USF Libraries
USF Digital Collections

Domestic violence within law enforcement families

MISSING IMAGE

Material Information

Title:
Domestic violence within law enforcement families the link between traditional police subculture and domestic violence among police
Physical Description:
Book
Language:
English
Creator:
Blumenstein, Lindsey
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Police stress
Intimate partner violence
Police personality
Gender
Tobit regression
Dissertations, Academic -- Criminology -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The most recent research in police domestic violence has shown that officers may perpetrate domestic violence at a higher rate than the general population, 28% versus 16%, respectively (Sgambelluri, 2000). Traditional police sub-culture has been identified, in several studies, as contributing to higher work stress, and using force on the job (Alexander et al., 1993; Drummond, 1976; Johnson et al, 2005; Kop and Euwema, 2001; Sgambelluri, 2000; Wetendorf, 2000). This research, however, has not fully examined the link between adherence to the traditional police sub-culture and officer involvement in domestic violence. This study attempts to identify whether officers who adhere to the aspects of the traditional police sub-culture are more likely to use violence against their intimate partner using two types of domestic violence-physical assault and psychological violence-as well as examine gender's moderating influence on police domestic violence and traditional police sub-culture. Using a survey created from existing scales, 250 officers were contacted within several departments in Central Florida, of these, 90 officers responded. Using Tobit and Logistic Regression the study found that officers who adhere to aspects of the traditional police subculture are more likely to engage in psychological domestic violence. There was no relationship found between traditional police culture and physical domestic violence. A thorough discussion of the results and future research directions is also included.
Thesis:
Thesis (M.A.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Lindsey Blumenstein.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 58 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002068258
oclc - 606592309
usfldc doi - E14-SFE0003106
usfldc handle - e14.3106
System ID:
SFS0027422:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002068258
005 20100412112408.0
007 cr bnu|||uuuuu
008 100412s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003106
035
(OCoLC)606592309
040
FHM
c FHM
049
FHMM
090
HV6025 (Online)
1 100
Blumenstein, Lindsey.
0 245
Domestic violence within law enforcement families :
b the link between traditional police subculture and domestic violence among police
h [electronic resource] /
by Lindsey Blumenstein.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 58 pages.
502
Thesis (M.A.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
3 520
ABSTRACT: The most recent research in police domestic violence has shown that officers may perpetrate domestic violence at a higher rate than the general population, 28% versus 16%, respectively (Sgambelluri, 2000). Traditional police sub-culture has been identified, in several studies, as contributing to higher work stress, and using force on the job (Alexander et al., 1993; Drummond, 1976; Johnson et al, 2005; Kop and Euwema, 2001; Sgambelluri, 2000; Wetendorf, 2000). This research, however, has not fully examined the link between adherence to the traditional police sub-culture and officer involvement in domestic violence. This study attempts to identify whether officers who adhere to the aspects of the traditional police sub-culture are more likely to use violence against their intimate partner using two types of domestic violence-physical assault and psychological violence-as well as examine gender's moderating influence on police domestic violence and traditional police sub-culture. Using a survey created from existing scales, 250 officers were contacted within several departments in Central Florida, of these, 90 officers responded. Using Tobit and Logistic Regression the study found that officers who adhere to aspects of the traditional police subculture are more likely to engage in psychological domestic violence. There was no relationship found between traditional police culture and physical domestic violence. A thorough discussion of the results and future research directions is also included.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Lorie Fridell, Ph.D.
653
Police stress
Intimate partner violence
Police personality
Gender
Tobit regression
690
Dissertations, Academic
z USF
x Criminology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3106



PAGE 1

Domestic Violence Within Law Enforcement Families: The Link Between Traditional Police Subculture and Domestic Violence Among Police by Lindsey Blumenstein A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts Department of Criminology College of Behavioral and Community Sciences University of South Florida Major Professor: Lorie Fridell, Ph.D. Christopher Sullivan, Ph.D. Shayne Jones, Ph.D. Date of Approval July 13, 2009 Keywords: police stress, intimate partner vi olence, police personality, gender, Tobit regression Copyright 2009, Lindsey Blumenstein

PAGE 2

i Table of Contents List of Tables ii Abstract iii Chapter One Introduction 1 Chapter Two Literature Review 3 Existence of Traditional Police Sub-Culture 4 Constructs of Traditional Police Sub-Culture 5 Negative Consequences of the Tr aditional Police Sub-Culture 7 Police Domestic Violence and Tr aditional Police Sub-Culture 9 Gender, Traditional Police Sub-Culture and Domestic Violence 11 Chapter Three Research Questions and Methodology 12 Research Questions 13 Sample and Data Collection 13 Measures 15 Dependent Variable: In timate Partner Violence 15 Independent Variables 18 Burnout 18 Authoritarianism 22 Cynicism 23 Demographic Variables 24 Chapter Four Analysis and Results 25 Descriptive Statistics 25 Bivariate Analysis 27 Multivariate Analysis 32 Chapter Five Discussion and Conclusions 38 Results 38 Limitations 42 Policy Implications and Future Research 45 References 48

PAGE 3

ii Appendices 52 Appendix A: Cover Letter for Survey 53 Appendix B: Law Enforcement Survey 54

PAGE 4

iii List of Tables Table 1 Descriptive Statistics of Dependent, Independent, and Control Variables 26 Table 2 Inter-Item Correlation Matr ix for Authoritarianism Scale 27 Table 3 Inter-Item Correlation Matrix for Cynicism Scale 28 Table 4 Inter-Item Correlation Matrix for Emotional Exhaustion Scale 29 Table 5 Bivariate Correlations of Physical and Psychological Domestic Violence 30 Table 6 Bivariate Correlations of Psychological Domestic Violence Relationships 32 Table 7 Tobit Regression Results of Psychological Domestic Violence 34 Table 8 Logistic Regression Results of Physical Domestic Violence 35 Table 9 Relationships of Variables by Gender 37

PAGE 5

iv Domestic Violence Within Law Enforce ment Families: The Link Between Traditional Police Subculture and Domestic Violence Among Police Lindsey Blumenstein ABSTRACT The most recent research in police domestic violence has shown that officers may perpetrate domestic violence at a higher rate than the general population, 28% versus 16%, respectively (Sgambelluri, 2000). Traditio nal police sub-culture has been identified, in several studies, as contri buting to higher work stress and using force on the job (Alexander et al., 1993; Drum mond, 1976; Johnson et al, 2005; Kop and Euwema, 2001; Sgambelluri, 2000; Wetendorf, 2000). This research, however, has not fully examined the link between adherence to the traditional police sub-culture and officer involvement in domestic violence. This study attempts to id entify whether officers who adhere to the aspects of the traditional police sub-culture ar e more likely to use violence against their intimate partner using two types of do mestic violence—physical assault and psychological violence—as well as examin e gender’s moderating influence on police domestic violence and traditi onal police sub-cultur e. Using a survey created from existing scales, 250 officers were contacted within several departments in Central Florida, of these, 90 officers responded. Us ing Tobit and Logistic Regression the study found that officers who adhere to aspects of the traditional police subculture are more likely to engage in psychological domes tic violence. There was no relationship found between traditional police culture and physical domestic violence. A thorough discussion of the results and future research directions is also included.

PAGE 6

1 Chapter One Introduction The Women’s Movement in the 1970’s infl uenced the public’s awareness of the victims and offenders of domestic viol ence, giving women a voice that had not previously been heard. Police were called upon to take these crimes more seriously; the public came to rely on the police to protect potential victims of domestic violence and arrest offenders. Yet some of those same of ficers were themselves “batterers with a badge.” The number of research articles on dome stic violence skyrocketed from less than a dozen in 1974 to more than 200 in the 1980’s (Johnson et al., 2005; Stith and Straus, 1995; Walton and Zigley, 2000). Yet, the police officers that were batterers—the same police who were responding to domestic abus e disputes and keeping victims safe—were essentially ignored in this literature as perpetrators. Beginning in the 1990’s, researchers bega n to look at violence within police families. Two key studies helped to raise public and organizational awareness of the high number of officers involved in domestic di sputes. Johnson (1991), w ho testified before the US Congress, reported that over 40% of the officers surveyed in her study reported being violent with th eir spouse. Neidig et al. (1992) conducted a subsequent study and found similar results ; about 40% of the officers in thei r survey inflicted violence on their spouses. The behaviors cited included slap ping, kicking, hitting, c hocking, or using a gun. The most recent research has shown that officers may perpetrate domestic violence

PAGE 7

2 at a higher rate than the general population, 28% versus 16%, respectively (Sgambelluri, 2000). It is important to identify factors that may make officers prone to commit violent acts against their significant others, as well as ascertain the differences between officers who engage in domestic violence and thos e who do not. Traditional police sub-culture has been identified, in severa l studies, as contributing to higher work stress and using force on the job (Alexander et al., 1993; Drummond, 1976; Johnson et al., 2005; Kop and Euwema, 2001; Sgambelluri, 2000; Wetendorf, 2000). This research, however, has not fully examined the link between adherence to the traditional police sub-culture and officer involvement in domestic violence. Al so, only one study has examined gender as it relates to intimate partner violence among police officers. Th is study found that burnout has an effect on violence use at different rates for males and females (Johnson, 2000). Effective responses cannot be produced unless there is an accurate understanding of this behavior. Therefore the following study will try to answer the following questions: 1) Are officers who adhere to the components of traditional police subculture more likely to enga ge in domestic violence? 2) Does gender moderate the influence of traditional pol ice sub-culture on police domestic violence? Chapter Two discusses the previous and current literature as it relates to traditional police culture, police domestic violence, and gender. Chapter Three gives a detailed explanation of the research questi ons and hypotheses. Chapte r Four discusses the data collection techniques and the measures used in the survey. Chap ter Five presents the results. Chapter Six discusses the results in the context of the research questions and literature, study limitations, policy imp lications and future research.

