‘—-J :j LIBRARY 1,.) . 1: Michigan State (0 8 I“; 30 3 Q University This is to certify that the thesis entitled SOUTH KOREAN POLICE OFFICERS’ JOB ATTITUDES AND JOB PERFORMANCE presented by Jongbo Kim has been accepted towards fulfillment of the requirements for the MS. degree in Criminal Justice {Li 9 ML... v Major Professor’s Signature 7h? 2/; 2135’ Date MSU is an Affinnativa Action/Equal Opportunity Institution U-.-.---.-.-.-.mC-----.-.-.--.-.-.-.-.-I-.--_.-“‘-O-‘-._-.-.‘4"9.-.-.-.-.-.--.M-.-.-.-.---.-.- PLACE IN RETURN Box to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE I} "~ JAN 0A6 2003 SOUTH KOREAN POLICE OFFICERS’ JOB ATTITUDES AND JOB PERFORMANCE By Jongbo Kim A THESIS Submitted to Michigan State University In partial fulfillment of the requirements For the degree of MASTER OF SIENCE School of Criminal Justice 2005 ABSTRACT SOUTH KOREAN POLICE OFFICERS’ JOB ATTITUDES AND JOB PERFORMANCE By Jongbo Kim The main purpose of this study was to examine the influence of police officers’ attitudes on their job performance that includes three areas: 1) law enforcement, 2) order maintenance, and 3) service provision. Following a model suggested by previous research, organizational and situational factors’ influence on the performance was also examined. Hypotheses are derived from the model found in the literature. The data were collected through self-administered questionnaires completed by a sample of 737/ 10020 (7 percent) police officers in Seoul Metropolitan Police Agency (SMPA), South Korea. Results show that officers’ attitudes regarding police role are mostly unrelated with their behavior, which suggests the same difficulties of social psychologists’ understanding the attitude-behavior relationship. The results also show that several organizational factors such as special order of focusing on issuing many police tickets and training experience respectively affect order-maintenance and service-provision activities. Probability of encountering suspects of arrest as a situational factor affects law- enforcement activities. Beyond the model concerned with attitudes-behavior relationship, several groups of rank and length of total service are correlated with their job performance. The non-finding results of attitudes-behavior relationship are consistent with those of much previous research in the United States. Measurement trials of this study, which indicated some limitations as well as merits, were discussed for future studies. Copyright by JONGBO KIM 2005 DEDICATION I would like to dedicate this thesis to my wife Chungheui and my children Intae and Hansol. I will dearly miss the past two years in the United States. I love you. iv ACKNOWLEDGEMENTS Writing this thesis was truly a labor of love and one of the hardest challenges of my life. I would not have completed this project without the {steadfast support of several people. I would like to give a very special acknowledgement to Dr. Vincent Hoffman, chairman of my graduate committee. With proper guidance and wisdom, Dr. Hoffman encouraged me to go further whenever I faced difficulties. I would like to acknowledge my graduate committee: Dr. Bruce Benson and Dr. Sheila Maxwell. Dr. Benson, your insightful guidance aided greatly in constructing this study. Dr. Maxwell, you have given me valuable comments during the course of my study. I greatly acknowledge my graduate friends for their contribution to making questionnaires, analyzing data, and discussing results: Jinseong Cheong, Daehoon Kwak and Euigab Hwang. A special thank you to Euigab Hwang for advice, Support, and encouragement. I would like to thank Susan Gade for elaborate revisions of my survey questionnaires and thesis. I would like to thank my colleague officers in Korea for voluntarily participating my survey. I would like to thank Byoungrock Kim, Dongun Kim, Seongyeob Cheong, and Jinwoo In for distributing and collecting survey questionnaires. I would also like to thank Jinbok Kim, my brother-in-law, for providing technical and laborious support. Finally, I would like to express my gratitude and appreciation to my parents and parents-in-law. Finishing my thesis was possible because you encouraged and advised me. I am blessed to share my life with all of you. TABLE OF CONTENTS LIST OF TABLES .................................................................................. viii LIST OF FIGURES ................................................ - ................................ ix CHAPTER I : INTRODUCTION ............................................................... 1 Statement of the Problem ....................................................................... 1 Purpose of this Study ........................................................................... 4 Overview of the South Korean Police ........................................................ 7 CHAPTER H : LITERATURE REVIEW ....................................................... l 1 Theory of Attitude-Behavior Relationship ................................................. 11 Attitudes and Job Performance ............................................................... 14 Organizational Factors, Situational Factors and Job performance ...................... l9 CHAPTERHI: METHODOLOGY ............................................................... 21 Research Hypotheses .......................................................................... 21 Null Hypotheses ........................................................................... 22 Data Collection and Sampling Procedure ................................................... 24 Survey Populations and Sampling Procedure .......................................... 24 Development of Survey Instrument ..................................................... 25 Data Collection ............................................................................ 26 Operationalization .............................................................................. 27 Measure of Independent Variables ....................................................... 27 1. Attitudes Toward Police Role ................................................... 30 2. Perceptions of Organizational Factors ........................................... 31 3. Perceptions of Situational Factors ................................................. 32 4. Demographic Factors ............................................................... 33 Measure of Dependent Variables ........................................................ 34 1. Job Performance of Law Enforcement ........................................... 36 2. Job Performance of Order Maintenance ......................................... 36 3. Job Performance of Service Provision ........................................... 37 CHAPTER IV: ANALYSIS ...................................................................... 38 Explanations of Demographic Factors ...................................................... 38 Descriptive and Univariate Statistics .......................................................... 40 Bivariate Statistics .............................................................................. 41 Test of Hypotheses ............................................................................. 48 Multivariate Analysis of Job Performance Involving Demographic Factors .......... 52 CHAPTER V: CONCLUSION .................................................................. 55 Summary of Findings ......................................................................... 55 Discussion ....................................................................................... 55 Research Implications ......................................................................... 60 vi Policy Implications ............................................................................ 61 Conclusion ...................................................................................... 63 APPENDICES ...................................................................................... 65 Appendix A. Factor Analyses of Independent Variables . ................................ 66 Appendix B. Questionnaires .................................................................. 67 BIBLIOGRAPHY .................................................................................. 78 vii LIST OF TABLES Table 1. Independent Variables, Coding, Frequencies, Mean, and .......................... 31 Standard Deviation Table 2. Dependent Variables, Coding, Mean and Standard Deviation ..................... 37 Table 3. Sample Characteristics .................................................................. 44 Table 4. Comparison of Mean differences (ANOVA) for Respondents .................... 47 Characteristics Table 5. Zero Order Correlations Matrix ....................................................... 49 Table 6. Multivariate Correlates of Officers’ Job Performance .............................. 53 Table 7. Multivariate Correlates of Officers’ Job Performance .............................. 54 (Including Demographic Factors) viii LIST OF FIGURES Figure 1. Model of Attitude-Behavior Relationship ........................................... 20 ix CHAPTER I INTRODUCTION Statement of the Problem When police mangers give an order to their subordinates, the goal of police managers should be twofold: for subordinates to clearly understand the expectations of administrators and to be responsive to organizational directives. Due to the uncertainty of the environment, however, this can often be problematic. This ambiguity stems from both the actual complexities intrinsic to the environment (citizen encounters) as well as the organizational change in police (Brown, 1988). Although laws and bureaucratic regulations are largely clearly defined, in some instances, police officers face conditions or circumstances where the rules or boundaries appear vague. However, organizational change in law enforcement is likely to confound police officers, since interpretation of new polices may be subjective. Police officers in the United States have addressed the aforementioned issues and continue to evolve. For example, community policing has largely dominated the United States since the mid-19903. According to the Bureau of Justice Statistics (2001), sixty- four percent of departments, representing eighty-six percent of the US. population served by local police had full-time officers engaged in community policing initiatives during 1999. Community policing is a strategic concept which seeks to redefine the ends and the means of policing. In contrast to traditional policing methods, community policing focuses attention on the problems that lie behind incidents, rather than on the incidents alone (Moore, 1992). The situation is quite similar in countries where organizational change of police is progressing. In Korea, for example, police officers have recently faced organizational and philosophical changes referred to as “J i goo-Dae”-—the community police corps (Korean National Police Agency, 2004). It is suggested that organizational change may confound officers’ job performance. These factors allow street-level officers to make discretionary decisions. According to Wortley (2003), of the entire criminal justice system, police are considered to have the greatest opportunity to exercise discretionary judgment. Most agree that discretion is an inevitable part of policing. However, there have been serious concerns about the exercise and amount of discretion. For example, selective enforcement of laws allows police to redefine justice in terms of their own priorities, which may not correspond with the administration’s mission. Thus, Brown considers discretion as “the dilemma of administrative control” (Brown, 