Expanding the measurement of police integrity

Policing: An International Journal of Police Strategies & Management Expanding t he measurement of police int egrit y Matthew J. Hickman Alex R. Piquero Zachary A. Powell Jack Greene Article information: Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) To cite this document: Matthew J. Hickman Alex R. Piquero Zachary A. Powell Jack Greene , (2016),"Expanding the measurement of police integrity", Policing: An International Journal of Police Strategies & Management, Vol. 39 Iss 2 pp. 246 - 267 Permanent link t o t his document : http://dx.doi.org/10.1108/PIJPSM-09-2015-0104 Downloaded on: 10 May 2016, At : 17: 13 (PT) Ref erences: t his document cont ains ref erences t o 118 ot her document s. To copy t his document : permissions@emeraldinsight . com The f ullt ext of t his document has been downloaded 54 t imes since 2016* Users who downloaded this article also downloaded: (2016),"The code of silence and ethical perceptions: Exploring police officer unwillingness to report misconduct", Policing: An International Journal of Police Strategies & Management, Vol. 39 Iss 2 pp. 370-386 http://dx.doi.org/10.1108/PIJPSM-10-2015-0108 (2016),"Exploring the viability of an attitudes toward ethical behavior scale in understanding police integrity outcomes", Policing: An International Journal of Police Strategies & Management, Vol. 39 Iss 2 pp. 319-337 http://dx.doi.org/10.1108/PIJPSM-11-2015-0130 (2016),"Special issue on police integrity: an introduction", Policing: An International Journal of Police Strategies & Management, Vol. 39 Iss 2 pp. - http://dx.doi.org/10.1108/PIJPSM-03-2016-0039 Access t o t his document was grant ed t hrough an Emerald subscript ion provided by emeraldsrm: 409222 [ ] For Authors If you would like t o writ e f or t his, or any ot her Emerald publicat ion, t hen please use our Emerald f or Aut hors service inf ormat ion about how t o choose which publicat ion t o writ e f or and submission guidelines are available f or all. Please visit www. emeraldinsight . com/ aut hors f or more inf ormat ion. About Emerald www.emeraldinsight.com Emerald is a global publisher linking research and pract ice t o t he benef it of societ y. The company manages a port f olio of more t han 290 j ournals and over 2, 350 books and book series volumes, as well as providing an ext ensive range of online product s and addit ional cust omer resources and services. Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) *Relat ed cont ent and download inf ormat ion correct at t ime of download. The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/1363-951X.htm PIJPSM 39,2 Expanding the measurement of police integrity Matthew J. Hickman 246 Received 11 September 2015 Revised 14 December 2015 Accepted 1 February 2016 Department of Criminal Justice, Seattle University, Seattle, Washington, USA Alex R. Piquero Program in Criminology, School of Economic, Political and Policy Sciences, University of Texas at Dallas, Richardson, Texas, USA Zachary A. Powell Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) Program in Criminology, University of Texas at Dallas, Richardson, Texas, USA, and Jack Greene School of Criminology and Criminal Justice, Northeastern University, Boston, Massachusetts, USA Abstract Purpose – Klockars et al. use scenario methodology to measure perceived seriousness, level of discipline warranted, and willingness to report fellow officers engaged in various negative behaviors. These data are used to characterize the occupational culture of integrity in a given agency, relative to other agencies. What remains unclear is whether these agency-level findings mask important meso- and micro-level variation in the data (i.e. at the precinct/district and officer levels) that may contribute to a more complete understanding of an agency’s culture of integrity. The paper aims to discuss these issues. Design/methodology/approach – This study replicates and extends Klockars et al.’s work using data from a survey administered to 499 Philadelphia police officers, with the goal of both validating their methodological approach and exploring the need for multi-level theory in the study of police integrity. In addition to comparing the results from Philadelphia to those obtained by Klockars et al., the authors test for differences across officer demographics, and explore variance in the willingness to report various behaviors at both the officer- and district-levels. Findings – Results indicate that bivariate relationships between officer-level demographics and willingness to report fellow officers are negated when controlling for theoretically relevant attitudinal variables such as cynicism and, consistent with Klockars et al., perceived seriousness of the underlying behavior. In addition, there is significant district-level variation in the average willingness to report fellow officers, and this variation can be explained by both organizational and environmental variables. On balance, the findings provide support for a multi-level approach to the study of police integrity. Originality/value – While the Klockars et al. approach addresses macro-level variation in police integrity, this study contributes important findings at the meso- and micro-levels. Keywords Police, Police culture Paper type Research paper Policing: An International Journal of Police Strategies & Management Vol. 39 No. 2, 2016 pp. 246-267 © Emerald Group Publishing Limited 1363-951X DOI 10.1108/PIJPSM-09-2015-0104 Introduction Police corruption, or the abuse of authority for personal gain (National Research Council, 2004, p. 268), has been an issue of central importance within the police profession and within the academic community (Goldstein, 1975; Klockars, 1999; Sherman, 1974). While the term corruption has morphed into other vernacular, such as integrity and most recently legitimacy and trust, the need for police departments that are trustworthy remains a critically important feature of police-citizen cooperation. Yet, there has been widespread debate about what integrity is, what behaviors would be Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) characterized as improper and under what circumstances, and how these concepts should be measured. Unsurprisingly, in their 2004 report “fairness and effectiveness in policing” the National Research Council observed an array of mixed findings regarding police integrity issues (p. 268), noting that the lack of an adequate measure of corruption is the culprit for this state of affairs (p. 269). In an effort to move beyond many of the limitations of previous work, Klockars et al. (2000) shifted the focus of integrity research to the officer level by asking about their attitudes and intentions within the context of morally ambiguous, potentially deviant situations. Their 30-department study found wide variability in integrity across departments and that officers were more likely to report misconduct for more – as opposed to less – serious behaviors. A recent news headline reinforces the importance of police integrity as a research topic: “Rat Cop” Joe Crystal shunned from Baltimore Police Department after reporting officer brutality (Murdock, 2015). Joe Crystal became a Baltimore police officer in 2008, and quickly advanced to a position as detective assigned to the Violent Crimes Impact Section. In October of 2011, following the arrest of a suspected drug dealer, Crystal’s two partners violently beat the suspect. Crystal reported this incident to an assistant state’s attorney, who in turn reported it to the state attorney’s police integrity unit. An investigation was opened that ultimately led to convictions for both of the involved officers in 2014. However, in the interim Crystal was labeled a “rat” and suffered serious harassment from within the agency, including his calls for backup being ignored by other officers, and an actual dead rat being placed under the windshield wiper on his car. Even after the convictions, Crystal reported continuing to suffer harassment that led him to resign from the Baltimore PD in late 2014. Taken in the context of the recent crisis of police legitimacy in the USA tied to concerns over the use of force (e.g. the deaths of Michael Brown, Eric Garner, Freddie Gray, and others), and the subsequent “President’s Task Force on 21st Century Policing” (Office of Community Oriented Policing Services, 2015) that was tasked with making recommendations to improve the quality of policing in the USA, understanding an agency’s culture(s) of integrity and the willingness of officers to report problematic behaviors takes on renewed importance. While the Klockars et al. study and its progeny are important, there is a need to continue research on the measurement of police integrity as well as to extend such research across departments, officers, and particularly sub-organizational units such as the police precinct/district, since these immediate work environments may be expected to substantively influence officer attitudes and behavior (Schafer, 2010a, b). Accordingly, this study uses data collected from a large sample of police officers in the Philadelphia Police Department (PPD) who were given the Klockars et al. integrity assessment. The goal is to replicate and extend their important work by assessing whether agency-level findings mask important meso- and micro-level variation (i.e. at the precinct/district and officer levels) that may be relevant for understanding a police agency’s culture of integrity. Our work begins to explore the need for multi-level theory in the study of police integrity. Before we present the results of our analysis, we provide a brief review of the integrity literature focussing primarily on the Klockars et al. study that we build upon herein. Prior research Police integrity The issue of police misconduct is critical to the policing profession and in particular to establishing the legitimacy of the police as a formal social control agent whose task is to