PAGE 8

3 Chapter Two Literature Review This chapter will review the extant litera ture within the realm of police culture, covering the constructs of the traditional police culture, negative consequences of the traditional culture, police domestic violence and the relationship between gender and police domestic violence. It is from this literature that the working hypotheses were developed. Since the seminal work of William Westley (1970), scientists have studied the existence of a sub-culture within the police organization. In that ti me, researchers have explored the characteristics of this sub-cultur e, and whether more than one culture exists (Chan, 1997; Cochran and Bromley, 2003; He rbert, 1998; Kappeler et al., 1994; Manning, 1995; Paoline 2004; Reiner, 1985; Reu ss-Ianni, 1983; Skol nick, 1966; Terrill et al., 2003; Van Maanen, 1974; Westley 1970). Westley (1970) studied police in Indiana in the 1950’s in order to identify a “si ngle” police culture. He focused on the shared norms and values among police that serve to control the strains that are created while working in a dangerous and hostile work envi ronment. Westley highlighted the loyalty and secrecy among officers. He identified the shared norms and values—including a coercive authority over citizens, or an author itarian personality; distrust and suspicion of the public including a cynical attitude; a strong emphasis on law enforcement tactics; assessing people and situations in terms of their potential threat; and burnout. All of these features are together known as the monolithic police culture. Research has evaluated the effects of traditional police sub-culture on various outcomes such as work stresses (Gershon, 2000; Jackson and Maslac h, 1982; Roberts and

PAGE 9

4 Levenson, 2001) and use of force on the job (Henkel, et al., 1997; Kop and Euwema, 2000; Terrill et al., 2003). Very few studies have assessed the effects of this sub-culture on domestic violence in police families (E rwin et al., 2005; Johns on, 2000; Johnson et al., 2005 (Sgambelluri, 2000), and only one of these studies has included gender as a variable of interest (Johnson, 2000). Each of these stud ies has shown how a specific construct of the traditional police sub-culture can adversely affect officers. Most of these studies have included only one construct of the traditional sub-culture (Johnson et al., 2005; Sgambelluri, 2000); no study has yet to incorpor ate a comprehensive model of traditional police sub-culture to study the various negative outcomes that accompany traditional officers. The current study includes three recognized characteristics of the traditional police sub-culture—authoritarianism, cyni cism and burnout—creating a comprehensive model and thus a better unders tanding of its effect on poli ce officer domestic violence. Existence of a Traditional Police Sub-Culture The notion of a single police culture exis ted for a number of years until changes in the demographics of police personnel o ccurred. Women, as well as African Americans, Latinos, and other racial groups were hired by law enforcem ent agencies. As the makeup of police changed, so did their collective no rms and values producing a subculture or subcultures that were different from the “t raditional” one identified by Westley (Chan, 1996; Cochran and Bromley, 2003; Herbert, 199 8; Manning, 1995; Paoline et al., 2000; Paoline, 2004). It appears there is no longer a single police culture; instead researchers are identifying multiple cultures that are adhered to by subgroups of officers. There is not yet a consensus among researchers regarding the na ture of these modern police cultures, but every study on police culture has identified a sub-culture that is consistent with the

PAGE 10

5 characteristics of the “traditional” police s ub-culture (Paoline, 2004). Examples of these sub-cultures—that embody the same charac teristics of the “traditional” police subculture—are Broderick’s (1977) “Enforcer,” Brown’ s (1988) “Old Style Crime Fighter,” Cochran and Bromley’s (2003) “SubCultural Adherents,” and White’s (1972) “Tough Cops.” Although there is no longer a single police culture, rese arch has shown that the traditional police sub-culture still exists. Some of the characteristics of the “traditional” subculture include negative views towa rds citizens, favorable outlooks towards aggressive police tactics, and an emotional exhaustion of o fficers over a short period of time. The traditional police sub-culture creat es a harmful environment in which officers are more likely to engage in negative behavi ors, such as violence (Erwin et al., 2005; Henkel et al., 1997; Johnson, 2000; Johns on et al., 2005; Kop and Euwema, 2001; Sgambelluri, 2000; Terrill et al., 2003). In past decades, a majority of officers a dhered to the “traditio nal” subculture of police. Many of these officers have retired and fewer of the new generation of officers adhere to the norms and valu es of the traditi onal police culture. There are no longer a large number of officers considered to be “tra ditional.” Despite this research has shown the traditional police sub-culture continues to exist despite changes in demographics, as well as changes in the struct ure and goals of policing (Cochran and Bromley, 2003). This suggests that its effect is strong. Constructs of Traditional Police Sub-Culture Over the years there have been a number of studies on the traditional police subculture. Although researchers have not yet co me to a consensus on the proportion of officers that adhere to this traditional cultu re, they have agreed that this sub-culture

PAGE 11

6 exists, and have identified the key characte ristics that comprise it. Three accepted characteristics of this culture are—au thoritarianism, cynicism and burnout. The occupational environment of the police officer is often a risky and dangerous one. In order to cope with this atmosphere, officers adopt certain attitudinal beliefs and behaviors (Brown, 1988; Herbert, 1998; Manning, 1995; Paoline, 2004; Reiner, 1985; Reuss-Ianni, 1983; Skolnick, 1966; Terrill et al., 2003; Westley, 1970). To combat the dangerous violence that may occur, officer s use aggressive law enforcement tactics (Brown, 1988; Reuss-Ianni, 1983). The occupati onal description of a police officer can be ambiguous; officers adhering to the tradit ional police culture will adopt a strictly crime fighting view of their role. This attitude constitutes selectively performing duties, using coercion over and aggression towards citizens, and rejecting any other job descriptions such as those that might refl ect community policing. The use of aggressive law enforcement tactics and a coercive authority over citizens is linked to an authoritarian personality of officers that spi lls over into other parts of th eir lives (Terrill et al., 2003). Cynicism is a characteristic that has b een traced to the traditional police culture (Cochran and Bromley, 2003; Paoline, 2004; Re uss-Ianni, 1983; S kolnick, 1966; Terrill et al., 2003; Westley, 1970). Histor ically, officers have believed that citizens outside of the police profession would be of no assistance to them in their duties, and even if they were to help they would do more harm th an good (Paoline, 2004; Sparrow et al. 1990). This coupled with a common “we-versus-them” attitude held by officer s in the traditional police culture, has created an environment of su spicion and distrust of the public. Police officers carry these attitudes with them in thei r daily interactions with citizens, as well as with fellow officers who may not share the sa me belief system, making them feel they must “maintain an edge” while fulfilling their duties (Cochran and Bromley, 2003; Van

PAGE 12

7 Maanen, 1974). This creates tension between officers and a feeling that everyone is against them, leading to a destructive wo rk environment, and therefore creating a situation for violence to occur. Police officers experience prolonged periods of stress daily, as compared to the general population. The very natu re of policing, including thr eats to life and exposure to violence, can create high levels of emoti onal exhaustion This emotional exhaustion, in turn, can produce harmful consequences in the officers’ occupati onal and personal lives. Officers who adhere to the traditional pol ice culture are more likely to experience burnout due to aggressive law enforcement tac tics and the cynical at titude they adopt. These dimensions create a highly demanding and taxing environment in which police officers are more likely to internalize daily inte ractions that create gr eater levels of stress and pressure which in turn lead to burnout. These three constructs of traditiona l police sub-culture described above— authoritarianism, cynicism and burnout—are freq uently referenced in the literature. Each of these aspects of the traditional police subculture has been linked to particular officer attitudes and behaviors esp ecially those related to aggression and violence. The current research will use these three constructs as a proxy measure of the traditional police sub-culture, and will study the effect of this culture on officer’s involvement in domestic violence. The next section describes theory and research that links officer adherence to these constructs to negative consequences. Negative Consequences of th e Traditional Police Sub-Culture Recent research has evaluated the effect s of traditional police sub-culture on various outcomes, such as work stresse s (Gershon, 2000; Jackson and Maslach, 1982; Roberts and Levenson, 2001) and use of force on the job (Henkel, et al., 1997; Kop and

PAGE 13

8 Euwema, 2001; Terrill et al., 2003). These stud ies show that adhere nce to the various components of the traditional culture produces negative consequences for officer. The unique stress experienced by those in a helping profession has been documented within the past 20 years (Jackson and Masl ach, 1982). The close co ntact that officers have with other people’s problems can lead to frustration, tension a nd fatigue, as well as emotional exhaustion, which the officers often bring home with them. Several studies have been conducted examining the amount of emotional exhaustion or burnout and officer experiences, and the effect on his or her family life (Gershon, 2000; Jackson and Maslach, 1982; Roberts and Levenson, 2001). Each of these studies found that officers with high levels of stress are more likely to display anger (Gershon, 2000; Jackson and Maslach, 1982; Roberts and Levenson, 2001), wi thdraw from social contact (Jackson and Maslach, 1982), experience physical exhaus tion (Roberts and Levenson, 2001), and use negative outlets for coping such as alcohol and gambling (Gershon, 2000). Officers with high levels of stress are also at highe r risk for poor marital outcomes. Police officers often find themselves in situations where it becomes necessary to use force. Researchers have become intere sted in the reasons one chooses to use or misuse force in a particular situation. Two studi es in particular have examined the impact of the “traditional” police sub-culture on the use or misuse of force by officers (Kop and Euwema, 2001; Terrill et al., 2003). Kop a nd Euwema (2001) studied Dutch officers’ burnout and depersonalization, or isolation, from the public and use of force. Using qualitative and quantitative data, Ko p and Euwema (2001) found that as depersonalization increased, officers felt more positively about using force, and were less interested in the problems of citizens.

PAGE 14

9 Terrill et al. (2003) used data from th e Project on Policing Neighborhoods to test the assertion that differences across officers in the use of coercive authority (including force) are a result of variation in cultu re alignments. The researchers developed a classification scheme based on officers’ attit udes. They found that o fficers who adhered to the “traditional” police culture are more likely to display coercive authority, using behavior such as pushing or hitting, more frequently, and at higher levels. Their findings indicate that officer’s use of force is related to officer’s em bodiment of the “traditional” police culture. Terrill et al’s (2003) research demonstrates that the values within the “traditional” culture of policing are tied to use of force. Previous research on police culture has indicated some important relationships between using force, or aggressiveness, and personifying the “traditi onal” characteristics of police culture. Though these studies have specified important relationships, none of these studies have linked adherence to the “t raditional” culture of policing to domestic violence within the home. Police Domestic Violence and Traditional Police Sub-Culture Only a few researchers have studied do mestic violence within police families (Erwin et al., 2005; Gershon, 2000; Johnson, 1991, 2000; Johnson et al., 2005; Neidig et al., 1992; Sgambelluri, 2000). These studies have documented a high frequency of officer-involved domestic violence. Even fewe r studies have taken on the challenge of identifying the effects of police culture on officer-involved domestic violence. Gershon (2000) conducted a study to assess the relationship between police work stress and domestic violence by police offi cers. Gershon (2000) developed a new police stress questionnaire and administered it to a large sample of swor n law enforcement in Baltimore, Maryland. By doing this, she was able to identify the prevalence of stress and

PAGE 15

10 its outcomes in male and female officers, as well as the coping mechanisms used to control stress. Although this study looked at domestic violence, it was only concerned with how stress was related to domestic violen ce; this study did not look at stress in the context of the “traditional” pol ice culture in its testing of officers’ use of domestic violence. Erwin et al. (2005) compared officers ch arged with intimate partner violence to those who had not been charged to identify risk factors. The fi ndings indicated that minority officers were more likely to be accu sed of domestic violence, accused officers were treated similar to civilians in the pro cessing of the case, and accused officers were more likely to be assigned to high crime distri cts. The authors did not measure the impact of any risk factors that refl ect the “traditional” police subculture, including work stress. Johnson et al. (2005) examined the rela tionship between violence exposure and domestic violence by police. They created meas ures of burnout, auth oritarian spillover, alcohol use, department withdr awal, and violence exposure. The findings suggest that the relationship between violence exposure and dom estic violence is a mediated process. When spousal violence occurs due to external burnout or authoritaria nism, the amount of violence exposure becomes important. This study is significant because it identifies burnout and authoritarianism, two central aspect s of traditional poli ce culture, as factors leading to domestic violence. This study is only one of a minority th at actually looked at the relationship between f eatures of police culture and domestic violence. These few studies are central to unde rstanding domestic violence in law enforcement. They show a connection betw een traditional police culture and spousal abuse by officers. The research of Johnson et al. (2005), desc ribed above, is the most relevant to the current study because they s how a relationship between characteristics of