1988, p.96). Yet, despite the interest and importance of discretion, there have been relatively few studies that explain the variables which affect the discretionary judgment. Some believe that decision-making and behavior may be affected by occupational attitudes and values. In social psychology, the general origin of the attitude-behavior relationship can be traced to a definition that attitudes are thought of both as a predisposition causing consistency in behavior, and also as a general evaluation of an object (Cushman and Mcphee, 1980). This relationship has been of particular interest to both social psychologists and police scholars for several decades. For example, Smith and Klein (1984) maintain that much of the variation in officers’ behavior which was not explained by situational characteristics could be explained by officers’ attitudes, values, and beliefs. Consistent with this ideology, many scholars and practitioners assume that administrators must change officers’ attitudes toward community policing if the strategy is to be successful. Evidence for strong and consistent relationships between attitudes and behavior has, however, proved somewhat elusive (Smith and Klein, 1983; Mastrofski et al., 1994; Worden, 1989). Moreover, little research illustrates a significant relationship between officers’ attitudes and their behavior (Mastrofski et al., 1995; Terrill et al., 2003). Contrary to the growing concerns about the link between attitudes and behavior, research findings in the United States have not yet proved the relationship conclusively. As previously stated, researchers have not provided hard evidence for the relationship between officers’ attitudes and behavior. Thus, it is worthwhile to address this issue in South Korea, since the collective Asian culture may lend additional support and generalizability for the results of previous research; and may provide additional insight for police managers. South Korea is unique for police managers, since the new police practice of J i goo-Dae was implemented (KNPA, 2004) and is expected to be implemented in local autonomous police departments in the future (Munhwa-Ilbo, 2004). Purpose of this Study The purpose of this study is to examine the relationship between police officers’ attitudes and their job performance by analyzing survey datacollected from four police departments in Seoul, South Korea. The sample consists of police officers who are currently working for the community policing corps. Four police stations among 31 police stations in Seoul were selected by a stratified sample which contains approximately 10,000 community policing officers in Seoul. This study attempts to identify attitudinal factors of police officers affecting job performance. Since several scholars have suggested that police officers exercise high amounts of discretion, police officers in Seoul, Korea also fit a similar profile in their job performance. The researcher suggests that the officers’ discretionary activities are affected by their attitudes. This proposition is based on the attitudinal theory in social psychology. A review of the literature suggests that police officers’ behavior may be affected by their attitudes. The hypotheses presented herein were developed on the basis of previous research and theoretical frameworks regarding the attitude-behavior relationship. Research indicates the methodological and conceptual considerations for the police attitude- behavior relationship which emphasizes consistency of the attitude-behavior relationship (Frank and Brandl, 1991). Other literature suggests there are three functions among patrol officers: law enforcement, order maintenance, and service provisions (Wilson, 1968). In light of these considerations, the author constructed three hypotheses that demonstrate attitudes-behavior relationship of police officers, adding organizational and situational factors’ influence upon officers’ activities. Thus, officers’ attitudes predict their behavior, from three perspectives: l) officers’ attitudes predict officers’ behaviors; 2) officers’ perception of organizational factors affects officers’ behaviors; and 3) officers’ perception of situational factors affect officers’ behaviors. Previous studies have generally focused on one dimension of the police role in defining officers’ behavior, while others have examined three dimensions of attitudes toward the police role (Smith & Klein, 1983; Mastrofski et al., 1994; Worden, 1989; Engel and Worden, 2003; Mastrofski et al., 1995; Terrill et al., 2003). This study, however, attempted to examine the dimensions of attitudes and behaviors on the police role from a holistic perspective. In regard to the relationship between attitudes and behavior, “the attitude and the behaviors (initiating officer activity and enforcing certain laws) are both measured at the same level, thus providing correspondence” Frank and Brand] (1991, p.87). The examination of police officers’ performance can advance our understanding of police actions more clearly. This attempt can also serve as a catalyst for future research concerning how attitudes and behaviors have changed through their careers, on the basis of longitudinal examination. Finally, this may assist police managers by providing the necessary information for recruitment, training, and evaluation concerning police activities. In addition to attitudes toward the police role, this research examines several demographic, organizational and situational factors. The literature review suggested these predictors may affect officers’ behaviors (Worden, 1989; Mastrofski et al., 1995; Terrill et al., 2003). Still, they often lack several organizational factors that are presumed to directly affect officers’ behavior. In regard to organizational factors, Frank and Brand] (1991) stated that officers who believe that increased arrest activity will lead to a job promotion may engage in this behavior irrespective of their attitudes concerning the propriety of such behavior. Thus, this study examines several organizational factors as well as the attitude-behavior relationship. This may also help police managers in recruiting and training police officers and in evaluating their activities by providing information about which factors and to what extent the factors affect their activities. Furthermore, a multi-factorial approach is necessary to investigate to what extent attitudes predict behaviors. This perspective should contribute knowledge regarding occupational activities as police officers. Overview of the South Korean Police With the advent of the new South Korean government in 1948, the modern South Korean police force began as the Bureau of National Security under the Ministry of the Internal Affairs. The bureau had nine departments including police administration, security, and economic affairs, etc. The bureau began as a subordinate organization which was considered a reflection of theidea that people and national assemblymen did not want a powerful police organization in South Korea, since during the Japanese colonial era, police had persecuted the independence movement. Since then, independence and neutralization of police has been the center of controversy among the government and political parties. In 1974, The Bureau of National Security was reorganized into the National Security Headquarters, which was upgraded from the bureau but still remained under the Ministry of Internal Affairs. In 1991, the National Police Agency and the Affiliated Organizations were again reorganized, replacing the National Security Headquarters. The National Police Agency is considered more independent and neutral than previous administrations, insofar it is independent from the Ministry of Domestic Affairs, which is in charge of political elections. Still now, however, the neutralization and independence of police organization remains controversial (KNPA, 2004; Je, 1998). The present national Korean police system can be characterized as highly centralized as well as democratic (KNPA, 2004). The National Korean Police Force is comprised of the Headquarters of the National Police Agency, the Central Police Organization, 14 provincial police agencies, 231 police stations, 2930 branch offices and other affiliated institutes including the National Police University, Police Comprehensive Academy, Central Police Training School, Driver’s Licensing Agency, and the National Police Hospital. The National Police Agency is divided into six bureaus, five offices and one Assistant Director General, and the Seoul Metropolitan Police Agency consists of seven departments, two offices and 16 divisions. The Busan Metropolitan Police Agency is divided into seven departments, three offices and nine divisions, and the Kyeonggi Provincial Police Agency is made up of three divisions, two offices and nine sub- divisions. Other provincial police agencies consist of three offices and six divisions. Every police station is divided into one to three levels with four to nine divisions. The Korean police has an 11 rank structure from bottom to top: the Patrol Officer, Senior Patrol Officer, Assistant Inspector, Inspector, Senior Inspector, Superintendent, Senior Superintendent, Superintendent General, Senior Superintendent General, Chief Superintendent General, and Commissioner General. The chief of the National Police Agency is the Commissioner General; the chiefs of provincial police agencies are the Chief Superintendent Generals and the Senior Superintendent Generals; the chiefs of police station are the Senior Superintendents and the Superintendents; and the chief of Jigoo-Dae is the Senior Inspector. As of 2003, there was 92,165 South Korean Police officers. Of these, 34,522 are Patrol Officers; 29,367 are Senior Patrol Officers; 14,436 are Assistant Inspectors; 9,188 are Inspectors; 2,926 are Senior Inspectors; 1,265 are Superintendents; 404 are Senior Superintendents; 33 are Superintendent Generals; 19 are Senior Superintendent Generals; four are Chief Superintendent Generals; and there is one Commissioner General. Of the police officers, about 15 percent occupy the rank of Inspector and higher ranking officials are considered to be executives. The higher the rank is, the more extensive the administrative power is. The recruitment of the Korean police officers from the rank of patrol officer to superintendent is divided in two ways: employment from a general service exam and special employment. The promotion of a police official below Senior Superintendent can be achieved in two ways: via general examination or through the screening of special board. After a certain number of required years at each rank, officers are eligible for promotion. The Performance of Police Functions Act, instituted in 1953 and recently amended in 1981, describes the scope of police function: preventing and investigating crimes; maintaining public safety, the protection of high officials, counteracting espionage; collecting and distributing public safety information; traffic control; and maintaining public order. Since 2003, South Korean police have tried to reform their organization by implementing the “Police Reform Committee” which is composed of 18 external high figures who are considered experts and reformers (KNPA, 2004). Served by the Police Reform Planning Corps, this committee has three subordinate committees: the Role, the Autonomous Police, and the Investigation System committee. The first substantial result of this effort was increasing the number of the rank structure. Before 2003, there were numerous lower ranked personnel of the total number of police officers. The structure severely discouraged low—rank personnel, since promotion was often too difficult for them to achieve. Finally, at