uphold the rule of law. Not surprisingly there has been a voluminous literature on Measurement of police integrity 247 PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 248 police misconduct, yet this research is generally troubled by methodological concerns given the sensitivity of the subject, participation by research subjects, and other data concerns. Researchers try to work around these concerns by examining what might be considered the opposite of misconduct – integrity – which is largely focussed on assessing officer attitudes toward misconduct. As defined by Klockars and Kutnjak Ivković (2004), police integrity is “the normative inclination to resist the temptations to abuse the rights and privileges of their occupation” (p. 5). An expansion of this concept views police integrity as the product of policing behavior and a citizenry’s perspective of their police department’s behavior (Gurevitch et al., 1971; Hickman et al., 2004a; Lim et al., 2000; Meyer et al., 2013; Wu and Sun, 2010). Therefore, integrity can be construed as an end result of sorts, encompassing prior ethical experiences and the summation of new experiences that influence a police officer’s perspective. Research on police integrity arose in large part due to publicized instances of police corruption that received media attention in the 1960s and 1970s (Sherman, 1978). Stories of police officers receiving payoffs from prostitutes (Sherman, 1978), operating illegal gambling “pads” (Knapp Commission, 1972), and the diversion of cocaine away from drug raids (Sechrest and Burns, 1992) are indicative of the temptations inherent in positions of power. As such, studies on police integrity attempt to detect attitudes toward corruption through responses on police integrity measures (Klockars and Kutnjak Ivković, 2004). This methodology decreases the likelihood that a respondent will perceive a question as threatening, yielding a more truthful response. Through an exploration of police integrity, one can more readily assess how prevalent and persistent corruption is in a law enforcement organization by measuring the strength of an organization’s integrity (Kutnjak Ivković, 2009; Kutnjak Ivković and Kang, 2012). While it may be sufficient to examine official data on police misconduct through civilian complaints and court records (Prenzler, 2011), this approach only considers the organizational level and thus fails to account for individual-level variations in self-perceptions of police misconduct and integrity. In addition, data on civilian complaints come with long standing concerns about both under- and over-reporting, bias due to the fact that they are generated from a citizen’s perspective, and are subject to low sustain rates from review commissions (e.g. Chappell and Piquero, 2004; Pate and Fridell, 1993). Research in the USA on police integrity has declined in recent years, largely due to the sheer number of police agencies and the variation in organizational attitudes towards research in general (Klockars and Kutnjak Ivković, 2004; Lum et al., 2011; Telep and Lum, 2014). Thus, comparisons of police integrity across the nation are subject to substantial differences, making a national “picture” of police integrity difficult to illustrate. There has been greater success in studying police integrity internationally due to a larger number of national police forces and universal training academies for police officers (Gottschalk, 2010; Klockars et al., 1997; Kutnjak Ivković and Kang, 2012; Kutnjak Ivković and Khechymyan, 2013; Kutnjak Ivković and Sauerman, 2015; Kutnjak Ivković and Shelley, 2005; de Meijer et al., 2010; Prenzler, 2006, 2011). Measuring police integrity Police integrity research is often executed through the use of vignettes portraying a particular behavior, where police officers are then asked to read and respond to questions about their attitudes and intentions within the context of the hypothetical Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) scenario. In their large-scale, cross-national study, Klockars et al. (2000) undertook an organizational/occupational culture approach to the study of police integrity, relying on this scenario methodology. They focussed on officer estimations of the seriousness of behaviors, level of discipline warranted, and willingness to report fellow officers engaged in the behaviors portrayed in the scenarios. A key finding of their research was that the seriousness of the behavior was strongly related to the willingness to report (as well as the level of discipline), and that these data can be used to characterize an agency’s occupational culture of integrity relative to other agencies, both nationally and internationally. Studies that have used the Klockars et al. approach have consistently found that police officers who regard behavior as serious are more likely to report their fellow officers (Klockars et al., 1997, 2000, 2004; Klockars and Kutnjak Ivković, 2004). The validity of the Klockars et al. (2000) questionnaire has also been replicated in economic models of police integrity, with findings consistent to the original survey instrument (Marche, 2009). What remains unaddressed and under-researched is whether these agency-level findings mask important meso-level variation (i.e. at the precinct or district) as well as micro-level variation (i.e. officer) that may contribute to a more complete understanding of an agency’s culture of integrity. The majority of theoretical work in the study of police attitudes and behaviors can be classified within the psychological, sociological, anthropological, or organizational traditions (Kappeler et al., 1994; Worden, 1995), although a handful of scholars have recently explored multi-level theories addressing police behavior in general as well as police misconduct specifically (e.g. Herbert, 1998; Kane, 2002; Klinger, 1997). Kane (2002), for example, demonstrated that structural disadvantage, population mobility, and change in Latino population predicted variation in police misconduct among New York City precincts and districts. The accumulated evidence seems to suggest that police attitudes and behaviors cannot be fully understood without considering the interactions between officers, their work groups, and their work environment, and thus multi-level approaches will be useful (see Smith, 1984, 1986). Unfortunately, because of the methodological approach taken by Klockars et al., which focussed at the agency-level and was designed to avoid potential measurement problems associated with the “code of silence,” they did not collect additional data that would permit such analyses. As such, it is critical to build on their important research by collecting data for this purpose. Such an effort would serve to both validate their methodological approach (i.e. support for the organizational approach is bolstered if results consistent with Klockars et al. are obtained and there is no meso- and micro-level variation evidenced in the data), and aid in the development and construction of multi-level theory of police integrity. New directions in police integrity Police integrity research typically aggregates responses from individual police officers in order to infer the overall level of integrity within a police department. While there is nothing inherently wrong with this approach, studies designed in this manner do not account for possible mid-level variation in police integrity that may result from possible variation in officer attitudes across a police department (Smith, 1986), based in part on functional assignment or in larger cities on where officers are assigned spatially. Large police departments often subdivide a city into different patrol precincts, responsible for different areas of the city. Each area is likely to have contextual, environmental differences that affect the people who live there, the type of criminal activity most Measurement of police integrity 249 PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 250 prevalent in those areas, as well as the individuals who compose each area. In turn, this may also drive police officer decision-making or type of service provided, which may vary based on crime frequency and racial makeup of a neighborhood (Mastrofski and Willis, 2010; Smith, 1984, 1986). There is some evidence that crime may have an effect on police misconduct and in turn, police integrity (Eitle et al., 2014). Additionally, precincts may vary in their own organizational behavior. Mid-level managers often have some control over how they allocate and deploy police resources in their area, separate from main administrative requirements. Differing organizational (Smith, 1984) and managerial styles can have an effect on officers and shape not only their view of their work, but the views of officers around them (Ingram, 2013; Ingram and Lee, 2015; Kutnjak Ivković and Shelley, 2010; Mastrofski and Willis, 2010; Reiss, 1992; Sarver and Miller, 2014; Schafer, 2010a, b). In this way, moving from the macro to the meso-level of analysis within larger police agencies can help reveal variation in the cultures of these organizational subunits, particularly when integrity is of central concern. Toward this end, an examination of police integrity that combines multiple levels of analysis, including the neighborhood level (Smith, 1986), may provide insight that heretofore has not been identified in existing literature because of its primary focus on the individual-level response. In addition, since differences across basic demographics such as gender and race, and important attitudinal variables such as cynicism, concerning police integrity remain virtually unexplored, we also seek to expand police integrity research across these sources of variability. Gender Research into morality differences between men and women has primarily focussed on the extent to which there are gender differences in ethical decision-making. Several researchers argue that there are differences in degree (Gilligan, 