PAGE 16

11 police culture and violence in the home. The current study builds upon Johnson et al’s (2005) study not only by adding to the literature on police cu lture and domestic violence, but also by researching a more comprehens ive model of traditiona l police sub-culture including three recognized charac teristics of this culture. Gender, Traditional Police Sub-Culture, and Domestic Violence The previous studies explored the rela tionship between characteristics of the police sub-culture and domestic violence. Only one study to date has looked at the effect of gender on this relationship (Johnson, 2000). With more women entering the police force, it becomes important to study the in fluence of gender on these relationships. Johnson (2000) examined the impact of gender and burnout on violence at home. Overall, she found that male officers, who experience high external burnout, are more likely to use violence. Female officers, w ho experience high internal burnout, are less likely to use violence. Importantly, Johnson (2 000) found that the women who experience high levels of external burnout (as above, males are generally more likely than females to experience high levels of exte rnal burnout) are more likely to be involved in various types of violence. When this female external burnout occurs, female violence equals or exceeds that of male officers. The study by Johnson indicates that gende r has an effect on aspects of the traditional sub-culture and police domestic violence. The current study builds on the findings of Johnson’s study by examining the relationship among three facets of traditional police sub-culture and domestic violence.

PAGE 17

12 Chapter Three Research Questions and Methodology This chapter sets forth the research qu estions addressed by this study and outlines the methods used. Information is provi ded regarding subject selection, survey administration, and measurement of the variables. Domestic violence within law enforcement families is rarely studied due to its delicate nature and the difficulty of accessi ng data within police departments (Johnson, 2000). That violence within police families is a problem has been documented, but no consensus exists regarding why this is the case. Researchers have made some significant steps in studying domestic violen ce in law enforcement, but more investigation is needed to understand why some officers are more prone to commit violence in the home compared to other officers. This study attempts to fill the gap in the literature regard ing the relationship between aspects of the traditional police sub-culture and domestic violence against intimate partners by police. The study will assess the link between adherence to specific aspects of the traditional police sub-culture—authoritarianism, cynicism, and burnout— and officer-involved domestic violence. The cu rrent study will also try to include gender as a moderating variable between traditional police sub-culture and police domestic violence.

PAGE 18

13 Research Questions Based on previous literature the current study will seek to answer the following questions in order to assess the relationship between tradit ional police sub-culture, as well as gender, on police officer domestic violence: 1) Are officers who adhere to the components of traditional police sub-culture more likely to engage in domestic violence? 2) Does gender moderate the influence of traditional police s ub-culture on police domestic violence? Sample and Data Collection A survey was used to measure each s ubject’s adherence to the three common constructs of traditional police sub-cultu re—authoritarianism, cynicism, and burnout— and subject involvement in physical and psyc hological abuse of an intimate partner. Items solicited demographic information incl uding gender, which will be used as a hypothesized moderating variable. A convenience sample of police officers was used. A group of local municipal and county police departments was contacte d to ascertain which agencies had a mechanism in place to distribute surveys to their officers (e.g. physical mailboxes for employees) and would cooperate with the res earch. The seventeen following agencies in the Tampa, FL area were contacted: Braden ton Police Department, Clearwater Police Department, Dade City Police Department Gulfport Police Department, Hillsborough County Sheriffs Office, Lakeland Police Depa rtment, Manatee County Sheriffs Office, Orlando Police Department, Pi nellas County Sheriffs Office, Pinellas Park Police Department, Sarasota County Sheriffs Office, Sarasota Police Department, St. Petersburg Beach Police Department, St. Petersburg Po lice Department, Tampa Police Department,

PAGE 19

14 Tarpon Springs Police Department, and Temple Terrace Police Department. Ultimately, four of these departments agreed to coopera te. Although it is impossible to know from which department the surveys came, the departments surveyed were varied in size; they included two small, one medium and one large agency (They are not named as they were promised agency-level confidentiality.) Surveys were distributed to the total population of 250 officers who worked for these four agencies through departmental ma ilboxes. (The executives at each agency provided information on the total number of sworn personnel.) Officers were assured complete confidentiality. Officers were asked to complete the survey and either place the completed instrument in a drop box near the department mailboxes or return the survey using self-addressed stamped envelopes. Ni nety officers returned the surveys for a response rate of 36%. This type of sampling comes with both w eaknesses and strengths. One important weakness of using this convenience sample is th at the results generate d on the nature of the police sub-culture and the frequency of interpersonal vi olence on the part of police will not necessarily be generalizable. Alt hough these results may not be generalizable, this sample is satisfactory for testing re lationships among the va riables—traditional police sub-culture, police domestic violence. This sample comes entirely from Central Florida, which further limits generalizability. On the positive side, the only other data set on this subject came from the northern United St ates, and the current dataset will give a picture of southern United St ates. Another weakness is the lack of survey tracking. There were no identifying characteristics on the survey, such as names or numbers, and therefore it was not possible to follow up with officers. However, because this topic is extremely sensitive, the assurance of confidentiality was crucial.

PAGE 20

15 Measures The following variables were measured: Intimate Partner Violence, Burnout, Cynicism and Authoritarianism, as well as ge nder, race, age, and years on the force. The following section details the scales, and que stions used to measure each of these variables. The instrument is included as Appendix A. Dependent Variable: Intim ate Partner Violence The dependent variable for this study is police domestic violence. In order to ascertain the frequency of interpersonal viol ence, a revised version of one of the most widely used scales, the Revised Conflict Ta ctics Scale (CTS2; Stra us et al., 1996) was utilized. The CTS2 was developed to measure the extent to which partners in dating, cohabitating, or marital rela tionships engage in different forms of aggression, and to examine the reasoning and negotia tion tactics used in a confli ct (Straus et al., 1996). The CTS2 is the second version of the Conflict Tactics Scale (C TS), which was developed in response to recommendations and critiques to enhance the CTS (Connelly et al., 2005; Newton et al., 2001). The CTS2 has been updated and improved to include a simplified format, new scales and improved items (Straus et al., 1996). The CTS2 has 5 subscales—Negotiation, Psychological Aggression, Physical Assault, Sexual Coercion, and Injury. E ach subscale is scored independently, and therefore it is possible to choose which s ubscales to use (Straus et al., 1996). The Psychological Aggression and Physical Assaul t scales were used in this study. The Psychological Aggression scale measures type s of verbal and nonverb al aggressive acts towards one’s partner (Straus et al., 1996). Th is scale references 8 behaviors including “Insulted or swore at part ner” and “Destroyed somethi ng of partners,” and has a Cronbach’s alpha of .79 in previous studies (Straus et al., 1996). The responses range

PAGE 21

16 from “Never” to “More than 20 times” and re flect prevalence of each behavior over the officer’s lifetime. There has been little to no research conducted on psychological abuse within the law enforcement population, and so th e inclusion of this measure in this study was important for advancing the study of poli ce domestic violence. The Physical Assault scale measures types of physical violence by a partner (Straus et al., 1996). This scale references 12 behaviors including “Pushed or shoved partner” and “Used knife or gun on partner,” and has a Cronbach’s alpha of .86 in previous stud ies (Straus et al., 1996). This measure of domestic violence provides more detail than a single question—because it inquires about a range of behaviors; such a measure has not been used in previous studies. Due to the sensitive nature of the subject, and the populat ion being studied, the sub-scale Sexual Coercion was not chosen becau se the questions in this scale may more highly affect the willingness of officers to re spond as compared to the other two scales included. The Negotiation sub-scale, which m easures emotions towards one’s partner (such as caring), was not used because it is not directly relevant to the current study. Also, the Injury sub-scale was not used because it measures the actual types of injuries that have been inflicted on the partner, su ch as bone or tissue damage, which is not relevant to the objectives of this study. Straus et al. (1996) performed the pre liminary psychometric analyses for the CTS2. The alpha reliability coefficients s how all of the scales have good internal consistencies. The coefficients ranged from .79 to .95 and are as high as or higher than the coefficients reported for the original CTS (Straus et al., 1996). Also, preliminary evidence has been found for construct and disc riminant validity. The correlation between the Psychological Aggression and Physical Assau lt scales was assessed to test construct validity. The correlations were high ranging from .67 to .71, which contributed to the

PAGE 22

17 validity of the CTS2 (Straus et al., 1996). By assessing correlations of scales for which there is no theoretical basis to expect correlation, Straus et al. (1996) were able to show evidence of discriminant validity. Straus et al. (1996) argued that because the CTS2 is fundamentally the same as the or iginal version of the CTS, th e evidence of the reliability and validity might be applied to the CTS2. Some additional studies have been conduc ted on the psychometric properties of the CTS2 since the early work of Straus et al. (1996). Newton et al. (2001) found results similar to that of Straus et al. concerning factor validity using a sample of high-risk postpartum women. Straus (2004) publishe d the first study on the cross-cultural reliability and validity of the CTS2. His results included high alpha coefficients of internal consistency as well as construct validi ty, which was present in at least two of the five sub-scales. These results are adequate to promote use of the CTS2 in a variety of settings and populations (Straus, 2004). C onnelly et al. (2005) also examined the psychometric properties of the CTS2. They f ound acceptable internal consistency of the total scale, which included all five s ub-scales, with alphas ranging from .70 to .84, although the subscale scores ranged from .46 to .80, which is slightly different from the original findings, yet still satisfactory. Each of these studies indicates that the psychometric properties of the CTS2 are satisfactory. As described above, the CTS2 was utili zed to measure the de pendent variable Domestic Violence. This instrument asked th e participants to indicate the number of times over their lifetime they have engaged in listed behaviors. After the participants chose a response category, the CTS2 was then scored by adding the midpoints of the response categories for each item chosen by the participant (Straus et. al., 1996). There are a total of 7 response categ ories and the midpoints for e ach response category are as

