An International Journal of Police Strategies & Management, 29, 625-642.

Barker

(1978). An empirical study of police deviance other than corruption. Journal of Police Science and Administration, 3, 264-272.

Brooks

L. W.

(1989). Police discretionary behavior: A study of style. In Dunham

R.G.

Alpert

G.A

(Eds.), Critical issues in policing: Contemporary readings (pp. 121-145). Prospect Heights, IL: Waveland Press.

Bureau of Justice Statistics. (2011). Contacts between police and public. Washington, DC: Government Printing Office.

Chermak

S. M.

McGarrell

Gruenewald

(2006). Media coverage of police misconduct and attitudes toward police. Policing: An International Journal of Police Strategies & Management, 29, 261-281.

Christopher

(1991). Report of the independent commission of the Los Angeles Police Department. Los Angeles, CA: City of Los Angeles.

Collins

S. C.

(2003). Sexual harassment and police discipline: Who’s policing the police? Policing: An International Journal of Police Strategies & Management, 27, 512-538.

Collinson

D. L.

Hearn

(1994). Naming men as men: Implications for work, organization, and management. Gender, Work and Organization, 1(1) 2-22.

Connell

R. W.

(1995). Masculinities. Berkley: University of California Press.

10.

Cornell students protest hiring of police officer. (1999, April 20). Cedar Rapids Gazette, p. 1.

11.

Du Mont

Myhr

T. L.

(2000). So few convictions: The role of client-related characteristics in the legal processing of sexual assaults. Violence Against Women, 6, 1109-1136.

12.

Eschholz

Vaughn

M. S.

(2001). Police sexual violence and rape myths: Civil liability under section 1983. Journal of Criminal Justice, 29, 389-405.

13.

Ex-Chief admits to sex charge. (2005, October 13). South Bend Tribune, B1.

14.

Ex-police officer starts serving jail sentence. (2004, February 11). Chicago Daily Herald, p. 3.

15.

Franklin

R. A.

(2005). 2005 survey of POST agencies regarding certification practices. Sykesville, MA: International Association of Directors of Law Enforcement Standards and Training.

16.

Geller

W. A.

Toch

(1995). And justice for all: Understanding and controlling police abuse of force. Washington, DC: Police Executive Research Forum.

17.

Goldman

R. L.

(2003). State revocation of law enforcement officers’ licenses and federal criminal prosecution: An opportunity for cooperative federalism. Saint Louis University Public Law Review XXII, 121-150.

18.

Goldman

R. L.

Puro

(1987, Fall). Decertification of police: An alternative to traditional remedies for police misconduct. Hastings Constitutional Law Quarterly, 15, 45-79.

19.

Goldman

R. L.

Puro

(2001). Revocation of police officer certification: A viable remedy for police violence? Saint Louis University Law Journal, 45, 541-580.

20.

Griswold

D. B.

(1994). Complaints against the police: Predicting dispositions. Journal of Criminal Justice, 22, 215-221.

21.

Harris

(2009). Exploring the relationship between experience and problem behaviors: A longitudinal analysis of officers from a large cohort. Police Quarterly, 12, 192-213.

22.

Hazelwood

Warren

(1990, February). The criminal behavior of the serial rapist. FBI Law Enforcement Bulletin, pp. 11-17.

23.

Hirschfield

P. J.

Simon

(2010). Legitimizing police violence: Newspaper narratives of deadly force. Theoretical Criminology, 14, 155-182.

24.

Hodgson

J. F.

(2001). Police violence in Canada and the USA: Analysis and management. Policing: An International Journal of Police Strategies &Management, 24, 520-549.

25.

Human Rights Watch. (1998). Shielded from justice: Police brutality and accountability in the United States. New York, NY: Author.

26.

Ivkovic

S. K.

(2005). Fallen Blue Knights. Oxford University Press.

27.

Jarriel

(1997, October 24). ABC News 20/20. New York, NY: ABC.

28.

Kappeler

V. E.

Sluder

R. D.

Alpert

G. P.

(1998). Forces of deviance: Understanding the dark side of policing (2nd ed.). Prospect Heights, IL: Waveland.

29.

Kappeler

V. E.

Vaughn

M. S.

(1997). Law enforcement: When the pursuit becomes criminal—Municipal liability for police sexual violence. Criminal Law Bulletin, 33, 352-376.

30.

Kraska

P. B.

Kappeler

V. E.

(1995). To serve and pursue: Exploring police sexual violence against women. Justice Quarterly, 12(1), 85-111.

31.

Kruttschnitt

Uggen

Shelton

(2000). Predictors of desistance among sex offenders: The interaction of formal and informal controls. Justice Quarterly, 17, 61-87.

32.

Lawrence

(2000). The politics of force: Media and the construction of police brutality. Berkeley: University of California Press.

33.