1982, 1983; Pollock, 1995, 1999; Spader, 2002) but others claim that such differences are inflated and are non-existent (Baumrind, 1986; Walker, 1984). As Wilson (1993) notes, there is the possibility that morality differences do exist between men and women, but it is impossible to say that the causes of these differences are due to biology, social upbringing, or some combination thereof. Police departments that emphasize gender diversity in the workplace tend to embrace programs and accreditation consistent with police integrity ideals (Cordner and Cordner, 2011; Schuck, 2014). At the same time, research on women entering police work suggests that women are often pressured to choose between their occupational role (police women) and their gender identity (police women) (Martin, 1980). Therefore, one purpose of this study is to examine whether there are moral differences across gender within a police department. Race Prior research indicates that there is variation regarding perceptions of police work across race and ethnicity. Personal experiences with police officers often affect one’s purview of policing behavior; an individual is more inclined to change and shape their behavior based on prior experience (Wu, 2010; Wu and Sun, 2010; Wu et al., 2011). Whites tend to have more positive perceptions of police than persons of color, perceiving police activity as more legitimate (Bobo and Tuan, 2006; Weitzer and Tuch, 2004; Wu et al., 2009). Low perceptions of authority figures are often correlated with a decreased satisfaction with policing in general (Brown and Benedict, 2002; Garcia and Cao, 2005; Lai and Zhao, 2010; Wu et al., 2011). One study however, contradicts this Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) pattern of results, finding that persons of color often look towards police with respect because the circumstances of their life often lead them to rely on police services (Radelet and Carter, 1994). This finding is of interest to our work as there may be a heightened level of morality in minority police officers, as they may take their ethical violations more seriously because they wish to be a role model for people of color in their community (cf. Piliavin and Briar, 1964). At the same time, like women in policing, minorities also have to conform to the police occupational norms (Bolton and Feagin, 2004; Cashmore, 2001). Cynicism The study of officer cynicism has deep roots, and is generally framed as a type of strain theory (e.g. Niederhoffer, 1967). In brief, it is typically argued that as officers experience a disconnect between the ideal expectations of policing and the reality of policing, frustration builds as they try to maintain professional commitment. This frustration leads to cynicism: a sense of hopelessness in fulfilling the police role; contempt for police administrators who impose unrealistic goals and objectives; general distrust in the public; and frustration with the criminal justice system. Cynicism is thought to build over the course of officer careers, and officers adapt by either overcoming cynicism and returning to professional commitment; becoming part of an occupational subculture dedicated to cynicism; or (experiencing increasing apathy and alienation, escalating to a condition of anomie (see generally Niederhoffer, 1967). Cynicism has been shown to be related to poor relationships with other officers and supervisors, problematic behaviors and citizen complaints, as well low job satisfaction (Bennett and Schmitt, 2002; Hickman et al., 2004a, b; Regoli, 1977; Regoli et al., 1989, 1990a, b), and it might reasonably be expected that officers exhibiting greater cynicism would be less willing to report the negative behavior of fellow officers. Current focus The present study builds on the work of Klockars et al. in measuring police integrity by replicating and extending their data collection in one agency, the PPD (Greene et al., 2000). The survey instrument included six of the scenarios developed and used by Klockars et al. to represent various levels of seriousness. The selection of the six scenarios was driven by both the seriousness rankings reflected in the original Klockars et al. data, as well as considerations of the specific study location as detailed in Greene et al. (2000). We begin by first comparing our results in Philadelphia to those obtained by Klockars et al. in their national study. We then test for differences in the response data by gender and race of officer. Lastly, we explore variance in the willingness to report various behaviors at the officer and district levels. Sample The data are drawn from a survey administered to a sample of 499 police officers from the PPD. The population of study is all Philadelphia police officers assigned to patrol, including ranks police officer, sergeant, and lieutenant, as of January 2000. This amounts to 3,810 officers from the 23 Philadelphia patrol districts. A random sample of 504 officers was drawn from this population. Five officers refused to participate, leaving 499 cases available for analysis. Table I presents demographic statistics for the population of 3,810 officers and the final sample of 499 officers. As can be seen, the demographic characteristics of the sample closely resemble those of the larger PPD officer population. Measurement of police integrity 251 PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 252 Population (n ¼ 3,810) Sex Male Female 2,720 (71.4) 1,090 (28.6) 341 (68.3) 158 (31.7) Race White Black Latino Asian American Indian Other 1,915 1,614 238 31 8 4 232 228 31 7 1 0 Age Mean (SD) Min-max 35.22 (8.37) 20-75 35.14 (8.24) 20-61 8.04 (7.14) 0-48 7.46 (6.93) 0-37 Years’ service Mean (SD) Min-max Table I. Demographic statistics, survey population and sample Final sample (n ¼ 499) (50.3) (42.4) (6.2) (0.8) (0.2) (0.1) (46.5) (45.7) (6.2) (1.4) (0.2) (0.0) Rank Police officer 3,418 (89.7) 455 (91.2) Sergeant 302 (7.9) 35 (7.0) Lieutenant 90 (2.4) 9 (1.8) Notes: Age and years of service are reported here as measured at the time of sampling; years of service equal to zero indicates an officer with less than one year of service Scenarios The survey instrument included six of the scenarios used by Klockars et al.; as previously noted, these were selected to represent low, medium, and high degrees of seriousness. The wording of the scenarios were slightly modified to be gender non-specific, and the wording of the questions following the scenarios were slightly altered to be particular to the PPD. The six scenarios were as follows: (1) A police officer routinely accepts free meals, cigarettes, and other items of small value from merchants on his beat. The officer does not solicit these gifts and is careful not to abuse the generosity of those who gave the gifts to him. (2) A police officer is widely liked in the community, and on holidays local merchants and restaurant and bar owners show their appreciation for the officer’s attention by giving the officer gifts of food and liquor. (3) At 2 a.m. a police officer, who is on-duty, is driving his patrol car on a deserted road. The officer sees a vehicle that has been driven off the road and is stuck in a ditch. The officer approaches the vehicle and observes that the driver is not hurt but is obviously intoxicated. The officer also finds that the driver is a police officer. Instead of reporting this accident and offense the officer transports the driver to his home. (4) A police officer on foot patrol surprises a man who is attempting to break into an automobile. The man flees. The officer chases him for about two blocks before apprehending him by tackling him and wrestling him to the ground. Measurement of police (5) A police officer discovers a burglary of a jewelry shop. The display cases are integrity After he is under control the officer punches him a couple of times in the stomach as punishment for fleeing. smashed and it is obvious that many items have been taken. While searching the shop, the officer takes a watch, worth about two days’ pay for that officer. The officer reports the watch had been stolen during the burglary. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) (6) While on-duty, a police officer finds a wallet in a parking lot. It contains the amount of money equivalent to a full-day’s pay for that officer. The officer reports the wallet as lost property, but keeps the money. Variables For each of the scenarios, respondents were asked six questions. The first two questions were, “how serious do you consider this behavior to be?” and “how serious do most police officers in the PPD consider this behavior to be?”, with responses indicated on a five-point Likert scale ranging from “Not at all serious” (1) to “Very serious” (5). The third question asked, “would this behavior be regarded as a violation of official PPD policy?”, with responses of “Yes,” “No,” or “Not sure.” The fourth question asked, “if another officer engaged in this behavior and was discovered doing so, what if any discipline do you think should follow, and what if any discipline do you think would follow?” with responses including “None” (1), “Verbal Reprimand” (2), “Written Reprimand” (3), “Suspension without Pay” (4), “Demotion in Rank” (5), and “Dismissal” (6). The fifth and sixth questions were, “do you think you would report a fellow officer who engaged in this behavior?” and “do you think most officers in the PPD would report a fellow officer who engaged in this behavior?”, with responses indicated on a five-point Likert scale ranging from “Definitely not” (1) to “Definitely yes” (5). The dependent variable of interest for all six scenarios is the respondents’ estimated likelihood of reporting a fellow officer who engaged in the behavior portrayed. Following the work of Klockars et al., the primary independent variable of interestfor all six scenarios is the respondents’ estimation of the seriousness of the behavior portrayed. Additional officer level independent variables include race (black ¼ 