PAGE 23

18 follows: Category 0 is 0, Category 1 is 1, Cate gory 2 is 2 Category 3 (3-5 times) is 4, Category 4 (6-10 times) is 8, Category 5 (1120 times) is 15, and Category 6 (20 or more) is 25 (Straus et. al., 1996). The CTS2 produced two scores for each participant concerning domestic violence, one for Psyc hological Aggression and one for Physical Assault. The score for Psychological Aggre ssion was the result of 8 midpoints added together. Higher scores represented a hi gher incidence of ps ychological domestic violence. Physical Assault was dichotomized and coded 0 and 1. Offi cers that responded to not engaging in some form of physical domestic violence were given a zero, and one was given to officers that admitted to enga ging in some form of physical violence. Independent Variables The independent variables for the st udy include burnout, authoritarianism, cynicism and various demographic variable s—each of which is discussed below. Burnout To measure the independent variable burnout, a popular and widely used scale was employed (Richardsen and Martinusse n, 2004)—the Maslach Burnout Inventory (MBI; Maslach and Jackson, 1981). The MBI was developed out of a need for an instrument that evaluated burnout in a wide ra nge of human service professions. With this outcome measure, researchers can understa nd the dimensions—personal, social, or institutional—that may elevate or dimi nish burnout (Maslach and Jackson, 1981). Maslach and Jackson (1981) modeled the MBI after the Hassles Scale (Lazarus and Cohen, 1977), in which each statement is rated on two separate dimensions— frequency and intensity. Frequency has seve n response categories ra nging from “Never” to “Everyday” (Maslach and Jackson, 1981). Intensity also has se ven response categories ranging from “Very mild, barely noticeable” to “Very strong, major” (Maslach and

PAGE 24

19 Jackson, 1981). If the frequency of a question is “Never,” there will be no answer for intensity. The MBI initially used a four-factor soluti on for the 25-item factor analysis; three of the factors had eigenvalue s greater than one, emotiona l exhaustion, depersonalization, and personal achievement, and those are cons idered the three subscales of the MBI (Maslach and Jackson, 1981). The fourth item, involvement with people, did not have an eigenvalue greater than one and therefore became an optional sub-scale on the MBI (Maslach and Jackson, 1981). The Emotional Ex haustion scale consists of nine items; it describes feeling emotionally overextended by one’s work as well as exhausted (Maslach and Jackson, 1981). This scale in cludes items such as “I f eel burned out from my work” and “I feel frustrated by my job,” and has a Cronbach’s alpha of .89 for frequency and .86 for intensity in previous st udies (Maslach and Jackson, 1981) Higher mean scores on the Emotional Exhaustion subscale correspond to higher degrees of burnout. The Depersonalization scale describes unfeeling a nd impersonal responses towards recipients of one’s service (Maslach and Jackson, 1981) This is a 5-item scale including “I’ve become more callous toward people since I took this job” and “I don’t really care what happens to some recipients.” It has a Cr onbach’s alpha of .77 for frequency and .72 for intensity in previous studies (Maslach and Jackson, 1981). The Personal Accomplishment scale describes feelings of successful achieve ment in one’s work (Maslach and Jackson, 1981). The Personal Accomplishment scale c onsists of eight items and includes statements such as “I feel I’m positively influencing other people’s lives through my work,” and “I feel exhilarated after worki ng closely with my recipients,” and has a Cronbach’s alpha of .74 for frequency and .74 fo r intensity in previous studies (Maslach and Jackson, 1981).

PAGE 25

20 In this study, only Emotional Exhaustion wa s used; because each scale is scored individually it is possible to do this. The Emo tional Exhaustion scale measures items that are closest to the desired construct, burnout This scale has the highest Cronbach’s Alpha of all the scales and the indivi dual factor loadings are the high est of the three scales, with values ranging from .54 to .84 (Maslach and Jackson, 1981). This scale includes the item with the highest factor loading (.84 for fre quency and .81 for intensity), “I feel burnout out from my work,” which refe rs directly to burnout. Due to the population being used in this study, to promote a higher response rate, it is important to keep the number of questions low; the Depersona lization and Personal Achiev ement scales would add too many questions that are not sufficiently related to the desired construct. Maslach and Jackson (1981) completed th e preliminary psychometric analyses. Concerning reliability, the internal consistency of the total 25-item scale was an alpha of .83 for frequency and .84 for intensity. Fo r the sub-scale Emotional Exhaustion, the coefficient was .74 for frequency and .68 fo r intensity. The convergent validity was demonstrated in three ways. The MBI scores were correlated with: 1) a behavior rating, 2) the presence of job characte ristics expected to contribute to burnout, and 3) the various outcomes hypothesized to be related to burnout All three correlations provided ample evidence of the construct validity of the MBI (Jackson and Maslach, 1981). The discriminant validity was established by dis tinguishing the measures of the MBI from other psychological constructs that may be related to burnout. Maslach and Jackson (1981) found less than 6% of the variance wa s accounted for by thes e correlations, and therefore the notion that bur nout is simply a synonym for job dissatisfaction can be rejected. Maslach and Jackson (1986) re ported high test-retest reliability.

PAGE 26

21 Other researchers have tested the reliability and validity of the MBI as well. Schaufeli et al. (1996) investig ated the internal consistency, the factor validity, and the discriminant validity of the MBI. They found the internal consistency of the three subscales to be relatively high. Also, Schaufeli et al. (1996) found the three-factor model fits relatively well compared to other models, such as the original f our-factor model. The authors found the MBI to be a valid and reli able self-report scal e in assessing burnout (Schaufeli et. al., 1996). Another study by Bakke r et al. (2002) assessed the factorial validity of the MBI. Using eight differen t occupational groups recruited through the Internet, they found the three-factor model to be favored. Richar dsen and Martinussen (2004) also tested the factorial validity us ing human service workers. They found the original three-factor model of the MBI fit reasonably well wi th the data. They also found all internal consistencies at or above .70. Each of these st udies has shown the MBI is a reliable and valid scale to measure burnout. The Emotional Exhaustion scale rate s each statement on two dimensions, frequency and intensity (Maslach and J ackson, 1981). Frequency was measured on a range from 1 (a few times a year or less) to 6 (everyday), if the re spondent indicated that the emotion, feeling or behavior never happene d a value of zero was given. Intensity was measured on a range from 1 (very mild, barely no ticeable) to 7 (major, very strong); if the frequency was zero, then intensity was zero as well. After the participant choose a response category for each item, the mean of the items in the sub-scale Emotional Exhaustion was computed separately for fre quency and intensity (Maslach and Jackson, 1981). This scale produced nine item responses which were added to create a mean for frequency and for intensity. This measure was eventually reduced to one measure, yielding only one score per person. The intensit y scale was not used during the analysis

PAGE 27

22 of the Emotional Exhaustion scale; only the frequency scale was utilized. Higher means referred to higher levels of burnout. Authoritarianism For this study, a measure known as Author itarian Spillover was used to measure characteristics of the authoritarian persona lity in law enforcement personnel. This measure was used in a previous study on po lice domestic violence called “Violence in Police Families: Work Family Spillover” (Johnson et. al., 2005) and is arguably the best measure available to study authoritarianism in police officers. Authoritarianism Spillover is the inability to leave the job at work and includes treating the family as citizens, being overly fault-finding, and doing things by the book (Johnson et. al., 2005). The Authoritarianism Spillover scale is a 6-item Likert-type scale with seven points ranging from “strongly disagree” to “strongly agree.” It includes items such as “I can’t shake the feeling of being a police officer when at hom e” and “I catch myself treating my family the way I treat civilians.” This scale was c hosen because it was used in one of the few police domestic violence studies and produced a Cronbach’s alpha of .74 in a previous study (Johnson et. al., 2005). There is no other measure of police author itarianism referenced in the literature. The instrument selected has shown preliminary reliability and validity in the study it was previously used in (Johnson et. al., 2005), but it was only used one time prior. This is a known weakness of the current study. In the Authoritarian Spillover scale respondents indicated the degree to which they agreed or disagreed with statements about authoritarian be havior and attitudes (Johnson et. al., 2005). This is a 6-item add itive scale with response options ranging from 1 (strongly disagree) to 7 (strongly agr ee). After the respondent chose a response

PAGE 28

23 category, each of the 6 item responses we re added together, yielding one score per participant. Higher scores i ndicated higher authoritarian be haviors and attitudes. Cynicism There is no scale measuring cynicism that has been extensively used and validated; for the purposes of this study a c ontemporary scale, which is relevant to the current study, was employed. This scale was used in research conducted by Cochran and Bromley (2003); these researchers assessed the traditional police sub-culture, and included cynicism as a measure. Some pr eliminary assessments have shown the scale that they used to be reliable and valid, a nd therefore is the best option available to measure this construct. Cynicism is defined as being highly susp icious of citizens, having an “us versus them” attitude, and “maintaining an edge” that prevents society from slipping into decay and unrest (Cochran and Bromley, 2003). The C ynicism scale is a nine-item Likert-type scale with five points ranging from “strongly disagree” to “strongly agree.” This scale includes items such as “Most people are unt rustworthy and dishonest” and “Most people lie when answering questions posed by law en forcement officers.” The Cynicism scale in Cochran and Bromley’s (2003) study had a Cronb ach’s alpha of .83 a nd its individual factor loadings range from .49 to .73. One wea kness of this scale is that it has not been fully validated; but, beca use it is relevant to the current study and its measures are most like the desired construct, it is the best option available. Cochran and Bromley’s (2003) Cynicism Scale asked each par ticipant the degree to which they agreed or disagreed with stat ements about cynical perceptions of the public (Cochran and Bromley, 2003). This was a 9-it em additive scale, measured on a range from 1 (strongly disagree) to 5 (strongly ag ree). After the responde nt chose a response

PAGE 29

24 category, each of the 9 item responses we re added together, yielding one score per person. After reverse coding questions 2, 5 and 8, high values demonstrate cynical attitudes toward the public. Demographic Variables There are four demographic variables of interest—gender, race/ethnicity, age, and years of law enforcement experience. Gender was measured using the question “What is your gender?” with the response choices male and female. Males were coded as 0 and females were coded as 1. Race and ethnicity were measured using response options that match the U.S. Census. Response options to the question, “What is your race?” are White, Black/African American, Asian/Pacific Islander, Native American/Eskimo/Aleut, and Middle Eastern/East Indian. Whites were coded as 0, Black/African Americans were coded as 1, Asian/Pacific Islanders were coded as 2, Native American/Eskimo/Aleuts were coded as 3, and Middle Eastern/East Indians were coded as 4. “Are you Hispanic/Latino?” has response choices of ye s and no. These responses were coded as 0 for no and 1 for yes. Age and Years of La w Enforcement Experience were measured by asking respondents to fill in their actual age and the number of years they have been involved in law enforcement. In the next chapter, the analyses and results are described.