the end of 2003, the government confirmed the “Three-Year Project for Improving Police Rank Structure” where each ranking member of the 10,364 employees would be upgraded to a higher rank during a period of three years. The second product was the reform of a police box system that had been utilized since the birth of the South Korean police without improvements or modifications. Considering the deficiencies of the police force and little support from citizens, the South Korean Police implemented the “Community Police Operating System”. The system is explained, in short, by that three to five police boxes are reorganized into one Patrol Jigoo—Dae (below Jigoo-Dae) which is reshuffled according to its geographic location. The chief of Jigoo-Dae is supported by three inspectors who take turns working 24 hour shifts. Thus, the system is intended to effectively combat street crime by mobilizing three to five patrol cars with 10 or more officers. Several police boxes are reorganized into the “Police Service Center” which contains officers in charge of civil appeals. The Police Service Center is intended to intensify the service provision function of police. Moreover, several efforts to reform the police have been attempted, such as a citizen corps for monitoring civil rights in the police organization, the recording and videotaping of child victims of sexual abuse, and changing the method of crackdowns. In the present study, police officers include four ranks: Patrol officer, Senior Patrol officer, Assistant Inspector, and Inspector, since the above officers work as field officers in Jigoo-Dae. Although the rank of inspector is generally regarded as an executive function in Korea, inspector-ranked officers in J i goo-Dae perform their job similar to lower level officials. Therefore, in the community policing corps, only the rank of Senior Inspector (Chief of Jigoo—Dae) was excluded from this study. 10 CHAPTER II LITERATURE REVIEW Theory of Attitude-Behamr Relationship People generate attitudes toward almost everything they encounter within the world. If people perceive something as new, they tend to form an attitude or opinion towards it (Cushman and McPhee, 1980). In other words, people’s attitudes may be inferred by their subsequent behavior. These assumptions have been the object of social scientific research. Initially, theorists assumed that if the attitude is defined as a predisposition to behavior, it is at least a necessary condition for that behavior (Cushman and McPhee, 1980). On the basis of this assumption, some researchers have criticized the attitude-behavior framework by building theoretical models for behavioral prediction. Rejecting a simple view such as massage, attitudes, and behavior, Fishbein (1975) initially defined attitude as the amount of effect for or against a psychological object. The usual cognitive component is redefined as belief. The cognitive component of attitude is termed behavioral intention, defined as a person’s location on a subjective probability dimension involving a relationship between the individual and some action. Fishbein separates these components of attitude in order to stipulate a causal chain with behavior. Behaviors are determined by behavioral intentions, which are functions of attitudes toward the acts in question. The set of generative mechanisms cited in the attitude-behavior literature is found in three ways. Many such models exist to explain behavior in specific domains including consumer buying (Garling et al., 1998), altruistic behavior (Schwartz, 1977), and health- 11 care behavior (Seibold and R0per, 1979). A small amount of research, however in public sector, has reported a significant relationship. For example, Brehm and Gates (1997) reported that public officers’ rate of “shrinking,” or time spent “goofing off” or “loafing,” were influenced by their professionalism, attitudes toward their jobs, and satisfaction with their supervisors. The first traditional mechanism underlying the attitude—behavior relationship was previously discussed and contains a “definitional explanation” of consistency. According to this mechanism, if a person has an attitude at all, his or her behavior must generally be consistent with it by definition (Shuman and Johnson, 1976). A second common mechanism underlying the attitude-behavior consistency is the “latent process” notion of DeFleur and Westie (1963)—an attitude is an unobservable cognitive state underlying various modes of verbal and behavioral response, forcing them to be consistent. The final mechanism was popularized most by McGuire’s (1969) influential essay. The mechanism can be labeled as “information processing”. That is, proponents try to identify precisely various cognitive states, mechanisms, and processes, and to characterize attitudes and behaviors in light of their informational determinant. Many social psychologists have discovered that there are several conditions that lead to a strong link between attitudes and behavior. First, some attitudes are extremely important to various aspects of our life, while other attitudes are less central to our being. Thus, stronger attitudes are likely to predict behavior (McCroskey, 1973). Second, the issue is of correspondence or the degree to which an attitude and some action is highest is when the focus on identical objects are in the same context, and at the same time (Fishbein and Ajzen, 1975). Third, attitudes are often formed from our direct experience 12 of issues and events. Attitudes formed in this capacity, as opposed to being learned through observing role models, are often more consistent with behavior (Fazio and Williams, 1986). Finally, the ease with which information can be located in one’s mind plays a key role in the attitude-behavior link. Fazio and Williams (1986) provide a model of attitude accessibility that suggests that the presentation of an attitude object elicits a memory of that object, which leads to a perception of the object, and finally a consistent behavior toward the object. With the hope of inducing increased job performance, many businesses, nonprofit organizations, and the government have attempted to encourage its employees’ to have positive work-related attitudes and subsequently surveyed them (Rainey, 2003). In the same context, some qualitative research on police has reported that officers’ behavior is shaped by their work-related outlooks (Brown, 1988; Worden, 1995; Terrill and Mastrofski, 2002). 13 Attitudes and Job Performflcg Worden (1995) synthesized officers’ belief systems into five typologies: professional, tough-cop, clean-beat crime fighter, problem-selver, and avoider. This research suggested that officers varied in their outlooks in terms of human nature (e. g. cynicism), processes or outcomes, and coercive authority. Wilson (1968) classified patrol officers’ role typology as follows: the role of law enforcement, role of order maintenance, and the role of service provisions. Among them, he defined maintaining order as the central role for patrolmen more than law enforcement and service provisions, since patrol officers encounter far more problems of order maintenance than opportunities for law enforcement. The other reason is that the service- provision role could easily be provided by a different government agency or private firm. Frank and Brand] (1991) reviewed the social psychological research on the measurement of attitudes and behavior. On the basis of their review, a tentative model by which a more complete assessment of the attitude-behavior relationship could be created. Figure 1 shows their model that suggests some factors that may affect the attitude- behavior relationship. Specifically, they explained the model definitions as follows: Behavior: Police behavior refers to the actions of officers. Environment: The environment is conceptualized as a network of factors that directly or indirectly influence police behavior (e.g. organizational, political, legal, and situational factors). Behavioral Intentions: Behavioral intention is the subjective probability that a person will perform a particular act of interest. 14 Figure 1. Model of Attitude-Behavior Relationship LE>OF———>A"IT——"BI—‘——’13 PC/ SF Where: LE = legal environment, PC = political culture, OF = organizational factors, ATT = attitude, BI = behavioral intention, B = behavior, SF = situational factors, E = environment (Frank and Brand], 1991, p.92) Attitudes: Individuals may possess attitudes toward a specific object (i.e., blacks, women, role of patrol) or toward the performance of a specific behavior (i.e., arrests, suspicious stops). Organizational Factors: Organizational factors include organizational demand for production, formal organizational rules, and organizational culture influences, and expectations established in the work group. The influence of organizational factors on behavior may occur in one of two ways: organizational factors may directly influence behavioral intentions and behavior (e. g. hope for promotion cause increased arrests); organizational factors may indirectly influence behavioral intentions by influencing officer attitudes (e.g. organizational rule requiring arrests in domestic violence cases may have the effect of changing officer attitudes concerning the appropriateness of this response). 15 Political Culture: Political culture encompasses the values and expectations of the community concerning the types of behavior the police should perform. Legal Environment: The legal environment includes a network of formal rules that are intended to structure police behavior and are backed by governmental sanctions. Situational Factors: Situational factors include the specific characteristics of the situations in which police encounter citizens (e.g., age, sex, race, demeanor, etc.). (Frank and Brand], 1991, pp. 93-97) Brown (1988) indicated that officers’ behavioral tendencies varied respectively. By combining the attitudinal dimensions of aggressiveness and selectivity, he derived a four-fold typology of operational styles: old style crime fighter, clean beat crime fighter, service style, and professional style. According to his typology, officers who defined the police role as one of crime-fighting and law enforcement were more aggressive on the street. His analysis further suggests that officers whose role conceptions differed in these respects also differed in their approaches to minor disorders and service requests. However, most quantitative research has reported that there is only a weak relationship between officers’ attitudes and their behavior (Smith and Klein, 1983; Mastrofski et al., 1994; Worden, 1989; Engel et al., 2003). Smith and Klein (1983) examined the extent to which organizational properties influence the propensity of police officers to invoke the law of arrest. They reported that police decisions to arrest are 16 influenced by the demand of the situation, certain organizational properties of police departments, and the organizational context in which police-citizen encounters occur. Furthermore, the most important factors influencing arrest decisions are certain aspects of the police-citizen encounter itself (e.g., offense seriousness, victims’ requests, acquaintance of the victim, etc.), while the effect of the organizational properties (e.g., supervisor, officer’s job satisfaction, officer’s education) of police departments has little direct influence (Smith and Klein, 1983). Mastrofski et a]. (1994) assessed arrest productivity for Driving Under the Influence (DUI) utilizing a sample of Pennsylvania police officers. They found that officers’ personal feelings about the seriousness of drunk driving are weakly related to performance, while commitment to a legalistic style is a