1, non-black ¼ 0), sex (male ¼ 1, female ¼ 0), and years of service, as well as cynicism. Officer cynicism was measured using Regoli’s (1976) 20-item modified version of Niederhoffer’s (1967) cynicism scale. Responses to the cynicism items were coded as 1 (Strongly Disagree), 2 (Disagree), 3 (Neutral), 4 (Agree), and 5 (Strongly Agree), with 8 of the 20 items reverse coded. Item scores were summed, with lower total scores indicating greater levels of cynicism. The scale scores ranged from a low of 30 to a high of 85, and the mean value was equal to 59.08 (SD ¼ 7.60). Consistent with Regoli (1976), α for the cynicism scale was equal to 0.67, and a principal components analysis with Varimax rotation revealed evidence supporting both uni- and multi-dimensional solutions. Consistent with previous research, with only general predictions for cynicism (rather than for specific sub-scales) the scale used in the current study is a unidimensional measure (for additional detail see Greene et al., 2000; Hickman et al., 2004b). District-level characteristics for calendar year 2000 were included as potential predictors of the mean likelihood of reporting a fellow officer, including: violent crime rate (VCR); socio-economic status (SES); residential stability; and population heterogeneity. District-level VCR was measured as the total number of Part I violent offenses reported to the FBI per 1,000 residents, and varies from a low of 1.72 to a high of 25.11, with a mean of 12.92 (SD ¼ 6.39). Data for other district-level characteristics 253 PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 254 were obtained from the US Census Bureau and geoprocessed for Philadelphia police districts using ArcView (ESRI, Inc.), and are detailed below. District-level SES was measured via an additive index combining the percent of district residents unemployed, percent of adults and children living in poverty, percent without a high school education, and percent of households with incomes less than $10,000. Scores on the SES scale ranged from 35.6 to 121.8 (with higher scores indicating lower SES), with a mean of 76.3 (SD ¼ 27.8). Principal components analysis with Varimax rotation indicated unidimensionality and Cronbach’s α was 0.89. Residential stability (STABLE) was measured by adding the percentage of occupied housing units (range ¼ 74.2 to 96.8 percent) and the percentage of owner occupied households (range ¼ 26.2 to 73.6 percent). These variables correlate at 0.79 ( p o 0.01), and scores ranged from 106.6 to 168.2 (with higher scores indicating greater stability), with a mean of 140.6 (SD ¼ 19.7). Population heterogeneity (POP_HET) was measured as the proportion of residents who are white multiplied by the proportion of residents who are not white. This variable ranges from values of 0.03 to 0.25 (with higher scores indicating greater racial heterogeneity), with a mean of 0.14 (SD ¼ 0.07). Finally, we also explore some district-level demographic characteristics including the proportion of officers who are male (mean ¼ 0.68, SD ¼ 0.07, min ¼ 0.58, max ¼ 0.84), the proportion non-white (mean ¼ 0.54, SD ¼ 0.20, min ¼ 0.04, max ¼ 0.81), and the average years of service (mean ¼ 7.44, SD ¼ 1.92, min ¼ 4.13, max ¼ 11.50). Analytic plan We begin with descriptive results by comparing the Philadelphia data with the Klockars et al. data in terms of the mean responses to the scenario questions. Next, we will explore demographic variation in the Philadelphia data, using t-tests to determine whether statistically significant differences exist across officer gender and race in the scenario responses. We will then proceed with multivariate modeling by exploring whether there is district-level variation in the scenario responses. We following the general modeling process suggested by Bryk and Raudenbush (1992). A fully unconditional ANOVA model will be used to determine whether there is a statistically significant amount of between-group variance in the dependent variables of interest. Provided that unconditional models reveal a significant amount of level-2 variance in the dependent variables, we will select the variable(s) having the greatest proportion of level-2 variance to be explained. For this variable, the next step will be to use this value as a base measure against which to compare the remaining level-2 variance when additional predictors of interest are included in a hierarchical linear model. All level-1 predictors will be group-mean centered, which changes the interpretation of the level-1 intercept from the expected score on the dependent variable for an officer in district k when all independent variables are equal to zero (a condition which is not possible in the present case and thus renders the intercepts meaningless), to the district k (unadjusted) mean scores on self-reported likelihood of reporting a fellow officer. Group-mean centering the level-1 predictors also pools variance within the groups (i.e. districts) and thus isolates level-1 and level-2 variation, which has the effect of maximizing level-2 (district) variation (Bryk and Raudenbush, 1992). Next, the slopes for level-1 predictors will be allowed to vary (i.e. police districts will take on their own slope coefficients for independent variables) in order to explore whether their individual effects evince ecological variation. If varying slopes add Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) significantly to the model (i.e. if there is a significant amount of variation and level-2 variance is reduced) they will be retained; otherwise, the slopes will be constrained. Finally, level-2 predictors will be introduced for the level-1 intercept while monitoring variance results. Predictors include district-level crime rate, SES, residential stability, and population heterogeneity. Each level-2 variable will be entered alone, and then all significant level-2 variables in combination. The same process will be followed for any varying level-1 slope coefficients. Measurement of police integrity 255 Results Philadelphia data compared to national sample Table II presents the scenario data obtained in Philadelphia, as well as the corresponding data obtained by Klockars et al. An important point about the data obtained in this study is that it does not appear to differ substantially from that obtained by Klockars et al. in their multi-site study. With regard to officer estimations of the seriousness of behaviors, the mean values across scenarios tend to be slightly lower in the Philadelphia sample as compared to Klockars et al.’s national sample (the only exception is with regard to personal judgment of seriousness in scenario 3, the DUI cover-up). Officer estimations of the likelihood of reporting a fellow officer also tended to be slightly lower in the Philadelphia sample, with the exception of scenarios 1 (free meals, cigarettes, and items of small value) and 3 (DUI cover-up). With regard to appropriate and likely discipline, officers in the Philadelphia sample tended to indicate Survey question Scenario 1 (free Scenario meals, Scenario 5 (theft cigarettes, Scenario from 4 and items 2 (free Scenario crime 3 (DUI (physical holiday of small scene) gifts) cover-up) abuse) value) ppd k ppd k ppd k ppd k ppd k Scenario 6 (takes $ from lost wallet) ppd k 1. How serious do you consider this behavior to be? 2.47 2.60 2.70 2.84 3.58 3.03 3.94 4.05 4.87 4.95 4.80 4.85 2. How serious do most police officers in the PPD consider this behavior to be? 2.17 2.31 2.33 2.64 3.25 2.86 3.43 3.70 4.66 4.88 4.47 4.69 3. If another officer engaged in this behavior and was discovered doing so, what if any discipline do you think should follow? 1.92 2.13 2.06 2.53 3.01 2.81 3.18 3.76 5.25 5.66 4.95 5.09 And what if any discipline do you think would follow? 2.43 2.37 2.57 2.82 3.56 3.21 3.61 4.00 5.39 5.57 5.12 5.03 4. Do you think you would report a fellow officer who engaged in this behavior? 2.14 1.94 2.24 2.36 2.87 2.34 3.16 3.39 4.12 4.54 4.02 4.23 5. Do you think most police officers in the PPD would report Table II. a fellow officer who engaged in Descriptive statistics, this behavior? 1.92 1.82 2.02 2.28 2.62 2.28 2.77 3.07 3.69 4.34 3.62 3.96 Philadelphia sample and Klockars et al. Notes: ppd, refers to the mean for the Philadelphia sample (499 officers); and k, refers to the mean for national sample Klockars et al.’s US sample (30 agencies, 3,235 officers) PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 256 lower levels of discipline that should follow (again, the exception being the DUI coverup scenario), and a mixed range of discipline that would follow (slightly higher on scenarios 1, 3, and 6; slightly lower on scenarios 2, 4, and 5). Officer race and gender With regard to gender, female officers on average considered the behavior portrayed in scenarios 3 (DUI cover-up) and 4 (physical abuse) to be more serious than did male officers (t ¼ 2.65, 2.50, respectively). Female officers on average thought that lower levels of punishment should follow for the behavior portrayed in scenario 1 (free meals, cigarettes, and items of small value), as compared to male officers (t ¼ −2.10). Female officers on average thought that lower levels of punishment would follow for the behavior portrayed in scenario 4 (physical abuse) (t ¼ −2.80). Finally, female officers on average indicated they would be less likely to report a fellow officer for the behavior portrayed in scenario 1 (free meals, cigarettes, and items of small value) (t ¼ −2.39), and female officers also indicated that most officers in the PPD would be less likely to report another officer for this behavior (t ¼ −2.40). With regard to race, black officers on average considered the behavior portrayed in scenarios 3 (DUI cover-up) and 4 (physical abuse) to be more serious than did white officers (t ¼ 2.93, 4.83, respectively). Black officers on average also thought that other officers in the PPD would consider the behavior in scenario 3 (DUI cover-up) to be more serious, as compared