PAGE 30

25 Chapter Four Analyses and Results As discussed above the research que stions for the current study are: 1) Are officers who adhere to the components of traditional police sub-culture more likely to engage in domestic violence? 2) Does gender moderate the influence of traditional police s ub-culture on police domestic violence? This chapter discusses the results of the analysis for the first question. The second question could not be answer ed, but a gender profile will reveal the differences among males and females at a bivariate level. The descriptive statistics are discussed first, followed by presentations of the bivariate and multivariate regression results. Descriptive Statistics The variables were first examined at a univariate level. Basic descriptive level statistics such as mean, mode, and standard deviation were produced to gain a better understanding of the sample as a whole. Table 1 provides the means, medians, modes and standard deviations for the independent variables, control variables, and dependent variables. The sample as a whole was 83.3 percent male, 96.7 percent Caucasian, and only 12.2 percent Latino. The mean age of of ficers was 36.8 years, and the mean number of years within law enforcement was 11.7 years. Physical domestic violence and psychologi cal domestic violence were analyzed separately. Over 87 percent of officers repo rted never having e ngaged in physical

PAGE 31

26 Table 1: Descriptive Statistics of Dependent, Independent, and Control Variables (N=90) Variable Mean (S.D.) Percent Range M in. Max. Psychological DV 24.6 (31.60) 0 159 Physical DV Yes 12.2 No 87.8 Gender Male 83.3 Female 16.7 Race White 96.7 Nonwhite 3.3 Latino Yes 12.2 No 87.8 Age 36.79 (9.69) 22 69 Years in L.E. 11.68 (9.28) 1 39 Emotional Exhaustion 1.33 (1.08) 0 5.11 Authoritarianism Score 19.22 (7.00) 0 39 Cynicism Score 24.57 (7.07) 0 44 domestic violence in their lifetime. The aver age number of times officers reported having engaged in psychological domestic viol ence within their lifetime was 24.6 times. The mean score of the Emotional Exhaustion frequency scale was 1.33, with a range from 0 to 5.11. For the Cynicism scale, the mean score was 24.6, with a range from 0 to 44. Last, the mean score for the Authorit arianism scale was 19.2, with a range from 0 to 39.

PAGE 32

27 Bivariate Analysis In order to determine the internal consis tency of each scale, bivariate correlations were examined. Bivariate correlations es timate the direction a nd strength of the relationship between variables. These Biva riate correlations have a range from –1.00 indicating a perfect inverse asso ciation, to +1.00, indicating a pe rfect direct association. If the correlation is equal to zero then th e variables are unrelated. Cronbach’s alphas were determined to make sure that each it em on the scale was measuring the same thing. Inter-Item Correlation Matrices for the three sc ales are provided in Tables 2 through 4. Table 2: Inter-Item Correlation Matrix for Authoritarianism Scale Variable AUTHbook AUTHcondition AUTHshake AUTHcritical AUTHholdbev AUTHtreatciv I like to do things by the book at home. My job conditioned me to expect to have the .116 final say on how things are done in my household I can't shake the feeling of being a police officer .130 .521 when at home. I have become overly critical at home due to .050 .577 .630 the police job I hold my family’s behavior to a high standard because .111 .239 .313 .361 I am a police officer I catch myself treating my family the way I treat .072 .488 .428 .574 .363 civilians Cronbach’s alpha = .744, N = 89, N of items = 6 The Authoritarianism scale produced a Cronbach ’s alpha of .744, which is the same alpha produced in the previous study by Johnson et. al (2005). This alpha indicates a relatively high level of internal consistency within this scale. The Cynicism Scale produced a Cronbach’s alpha of .857. This alpha is sli ghtly higher than Cochran and Bromley’s

PAGE 33

28 (2003) alpha of .83. The Cronbach’s alpha signifies a high leve l of internal consistency within this scale. Table 3: Inter-Item Correlation Matrix for Cynicism Scale Variable CNYlie CNYhelp CNYtrust CNYsteal CNYresp ect CNYlack CNYletrust CNYopen CNYc itztrust Most people lie when answering questions posed by law enforcement officers. Most people do not .363 hesitate to go out of their way to help someone in trouble Most people are .393 .337 untrustworthy and dishonest Most people would .482 .377 .382 steal if they knew they would not get caught Most people .233 .339 .229 .407 respect the authority of law enforcement officers. Most people lack .441 .392 .351 .586 .632 the proper level of respect for law enforcement officers Law enforcement .315 .247 .681 .196 .186 .267 officers will never trust citizens enough to work together effectively Most citizens are .285 .366 .493 .286 .357 .407 .540 open to the opinions and suggestions of law enforcement officers Citizens will not trust law .588 .512 .397 .707 .385 .536 .277 .540 Law enforcement officers enough to work together effectively Cronbach’s alpha = .857, N = 88, N of items = 9 The Emotional Exhaustion scale produced a Cronbach’s alpha of .893. This alpha is the same as the Cronbach’s alpha produ ced in Jackson and Maslach’s study (1982).

PAGE 34

29 This alpha indicates a high level of internal consistency among the items in the scale. Each of the three scales has shown a high leve l of internal consistency indicating that the items on each scale are measuring the constr uct. The alphas of each scale are also consistent with each of the previous studies affirming the reliabil ity of these scales. Table 4: Inter-Item Correlation Matrix for Emotional Exhaustion Scale Variable BRNemotion BRNused BRNfatigue BRN strain BRNburnout BRNfrustrate BRNhard BRNpeople BRNrope I feel emotionally drained from work I feel used up at the .788 end of the day I feel Fatigued when I .621 .713 get up in the morning and have to face another day Working with people All day .529 .528 .561 Is really A strain for me I feel burned out from my job .568 .460 .568 .594 I feel frustrated by my job .552 .601 .570 .523 .689 I feel Im Working too hard 339 .265 .310 .314 .508 .482 on my job Working with people Directly puts. too .526 .469 .378 .511 .543 .402 .240 much stress on me I feel like Im at the end of my rope .445 .383 .383 .545 .623 .399 .246 .637 Cronbachs alpha = .893, N = 89, N of items = 9

PAGE 35

30 Table 5 examines the bivariate relationshi ps between the dependent variables in order to assess the relations hip between psychological dom estic violence and physical domestic violence. As shown in Table 5, these constructs were moderately related to one another. Despite this positive, significant re lationship, the correlati on also suggests there is only modest overlap between the two depende nt variables. This, in turn, suggests each dependent variable should be assessed independently. Table 5: Bivariate Correlation of Physical and Psychological Domestic Violence (N = 90) Variable Physical Domestic Violence Scale Psychological Domestic Violence .301** Scale **p< 0.01 Bivariate correlations were also utili zed to examine the relationship between psychological domestic violence and the three in dependent variable scal es, as well as the control variables (See Table 6). Pearson corre lations show that two of the independent variables, Burnout and Authoritarianism, were significantly correlate d with psychological domestic violence. The Pearson correlati on for Burnout and psychological domestic violence was 0.327, which was significant at th e .01 level. This indicates a positive, moderate relationship between th e two variables, suggesting th at as Burnout increases so to does psychological domestic violence. Th e Pearson correlation for Authoritarianism and psychological domestic violence was 0.415, which was significant at the .01 level. This specifies a positive, moderate relations hip between the two vari ables signifying that as Authoritarianism increases so does psyc hological domestic violence. The relationship of Authoritarianism to psychol ogical domestic violence had th e strongest relationship of any of the other variables to psychological dom estic violence. The Pearson correlation for psychological domestic violence and Cyni cism was 0.039, which was not significant.

PAGE 36

31 Cynicism was the only indepe ndent variable that was not significantly related to psychological domestic violence. There was no statistically significant relationship between any of the control variables and psychological domestic violence. All three of the independent variables were significantly relate d to one another. The Pearson correlation for Emotional E xhaustion and Authoritarianism was 0.348, which was statistically significant at the .01 level. This suggests a moderate, positive relationship between the tw o variables; as Emotional Exhaustion increases Authoritarianism increases. The Pearson correlation for Emotional Exhaustion and Cynicism was 0.313, which was also significan t at the .01 level. This is a positive moderate relationship indicating as Emotiona l Exhaustion increases so does Cynicism. The Pearson correlation for Authoritarianism and Cynicism was 0.301, also significant at the .01 level. The coefficient designates a pos itive moderate relationship in which the increase of Authoritarianism is associ ated with an increase in Cynicism. Some of the control variables were also significantly related to one another. The Pearson correlation coefficient for Latino a nd gender was .288; this is statistically significant at the .01 level. This relations hip is a moderately positive relationship signifying that more females report bei ng Latino. Age was significantly related to Cynicism at the .05 level. The Pearson co efficient was –0.298. This suggests a moderate negative relationship, indicating as an officer’s age increases Cynicism decreases. The number of years in law enfor cement was also significantly related to Cynicism at the .05 level. The Pearson correlation was –0.260. Th is is a moderate negative relationship signifying that, as the number of years in la w enforcement increases, Cynicism decreases. Age was statistically signifi cant at the .05 level, with Latino. The Pearson correlation was –0.238 indicating a moderate negative relationship; older officers in the sample were less

PAGE 37

32 likely to be Latino. The number of years in law enforcem ent was also significantly related to Latino at the .05 level. Th e Pearson coefficient was .265 suggesting a negatively moderate relationshi p; officers with more years in law enforcement were less likely to be Latino. The la st significant relationship was age and years in law enforcement. The Pearson correlation wa s 0.862, which was significant at the .01 level indicating a strong positive relationship. The older officers had more years of experience in law enforcement. Bivariate correlations were also used to examine the relationships between physical domestic violence and the three indepe ndent variables, as well as the control variables. Pearson correlations reveal none of the three scal es, nor the control variables, were significantly related to physical domestic violence. Table 6: Bivariate Correlations of Psychological Domestic Violence Relationships Variable Psychological Emotional Authoritarianism Cynicism Gender Race Lati no Age Years Domestic Exhaustion in L.E. Violence Psychological Domestic Violence Emotional Exhaustion .327** Authoritarianism .415** .348** Cynicism .039 .313** .301** Gender -.076 .062 -.066 .087 Race .075 -.093 -.033 .026 -.0 57 Latino -.153 -.151 -.066 .081 .28 8** .273** Age .112 -.007 .041 -.298* .01 3 -.133 -.238* Years in Law Enforcement .103 -.044 .032 -.260* -.056 -.116 -.265* .862** **p< 01, *p< .05, N=90 Multivariate Analyses Two types of analyses were employed to examine the effects of traditional police culture on domestic violence, Tobit Regression and Logistic Regre ssion. Tobit regression

PAGE 38

33 was used to examine the impact of traditional police culture on psychological violence. In criminology we deal with variables that are not always fully observed; they may contain a lot of zeros. By dichotomizing these variables we may lose valuable information. In order to avoid this problem, alternative models may be used that incorporate the censored nature of the variable, such as the Tob it Model. Tobit regression coefficients are interpreted similarly to OLS regression coeffici ents. Logistic regression, which is used to examine relationships of dichotomous outcome variables, was used to examine the influence of traditional police culture on phys ical violence. Because so few of the respondents reported physical domestic violen ce, the variable wa s dichotomized. Table 7 represents the Tobit analysis of psychological domestic violence and its relationship with traditional police culture The model as a whole was statistically significant (chi square = 23.52, df = 6, p<0.01, Pseudo R2 = 0.0297). Of the six covariates included in the analysis, only tw o were significant, including the Emotional Exhaustion frequency scale and the Authoritar ianism scale. Emotional Exhaustion had a t = 2.04, std. error = 3.38, and beta = .211, indicating a st atistically significant relationship with psychological domestic vi olence at the .05 level. The standardized coefficient indicates a moderate positive relationship between Emotional Exhaustion and psychological domestic violence. The positive coefficient suggests that, as officers’ scores increase on the emotional exhaustion scale, they indicate a higher frequency of engaging in psychological domes tic violence. Authoritariani sm had a t = 3.41, std. error = .520, and beta = .353, indicating a statistically significant relationship with psychological domestic violence at the .01 level. The sta ndardized coefficient signifies a moderate positive relationship between Authoritarianism and psychological domestic violence. The