significant but less powerful intrinsic reinforcement (Mastrofski et al., 1994). Worden (1989) found that styles of performing their jobs bear a weak relationship to their occupational attitudes (e.g., role orientation, citizen respect, and legal restriction, etc.), while officers might individually ad0pt these styles that manifest themselves in behavioral patterns across similar situations. Engel et a]. (2003) examined the influence of officers’ attitude and priorities toward community policing over the time officers spend conducting problem-solving activities. They also found that officers’ attitudes toward their roles are weakly correlated with their activities. Recently, small research has found a significant relationship between officers’ attitudes and their behavior (Mastrofski et al., 1995; Terrill et al., 2003). Mastrofski et al. (1995) indicated that in a time of community policing, officers who support it do 17 manifest some arrest decision patterns distinguishable from those of colleagues who adhere to a more traditional view of law enforcement. Officers classified as “positive” toward community policing arrested five percent of the suspects they encountered, while the “negative” toward community policing arrested seventeen percent of the suspects. In their study, legal variables illustrated stronger effects than extra-legal ones, but this depended on the officers’ attitudes toward community policing. Terrill et a1. (2003) found that officers who closely embody the values of the police culture are more coercive compared to those who differentially align with the culture. Encounters involving pro-cultural officers had nearly a twenty percent chance of physical restraint, compared with fifteen percent in those encounters involving non-pro- cultural officers. He concluded with “Police use of force is a function of officers’ varying attitudinal commitments to the traditional view of police culture (Terrill et al., 2003, p.1101)”. 18 Organizational Factors, Situationa_l_F_actors and Job Perform_a_n§£ Previous research examined the factors that influence officers’ decision making. In addition to the influence of officers’ attitudes, organizational and situational factors, officers’ demographic factors may affect officers’ behavior (Novak et al., 2002; Frank and Brand], 1991; Worden, 1989). Much research has constructed its own theoretical frame and suggested the influence of the organizational, situational, and demographic factors on officers’ behavior. Worden (1989) reassessed the value of situational and attitudinal variables as parts of a theory of police behavior. He examined effects of situational factors and officers’ attitudes on traffic stops. He reported that situational factors (e.g., discretionary time, graveyard shift, evening shift, and seriousness of offense) have a significant effect on officers’ decisions to make traffic stops, as previous research also indicated. From an organizational and situational perspective, Novak et a]. (2002) found that there is no significant direct influence of assignment on arrest decision, while situational factors such as evidence, citizen age, and citizen noncompliance significantly affect arrest decision. Furthermore, regarding the situational, psychological, and organizational context, Worden (1995) found that to some extent, the use and abuse of force by police are influenced by characteristics of the situation in which officers and citizens interact; neither officers’ characteristics nor attitudes had a substantively significant effect on the use of force; bureaucratic agencies use more force. Specifically, he found that officers are more likely to use force against those who offer physical resistance, and who are blacks; compared with officers in more bureaucratized departments, officers in less bureaucratized departments either are less likely to use force when it would be justified or 19 are less likely take actions early in an encounter that has an high probability of force later (Worden, 1995). Engel et a]. (2000) exarrrined the relationship between the demeanor of suspects and the police behavior, and found that demeanor of suspects did not affect police behavior. Considering the findings collectively, however, the researchers thought that in most circumstances officers do not treat hostile juveniles or adults, males and females, or blacks and whites differently. Utilizing a situational, psychological, and organizational theoretical framework, Terrill and Mastrofski (2002) found higher rates of force in Indianapolis than St. Petersburg, which attributed to more aggressive policies. Their finding indicates that force was more likely in encounters involving inexperienced officers; encounters initiated by officers themselves involved a great likelihood of force. From the neighborhood context, Smith (1986) found that police are more likely to exercise coercion toward suspects toward non-whites and racially mixed neighborhoods. Moreover, he found that these effects are independent of the race, sex, and demeanor of the suspect, the type of problem involved, and whether the encounter occurs in a public or private setting. In sum, previous research illustrates there is no firm conclusion that organizational factors and situational factors affect officers’ job performance. 20 CHAPTER III METHODOLOGY Research HvDOtheses The primary hypothesis in this study is that police officers’ actions are based according to previous perceptions. This, of course, sets a discretionary environment forth as a premise. When police officers have high amounts of discretion, their behaviors may be consistent with ordinary citizens exercising discretion. As an explanation of behavior, many social psychologists claim that attitudes can predict behavior (e.g. Cushman and McPhee, 1980; Fishbein and Ajzen, 1975). Moreover, many researchers, including police scholars have argued there is a significant relationship between attitudes and behaviors (Brehm & Gates, 1997; Mastrofski et al., 1995; Terrill et al., 2003). Thus, these findings lead to the hypotheses in this study. One key factor in forming the hypotheses was the consistency of attitudes and behaviors. Police researchers have often failed to identify the relationship between officers’ attitudes and their behaviors. This failure has been attributed to problems in the measurement of police attitudes, behaviors, or both (Frank and Brand], 1991; Worden, 1995). Previous literature has shown that often researchers have measured officers’ behaviors not corresponding to their attitudes, but merely focusing on their attitudes. In order to properly assess the relationship between the attitudes and behaviors, it is important that the attitude and behavior be theoretically related (Frank and Brand], 1991). Because attitudes refer to predisposition of behaviors, it is reasonable that specific attitudes predict specific behaviors (Ajzen, 1988). For example, a research hypothesis is 21 considered logical when an officer who has positive attitude toward arrest is expected to make many arrests. Thus, hypotheses in this study were constructed by the logic where each attitudes and behaviors are consistently related. Null Hypotheses Hypothesis 1: There is no relationship between officers’ attitudes toward police work and job performance. A: There is no relationship between positive attitudes toward law enforcement and law-enforcement activities. B: There is no relationship between positive attitudes toward order maintenance and order-maintenance activities. C: There is no relationship between positive attitudes toward service provisions and service-provision activities. Another point of consideration is the systematic ordering of variables. Actually, the systematic ordering of propositions regarding attitude-behavior relationship is still controversial at least in two areas: the closure of the system of propositions, and the stipulation of relationships as direct, indirect, or spurious (Cushman and Mcphee, 1980). Thus, the assumption that there is a direct causal relationship between attitudes and behavior is an oversimplification of a very complex process. Frank and Brand] (1991) suggest that police behaviors are affected by many factors systematically. Although this model does not provide the rationale of ordering, it attempts to explain and predict police behavior systematically through recruiting diverse factors. Some hypotheses were 22 constructed on the basis of this model, which emphasized the importance of investigation of organizational and situational factors’ influence on officers’ behavior. Therefore, this model suggests that organizational or situational factors predict officers’ behavior. Hypothesis II: There is no relationship between perception of organizational factors and officers’ job performance. Hypothesis 111: There is no relationship between perception of situational factors and officers’ job performance. 23 DataCollection and Sampling Procedure Survey Population and Sampling Procedure The target population of this study is police officers Who are currently working at J igoo-Dae (community police corps) in police departments in Seoul, South Korea. Seoul’s population was estimated at 10,276,968 at the end of 2003. This accounts for about a quarter of the total national population. Seoul is known as a center of politics, economy, society, and culture since the 1300s. Thus, examining police officers from Seoul is an appropriate sample and provides an understanding South Korean police officers. In South Korea, there are 14 provincial police agencies, 233 police stations, 866 Jigoo-Daes and 92,488 police officers who are working for provincial police agencies. Among the provincial police agencies, the Seoul Metropolitan Police Agency (SMPA) has 31 police stations, 14] Jigoo-Daes, and 24,560 police officers, which account for approximately a quarter of South Korean police officers working for the SMPA. The number of police officers who work at Jigoo-Dae in the SMPA is 10,020 as of 2003. If the respondent was not a police officer from SMPA in South Korea, he or she was excluded from the sample. Chiefs of Jigoo-Dae were also excluded since several questions were not appropriate for Chiefs of Jigoo-Dae. In Seoul, each police station has 2 to 6 Jigoo-Daes, and has 149 to 442 personnel, so police departments were stratified. Four police stations were selected from a convenience sample. According to the KNPA (2004), Police station A has five Jigoo-Daes and 432 police officers; Police station B has five Jigoo-Daes and 343 police officers; Police station C has six Jigoo-Daes and 408; and Police station D has 4 Jigoo—Daes and 245 police officers. The expected number of 24 sampled officers was approximately 1,300 persons, which represented 13 percent of the parameter of Jigoo-Dae officers in Seoul. Development of Survey Instrument The purpose of the study is to assess the relationship between officers’ attitudes and their job performance. To measure attitudes, most social psychologists and police scholars have used the survey method because it is considered both reliable and attainable. To measure behaviors, many psychologists also used this method, while recent police scholars have attempted to use observation method. Despite its strength, the observation method typically involves additional time, personnel, and costs. For these reasons, the survey method was used to measure behavior as an alternative. The questionnaire was developed on basis of a thorough review of the literature. The overall frame of the instrument followed Frank and Brandl’s model. According to Frank and Brand] (1991), it is necessary to measure organizational and situational factors as well as attitudinal factors to assess the attitude-behavior