to white officers (t ¼ 2.33). Black officers on average thought that lower levels of punishment should follow for the behaviors portrayed in scenarios 2 (free holiday gifts), 5 (theft from crime scene), and 6 (takes money from lost wallet), as compared to white officers (t ¼ −2.19, −2.69, −2.49). Black officers on average thought that lower levels of punishment would follow for the behavior portrayed in scenario 6 (takes money from lost wallet) (t ¼ −2.29). Finally, black officers on average indicated they would be more likely to report a fellow officer for the behaviors portrayed in scenarios 3 (DUI cover-up) and 4 (physical abuse) (t ¼ 2.65, 2.93). Officer- and district-level variance The results from the initial ANOVA models are presented in Table III. As can be seen, there is no significant level-2 variance in reporting for scenario 6 (taking a large amount of money from a lost wallet). Officers universally indicated that they would be willing to report a fellow officer who engaged in this behavior. Level-2 variance in the remaining scenarios/responses (i.e. the interclass correlation coefficient) varies from 2.7 percent (scenario 1) to 5.7 percent (scenario 4). Because it registered the highest between officer variability (and is of particular interest due to recent and increased media attention), we selected scenario 4 (physical abuse scenario) for subsequent analyses. Table III. Variance decomposition (estimates of reporting) S1 S2 S3 S4 S5 S6 (free meals, cigarettes) (free holiday gifts) (DUI cover-up) (physical abuse) (theft from crime scene) (takes $ from lost wallet) Between Within 0.027 0.034 0.037 0.057 0.028 0.013 0.974 0.967 0.966 0.946 0.973 0.988 χ2 between 34.99, 38.89, 40.25, 50.62, 35.56, 28.32, p ¼ 0.039 p ¼ 0.015 p ¼ 0.010 p ¼ 0.001 p ¼ 0.034 p ¼ 0.165 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) We began with the fully unconditional model with no level-1 predictors and a random intercept term as described above. We refer to this as the “base model” and the coefficients and variance components are reported in the first column of Table IV. While the fixed effect of the intercept is not significantly different from zero (to be expected since the dependent variable is z-scored), the variance component for the intercept is significant ( χ2 ¼ 50.62, p o 0.001). As previously noted, the intraclass correlation coefficient is equal to 0.057, indicating that the random effect of police district on the mean likelihood of reporting accounts for about 6 percent of the total effects. The deviance statistic is equal to 1,409.80, which will serve as the basis of comparison for improvement in fit in subsequent models. We next introduced the level-1 predictors (sex, race, years of service, cynicism, and seriousness of the scenario, all group-mean centered), and explored variance in each of the level-1 slope coefficients by adding a random effect for each coefficient in turn. We found significant variation in the coefficient for officer sex, indicating that the effect of officer sex on the likelihood of reporting a fellow officer varied across police districts, and retained the random effect for the slope coefficient in the model. We refer to this as Model 1, and the resulting regression coefficients and variance components are presented in the second column of Table IV. As can be seen, only cynicism and the degree of seriousness attributed to the scenario were significant predictors of the likelihood of reporting a fellow officer for physical abuse. Specifically, cynicism had a positive effect indicating that officers who scored higher on the cynicism scale (meaning they are less cynical, as higher scores correspond to lower cynicism) reported a higher likelihood of reporting a fellow officer Fixed effect coeff Base model SE t-ratio coeff For INTRCPT1, β0 INTRCPT2, γ00 0.004 SEX_P, γ01 STABLE, γ02 0.065 0.060 Model 1 SE t-ratio coeff Model 2 SE t-ratio 0.006 0.066 0.093 1.900 −1.946 −0.004 0.323 0.531 0.002 5.883* −3.667** −2.432*** −0.031 0.102 −0.300 −0.015 0.102 −0.148 For NON-WHITE slope, β2 INTRCPT2, γ20 0.088 0.083 1.058 0.096 0.083 1.149 For YEARS slope, β3 INTRCPT2, γ30 0.006 0.004 1.423 0.006 0.004 1.349 For CYNICISM slope, β4 INTRCPT2, γ40 0.017 0.005 3.703* 0.017 0.005 3.653* 0.509 0.050 10.203* 0.509 0.050 10.125* For SEX slope, β1 INTRCPT2, γ10 For SERIOUS slope, β5 INTRCPT2, γ50 Random effect SD Variance component 0.239 0.057 INTRCPT1, u0 SEX slope, u1 Level-1, r 0.972 χ 2 50.62* 0.946 Notes: *p o 0.001; **p o 0.01; ***p o0.05 SD Variance component 0.271 0.073 0.321 0.103 0.796 0.634 χ 2 75.58* 38.71*** SD Variance component 0.203 0.041 0.318 0.101 0.795 0.633 Measurement of police integrity 257 χ2 47.42* 38.78*** Table IV. Hierarchical regression results for S4 (physical abuse) PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 258 for physical abuse. Seriousness had a positive effect indicating that officers who rated the scenario as more serious were also more likely to report a fellow officer. Officer sex, race, and years of service were not significant predictors. This model explains about 33 percent of the level-1 variance in the likelihood of reporting a fellow officer for physical abuse. The proportion of total remaining variance attributable to the random intercept is equal to 9 percent and is statistically significant ( χ2 ¼ 75.58, p o 0.001), and the proportion attributable to the random slope coefficient is equal to 13 percent and is statistically significant ( χ2 ¼ 38.71, p ¼ 0.015). The deviance statistic dropped from 1409.80 to 1248.17 ( χ2 ¼ 161.63, p o 0.001), indicating significant improvement in fit over the base model. The next step was to explore level-2 predictors for the intercept, one at a time. These included the district-level VCR, an indicator of SES, residential stability, population heterogeneity, as well as the proportion of officers who are male, proportion non-white, and average years of service. Individually, VCR was significant and positive (indicating that districts with higher VCRs had higher average likelihoods of reporting fellow officers); SES was significant and positive (indicating that districts characterized by higher SES had higher average likelihoods of reporting fellow officers); residential stability was significant and negative (indicating that districts characterized by higher residential stability had lower average likelihoods of reporting fellow officers); and the proportion of officers who are male was significant and negative (indicating that districts with a higher proportion of male officers had lower average likelihoods of reporting fellow officers). Population heterogeneity, the proportion of officers who are non-white, and the average years of service were not significant predictors of the intercept. In combining the level-2 predictors, we ultimately retained district-level residential stability and the proportion of male officers in the district. We refer to this as Model 2 and the resulting regression coefficients and variance components are presented in the third column of Table IV. Compared to the previous model, this model explains 44 percent of the level-2 variance in the intercept. The proportion of total remaining variance attributable to the random intercept is equal to 5 percent and is statistically significant ( χ2 ¼ 47.42, p o 0.001), and the proportion attributable to the random slope coefficient is equal to 13 percent and is statistically significant ( χ2 ¼ 38.78, p ¼ 0.015). While the deviance statistic did not change significantly (deviance ¼ 1,246.07; χ2 ¼ 2.10, p o 0.500) indicating no better fit than the previous model, these level-2 findings are nonetheless of significant theoretical importance. Finally, we explored level-2 predictors for the varying slope coefficient for sex, again one at a time. We found that none of the district-level variables explained variation in the slope coefficient for officer sex. This leaves us with a significant random effect for officer sex on the likelihood of reporting, even though the fixed effect for officer sex is not statistically significant. What this means is that the effect of officer sex on the likelihood of reporting varies across police districts (i.e. the random effect), but the average effect of officer sex on reporting (i.e. the fixed effect) is zero. Discussion The goal of this effort was to build on the prior research of Klockars et al. by collecting and presenting data that would serve to both validate their approach to the measurement of police integrity, as well as to explore the need for construction and development of multi-level theory in police integrity research. Using survey data obtained from a large sample of police officers in the PDD, we found results similar to Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) those obtained by Klockars et al., and only a few response differences across gender and race of officer that were negated when controlling for seriousness. This suggests that the methodological approach used by Klockars et al., designed to avoid effects of the “code of silence” by not soliciting additional identifying information about respondents, provides a reasonable picture of agency integrity. At the same time, and extending the important work of Klockars et al., our models indicated a statistically significant amount of level-2 (district-level) variance in the likelihood of reporting a fellow officer for most of the scenarios. The greatest proportion of level-2 variance to be explained was found for a physical abuse scenario. We focussed on this scenario for subsequent analyses, and found that, in addition to estimations of the seriousness of the behavior and cynicism at the officer level, several district-level characteristics including the “maleness” of a police district (i.e. the proportion of officers in a police district who are male) are related to the mean likelihood of reporting fellow officers who engage in physical abuse. Specifically, the more “male” a police