PAGE 39

34 relationship between Authoritarianism and ps ychological domestic violence is slightly stronger and, in fact, is the strongest relationship of all pa irs of variables. The positive coefficient indicates that, as officers increase on the Authoritarianism scale, they report a higher frequency of engaging in psychological domestic violence. None of the control variables—gender, age, Hispanic, and years in law enforcement—were significant. Race and Cynicism were left out of the analysis because their bivariate relationships showed no statistical significance. Table 7: Tobit Regression Results for Psychological Domestic Violence, Traditional Police Culture, and Control Variables Variable Coefficients Standard t bStdXY Error MBI Burnout Scale 6.89 3.38 2.04* .211 Authoritarianism Scale 1.77 0.52 3.41** .353 LE work 0.17 0.72 0.24 -.023 Gender -2.18 9.33 -0.23 -.038 Age 0.18 0.69 0.27 .05 1 Hispanic -4.10 11.02 -0.37 .045 Chi-square (df) 23.52 (6)** Pseudo R2 0.0297 Log Likelihood -383.9861 *p< .05, **p< .01, N=90 Table 8 illustrates the results from the s econd analysis, Logistic regression, which was employed to examine the relationshi p between physical domestic violence and traditional police culture. The model as a whole was not significant (-2LL = 61.680, chi square = 5.16, df = 8, p>0.01). None of the individual covariates were significant, including the traditional police culture scal es, and each of the control variables. Overall, the multivariate results suggest that officers who adhere to the components of the traditional police culture have a higher frequency of psychological domestic violence

PAGE 40

35 than officers who do not adhere to the traditi onal culture. The results, also suggest there is no relationship between physical domestic violence and the trad itional police culture within this sample. Table 8: Logistic Regression Results for Physical Domestic Violence, Traditional Police Culture and Control Variables Variable B Wald Exp(B) MBI Burnout Scale 0.341 1.110 1.406 Authoritarianism Scale -0.011 0.048 0.989 Cynicism Scale 0.052 0.858 1.053 Gender -0.226 0.057 0.798 Race -16.029 0.000 0.000 Hispanic 1.040 0.991 2.836 Age -0.054 0.507 0.947 LE work 0.780 1.022 1.081 Chi-square (df) 5.160 (8) -2 Log Likelihood 61.680 Nagelkerke R2 0.106 *p< .05, **p< .01, N=90 As previously mentioned, one of the foci in the current study was to assess the potential moderating role of gender on the rela tionship between police domestic violence and aspects of traditional police culture Unfortunately, because so few females completed the survey, in depth analysis was not possible. Nonetheless, using Crosstabulations, and an ANOVA, a gender profile was produced and the results are presented in Table 9. The male sample incl uded both whites and nonwhites; all of the female subjects were white. Males officer s were 96% white, 2.7% African American,

PAGE 41

36 and 1.3% other. Of the male officers, 8% reported being Latino; 33.3% of the female responses reported being Latino. The male officer sample reported yes to engaging in physical domestic violence at 12%. The fema le officer sample reported yes to engaging in physical domestic violence at 13.3%. The mean age for males was 36.73 years old, and females had a slightly higher mean age at 39.06 years. Males reported their years in law enforcement to be slightly higher than females; the male and female means were 11.91 years and 10.53 years, respectively. The average number of times males reported engaging in psychological domes tic violence in their lifetime was about 26. Females reported engaging in psychological domestic violence about 19 times during their lifetime. Each of the traditional police culture scales had similar average findings for both males and females. The average score on the Authoritarianism scale for males was 19, and for females it was 18.2. For males, th e average score on the Emotional Exhaustion scale was 1.31, and for females it was 1.48. Th e average scores on the Cynicism scale was 24.29 and 25.93 for males and females, re spectively. Only one relationship was statistically significant; females were signi ficantly more likely to be Latino.

PAGE 42

37 Table 9: Relationships of Variables By Gender Variable Male Female F/Chi-square % (n) Mean (std.dv.) % (n) Mean (std.dv.) s tatistic Race 0.62 White 96.0 (72) 100.0 (15) Non-white 2.7 (2) 0.0 (0) Missing 1.3 (1) 0.0 (0) Hispanic 7.48* No 92.0 (69) 66.7 (10) Yes 8.0 (6) 33.3 (5) Age 36.73 (9.65) 39.06 (10.21) .015 Years in L.E. 11.91 (9.41) 10.53 (8.79) .277 Physical D.V. 0.02 No 88.0 (66) 86.7 (13) Yes 12.0 (6) 13.3 (2) Psychological D.V. 25.68 (32.60) 19.27 (26.37) .512 Authoritarianism 19.43 (7.43) 18.20 (5.05) .336 Emotional Exhaustion 1.30 (1.06) 1.48 (1.20) .381 Cynicism 24.29 (7.53) 25.93 (3.92) .670 *p< .05

PAGE 43

38 Chapter Five Conclusion and Discussion First, this chapter will provide an overvie w of the results, and discuss the findings from the analysis section. The limitations of this project will be considered next, and a discussion of the policy implications of th ese findings will follow. Finally, future research goals will be provided. Results This study examined the aspects of the traditional police s ub-culture and its effects on police involved domestic violence. The traditional poli ce culture has been linked to a number of negative consequences in the policing prof ession, including but not limited to use of force on the job (Henkel et al., 1997; Kop and Eu wema, 2001; Terrill et al., 2003), high work stress (Gershon, 2000; Jackson and Maslach, 1982; Roberts and Levenson, 2001), and burnout. Only a few st udies have evaluated the effects of traditional police culture on the use of domestic violence within police families (Erwin et al., 2005; Johnson, 2000; Johnson et al., 2005; Sg ambelluri, 2000), and to date only one study has assessed the moderating effects of gender on aspects of traditional police culture and police domestic violence (Johns on, 2000). Although there have been a few studies evaluating the effects of traditional police culture on police domestic violence, each of these studies has included no more than one aspect of traditional police subculture. This study is the firs t to examine the relationships between several aspects of traditional police culture and two types of domestic violence.

PAGE 44

39 For this research data were collected from several police a nd sheriff’s departments in the Tampa Bay area and a survey instrument wa s developed from existing measures. This study included three of the most common asp ects of the traditiona l police sub-culture— burnout, authoritarianism, and cynicism—in or der to build and expand upon the previous research and gain a better unders tanding of the factors that le ad police officers to engage in domestic violence. Tobit and Logistic Regression were used to examine the relationships derived from the previous lit erature and answer the current research questions. The first research question was, “Are o fficers who adhere to the components of traditional police sub-culture more likely to engage in domestic violence?” The results of the analyses partially suppor ted traditional police culture’s relationship with police domestic violence. In order to expand the literature two different types of domestic violence were examined psychological and phys ical violence. The results showed that two of the aspects of the traditional police sub-culture, burnout and authoritarianism, were significantly related to psychological domestic violence. As both burnout and authoritarianism increase, the frequency of engaging in psychologi cal domestic violence increases. Cynicism was not significantly re lated to psychological domestic violence. On the other hand, physical domestic violence was not significantly related to any of the aspects of traditional police sub-culture. None of the control variables in this analysis had significant relationships with either type of domestic violence. It is important to question why cynicism was the only aspect of traditional police culture that was not significantly related to psychological domestic violence. Kop and Euwema (2001) documented that as isolation, or cynicism, increase d, officers felt more favorably toward using violence on the j ob. Although they documented a relationship

PAGE 45

40 between cynicism and violence, the outcome was attitudes toward violence, and not the actual use of violence. Even though officers ma y have felt more favorable toward the use of violence that does not necessa rily translate to engaging in more violence. It may be that cynicism toward the public just increa ses positive attitudes toward using violence and not the actual engagement in violence, a nd therefore does not ha ve an effect on the actual level of violence officers have used throughout their lifetime. Also, cynicism of the public creates an au ra of suspicion and distrust. As other types of police cultures materialize and beco me popular, the fewer the number of officers who adhere to the traditional cu lture. It may be possible that officers who adhere to the traditional culture are gradually being influenced by other cultures such as community policing, and therefore they are slowly begi nning to change their idea of what the policing job entails. One of the main goals of community policing is that officers and the public can work together; the we-versus-them attitude does not exist. As community policing becomes more popular, cy nical attitudes may begin to dissipate. It is possible that pessimistic attitudes may slowly begin to disappear in order to not “rock the boat,” but when officers are out on the job they still adhere to the other aspects of the traditional police culture, such as aggressive law enforcem ent tactics, which lead to high levels of burnout. Authoritarianism was significantly related to psychological domestic violence. To combat the dangerous violence and hostility, o fficers who adhere to th e traditional police culture tend to use aggressive law enforcemen t tactics. They also believe that their occupational description is ambiguous, and ther efore they adopt a st rictly crime-fighting attitude, and authoritarian personality. Terrill et al. (2003) establis hed that officers who use coercive authority, which was the result of cultural alignmen t to traditional police

PAGE 46

41 culture, are more likely to display higher levels of force, such as pushing and hitting. This previous study has shown that an increase in actual violence occurs when officers adhere to the aspect of authoritarianism. The curre nt study has revealed that other types of violence also occur—psychological domestic violence—when officers adhere to the aspect of authoritarianism. Burnout was also signific antly related to psychological domestic violence. Jackson and Maslach (1982) have documented that people in helping professions, such as policing, experience unique stress. They internalize daily in teractions that create higher levels of stress and lead to burnout. It has been shown that these grea ter levels of stress lead to all types of negati ve consequences such as displaying anger (Gershon, 2000; Jackson and Maslach, 1982; Roberts and Le venson, 2001), using ne gative coping outlets (Gershon, 2000), and poor marital outcomes. The findings from the current study are in line with these previous studies showing that burnout leads to nega tive consequences, in this case higher levels of psychological domestic violence. None of the aspects of traditional poli ce culture were significantly related to physical violence. This may be due to the small number of officers who admitted to engaging in physical violence. It also may be that cultu re has no effect on physical violence. These findings will be discussed more thoroughly in the future research section. The second research question, “Does gender moderate the influence of traditional police sub-culture on police domestic violence?” could not be answere d. The analysis of this question required a sufficiently large number of female respondents. Without a sufficient number of females, it would not be possible to the moderating influence of gender. Unfortunately, of the 90 total respon ses only about 16 percent of the sample was female respondents, and therefore this analys is was not possible. Although the originally