relationship. To determine attitudinal factors, previous research instruments were utilized. Some instruments were modified to assess situational and organizational factors, as well as job performance in a Korean context. The first question items were open-ended. Upon informal discussion with Korean police officials, changes were implemented to ensure reliability of the instrument. The finalized version was closed-ended. The questionnaire was pre-tested by investigators at the Seoul Metropolitan Police Agency (SMPA). After examining the result of the pre-test, the ranges of each question item were modified. Items were added to assess policing issues. The finalized instrument was an eight page questionnaire 25 consisting of 93 items. For the Korean version, the items were translated by the author, who is proficient in the language. The translation was reviewed by several natives for reliability and applicability. All respondents were notified that participation was completely voluntary, and that the survey could be terminated at any time without consequence or penalty. Both the English and Korean questionnaires were submitted to the University Committee on Research Involving Human Subjects (UCRIHS) at Michigan State University (MSU) for approval to ensure the rights of human subjects in this study were upheld. Data Collection A letter of solicitation was sent the chiefs of four police stations to obtain survey respondents in January 2005. Personal identifying information was not requested and only the author viewed responses. The Korean version was printed and sent from Korea to the four police stations. Designated officers at each police station received the questionnaires and distributed copies to representatives of J i goo-Dae. A voluntary representative of the officers working for Jigoo-Dae served as a survey administrator in each police station. The representative distributed the surveys together with envelopes. During the distribution, the survey administrators stressed the importance of anonymity and voluntary participation. The respondents were asked to submit their completed surveys directly to the representatives within two weeks. The survey was administered in office of Jigoo-Dae, where respondents have privacy. The representatives subsequently mailed the collected surveys to the author. The overall response rate was 56.6 percent; among 10,020 officers in the four police stations, 737 usable responses were received. 26 Omrationalization Measure of Independent Variables The main independent variable in this study is police officers’ attitudes. In social psychology, an attitude is an individual’s predisposition to behave in a particular way in response to given stimuli within his or her world (McCroskey, 1973). That is, officers’ attitudes can be referred to officers’ predisposition to behavior. An attitude represents an evaluation of a particular focus. Therefore, the evaluation of police work represents an officer’s attitudes. For example, Paoline et a]. (2000) states that officers’ attitudes toward problem solving concerns “the degree to which officers’ role conceptions encompass a responsibility for handling problematic conditions (Paoline et al., 2000, p.589)”. The degree to which officers’ conception of police work encompasses their work represents attitudes toward police work. By asking officers the extent of their conceptions about police work, officers’ attitudes can be measured. Most items used to construct independent variables regarding attitudes were developed from previous research. Terrill et a]. (2003) and Engel and Worden (2003) tested items through factor analysis and assembled highly related items to construct a new variable. Most items were formatted on a Likert-type scale since it is a common way of measuring attitudes (Singleton and Straits, 1999). Officers were asked to indicate the degree of agreement or disagreement that they felt about each concept according to the following scale: “Disagree Strongly (1)”, “Disagree Somewhat (2)”, “Agree Somewhat (3)” and “Agree Strongly (4)”; and “Never (1)”, “Rarely (2)”, “Sometimes (3)”, and “Always (4)”. A high score indicates that an officer is supportive of a particular concept. Responses to items were summed to create a composed variable. 27 In addition to officers’ attitudes, organizational and situational factors were also operationalized as independent variables. According to Frank and Brand] (1991), except for the attitudes, organizational and situational factors are considered to affect officers’ behavior. Organizational factors include demand for production, formal organizational rules, and organizational culture influence. That is, some organizational factors such as the desire to get promoted may directly influence behavioral intentions and behavior such as increasing arrests. For this study, organizational factors refer to these kinds of direct influential organizational factors. The organizational factors can be divided into two groups: a group of motivation factors providing rewards when officers perform well, and a group of frustration factors that punish officers when officers do perform poorly. The organizational factors can be found in officers’ evaluation of these factors. Police theorists have emphasized the value of situational factors as part of a theory of police (Worden, 1989; Frank and Brand], 1991). Frank et al. (1991) noted that situational and organizational factors are expected to either increase or decrease correlations between attitudes and behaviors. Situational factors refer to the specific characteristics of the situations in which police encounter citizens. Situational factors can be found in the complainant or perpetrator (age, sex, race, demeanor, etc.) and the degree to which the situation is visible (Friedrich, 1980). Demographic characteristics may also be related to behaviors. Demographics refer to officers’ gender, age, marital status, rank, education, length of total service, and length at current position. These independent variables are utilized in order to measure the relationship between officers’ attitudes and their job performance. Table 1 illustrates the independent variables. 28 Table 1. Independent Variables, Coding, Frequencies, Percentages, Mean, and Standard Deviation. Independent Variables Coding N Percent a Mean (3D,) Total Number of Cases 737 ' 100.0 Attitudes toward Role of Police Law Enforcement a 2 - 8 6.71 (1.27) Order Maintenance 3 3 - 12 9.09 (1.82) Service Provision a 3 - 12 9.65 (1.84) Organizational Factors Arrest - Motivation Factors 3 3 - 12 8.52 (2.16) - Frustration Factors a 3 - 12 6.71 (1.95) - Order Yes = 1 506 68.7 N0 = O 215 29.2 - Training Yes 2 1 351 47.6 N0 = 0 375 50.9 Police Ticket - Motivation Factors 3 3 — 12 6.85 (2.16) - Frustration Factors 3 3 - 12 6.78 (1.96) - Order Yes = 1 469 63.6 N0 = 0 259 35.1 - Training Yes = 1 293 39.8 N0 = 0 432 58.6 Situational Factors Probability concerned arrest 1 — 4 2.81 (.82) Probability concerned police 1 - 4 ticket 2.73 (.77) 3 Refer to appendix 1 for factor analyses of aggregate variables Missing cases not reported 29 I. Attitudes toward Police Role According to Wilson (1980), the role of police is divided into three areas: law enforcement, order maintenance, and service provision. Correspondingly, the concept of attitudes toward police role can be divided into three-dimensional attitudes, such as attitudes toward law enforcement, order maintenance, and service provisions. Thus, utilizing factor analysis, three factors including attitudes toward law enforcement (See Appendix B, Items 12, 19), attitudes toward order maintenance (See Appendix B, Items 21, 22, 23), and attitudes toward service provision (See Appendix B, Items 14, 16, 17) were created to measure attitudes toward police role. Through factor analysis (See Appendix A), three types of attitudes emerged toward police role: attitudes toward law enforcement, order maintenance, and service provisions. Officers’ predisposition toward aggressive enforcement concerns “the extent to which officers endorse a proactive style of patrol involving frequent stops and field interrogations (Paoline et al., 2000, p.588)”. To measure attitudes toward law enforcement, officers were asked to respond to two statements: 1) The primary function of police is to handle crime-related problems; 2) Strict enforcement of the law is a patrol officers’ most important responsibility. These items were used in previous research in a similar manner (Terrill et al., 2003; Engel et al., 2003). Officers’ inclination toward order maintenance involves the degree to which officers include in their role conception a responsibility for handing routine order- maintenance situations. For measuring attitudes toward order maintenance, officers were asked by three questions: 1) How often do you think that patrol officers should respond to neighbor disputes? 2) How often do you think that patrol officers should respond to 30 domestic disputes? 3) How often do you think that patrol officers should respond to nuisance violations that may be problematic for community residents (e. g. abandoned cars and houses)? The former two question items were extracted previous research for measuring attitudes toward order maintenance. The final question items were applied to measure attitudes, since factor analysis revealed that officers regarded this item as the same as the former two question items, although Terrill et a]. (2003) used this item to measure attitudes toward problem—solving policing. In the same context, officers were asked to respond to three statements concerning attitudes toward service provisions: 1) The primary function of police is to provide community service; 2) The primary purpose of police is to serve the citizen; 3) the primary function of police is to safeguard citizens’ rights. The questions were developed to measure officers’ focus of attention. 2. Perceptions of Organizational Factors Frank and Brand] (1991) noted that organizational factors might directly influence behavioral intentions and behaviors. The authors indicated that officers who believe the way to get promoted is through increased arrests may engage in this behavior irrespective of their attitudes. Waegel (1981) states organizational factors include the organizational setting where the work is performed, which places distinctive constraints and demands on legal agents. This study follows Waegel’s conception of organizational factors. Organizational factors can be classified as: inducement of promotion and evaluation, frustration of transfer and undermining competency, and special directives and training. Thus, officers were asked the extent to which they agree or disagree with the following 31 statements “Making a lot of arrests/Issuing a lot of tickets helps obtain a job promotion” (See Appendix B, Items 45, 55); “Making many arrests/Issuing many tickets helps you receive a good evaluation” (See Appendix B, Item 46, 56); “Making few arrests/Issuing a few tickets leads to an unexpected transfer to another department” (See Appendix B, Items 49, 59); “Making few arrests/Issuing few tickets leads to the undermining of your competency among fellow police officers” (See Appendix B, Items 50, 60). Through factor analysis (See Appendix A), two organizational factors concerned with making arrests or issuing tickets emerged: motivation factors and frustration factors. Besides, experience of training and receiving orders was asked. They were the questions “During the past six months, did you receive an order from your supervisor to focus on making numerous arrests (See Appendix B, Item 53) or on issuing tickets (See Appendix B, Item 63)?”; “Did you receive special educational training focusing on making arrests (See Appendix B, Item 54) or issuing tickets (See Appendix B, Item 64)?”