district, the less likely officers are to report a fellow officer who engages in physical abuse. In addition, while the individual effect of gender is negated by seriousness, the individual effect of gender varies across districts. Overall, our findings suggest that an agency-level approach to the study of police integrity may be more appropriate in smaller departments. Larger departments having many precincts or districts will be more likely to exhibit district-level variation in the extent to which officers are willing to report the negative behaviors of other officers. This is an important element of police integrity that needs both theoretical development and further empirical investigation. Note that 60 percent of the respondents in Klockars et al.’s national sample were from “Very Large” agencies having 500+ sworn officers. It may be the case that homogeneity is in itself an important indicator of agency integrity; agencies exhibiting a large degree of districtlevel variation in willingness to report may have a greater degree or a different kind of integrity problem that would be masked by the agency-level approach. Our findings suggest that research along the lines initiated by Klockars et al. should be continued. While our results are generally consistent with Klockars et al., it does appear that continued theoretical development is warranted. Much work on this front remains to be done, as the present research only investigated demographic variables, one attitudinal variable, and only focussed on a physical abuse scenario. Future work should focus on all types of officer behavior, controls for additional measures should be explored at both the individual and district levels, and additional district-level variables should be investigated. The Klockars et al. survey instrument may require an update to capture a more complete image of police corruption, given that many police officers view the standard vignette scenarios as serious, resulting in a lack of variation in responses (Gottschalk, 2010; Kutnjak Ivković and Kang, 2012; Kutnjak Ivković and Khechymyan, 2013). With respect to policy implications, our results suggest that police departments could continue their efforts at not only targeting potential police misconduct but more important developing training and prevention strategies that begin upon entry into the academy and/or when officers transfer to a new agency. These strategies, which would require that departments initiate and maintain data collection systems and assign appropriate (internal affairs) staff to analyze such data, could involve efforts such as complaint profiling and early intervention systems that include police organizational units (see, e.g. Alpert and Walker, 2000; Prenzler and Porter, 2015; Walker et al., 2000). A related concern is that early intervention systems that do not consider the immediate Measurement of police integrity 259 PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 260 work context will tend to be somewhat inefficient, since behavioral norms, and integrity, may vary across these organizational units. Additionally, research in police integrity may benefit by including survey items that gauge a police officer’s responses on misconduct categories such as prisoner mistreatment, physical abuse, and evidence manipulation (Gottschalk, 2010). Others suggest moving towards using video-based situational judgment tests (de Meijer et al., 2010). Behavioral measures have also shown to be strong predictors of attitudes towards police integrity and misconduct, suggesting that these factors should be included in gauging opinions on integrity issues (Sellbom et al., 2007). Prenzler (2006) notes that some police departments may benefit by using random “integrity tests” to assess the level of corruptness of a police department. Integrity tests create a “real” misconduct scenario and unaware participants have their integrity assessed based on their reactions to these simulated misconduct scenarios (Her Majesty’s Inspectorate of Constabulary, 1999; Knapp Commission, 1972; Mollen, 1994). Further, it may be beneficial to gauge integrity not only with police officers, but civilian, unsworn staff who support police activity. Additionally, living in the information age has expanded the ways that the general public comes into contact with each other, especially through the use of social media. Future research into police integrity may wish to include items on social media indiscretions by police officers (Goldsmith, 2015). Social media provides the platform and opportunity to communicate with large numbers of people in just a few clicks of a keyboard (Lieberman et al., 2013); the ability to damage the reputation of a police officer is more accessible than ever and can have severe consequences on a police department’s interactions with the public. Keeping in mind that measuring police integrity offers one indirect inference into the potential level of problem behavior in a police agency, the Klockars et al. (1997, 2000) use of vignettes may not act as an ideal reliable proxy for police corruption, but it certainly offers an excellent baseline from which to build upon. One proposed alternative is the Police Ethical Behavior Scale (PEBS) which directly asks respondents to give opinions on police corruption (Aremu et al., 2011). We acknowledge that asking police officers directly about corruption in a police department is subject to the possibility of inaccurate and inconsistent answers; however, this scale is relatively new and should have its validity and reliability tested in other research studies and perhaps compared to the Klockars et al. methodological approach. Finally, it is also important to point out that, as is the case with most surveys of police integrity, there is always the potential for omitted variable bias as other variables that are relevant to the study of police integrity are not captured. Some candidate variables for future researchers to capture would include other demographic characteristics, such as education, as well as individual-level perceptions of the organization such as job satisfaction and organizational commitment, all of which have shown utility in studying police corruption and misconduct. Conclusion Research on police agency integrity can certainly benefit from broader assessments of police culture (for a review see, Crank, 2015; Foster, 2003, pp. 196-227). Whereas in the past police culture was often presented as monolithic, overly structured, and invariant to change, recent assessments of police culture find a multiplicity of “cultures” within police agencies, each more focussed on differing task environments confronting the police (management vs the street, investigations vs patrol, street crime units vs general patrol, community policing vs motorized patrol, and the like). Of course, larger Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) police agencies are likely to confront wide variation in the sub-sections of large cities they serve and protect, perhaps making assessments of police agency integrity more difficult in the face of the variation in police districts revealed in this Philadelphia study. At the same time it can be argued that the “crime control-crime attack” model of policing that has taken hold of policing worldwide, fueled by concerns with domestic security and crime, has become an overarching philosophy of many Western police agencies. Police overreach and the use of aggressive police tactics has helped shaped macro-police cultures, consistent with the New Institutionalism (DiMaggio and Powell, 1983) perspectives of organizational dynamics (see Willis, 2014). Concerns with crime and now terrorism have tended to narrow the perspective of the police, often introducing concerns about the integrity and legitimacy of police agencies. As a result, police integrity affairs must be cognizant of the macro-level pressures brought on police agencies, including institutional mimicry of police strategies and tactics, as well as considerable variation in the “communities” the police ultimately serve. Although the large-scale study of police integrity is a challenging endeavor that will require a substantial investment of resources, the current crisis of police legitimacy in the USA demands that this type of research be prioritized. References Alpert, G. and Walker, S. (2000), “Police accountability and early warning systems: developing policies and programs”, Justice Research and Policy, Vol. 2 No. 2, pp. 59-72. Aremu, A.O., Pakes, F. and Johnston, L. (2011), “The moderating effect of emotional intelligence on the reduction of corruption in the Nigerian police”, Police Practice and Research, Vol. 12 No. 3, pp. 195-208. Baumrind, D. (1986), “Sex differences in moral reasoning: response to Walker’s (1984) conclusion that there are none”, Child Development, Vol. 57 No. 2, pp. 511-521. Bennett, R. and Schmitt, E. (2002), “The effect of work environment on levels of police cynicism”, Police Quarterly, Vol. 5 No. 4, pp. 493-523. Bobo, L. and Tuan, M. (2006), Prejudice in Politics: Group Position, Public Opinion, and the Wisconsin Treaty Rights Dispute, Harvard University Press, Cambridge, MA. Bolton, K. and Feagin, J.R. (2004), Black in Blue: African-American Police Officers and Racism, Routledge, New York, NY. Brown, B. and Benedict, W.R. (2002), “Perceptions of police: past findings, methodological issues, conceptual issues and policy implications”, Policing: An International Journal of Police Strategies & Management, Vol. 25 No. 3, pp. 543-578. Bryk, A. and Raudenbush, S. (1992), Hierarchical Linear Models, Sage, Thousand Oaks, CA. Cashmore, E. (2001), “The experiences of ethnic minority police officers in Briton: under recruitment and racial profiling in a performance culture”, Ethnic & Racial Studies, Vol. 24 No. 4, pp. 642-659. Chappell, A. and Piquero, A. (2004), “Applying social learning theory to police misconduct”, Deviant Behavior, Vol. 25 No. 2, pp. 89-108. Cordner, G. and Cordner, A. (2011), “Stuck on a plateau? Obstacles to recruitment, selection, and retention of women police”, Police Quarterly, Vol. 14 No. 3, pp. 207-226. Crank, J.P. (2015), Understanding Police Culture, Routledge, New York, NY. Measurement of police integrity 261 PIJPSM 39,2 de Meijer, L.A., Born, M.P., van Zielst, J. and Henk, T. (2010), “Construct-driven development of a video-based situational judgment test for integrity: a study in a multi-ethnic police setting”, European Psychologist, Vol. 15 No. 3, pp. 229-236. DiMaggio, P.J. and Powell, W. (1983), “The iron cage revisited: institutional isomorphism and collective rationality in organizational fields”, American Sociological Review, Vol. 48 No. 2, pp. 147-160. 