PAGE 47

42 contemplated analysis was not possible, using Crosstabulations and ANOVA, a gender profile was created to determine if there we re any significant differences between males and females. The only significant result was being Latino. Neither of the means on the dependent variables was statis tically different, signifying th at males and females engaged in similar levels of domestic violence. Also none of the traditiona l police culture scale means were significantly different from one anot her. This shows that males’ and females’ means were similar on each traditio nal police culture measure. Limitations Although the current study ha s resulted in some importa nt findings, these results come with limitations. First and foremost the results of this study were distinctly different from previous studies. Earlier studies demonstr ated that about 40 perc ent of their samples had been physically violen t with their spouses (Johnson, 1991; Neidig et. al., 1992). The current study found that only 12 percent of the sample had be en physically violent with their spouse. It is possible that this difference is due to the fact that previous studies used different questions to define physical violence. Also it ma y be that the composition of officers from these studies was different. Psychological domestic violen ce was significantly related with two aspects of the traditional police culture. It is important to question why psychological domestic violence was significant and physical domestic violence was not. It may be that officers are more likely to be honest about psychological violen ce because to them it does not constitute actual violence; it does not involve leaving a mark on the other person. An additional drawback of the current study was the analyses illustrated no significant relationships with any of the as pects of traditional police culture, or the control variables to physical domestic violen ce. Due to the nature of policing it is

PAGE 48

43 extremely difficult to survey officers on sens itive subjects such as domestic violence, even when ensuring anonymity and confiden tiality; this may lproduce a low level of reported domestic violence, which in tu rn precludes relationship testing. Another possible explanation is that traditio nal police culture is not related at all to police officer domestic violence. It may be that those individuals that enter in to policing are inherently more likely to be authoritarian and cynical. People who are shy and quiet are probably not likely to enter in to the police academy as often as those who are more assertive and authoritarian. Certain officers may be bringing these aspects of the traditional police culture with them to the j ob, and therefore, these aspects are previous personality characteristics, rather than aspe cts of a police culture, leading to a selection effect. Those that enter into the police profession and stay may have unique personality characteristics (Drew et. al., 2008). Another possible explanation for physical domestic violen ce not being related to any aspects of the traditional police culture is that certain officers have been exposed to better coping mechanisms than other offi cers. Officers that use negative coping mechanisms such as drinking, gambling, or illic it drugs might be more likely to enter in to violence in the home because they are una ble to cope with the stresses of the job. Officers who have been taught how to deal w ith stress in appropriate ways may be less likely to engage in violence at home or on th e job. The current study did not take coping mechanisms into account and this may be an interesting avenue fo r future research. Another limitation of the study was th e operationalization of violence. The measure of violence was over the officers’ lifetim e. It is impossible to tell if the violence that was committed was while they were on the force, or before they were on the force. Also, the older a person is the more time th ey have had to engage in violence. The

PAGE 49

44 amount of time with one’s partner was not take n in to account. Thes e issues affect the analysis of physical violence, and whether they occurred while on the force, and this may affect the relationship with traditional police culture. The current study gave atten tion to the variable gender. Previous research has shown that female officers are more likely to engage in domestic violence than male officers (Gershon, 2000). One aim of this rese arch was to examine more in depth the variable gender to better unde rstand its role in police domes tic violence. Unfortunately this was not possible due to the low number of responding fe males. The small number of female subjects may have been due in pa rt to the sampling approach. The current study used a convenience sample to test the relationships am ong the variables. The final makeup of the sample may indicate why it ma y not be appropriate to assume that a convenience sample is satisfactory for the purposes of testing th e relationships among these variables. The low number of female of ficers resulted in the in ability to look at the gender variable in great depth. A better sampling approach may be a stratified sample in order to ensure enough fema le officers are included. Even though the race variable was not of great interest, the same problems arose with race, as with gender. The sample was ma de up of almost entirely Caucasian officers, and few African American officers. It is im portant when studying relationships that the samples are as complete as possible. A strati fied sample would seek to include specific groups including females and African Amer icans in order to generate a more representative sample. Even though there are limitations to th is research, important results were obtained. Significant relationships between psychological domestic violence and traditional police culture were revealed. The more an officer associates himself or herself

PAGE 50

45 with the aspects of traditional police culture, the more likely her or she is to engage in psychological domestic violence. Yet, future research should be concerned with these limitations. Policy Implications and Future Research Traditional police culture is no longer as widespread as it once was, but it appears that those who adhere to this culture are at more risk for engaging in domestic violence than those officers who do not adhere to thos e cultures. In order to create effective responses to police domestic violence, departme nts must be aware if there is something— for instance, related to the profession—that is increasing officers’ susceptibility to engage in violence at home. This research shows that traditional police sub-culture has an effect on police family violence. By understanding th e sources of violence, agencies may be able to create training programs to help combat this violence before it starts, or programs to intervene once this violence has occurre d. Police agencies already utilize preemployment screening exams; by understandi ng the risk factors of domestic violence within policing, questions may be included on those exams that will uncover officers who may be at risk. This research is only one of a few, a nd it is important that future research understand the limitations and the strengths—red ucing the former and incorporating the latter. First a different sampling approach, such as a stratified sampling approach, needs to be implemented. In this approach member s of a population are grouped together into relatively homogeneous subgroups before sa mpling. In this case the population of officers would be grouped into subgroups of i ndividuals including males and females. In order to do this a list of de partments would be collected to find out how many male and female officers there are in each department. This will help to determine how many

PAGE 51

46 officers would need to be surveyed in order to obtain a decent response rate. Oversampling would then be used to ensure enough female officers responded. The survey in the current study was a good representation of each of the three aspects of traditional police culture that were being examined. Also, the survey included two types of domestic violence instead of one, and even though the questions were particularly sensitive, they encompassed a co mplete range of behaviors that could occur in a violent relationship. Because of the null fi ndings in the current st udy it may be useful to look at something other than culture. Usua lly specific types of people enter into the policing profession. People who are more aggr essive may be more likely than someone who is very passive to become an officer. Future research should include measures of personality to understand whether this aggre ssiveness exists before a person enters the field of law enforcement or whether someth ing within law enforcement brings that aggressiveness out. Measures of stress and copi ng mechanisms also need to be included. Stress may be the precursor to the violence. Stress may be causing authoritarianism and burnout leading to police domestic violence, which may be one e xplanation of their relationship. By removing traditional poli ce culture measures and looking at other possible relationships, a better explanation of police domestic violence may emerge. The results in this study indicate that culture may have no effect on domestic violence, and therefore, future research shoul d look in other directions. In order to increase the total sample, more than one attempt to contact the officers needs to occur. In the current study, surveys were placed in the officers’ mailboxes with a cover letter explaining the survey and asked th e officer to send the survey back with the enclosed self addressed stamped envelope. This was the only attempt to contact the officer. Instead of a cross-sectional study, a l ongitudinal study may be better, in order to

PAGE 52

47 determine exactly what is triggering po lice domestic violence. One possible method would be to survey officers when they ente r the academy, again once they have been in the field for a few years, and then once more after they have been in law enforcement for awhile. By surveying the officer s at different times in their career, researchers will be able to understand whether culture, personalit y, stress, or maybe even something else is having an effect on officers engaging in domestic violence. Though a longitudinal study will take a long time, it may be the only way to separate personality and selection effects, from something that may be occurrin g once the officers are on the job. The current study used several different departments in Florida and was the first to use more than one department. Although it is not possible to know where each of the surveys came from, the attempt was made to ge t subjects from more than one department in order to expand the generali zability. Future research should try to expand the area of study in order to make the results more generalizable. The current research is one of only a few studies that have examined the impact of traditional police culture on domestic violence. Subsequent studies will be needed to confirm or refute the results. Although the study revealed not all of the aspects of traditional police culture were significantly related to domestic violence, some important relationships were observed. These relationshi ps will need to be studied further to confirm whether traditional police culture does have an impact on police domestic violence, or if it is so mething else altogether. The public trusts the police to protect them; officers who engage in domestic violence undermine this trust and therefore undermine the job of all police officers. It is important to continue this re search in order to fully understand the relationships that effect police domestic violence.

PAGE 53

48 References Alexander, D. A., Walker, L. G ., Innes, G., Irving, B. L. (1993), Police stress at work London: Police foundation. Bakker, A. B., Demerouti, E., Schaufeli, W. B. (2002), “Validation of the Maslach Burnout Inventory-General Su rvey: An Internet Study,” Anxiety, Stress, and Coping Vol. 15 No. 3, pp. 245. Broderick, J. J. (1977), Police in a time of change Morristown, NJ: General Learning Press. Brown, M. K. (1988), Working the street: Police discreti on and the dilemmas of reform (2nd ed.). New York: Russe ll Sage Foundation. Chan, J. (1997), Changing Police Culture: Polici ng a Multicultural Society Cambridge University Press, New York, NY. Chan, J. (1996), “Changing Police Culture,” British Journal of Criminology Vol. 36 No. 1, pp. 109. Cochran, J., Bromley, M. (2003), “The My th (?) of the Police Sub-culture,” Policing Vol. 26 No. 1, pp. 88. Connelly, C. D., Newton, R. R., Aarons, G. A. (2005), “A Psychometric Examination of English and Spanish Versions of th e Revised Conflict Tactics Scale,” Journal of Interpersonal Violence Vol. 20 No. 12, pp. 1560. Drew, J., Carless, S. A., Thompson, B. M. (2008), “Predicting Turnover of Police Officers Using the Sixteen Pers onality Factor Questionnaire,” Journal of Criminal Justice Vol. 36 I. 4, pp. 326. Drummond, D.S. (1976), Police Culture Sage, Beverly Hills, CA. Erwin, M.J., Gershon, M.R., Tibur zi, M., Lin, S. (2005), “Repor ts of Intimate Partner Violence Made Against Police Officers,” Journal of Family Violence Vol. 20 No. 1 pp. 13. Gershon, R. M. (2000), Domestic Violence in Police Families MidAtlantic Regional Community Policing Institute, Baltimore, MD.