. ~ 3. Perceptions of Situational Factors Worden (1989) noted that police behavior can be affected by situational factors. Features of social structure, such as the social status of victims, complainants, and suspects, represent extra-legal criteria on the basis of which sanctions are applied. In previous research, situational factors refer to the features of social structure, social status of victims, measured by the severity of the offense, the visibility of the encounter, characteristics of the suspects, characteristics of the victims, and the relationship between the parties (Worden, 1989). Situational factors in this study refer to the features of social structure when the police work is executed. Due to the research design, it is difficult to 32 elicit information concerning all situational factors. For convenience, officers were asked only about the visibility of the suspects with the statement “The probability of encountering suspects concerned with Five Indexed Crimes (FIC) is higher in your jurisdiction than in the other jurisdictions (See Appendix B, Items 51)” and “The probability of encountering suspects concerned with Basic Order Violation (BOV) is higher in your jurisdiction than in the other jurisdictions (See Appendix B, Items 62).” 4. Demographics Factors Six demographic characteristics, including gender (See Appendix B, Item 1), age (See Appendix B, Item 2), marital status (See Appendix B, Item 3), rank (See Appendix B, Item 4), education (See Appendix B, Item 5), length of total service (See Appendix B, Item 6), and length of current position (See Appendix B, Item 7) were asked to provide background information. Gender was measured on the basis of respondents’ self- identification. Age was measured according to respondents’ self-description. Marital status contained three categories: married, single or never married, and other (divorced, widowed, etc.). Rank was measured in four categories: Patrol officer, Senior Patrol officer, Assistant Inspector, and Inspector. Education was measured on a continuum of four categories from high school or below to graduate or above. Length of total service and length in current position were measured according to officers’ self-description. 33 Measure of Dependent Variables Wycoff and Manning (1983) stated that police officers’ job performance refers to the execution of police work that is independent of its consequences after their work is completed. Wilson (1968) stated that police have three major types of work: law enforcement, order maintenance, and service provisions. Therefore, the conceptual definition of officers’ job performance in this study is in regard to the performance of the three major types of work independent of consequences. Wilson (1968) suggested that law enforcement activities are crime-control activities such as making arrests, checking vehicles, and investigating a burglary in progress; order maintenance activities refer to control behavior that either disturb or threatens to disturb the public peace or involves a dispute over what is right or seemly conduct; service provisions are exemplified through such activities as assisting persons, dealing with drunkenness or public intoxication, and escorting vehicles. The difference between the law enforcement and order maintenance is that former is what he does when there is no dispute, while the later is what he does when there is a dispute with culpability. Table 2 illustrates the dependent variables regarding officers’ job performance. Frequencies of making an arrest represent officers’ job performance regarding law enforcement duties. Previous researchers measured officers’ law enforcement activities in the same manner (Terrill et al., 2003; Engel and Worden, 2003). In Korea, several crimes such as murder, robbery, rape, theft, and violence are accentuated in officers’ duties because these crimes are considered the most heinous by Korean police managers, and are also referred to as the Five Index Crimes (FIC). As a result, making a FIC arrest is valued more in officers’ evaluation, so officers are more likely to focus on making FIC 34 Table 2. Dependent Variables, Coding, Mean and Standard Deviation (N = 737) DeJendent Variables Coding Mean (SD) Law Enforcement (N a = 708) Arrest (Five Indexed Crime) 1 — 8 ' 2.39 (1.69) Order Maintenance (N a = 699) Police Ticket (Basic Order Violation) 1 — 8 4.29 (2.66) Service Provision (N a = 703) Time 1 — 8 4.00 (2.06) 3 Missing is not reported arrests rather than arrests for petty crimes. Yet, both FIC arrests and arrests for other crimes in Korea represent officers’ law enforcement activities/duties. In 2003, the Korean police made 399,119 FIC arrests. During the last decade, the number arrests by Korean police have slightly increased. In this study, frequencies of issuing tickets of Basic Order Violation represent officers’ job performance for order maintenance. In Korea, police issue a ticket for a violation of a Minor Criminal Act as of the Basic Order. The Minor Criminal Act was initiated in 1954 and amended in 1983. It is intended to prevent crime, to maintain public peace and order, and to achieve other purposes. Furthermore, it is intended to punish minor immoral acts that do not meet the criteria for violations defined in the Criminal Act. Some of the acts prescribed in the Minor Criminal Act are as follows: simple assault, jaywalking, begging, public indecency (exposure), dumping garbage in a public place, urinating in public, spitting, public drunkenness, and smoking where prohibited. Police can issue a ticket, called “Tonggo-Cheoboon,” which exempts violators from an ordinary expensive legal procedure unless they disobey the ‘Tonggo-Cheoboon”. If a violator accepts the ”Tonggo-Cheoboon”, he or she pays a fine without going onto one’s record. 35 In 2003, Korean police issued 800,392 tickets nationwide, 761,683 of which accepted the “Tonggo-Cheoboon” and 38,709 of which preferred an ordinary legal procedure, called “Jeokeyol,” a summary trial. During the last decade, the number of tickets issued by Korean police has fluctuated. Time spent providing citizens with police services represents officers’ job performance regarding service provisions. Previous qualitative researchers measured officers’ community policing activities through observing their time spent with citizens (Terrill et al., 2003; Engel & Worden, 2003; DeJong et al., 2001). These studies examined the time that officers spent on problem-solving activities as dependent variables for estimating the effects of officers’ opportunities, attitudes, and police culture. 1. Job Performance of law Enforcement An item was created to measure arrest activities concerning the FIC (Item 65). The item was generated for the Korean context. Respondents were asked to indicate how many arrests they had made during the past six months. Item (65) “arrest of FIC” was broken down into eight separate parts: (1)0 — 1, (2) 2 — 3, (3)4 - 5, (4) 6 — 7, (5) 8 - 9, (6) 10 - 11, (7) 12 — 13, (8) Over 14. The range of each item was decided after examining several officers’ arrest rates. High values indicate higher law enforcement activities. 2. Job Performance of Order Maintenance This study uses issuing tickets for a Basic Order Violation (BOV) as a dependent variable for order maintenance (Item 67). For Item 67, respondents were asked to indicate 36 the number of BOV tickets they issued during the past six months. Item (67) “police ticket of a BOV” was broken down into eight separate parts: (1) 0 - 6, (2) 7 — 13, ( 3) 14 — 20, (4) 21 — 27, (5) 28 — 34, (6) 35 — 41, (7) 42 — 48, (8) Over 49. The range of the item was decided after examining several officers’ ticketing rates of BOVs. High values indicate higher order maintenance activities. 3. Job Performance of Service Provision This study uses time spent providing citizens with services as a dependent . variable for service provisions (Item 69). Officers were asked to indicate the amount of time spent providing services to citizens during each shift (09:00-21:00), excluding time spent staying in the office, making arrests, and issuing tickets. The item “time spent- service provisions” is broken down into eight separate parts: (1) 0— 15 Min., (2) 16— 30 Min., (3) 31- 45 Min., (4) 46— 60 Min., (5) 61— 90 Min., (6) 91— 120 Min., (7) 121- 150 Min., and (8) Over 151 Min. The range of the item was decided after examining several officers’ working styles. High values indicate higher service provision activities. 37 CHAPTER IV ANALYSIS This paper investigates the relationship between officers’ attitudes toward the role of police and their job performance, using data collected from four police stations in Seoul, South Korea. Several hypotheses which examine the influence of three factorial attitudes, organizational factors, and one situational factor on officers’ job performance were tested. This chapter begins with a discussion of the descriptive statistics and the bivariate correlations among the variables. Next, multivariate analyses are reviewed. Explmtions of Demographic Factors Table 3 illustrates the demographic characteristics of police officers in Seoul, Korea (N=737). In regard to gender, 95.4 percent of officers were men, while 1.9 percent of officers were women. The age of officers ranged from 21 to 57 years old; both the middle-aged group (33-44 years old) and the older group (45 or more) were larger than the younger group (21-32 years old). Of the officers, 58.7 percent were less than 45 years old. Regarding marital status, 79.6 percent of officers were married, while divorced or widowed officers only amounted to .5 percent. In regard to rank, 92.2 percent of the officers were Assistant Inspectors or lower level officers, while 4.2 percent of the officers were Inspectors. Almost half of the officers had the equivalent of a high school education or less, while 25.8 percent of the officers had the equivalent of a four-year college degree or above. The total length of serve as a police officer ranged from zero to 388 months. Officers who had 10 years or less experience amounted to 31.1 percent, while 36.4 38 Table 3. Sample Characteristics Demographics N * Percent* Total Number of Cases 737 100.0 Gender Male 703 ‘ 95.4 Female 14 1.9 Age 21-32 152 20.6 33-44 281 38.1 45 or more 273 37.0 Marital status Married 587 79.6 Single, Unmarried 120 16.3 Others (Divorced, Widow, etc.) 4 .5 Rank Patrol officer 130 17.6 Senior Patrol officer 214 29.0 Assistant Inspector 336 45.6 Inspector 31 4.2 Education High School or Below 377 51.2 2-year College 133 18.0 4-year College or Above 190 . 25.8 Length of Total Service (Month) 0- 120 229 31.1 121-240 268 36.4 241 or more 203 27.5 Length of Current Position (Month) 0-12 270 36.6 13-24 303 41.1 25 or more 121 16.4 * Missing is not reported percent had 10 to 20 years of service, and 27.5 percent had more than 20 years of service. For length in current position, 77.7 percent of the officers served two years or less, while 16.4 percent of the officers had more than two years experience in the community policing corps. 