Lersch

K. M.

Feagin

J. R.

(1996). Violent police-citizen encounters: An analysis of major newspaper accounts. Critical Sociology, 22(2), 29-49.

34.

Lersch

K. M.

Mieczkowski

(1996). Who are the problem-prone officers? An analysis of citizen complaints. American Journal of Police, 15(3), 23-44.

35.

Maher

T. M.

(2003). Police sexual violence: Officers’ perceptions of its extent and causality. Criminal Justice Review, 28, 355-381.

36.

Maher

T. M.

(2008). Police chiefs’ views on police sexual misconduct. Police Practice & Research: An International Journal, 9, 239-250.

37.

Martin

(1998). Gender and organizations. In Myers

K. A.

Anderson

C. D.

Risman

B. J.

(Eds.), Feminist foundations; Towards transforming sociology (p. 324). Newbury Park, CA: SAGE.

38.

Martin

(1990). On the move: The status of women in policing. Washington, DC: Police Foundation.

39.

McGurrin

Kappeler

V. E.

(2002). Media accounts of police sexual violence: Rotten apples or state-supported violence? In Lersch

K. M.

(Ed.), Policing and misconduct (pp. 121-142). Upper Saddle River, NJ: Pearson Education.

40.

Parks

. (1982). Citizen surveys for police performance assessment: Some issues in their use. Urban Interest, 4, 17-26.

41.

Pate

A. M.

Fridell

L. A.

(1993). Police use of force: Official reports, citizen complaints, and legal consequences. Washington, DC: Police Foundation.

42.

Prokos

Padavic

(2002). “There oughtta be a law against bitches”: Masculinity lesson in police academy training. Gender, Work and Organization, 9, 440-459.

43.

Puro

Goldman

(1997). Police decertification: Changing patterns among the states, 1985-1995. Policing: An International Journal of Police Strategies & Management, 20, 481-496.

44.

Rabe-Hemp

. (2009). Survival in an all boys club: Policewomen and their fight for acceptance. Policing: An International Journal of Police Strategies and Management, 31(2), 251-270.

45.

Sapp

A. D.

(1997). Police officer sexual misconduct: A field research study. In Cromwell

P. F.

Dunham

R. G.

(Eds.), Crime and justice in America: Present realities and future prospects (pp. 139-151). Upper Saddle River, NJ: Prentice-Hall.

46.

Scully

(1990). Understanding sexual violence. New York, NY: HarperCollins.

47.

Scully

Marolla

(1984). Convicted rapists’ vocabulary of motive: Excuses and justifications. Social Problems, 31, 530-544.

48.

Seager

(1994, May 2). City pays Cochran $2 million in fees. Los Angeles Daily Journal, p. 2.

49.

Shay

(1999). Sexual abuse and civil rights: The impact of the PLRA physical injury requirement. National Prison Project Journal, 12(4), 1-5.

50.

Sherman

(1980). Causes of police behavior: The current state of quantitative research. Journal of Research in Crime and Delinquency, 17, 69-100.

51.

Shockey-Eckles

M. L.

(2011, August). Police culture and the perceptuation of the officer shuffle: The paradox of life behind “the blue wall.” Humanity & Society, 35, 290-309.

52.

$625 settlement in alleged cop rape. (2000, February 11). Chicago Tribune, Section 2, p. 2.

53.

Skolnick

J. H.

Fyfe

J. J.

(1993). Above the law: Police and the excessive use of force. New York, NY: Free Press.

54.

Spohn

Beichner

Davis-Frenzel

(2001). Prosecutorial justification for sexual assault case rejection: Guarding the gateway to justice. Social Problems, 48, 206-235.

55.

Stevens

D. J.

(1994). Predatory rapists and victim selection techniques. Social Science Journal, 31, 421-433.

56.

Terrill

McCluskey

J. D.

(2002). Citizen complaints and problem officers: Examining officer behavior. Journal of Criminal Justice, 30, 143-155.

57.

Third charge expected against police officer; Another woman alleges sex assault. (2006, August 10). Milwaukee Journal Sentinel, p. B1.

58.

Vaughn

M. S.

(1999). Police sexual violence: Civil liability under state tort law. Crime and Delinquency, 45, 334-357.

59.

Vaughn

M. S.

Coome

L. F.

(1995). Police civil liability under section 1983: When do police officers act under color of law? Journal of Criminal Justice, 23, 395-415.

60.

Walker

S. G.

Alpert

G. P.

Kenney

(2001). Early warning systems: Responding to the problem officer. Washington, DC: National Institute of Justice, Research in Brief.

61.

Walker

S. G.

Irlbeck

(2002). Driving while female: A national problem in police misconduct. Omaha: University of Nebraska at Omaha: Police Professionalism Initiative.

62.

Williamson

Baker

Jenkins

Cluse-Tolar

(2007). Police-prostitute interactions: Sometimes discretion, sometimes misconduct. Journal of Progressive Human Services, 18(2), 15-37