PAGE 54

49 Henkel, J., Sheehan, E., Reichel, D. ( 1997), “Relation of Po lice Misconduct to Authoritarianism,” Journal of Social Behavior and Personality Vol. 14, pp. 194. Herbert, S. (1998), “Police Sub-culture Reconsidered,” Criminology, Vol. 36 No. 2 pp. 343. Jackson, S.E., Maslach, C. (1982), “After-Eff ects of Job-Related Stress: Families as Victims,” Journal of Occupational Behavior Vol. 31, No.1, pp. 63. Johnson, L. B. (1991b). On the Front Lines: Police Stress and Family Well Being Hearing Before the Select Committee on Children, Youth, and Families House of Representatives: 102 Congress First Session May 20 U.S. Government Printing Office, Washington, DC, pp. 32. Johnson, L.B. (2000), “Burnout and Work a nd Family Violence Among Police: Gender Comparisons,” In Sheehan, D.C. (ed.) Domestic Violence by Police Officers U.S. Government, Washington, D.C., pp. 107. Johnson, L.B., Todd, M., Subramanian, G. ( 2005), “Violence in Police Families Work Family Spillover,” Journal of Family Violence Vol. 20 No. 1, pp. 3. Kappeler,V.E., Sluder,R.D., Alpert, G.P. (1994), Forces of Deviance: Understanding the Dark Side of Policing Waveland Press, Prospect Heights, IL. Kop, N., Euwema, M. (2001), “Occupational Stress and the Use of Force by Dutch Police Officers,” Criminal Justice and Behavior Vol.28, No.5, pp. 631. Lazarus, R. S., Cohen, J. B. (1977). 'The Ha ssles scale'. Unpublished scale measure, University of California at Berkeley. Manning, P.K. (1995), “The Police Occupati onal Culture”, in Bailey, W. (ed.), Encyclopedia of Police Science Garland, New York, NY, pp. 471. Maslach, C., Jackson, S. E. (1981), “The Measurement of Experienced Burnout,” Journal of Occupational Behaviour Vol. 2 No. 2, pp. 99. Maslach, C., Jackson, S. E. (1986), Maslach Burnout Inventory manual (2nd ed.). Palo Alto, CA: Consulting Psychologists Press. Neidig, P., Russel, H., Seng, A. (1992), Interspo usal Aggression in Law Enforcement Personnel Attending the OP Biennial Conference. Natl. Fraternal Order Police J Fall/Winter, pp. 25. Newton, R. R., Connelly, C. D., Landsver k, J. A. (2001), “An Examination of Measurement Characteristics and Factor Va lidity of the Revise d Conflicts Tactics Scale,” Educational and Psychological Measurement Vol. 61 No. 2, pp. 317.

PAGE 55

50 Paoline, E. (2004), “Shedding Light on Poli ce Culture: An examination of Officers Occupational Attitudes,” Police Quarterly Vol. 7 No.2, pp. 205. Paoline, E. A., III, Myers, S. M., Worden, R. E. (2000). “Police culture, individualism, and community policing: Evidence from two police departments,” Justice Quarterly Vol. 17, pp. 575. Reiner, R. (1985), The Politics of the Police St Martin’s Press, New York, NY. Reuss-Ianni, E. (1983), Two Cultures of Policing Transaction Books, New Brunswick, NJ. Richardsen, A. M., Martinussen, M. (2004), “T he Maslach Burnout I nventory: Factorial Validity and Consistency Across Occupational Groups in Norway,” Journal of Occupational and Organizational Psychology Vol. 77, pp. 377. Roberts, N.A., Levenson, R.W. (2001), “The Remains of the Workda y: Impact of Job Stress and Exhaustion on Marital Interaction in Police Couples,” Journal of Marriage and Family, Vol. 63, pp. 1052. Schaufeli, W. B., Daamen, J., Van Mierl o, H. (1994), “Burnout Among Dutch Teachers: An MBI-Validity Study,” Educational and Psychological Measurement Vol. 54 No. 3, pp. 803. Sgambelluri, R. (2000), “Police Culture, Po lice Training, and Poli ce Administration: Their Impact on Violence in Police Families,” In Sheehan, D.C (ed.) Domestic Violence by Police Officers U.S. Government, Washington, D.C., pp. 309. Skolnick, J. (1966), Justice Without Trial John Wiley and Sons, New York, NY. Sparrow, M. K., Moore, M. H., Kennedy, D. M. (1990), Beyond 911: A New Era for Policing. New York: Basic Books. Stith, S.M., and Straus, M. A. (1995). Understanding Partner Vi olence: Prevalence, Causes, Consequences, and Solutions National Council on Family Relations, Minneapolis. Straus, M. A., Hamby, S. L., Boney-McCoy, S., Sugarman, D. B. (1996), “The Revised Conflict Tactics Scales (CTS2): Deve lopment and Preliminary Psychometric Data,” Journal of Family Issues Vol. 17 No. 3, pp. 283. Straus, M. A. (2004), “Cross-Cultural Reliabi lity and Validity of the Revised Conflict Tactics Scales: A Study of University Student Dating Couples in 17 Nations,” Cross Cultural Research Vol. 38 No. 4, pp. 407.

PAGE 56

51 Terrill, W., Paoline, E. and Manning, P. (2003), “Police Cult ure and Coercion,” Criminology Vol. 41 No. 4, pp. 1003. Van Maanen, J. (1974), “Working the Stre et: A Developmental View of Police Behavior”, in Jacob, H (Ed.), The Potent ial for Reform of Criminal Justice, Sage, Beverly Hills, CA, pp. 83. Walton, S., and Zigley (2000). Whatever He Does, Don’t Fight Back or You’ll Lose Your Gun: Strategies police officer victims use to cope with spousal abuse. Domestic Violence By Police Officers United States Government, Washington, DC, pp. 365. Wetendorf, D. (2000). Police Domestic Violence: A Handbook for Victims Life Span Inc., Des Plaines, IL. Westley, W. (1970), Violence and the Police MIT Press, Cambridge, MA. White, S. O. (1972). “A Perspect ive on Police Professionalization,” Law & Society Review Vol. 7 pp. 61.

PAGE 57

52 Appendices

PAGE 58

53 Appendix A Cover Letter for Survey Dear Law Enforcement Professional, I request your assistance with a resear ch project that I’m conducting in various law enforcement agencies in the region for my masters thesis in Criminology at USF. I would be pleased if you would fill out the a ttached survey. This study is examining how thoughts, attitudes, and feelings on the part of law enforcement officers/deputies impact on their behaviors The survey—which is att ached to this letter—will ta ke you about 10 minutes to complete. Because there are some sensitive questions in the survey regarding your interactions with significant others in your life, our procedures ensure complete anonymity and confidentiality. You will NOT put your name on this survey. Your answers cannot be linked to you personally because there are no identifiers on the survey. I hope that you will complete this survey and return it to me in the self addressed stamped envelope included in this packet. There are no risks to you or your privacy if you decide to assist with this pr oject by filling out this survey. If you are uncomfortable with any of the questions feel free not to answer them. The Institutional Review Board (IRB) at the University of South Florida has appr oved this study. If there are any questions or concerns I can be reached at lblumens@mail.usf.edu I hope you will take the time to complete and return this questionnaire. Your participation is voluntary and there is no pena lty if you do not participate. If you have any questions or comments or you would like a c opy of the results (which you may receive regardless of whether you choose to participate) I can be contacted at my email address. Thank you so much for your cons ideration of th is request. Sincerely, Lindsey Blumenstein University of South Florida Department of Criminology lblumens@mail.usf.edu

PAGE 59

54 Appendix B: Law Enforcement Survey For the following questions, please check the answer that best represents you. 1. What is your gender? Male Female 2. What is your race? White Black/African American Asian/Pacific Islander Native American/Eskimo/Aleut Middle Eastern/East Indian 3. Are you Hispanic/Latino? Yes No For the following questions, please fill in the blank. 4. What is your age? ____________________ 5. How many years have you been working in the field of law enforcement? ____________________ Thank you for participating in this study on Police Culture. Your responses will help us understand how the culture of policing affects experiences of law enforcement personnel. All responses are completely anonymous, and confidential. No personnel, or department will be linked to the responses in this survey. I appreciate your contribution to th is extremely important project. Instructions: Please read and follow the directions at the beginning at each section. Thank you again.

PAGE 60

55 Appendix B: Law Enforcement Survey (Continued) For each of the following statements, please i ndicate which answer best represents your position. Strongly Agree Agree Neutral Disagree Strongly Disagree 6. Most people lie when answering questions posed by law enforcement officers. 7. Most people do not hesitate to go out of their way to help someone in trouble. 8. Most people are untrustworthy and dishonest. 9. Most people would steal if they knew they would not get caught. 10. Most people respect the authority of law enforcement officers. 11. Most people lack the proper le vel of respect for law enforcement officers. 12. Law enforcement officers will neve r trust citizens enough to work together effectively. 13. Most citizens are open to the opinions and suggestions of law enforcement officers. 14. Citizens will not trust law enforcement officers enough to work together effectively.

PAGE 61

56 Appendix B: Law Enforcement Survey (Continued) For the following statements, please indicate which answer best represents your position. Strongly Agree Agree Somewhat Agree Neutral Somewhat Disagree Disagree Strongly Disagree 15. I like to do things by the book at home. 16. My job conditioned me to expect to have the final say on how things are done in my household. 17. I can’t shake the feeling of being a police officer when at home. 18. I have become overly critical at home due to the police job. 19. I hold my family’s behavior to a high standard because I am a police officer. 20. I catch myself treating my family the way I treat civilians.

PAGE 62

57 Appendix B: Law Enforcement Survey (Continued) Items 21-29 list feelings/attitudes that some law enforcement pers onnel experience as a result of their job. For each item, pl ease indicate how frequently you feel this way, and how intense those feelings are. Please use the fo llowing scales to respond to i tems 2129: Frequency: 0 = never, 1 = a few times a year, 2 = monthly, 3 = a few times a month, 4 = every week, 5 = a few times a week, 6 = everyday Intensity : 1 = very mild barely noticeable 2 3 4 = moderate 5 6 7 = very strong major Note: If a statement has a frequency of zero, th ere is no need to chec k a box for intensity. Frequency 0 1 2 3 4 5 6 Intensity 1 2 3 4 5 6 7 21. I feel emotionally drained from work. 22. I feel used up at the end of the day. 23. I feel fatigued when I get up in the morning and have to face another day. 24. Working with people all day is really a strain for me. 25. I feel burned out from my job. 26. I feel frustrated by my job. 27. I feel I’m working too hard on my job. 28. Working with people directly puts too much stress on me. 29. I feel like I’m at the end of my rope.

PAGE 63

58 Appendix B: Law Enforcement Survey (Continued) For the following statements, please check the box rep resenting how many times over the span of your law enforcement career have done th is to a spouse, girlfriend/boyf riend, or significant other. Never 1 time 2 times 3-5 times 6-10 times 11-20 times More than 20 times 30. Insulted or swore at my partner. 31. Shouted or yelled at my partner. 32. Stomped out of the room, house, or yard during an argument. 33. Said something to spite my partner. 34. Called my partner fat or ugly. 35. Destroyed something belonging to my partner. 36. Accused my partner of being a lousy lover. 37. Threatened to hit or throw something at my partner. 38. Three something at my partner that could hurt. 39. Twisted my partners’ arm or hair 40. Pushed or shoved my partner. 41. Grabbed my partner. 42. Slapped my partner. 43. Used a knife or gun on my partner 44. Punched or hit my partner with something that could hurt. 45. Slammed my partner against the wall. 46. Beat up my partner. 47. Burned or scalded my partner. 48. Kicked my partner