39 Descriptive and Univariate Statistics Tables 1 and 2 illustrate the frequencies, percentages, means and standard deviations for all variables. In regard to attitudes toward the police role, the mean of positive attitude toward law-enforcement activities was 6.71 with a standard deviation of 1.27, while the mean of positive attitude toward order-maintenance and service-provision activities were respectively 9.09 with a standard deviation of 1.82 and 9.65 with a standard deviation of 1.84. For perceptions of organizational factors, the mean of motivation factors for making arrests was 8.52 with a standard deviation of 2.16, while the mean of frustration factors of making arrest was 6.71 with a standard deviation of 1.95. The mean of motivation factors for issuing tickets was 6.85 with a standard deviation of 2.16, while the mean of frustration factors for issuing tickets was 6.78 with a standard deviation of 1.96. Almost 70 percent of officers received an order from their supervisor to focus on making arrests, while less than half of the officers (48%) received special educational training concerned with arrests. More than half of the officers (64%) received an order from their supervisor to focus on issuing tickets, while less than half of the officers (40%) received special educational training concerned with issuing tickets. For officers’ perceptions of situational factors, the mean of the likelihood of encountering suspects for arrest was 2.81 with a standard deviation of .82, while the mean of the likelihood of encountering individuals for issuing tickets was 2.73 with the standard deviation of .77. For the dependent variables, the mean of arrests for a Five Index Crime (FIC) was 2.39 with a standard deviation of 1.69; the mean of tickets for a Basic Order Violation 40 was 4.29 with a standard deviation of 2.66; and the mean of the time spent on service provisions was 4.00 with a standard deviation of 2.06. Bivarifiate mtistics To see the bivariate relationship between demographic characteristics and arrest, issuing tickets and time spent on service provisions, One-Way Analysis of Variance (ANOVA) was conducted. When F value was statistically significant, the Turkey analysis was conducted for multiple comparisons. Table 4 illustrates the mean differences by demographic characteristics. For age on officers’ job performance, the middle-aged group made more arrests (2.53) than the young group (2.17). Interestingly, regarding time spent on service provisions, young officers more spent on service provision than the old group (4.25, 3.78, respectively). Married officers made more arrests (2.45) than single or unmarried officers (2.10), although the Turkey statistic indicated the mean difference between them was not statistically significant. In regard to rank, Senior Patrol officers made most arrests (2.69), while Patrol officers made fewest arrests (1.96). Regarding tickets, Senior Patrol officers issued the most police tickets (4.61), while the mean for Assistant Inspectors was 4.38, and Inspectors issued the fewest tickets (2.39). Regarding time spent on service provisions, Patrol officers (4.24) spent more time on service provisions than Inspectors (3.24). Higher-ranking officers were spent on less time on service provisions (Patrol officer-4.24, Senior Patrol officem4. 10, Assistant Inspector=3.88, Inspectom3.24), even though multiple comparisons of the Turkey statistic only showed the mean difference between 41 Table 4. Comparison of Mean Differences (ANOVA) for Respondent Characteristics Arrest a Police Ticket 8 Time a Age 21-32 2.17/150 3.98/148 4.25/148 (1.54) . (2.64) (2.15) 33-44 2.53/275 4.50/272 4.10/272 (1.79) (2.62) (2.12) 45 or more 2.36/255 4.18/252 3.78/256 (1.64) (2.70) (1.95) F Value 2.33* 2.04 2.88* Marital Status Married 2.45/565 4.33/559 3.97/560 (1.70) (2.66) (2.07) Single, Unmarried 2.10/116 3.96/114 4.11/116 (1.53) (2.60) (2.10) Others (Divorced, Widow, etc.) 3.50/4 4.50/4 3.75/4 (3.00) (2.89) (2.06) F Value 2.88* .93 .25 Rank Patrol officer 1.96/ 128 3.67/126 4.24/127 (1.45) (2.51) (2.15) Senior Patrol officer 2.69/206 4.61/204 4.10/203 (1.76) (2.64) (2.14) Assistant Inspector 2.37/324 4.38/323 3.88/320 (1.64) (2.65) (2.02) Inspector 2.38/26 2.39/23 3.24/29 (2.26) (2.68) (1.68) F Value 5.01** 7.47*** 239* Education _ High School or Below 2.39/36] 4.35/357 3.90/360 ( 1.69) (2.75) (2.07) 2-year College 2.56/131 4.26/128 3.96/128 (1.79) (2.49) (1.98) 4-year College or Above 2.27/ 182 4.05/182 4.25/183 (1.64) (2.59) (2.13) F Value 1.12 .756 1.78 Length of Total Service (Month) 0-120 2.26/226 4.14/224 4.24/223 (1.58) (2.56) (2.13) 121-240 2.61/259 4.59/256 3.90/256 (1.79) (2.70) (2.00) 241 or more 2.26/190 4.05/188 3.79/191 (1.64) (2.70) (2.03) F Value 352* 2.76* 2.78* Length of Current Position (Month) 0-12 2.22/258 3.85/254 4.12/257 (1.64) (2.54) (2.11) 13-24 2.46/294 4.66/292 3.87/291 (1.66) (2.74) (2.04) 25 or more 2.69/118 4.27/116 4.13/116 (1.86) (2.59) (2.07) F Value 3.31* 6.53“ 1.23 a . . Mean scores / number of cases; (standard devratron) * p S .10, ** p S .01, *** p S .001; *, **, ***: Turkey analysis was used for multiple comparisons. 42 Patrol officers and Inspectors was statistically significant. In education, no statistically significant mean differences were found. However, the results showed interesting patterns on officers’ job performance: more educated officers made less arrests and issued fewer tickets (arrests=2.39, 2.56, 2.27 respectively; tickets: 4.35, 4.26, 4.05 respectively); and more educated officers spent more time on service provisions (3.90, 3.96, 4.25 respectively). For the length of total service, statistically significant differences of the means were found in almost all job performance areas. Specifically, those with 10-20 years service made most arrests (2.61), compared to those with 10 years or less of service or officers with 20 or more years of service (2.26). Officers with 10-20 years of service issued the most tickets (4.59), while those with 20 years or more issued fewer tickets (4.05). Regarding the time spent on service provisions, those with 20 years or more of service (3.79) spent less time on service provision than those with 10 years or less of service (4.24). As for the length in current position, statistical mean differences were found in two job performance areas. Regarding arrest, officers who spent 2 years or more in current positions (2.69) made more arrests than those who spent 1 year or less in current positions (2.22). Regarding ticket, one to two years in the current position issued the most tickets (4.66) compared to other groups (3.85, 4.27, respectively), while the mean ' difference among them was not statistically significant, according to the Turkey statistic. The correlation coefficients (Pearson’s r) for the variables included in the analyses are displayed in Table 5. An examination of the coefficients reveals several significant relationships among the independent variables. More specifically, service provision- 43 .2336 3?: Se 2: a edge»: a eoesoeeu $25-9 _o>o_ 3o 2: a 28:2? 2 8.3280 .. :v0g. :02. :02. 35. :08. a8.- 30. :NE. ~83: no 523305 .tmao. :03. .am _. .08. .vao. 000.- 2-0. ..v0_. 3.2.8 00 com—Bangs” :2 _. :30. 0 _ o. .30. we. :aao. 08. 0:0.- wEESH .mwo. coo. a5. 08.- :3 M. . wmc. :30. v3. EEO :am0. woo. wmor amor 35.. v00.- .ms. 500. magnum :oambman— :ovm. :Eh. 08. N50. 340.- 08. Re. vmo. 392$ 503002 Soc—own. oozom 03. .mwo. 08. .30. 58. 000. m8.- mmo. wet—02H. :m ~ _. 03. 03. wmc. .30. too. :wao. mmo. 820 _0o. 03. to: mac. 2b.- N-wo. v.00. 382$ eon—Ewan :3” _. LC. :30. a3. owe. mvo. mcouonm coagaoE ~35 :Nvm. :03. 30. So.- 03. 53305 333m :w0~ . .boo. So. 03. coca—.8282 .505 who. moor 5N0. eeoEoocouem 33 .03. 30. 082. :Ohm. Hoe—o; com—om “mug Eouomm accomm E5305 ooemcuHEe—z Echocfiem :onebmam conguoz oo_>..om EEO 3 08¢. “cc—o:- oozom $2.2. “mug $2.3. . 3 £522 $5022.50 EEO ocoN .m 20a. 44 .6253 :32 So on. a ace-Ewe e eoesoeoc 623-8 as. 86 cs 3 Eocene a 83850 .. .84. :8. ..§. .oE. ins. e8. So. ease o 5:585 .98. .25. c8. ewe. as. one. nose cc 5538.. tmmm. .oB. rmfl. tone. :02. wees; :5. . N8: ...-.2. ram. 520 tam. So. $8. cocoa Seesaw n00. wvo. £98m 503002 coco:- 8:9. ran. wees; . HDUhO £98m :ozabmam €8er :oumzcoz 38.9. .5335 835m 3:38:82 60.5 EoEoocoEm 33 ca: cocoa. oozoe ~85 6on we “more we wage; EEO 958$ 398”— wemfiafi. 505 3:535 £335 5on 8:8 sec. case Seneca”. 82sec: :22 .85. coco; oo=£ cocoa. 8:8 Goneucouv .er2 9802850 EEO ocoN .m 20$. 45 oriented officers were more likely to be motivated in arresting than law enforcement- oriented officers (r = .134, .098, respectively). Service provision-oriented officers were more likely than law enforcement-oriented officers to regard the probability of encountering suspects for arrest and issuing tickets as higher than other jurisdictions (r = .159, .094; r = .105, .096, respectively). In terms of organizational factors, officers encouraged by motivation factors for arrest were more likely to also be encouraged by motivation factors for issuing tickets than officers encouraged by frustration factors of arrest (r = .447, .240, respectively). Finally, officers encouraged by motivation factors to issue tickets were less likely to perceive encounters as higher than other jurisdictions than officers encouraged by frustration factors (r = .125, .189). While significant relationships exist among a number of independent variables, there are also several independent variables related to officers’ job performance. An examination of the coefficients reveals several significant relationships between the independent and dependent variables. In regard to arrest, only situational factors were statistically associated with arrest. Specifically, the probabilities of encountering suspects for both arrests and tickets were significantly associated with arrest (r = .164, .102). For issuing tickets, several organizational factors were statistically associated with issuing ticket activities. Specifically, officers who received orders to make increased arrests were more likely to issue more tickets than those who did not (r = .098). One attitudinal factor was associated with the time spent on service provisions. Order maintenance-oriented officers were weakly associated with the time spent on service provision (r = .097). Training experience as an organizational factor was also associated with the time spent on service provisions (r = .099). 46 The results of the bivariate correlations illustrate that attitudes, organizational factors, and situational factors are associated with an officers’ job performance. In order to determine to what degree to which each variable influences outcomes at each stage, controlling for all other variables in the model, multivariate analysis is needed. 47 Test of Hypotheses The hypotheses tests were based on assessments of direction, statistical significance, and size of the individual level regression coefficients and the coefficient of determination (R2). The generalized linear regression analysis was used to test the hypotheses since all dependent variables consisted of more than eight levels, most of which were ordinal or ratio level. Typically, when two variables are an interval or ratio level, the relationship between the two variables is usually measured by linear regression analysis. However, this study used generalized linear regression analysis. In the generalized linear regression, the eta statistic is reported, which conveys the extent of the total explained variance (R2) contributed by individual predictors. In this study, p S .10 was used as an indicator that an effect was greater than zero. Multivariate linear regression analysis, including all demographic characteristics was conducted to analyze the relationship between all the independent variables including demographic factors and dependent variables. Table 6 displays the multivariate correlates on job performance. To determine the relative influence of each factor of attitudes, organizational and situational perceptions, linear regression analysis was performed. The coefficient of determination (R2 = .04) illustrates that four percent of the variance in making arrest activities is explained by all attitudinal, organizational, and situational factors. In addition, the R-squared value represents that four percent of the variance in both issuing tickets (.04) and time spent on service provisions (.04) is explained by all the factors. 