262 Eitle, D., D’Alessio, S.J. and Stolzenberg, L. (2014), “The effect of organizational and environmental factors on police misconduct”, Police Quarterly, Vol. 17 No. 2, pp. 103-126. Foster, J. (2003), “Police cultures”, in Newburn, T. (Ed.), Handbook of Policing, Willan, Devon, pp. 196-227. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) Garcia, V. and Cao, L. (2005), “Race and satisfaction with the police in a small city”, Journal of Criminal Justice, Vol. 33 No. 2, pp. 191-199. Gilligan, C. (1982), In a Different Voice, Harvard University Press, Cambridge, MA. Gilligan, C. (1983), “Do the social sciences have an adequate theory of moral development?”, in Haan, N. (Ed.), Social Sciences as Moral Inquiry, Columbia University Press, New York, NY, pp. 87-100. Goldsmith, A. (2015), “Disgracebook policing: social media and the rise of police indiscretion”, Policing & Society, Vol. 25 No. 3, pp. 249-267. Goldstein, H. (1975), Police Corruption: A Perspective on its Nature and Control, Police Foundation, Washington, DC. Gottschalk, P. (2010), “Crime-based survey instrument for police integrity measurement”, Policing: An International Journal of Police Strategies & Management, Vol. 33 No. 1, pp. 52-68. Greene, J.R., Piquero, A.R., Hickman, M.J. and Lawton, B. (2000), Police Integrity and Accountability in Philadelphia: Predicting and Assessing Police Misconduct, Department of Justice, National Institute of Justice, Washington, DC. Gurevitch, M., Danet, B. and Schwartz, G. (1971), “The image of the police in Israel”, Law and Society Review, Vol. 5 No. 3, pp. 367-388. Her Majesty’s Inspectorate of Constabulary (1999), Police Integrity, England, Wales, and Northern Ireland: Securing and Maintaining Public Confidence, Her Majesty’s Inspectorate of Constabulary, London. Herbert, S. (1998), “Police subculture reconsidered”, Criminology, Vol. 36 No. 2, pp. 343-370. Hickman, M.R., Piquero, A.R. and Greene, J.R. (2004a), “Police integrity: exploring the utility of a risk factor model”, in Hickman, M.J., Piquero, A.R. and Greene, J.R. (Eds), Police Integrity and Ethics, Wadsworth/Thomson Learning, Belmont, CA, pp. 67-83. Hickman, M.J., Piquero, N.L. and Piquero, A.R. (2004b), “The validity of Niederhoffer’s cynicism scale”, Journal of Criminal Justice, Vol. 32 No. 1, pp. 1-13. Ingram, J.R. (2013), “Supervisor-officer fit and role ambiguity: re-assessing the nature of the sergeant-officer attitudinal relationship”, Policing: An International Journal of Police Strategies & Management, Vol. 36 No. 2, pp. 375-398. Ingram, J.R. and Lee, S.U. (2015), “The effect of first-line supervision on patrol officer job satisfaction”, Police Quarterly, Vol. 18 No. 2, pp. 193-219. Kane, R.J. (2002), “The social ecology of police misconduct”, Criminology, Vol. 40 No. 4, pp. 867-896. Kappeler, V.E., Sluder, R.D. and Alpert, G.P. (1994), Forces of Deviance, Waveland Press, Long Grove, IL. Klinger, D.A. (1997), “Negotiating order in patrol work: an ecological theory of police response to deviance”, Criminology, Vol. 35 No. 2, pp. 277-306. Klockars, C. and Kutnjak Ivković, S. (2004), “Measuring police integrity”, in Hickman, M.J., Piquero, A.R. and Greene, J.R. (Eds), Police Integrity and Ethics, Wadsworth/Thomson Learning, Belmont, CA, pp. 3-20. Klockars, C.B. (1999), “Some really cheap ways of measuring what really matters”, in Langworthy, R.H. (Ed.), Measuring What Matters: Proceedings from the Policing Research Institute Meetings, Department of Justice, National Institute of Justice, Washington, DC, pp. 195-214. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) Klockars, C.B., Kutnjak Ivković, S. and Haberfeld, M.R. (1997), “The measurement of police integrity”, final report, Department of Justice, National Institute of Justice, Washington, DC. Klockars, C.B., Kutnjak Ivković, S. and Haberfeld, M.R. (2004), “The contours of police integrity”, in Klockars, C.B., Kutnjak Ivkovich, S.K. and Haberfeld, M.R. (Eds), The Contours of Police Integrity, Sage Publications, Thousand Oaks, CA, pp. 1-18. Klockars, C.B., Haberfeld, M.R., Kutnjak Ivković, S.K. and Uydess, A. (2002), “A minimum requirement for police corruption”, in Silverman, R.A., Thornberry, T.P., Cohen, B. and Krisberg, B. (Eds), Crime and Justice at the Millennium: Essays by and in Honor of Marvin E. Wolfgang, Springer Publishing Company, New York, NY, pp. 185-207. Klockars, C.B., Kutnjak Ivković, S., Harver, W.E. and Haberfeld, M.R. (2000), The Measurement of Police Integrity, NIJ Research in Brief, Department of Justice, National Institute of Justice, Washington, DC. Knapp Commission (1972), “The Knapp Commission Report on Police Corruption”, George Braziller, New York, NY. Kutnjak Ivković, S. (2009), “The croatian police, police integrity, and transition toward democratic policing”, Policing: An International Journal of Police Strategies & Management, Vol. 32 No. 3, pp. 459-488. Kutnjak Ivković, S. and Kang, W. (2012), “Police integrity in South Korea”, Policing: An International Journal of Police Strategies & Management, Vol. 35 No. 1, pp. 76-103. Kutnjak Ivković, S. and Khechymyan, A. (2013), “The state of police integrity in Armenia: findings from the police integrity survey”, Policing: An International Journal of Police Strategies & Management, Vol. 36 No. 1, pp. 70-90. Kutnjak Ivković, S. and Sauerman, A. (2015), “Threading the thin blue line: transition towards democratic policing and the integrity of the South African police service”, Policing and Society: An International Journal of Research and Policy, Vol. 25 No. 1, pp. 25-52. Kutnjak Ivković, S. and Shelley, T.O. (2005), “The Bosnian police and police integrity: a continuing story”, European Journal of Criminology, Vol. 2 No. 4, pp. 428-464. Kutnjak Ivković, S. and Shelley, T.O. (2010), “The code of silence and disciplinary fairness”, Policing: An International Journal of Police Strategies & Management, Vol. 33 No. 3, pp. 548-574. Lai, Y. and Zhao, J. (2010), “The impact of race/ethnicity, neighborhood context, and police/citizen interaction on residents’ attitudes toward the police”, Journal of Criminal Justice, Vol. 38 No. 4, pp. 685-692. Lieberman, J.D., Koetzle, D. and Sakiyama, M. (2013), “Police departments’ use of Facebook: patterns and policy issues”, Police Quarterly, Vol. 16 No. 4, pp. 438-462. Lim, V.K.G., Teo, T.S.H. and See, S.K. (2000), “Perceived job image among police officers in Singapore: factorial dimensions and differential effects”, Journal of Social Psychology, Vol. 140 No. 6, pp. 740-750. Measurement of police integrity 263 PIJPSM 39,2 264 Lum, C., Koper, C. and Telep, C.W. (2011), “The evidence-based policing matrix”, Journal of Experimental Criminology, Vol. 7 No. 3, pp. 3-26. Marche, G.E. (2009), “Integrity, culture, and scale: an empirical test of the big bad police agency”, Crime, Law, and Social Change, Vol. 51 No. 5, pp. 463-486. Martin, S.E. (1980), Breaking and Entering: Police Women on Patrol, University of California Press, Berkeley, CA. Mastrofski, S.D. and Willis, J.J. (2010), “Police organization continuity and change: into the twenty-first century”, in Tonry, M. (Ed.), Crime and Justice: A Review of Research, Vol. 39 No. 1, University of Chicago Press, Chicago, IL, pp. 55-144. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) Meyer, M.E., Steyn, J. and Gopal, N. (2013), “Exploring the public parameter of police integrity”, Policing: An International Journal of Police Strategies & Management, Vol. 36 No. 1, pp. 140-156. Mollen, M. (1994), “Commission to investigate allegations of police corruption and the anticorruption procedures of the police department”, Commission Report, New York, NY. Murdock, S. (2015), “‘Rat Cop’ Joe Crystal shunned from Baltimore Police Department after reporting officer brutality”, Huffington Post, June 16, available at: www.huffingtonpost. com/2015/06/16/baltimore-joe-crystal_n_7582374.html (accessed June 19, 2015). National Research Council (2004), Fairness and Effectiveness in Policing: The Evidence, Committee to Review Research on Police Policy and Practices, Committee on Law and Justice, Division of Behavioral and Social Sciences and Education, The National Academies Press, Washington, DC. Niederhoffer, A. (1967), Behind the Shield: The Police in Urban Society, Anchor Books, New York, NY. Office of Community Oriented Policing Services (2015), “President’s Task Force on 21st Century Policing”, Final Report, Office of Community Oriented Policing Services, Washington, DC. Pate, A.M. and Fridell, L.A. (1993), “Police use of force: official reports, citizen complaints, and legal consequences”, Police Foundation, Washington, DC. Piliavin, I. and Briar, S. (1964), “Police encounters with juveniles”, American Journal of Sociology, Vol. 70 No. 2, pp. 206-214. Pollock, J.M. (1995), “Women in corrections: custody and the caring ethic”, in Merlo, A.V. and Pollock, J.M. (Eds), Women, Law and Social Control, Allyn and Bacon, Boston, MA, pp. 97-116. Pollock, J.M. (1999), Criminal Women, Anderson, Cincinnati, OH. Prenzler, T. (2006), “Senior police managers’ views on integrity testing, and drug and alcohol testing”, Policing: An International Journal of Police Strategies & Management, Vol. 29 No. 3, pp. 394-407. Prenzler, T. (2011), “The evolution of police oversight in Australia”, Policing & Society, Vol. 21 No. 3, pp. 284-303. Prenzler, T. and Porter, L. (2015), “Accountability of policing”, in Rowe, M. and Lister, S. (Eds), Routledge, London, pp. 49-68. Radelet, L. and Carter, D. (1994), The Police and the Community, Macmillan, New York, NY. Regoli, R. (1976), “An empirical assessment of Niederhoffer’s police cynicism scale”, Journal of Criminal Justice, Vol. 4 No. 3, pp. 231-241. Regoli, R. (1977), Police in America, University Press of America, Washington, DC. Regoli, R., Crank, J. and Culbertson, R. (1989), “Police cynicism, job satisfaction, and work relations of police chiefs”, Sociological Focus, Vol. 22 No. 3, pp. 161-171. Regoli, R., Crank, J. and Rivera, G. (1990a), “The construction and implementation of an alternative measure of police cynicism”, Criminal Justice and Behavior, Vol. 17 No. 4, pp. 395-409. Regoli, R., Crank, J., Potgieter, P. and Powell, J. (1990b), “Police cynicism in Transkei”, Journal of Criminal Justice, Vol. 18 No. 2, pp. 133-146. Reiss, A.J. Jr (1992), “Police organization in the twentieth century”, in Tonry, M. and Morris, N. (Eds), Crime and Justice: A Review of Research, University of Chicago Press, Chicago, IL. Sarver, M.B. and Miller, H. (2014), “Police chief leadership: styles and effectiveness”, Policing: An International Journal of Police Strategies & Management, Vol. 37 No. 1, pp. 126-143. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) Schafer, J.A. (2010a), “The ineffective police leader: acts of commission and omission”, Journal of Criminal Justice, Vol. 38 No. 4, pp. 737-746. Schafer, J.A. (2010b), “Effective leaders and leadership in policing: traits, assessment, development, and expansion”, Policing: An International Journal of Police Strategies & Management, Vol. 33 No. 4, pp. 644-663. Schuck, A.M. (2014), “Female representation in law enforcement: the influence of screening, unions, incentives, community policing, CALEA, and size”, Police Quarterly, Vol. 17 No. 1, pp. 54-78. Sechrest, D.K. and Burns, P. (1992), “Police corruption: the Miami case”, Criminal Justice and Behavior, Vol. 19 No. 3, pp. 294-313. Sellbom, M., Fischler, G.L. and Ben-Porath, Y.S. (2007), “Identifying MMPI-2 predictors of police officer integrity and misconduct”, Criminal Justice and Behavior, Vol. 34 No. 8, pp. 985-1004. Sherman, L.W. (1974), Police Corruption: A Sociological Perspective, Anchor Press, Garden City, NY. Sherman, L.W. (1978), Scandal and Reform, University of California Press, Berkley, CA. Smith, D.A. (1984), “The organizational context of legal control”, Criminology, Vol. 22 No. 1, pp. 19-38. Smith, D.A. (1986), “The neighborhood context of police behavior”, in Reiss, A.J. Jr (Ed.), Crime and Justice: A Review of Research, Vol. 8, University of Chicago Press, Chicago, IL, pp. 313-341. Spader, D.J. (2002), “The morality of justice and the morality of care: are there distinct moral orientations for males and females?”, Criminal Justice Review, Vol. 27 No. 1, pp. 66-88. Telep, C.W. and Lum, C. (2014), “The receptivity of officers to empirical research and evidencebased policing: an examination of survey data from three agencies”, Police Quarterly, Vol. 17 No. 4, pp. 359-385. Walker, L. (1984), “Sex differences in the development of moral reasoning: a critical review”, Child Development, Vol. 55 No. 3, pp. 677-691. Walker, S., Alpert, G. and Kenney, D. (2000), “Early warning systems for police: concept, history, and issues”, Police Quarterly, Vol. 3 No. 2, pp. 132-152. Weitzer, R. and Tuch, S.A. (2004), “Race and perceptions of police misconduct”, Social Problems, Vol. 51 No. 3, pp. 305-325. Willis, J.J. (2014), “A recent history of the police”, in Reisig, M.E. and Kane, R.J. (Eds), The Oxford Handbook of Police and Policing, Oxford University Press, New York, NY, pp. 3-33. Wilson, J.Q. (1993), The Moral Sense, Free Press, New York, NY. Worden, R.E. (1995), “The ‘causes’ of police brutality: theory and evidence on police use of force”, in Geller, W.A. and Toch, H. (Eds), And Justice for All, Police Executive Research Forum, Washington, DC, pp. 31-60. Measurement of police integrity 265 PIJPSM 39,2 Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) 266 Wu, Y. (2010), “College students’ evaluation of police performance: a comparison of Chinese and Americans”, Journal of Criminal Justice, Vol. 38 No. 4, pp. 773-780. Wu, Y. and Sun, I.Y. (2010), “Perceptions of police: an empirical study of Chinese college students”, Policing: An International Journal of Police Strategies & Management, Vol. 33 No. 1, pp. 93-113. Wu, Y., Jiang, S. and Lambert, E. (2011), “Citizen support for community policing in China”, Policing: An International Journal of Police Strategies & Management, Vol. 34 No. 2, pp. 285-303. Wu, Y., Sun, I.Y. and Triplett, R. (2009), “Race, class or neighborhood context: which matter more in measuring satisfaction with police?”, Justice Quarterly, Vol. 26 No. 1, pp. 125-126. Wu, Y., Sun, I.Y. and Smith, B. (2011), “Race immigration, and policing: Chinese immigrants’ satisfaction with police”, Justice Quarterly, Vol. 28 No. 5, pp. 745-774. Further reading Alain, M. (2004), “An exploratory study of Quebec police officers’ attitudes toward ethical dilemmas”, in Klockars, C.B., Kutnjak Ivković, S. and Haberfeld, M.R. (Eds), The Contours of Police Integrity, Sage Publications, Thousand Oaks, CA, pp. 40-55. Andreescu, V., Keeling, D.G., Vito, G.F. and Voinic, M.C. (2012), “Romanian and American police officer’s perceptions of professional integrity and ethical behavior”, Revista Romanânā de Sociologie, Vol. 23 No. 3, pp. 185-207. Decker, S. (1981), “Citizens attitudes towards the police: the difference police contacts makes”, Journal of Police Science and Administration, Vol. 9 No. 1, pp. 80-87. Edelbacher, M. and Kutnjak Ivković, S. (2004), “Ethics and the police – studying police integrity in Austria”, in Klockars, C.B., Kutnjak Ivković, S.K. and Haberfeld, M.R. (Eds), The Contours of Police Integrity, Sage Publications, Thousand Oaks, CA, pp. 19-39. Frank, J., Brandl, S., Cullen, F. and Stichman, A. (1996), “Reassessing the impact of race on citizen’s attitudes toward the police: a research note”, Justice Quarterly, Vol. 13 No. 2, pp. 322-334. Garner, J.H., Maxwell, C.D. and Heraux, C. (2004), “Pattern of police use of force as a measure of police integrity”, in Hickman, M.J., Piquero, A.R. and Greene, J.R. (Eds), Police Integrity and Ethics, Wadsworth/Thomson Learning, Belmont, CA, pp. 109-125. Haberfeld, M.R., Klockars, C.B., Kutnjak Ivkovich, S. and Pagon, M. (2000), “Police officer perceptions of the disciplinary consequences of police corruption in Croatia, Poland, Slovenia, and the United States”, Police Practice & Research, Vol. 1 No. 1, pp. 41-72. Huberts, L.W.J.C., Kaptein, M. and Lasthuizen, K. (2007), “A study of the impact of three leadership styles on integrity violations committed by police officers”, Policing: An International Journal of Police Strategies & Management, Vol. 30 No. 4, pp. 587-607. Jiang, S., Sun, I.Y. and Wang, J. (2012), “Citizens’ satisfaction with police in Guangzhou, China”, Policing: An International Journal of Police Strategies & Management, Vol. 35 No. 4, pp. 801-821. Joyner, I. (1975), “Police repression in America”, Journal of Political Repression, Vol. 1 No. 1, pp. 41-47. Klockars, C.B., Kutnjak Ivković, S. and Haberfeld, M.R. (2006), Enhancing Police Integrity, Springer Academic, New York, NY. Kutnjak Ivković, S. (2003), “To serve and collect: measuring police corruption”, Journal of Criminal Law and Criminology, Vol. 93 Nos 2/3, pp. 593-650. Kutnjak Ivković, S., Haberfeld, M. and Peacock, R. (2012), “Rainless west: the integrity survey’s role in agency accountability”, Police Quarterly, Vol. 16 No. 2, pp. 148-176. Downloaded by University of Texas at Dallas At 17:13 10 May 2016 (PT) Martin, C. (1994), Illinois Municipal Officers’ Perceptions of Police Ethics, Department of Justice, National Institute of Justice, Washington, DC. Mazerolle, L., Darroch, S. and White, G. (2013), “Leadership in problem-oriented policing”, Policing: An International Journal of Police Strategies & Management, Vol. 36 No. 3, pp. 543-560. Meyer, J.C. Jr (1976), “Definitional and etiological issues in police corruption: assessment and synthesis of competing perspectives”, Journal of Police Science and Administration, Vol. 4 No. 1, pp. 46-55. Miller, L.S. and Brasswell, M.C. (1992), “Police perceptions of ethical decision-making: the ideal vs the real”, American Journal of Police, Vol. 11 No. 4, pp. 27-45. Prenzler, T. and Ransley, J. (2002), Police Reform: Building Integrity, Federations Press, Sydney. Regoli, R. and Poole, E. (1979), “Measurement of police cynicism: a factor scaling approach”, Journal of Criminal Justice, Vol. 7 No. 1, pp. 37-51. Schafer, J.A. and Martinelli, T.J. (2008), “First-line supervisor’s perceptions of police integrity – the measurement of police integrity revisited”, Policing: An International Journal of Police Strategies & Management, Vol. 31 No. 2, pp. 306-323. Sun, I., Hu, R. and Wu, Y. (2012), “Social capital, political participation, and trust in the police in urban China”, The Australian and New Zealand Journal of Criminology, Vol. 45 No. 1, pp. 87-105. Westmarland, L. (2005), “Police ethics and integrity: breaking the blue code of silence”, Policing & Society, Vol. 15 No. 2, pp. 145-165. About the authors Matthew J. Hickman is an Associate Professor of Criminal Justice at Seattle University. Matthew J. Hickman is the corresponding author and can be contacted at: hickmanm@seattleu.edu Alex R. Piquero is an Ashbel Smith Professor of Criminology and Associate Dean for Graduate Programs in the School of Economic, Political and Policy Sciences at The University of Texas at Dallas. Zachary A. Powell is a Doctoral Student in Criminology at The University of Texas at Dallas. Jack Greene is a Professor of Criminology and Criminal Justice at Northeastern University. For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com Measurement of police integrity 267