48 Table 6. Multivariate Correlates of Officers’ Job Performance Arrest Police Ticket Time b (S.E.) b (S.E.) b (S.E.) Attitudes . Law Enforcement .03 (.05) -.04 (.09) .09 (.07) Order Maintenance .03 (.04) .01 (.06) .06 (.05) Service Provision -.03 (.04) -.05 (.06) .00 (.05) Organizational Factors Arrest -Motivation Factors -.00 (.03) .06 (.06) .00 (.04) -Frustration Factors .02 (.04) .02 (.07) -.07 (.05) -Order .00 (.17) .17 (.27) .06 (.21) -Training '.17 (.18) -.23 (.29) -.13 (.23) Police Ticket -Motivation Factors .02 (.04) -.05 (.06) .08* (.05) —Frustration Factors .03 (.05) .11 (.07) -.07 (.06) -Order .05 (.16) .74** (.26) .13 (.20) -Training -.00 (.19) -.01 (.30) .45* (.23) Situational Factors Probability of arrest .29*** (.09) .22 (.14) -.20* (.11) Probability of ticket .08 (.09) -.15 (.15) .03 (.12) Constant .70 .58 3.16 .94 3.45 .73 R2 04 .04 .04 * p s .10, ** p s .01, *** p s .001 The Relationship between Ofiicers ’Attitudes and Their Job Performance The first hypothesis suggests there is a relationship between officers’ attitudes and job performance. Illustrated in Table 6, essentially no relationship exists between officers’ attitudes toward police role and job performance. Regarding the relationship between officers’ attitudes toward the police role and arrest, no statistically significant relationship was found. Specifically, although the law enforcement-oriented officers made more or less arrests (b = .03, .03, -.03, respectively), when officers’ orientation increased by 1 unit, there was no statistically significant relationship. Regarding the relationship between 49 officers’ attitudes toward police role and issuing tickets, no statistically significant relationship was found. Specifically, even if the order maintenance-oriented officers made more or less arrests (b = -.04, .01, -.05, respectively), when officers’ orientation increased by 1 unit, there was no statistically significant relationship. Finally, no significant relationship was found for the relationship between officers’ attitudes toward police role and spent time in service provisions. The Relationship between Organizational Factors and Oflicers’Job Performance It was also hypothesized that there may be a relationship between organizational factors and officers’ job performance. Illustrated in Table 6, there are weak relationships between several organizational factors and job performance. Specifically, regarding the relationship between organizational factors and making arrests, no statistically significant relationship was found. However, regarding the relationship between) organizational factors and both issuing tickets and time spent on service provisions, significant relationships were found. Moreover, officers who received orders to focus on making arrests issued more tickets (.74). Officers encouraged by motivational factors of issuing tickets spent more time on service provisions (.05) when officers’ encouragement increased by 1 unit. Lastly, officers who received special education about police ticket spent more time on service provisions (.45) when they received specialized training. The Relationship between Situational Factors and Oflicers’Job Performance This study investigated the relationship between situational factors and officers’ job performance. Table 6 indicates there are two relationships between officers’ 50 perception of situations and job performance. Specifically, in regard to the relationship between officers’ perception of the possibility of encountering suspects for arrest and arrest, a statistically strong relationship was found. Officers. who thought they would encounter more suspects to arrest than other jurisdictions made .29 more arrests when their perception increased by 1 unit. However, officers who thought they would encounter more suspects for arrest than other jurisdictions spent less time on service provisions (-.20). 51 Multivariate Analysis of Job performance Involving Demographic Factors To further explore the effects of all independent variables on officers’ job performance, multivariate analysis including demographic factors was performed. Linear regression was used for the multivariate analysis. Table 7 displays the multivariate analysis of all independent variables including demographic factors on officers’ job performance. An R2 value of .08 illustrates that eight percent of the variance in making arrests is explained by all attitudinal, organizational, situational factors, and several demographic factors. Furthermore, eight percent of the variance in issuing tickets is explained by all the attitudinal, organizational, situational factors, and several demographic factors (R2=.08). Finally, six percent of the variance in time spent on service provision is explained by all attitudinal, organizational, situational, and several demographic factors (R2=.06). The results of the analyses are mixed. Specifically, attitudinal (factors did not have statistically significant relationship with officers’ job performance. Regarding the relationship between organizational factors and officers’ job performance, two statistically significant relationships were found: Officers who received orders to focus on making arrests issued more tickets (.72); and officers who received special educational training regarding tickets spent increased time on service provisions (.44). Regarding situational factors and officers’ job performance, only one relationship was found. Officers who thought they would encounter more suspects to arrest than other jurisdictions made .29 more arrests when their perception increased by 1 unit. Several statistically significant relationships were found for demographic characteristics and officers’ job performance. In regard to the demographic factors and 52 Table 7. Multivariate Correlates of Job Performance (Including Demographics Factors) Arrest Police Ticket Time b (S.E.) b (S.E.) b (S.E.) Attitudes Law Enforcement .01 (.06) 7.10 (.09) .ll (.07) Order Maintenance .05 (.04) .OO (.06) .06 (.05) Service Provision -.01 (.04) -.OO (.06) .Ol (.05) Organizational Factors Arrest - Motivation Factors -.Ol (.04) .08 (.06) -.01 (.05) - Frustration Factors .0] (.04) .02 (.07) -.08 (.06) - Order -.01 (.17) .17 (.28) -.01 (.22) - Training -.10 (.19) -.12 (.31) -.02 (.24) Police Ticket - Motivation Factors .05 (.04) -.OO (.06) .07 (.05) - Frustration Factors .04 (.05) .ll (.07) -.04 (.06) - Order .09 (.17) .72** (.27) .13 (.21) - Training -.08 (.20) -.15 (.32) .44* (.25) Situational Factors Probability of arrest .31*** (.09) .14 (.15) -.18 (.11) Probability of ticket .08 (.10) -.07 (.16) -.OO (.13) Officers’ Characteristics Age - Young (21-33) .21 (.37) .82 (.59) .41 (.46) - Middle (3344) -.15 (.25) .13 (.40) .26 (.31) - Old (45-57) Reference Group Marital Status - Married Reference Group - Single, Unmarried .ll (.24) .19 (.38) -.19 (.30) - Other (Divorced, Widow, etc.) -.90 (1.19) -.16 (1.89) -1.00 (1.48) Rank - Patrol Officer -.27 (.35) -l.04* (.56) -.05 (.44) - Senior Patrol Officer 50* (.21) .05 (.34) .23 (.27) - Assistant Inspector Reference Group - Inspector .14 (.36) -2.07*** (.60) -.93* (.43) Education - High School or Below -.07 (.19) .04 (.30) -.07 (.24) - 2—year College -.05 (.21) .13 (.30) .19 (.24) - 4-year College or Higher Reference Group Length of Total Service (Months) - 0-10 years (0-120) - 10-20 years (121-240) .43* (.24) .25 (.39) -.O4 (.31) - 20 years or more (241-388) .14 (.35) .11 (.56) .12 (.44) Length of Current Position (Mon.) - 1 year or less (0-12) -.20 (.20) -.55* (.25) .31 (.20) - 1-2 years (13-24) Reference Group - 2 years or more (25-336) .14 (.19) -.33 (.30) .28 (.24) Constant .13 (.71) 3.30 (1.14) 2.92 (.89) R2 .08 .08 .06 * p5 .10, ** pf .01, *** p5 .001 53 arrests, two statistically significant relationships were found. Specifically, Senior Patrol officers made more arrests than Assistant Inspectors (.50) while those who served 10-20 years made more arrests than those with 10 years or less of service (.43). An examination of demographics and issuing tickets revealed two statistically significant relationships. Patrol officers issued fewer tickets than Assistant Inspectors (- 1.04) while Inspectors issued the fewest tickets compared to the Assistant Inspectors (- 2.07). With regard to the relationship between demographic factors and time spent on service provisions, only one statistically significant relationship was found. Inspectors spent less time on service provisions than Assistant Inspectors (-.93). 54 CHAPTER V CONCLUSION Summm of Findings This study examined the factors which influence officers’ job performance. More specifically, the attitudinal influence, including organizational and situational factors on officers’ job performance were assessed. The purpose of this paper was to determine if there was a relationship between officers’ attitudes and their job performance; between perceptions of organizational factors and job performance; and between perceptions of situational factors and officers’ job performance. Overall, the results indicate: 1) there was little, if any, relationship between officers’ attitudes and their job performance; 2) several organizational factors seemed to affect officers’ job performance; 3) one situational factor of probability of an arrest encounter in an officer’s jurisdiction significantly increased the probability of arrest. However, the relatively low amount of explained variance in the dependent variables described above indicates there are other predictors which influence job performance. Discussion The results are consistent with most previous research findings concerning the attitudinal influence on officers’ behavior. Since no relationship between attitudes and behavior emerged, this suggests a similar dilemma of social psychologists’ understanding of the attitude-behavior relationship. Social psychiatrists have identified complex relationships between attitudes and behavior. Ideally, positive attitudes manifest well- 55 adjusted behavior. However, in some cases healthy attitudes may result in destructive behavior. Similarly, in this study, service provision-oriented officers spent less time on service-provisions, made fewer arrests and issued fewer tickets, although this relationship was not significant. Studies have demonstrated that inconsistencies between attitudes and behavior can redirect behavior. Officers in this sample may perform their jobs differently, inconsistent with components of the attitude-behavior relationship. It is also possible that some of the measurement problems regarding the independent and dependent variables resulted in an inability to fully capture all of the elements of the attitude and behavior relationship. This author measured attitudes toward order maintenance by dividing into the item “how often do you think patrol officers should respond to nuisance violations which may be problematic for community residents (e.g. abandoned cars and houses)”? However, some researchers used this item to measure attitudes toward community policing. The result of the factor analysis of the attitudinal items illustrated that this item was considered the same to respondents as attitudes toward order maintenance. This may result from a difference in perceptions between US. police officers and South Korean police officers. Yet, clearly it illustrates the difficulty of measuring an attitudinal dimension. Moreover, measuring behavior had additional complications. Most research previously utilized one dependent variable as a behavioral dimension, such as arrest, time spent on problem solving, or use of force, etc. This type of approach may also result in an insufficient measure of the dependent variable. That is, in investigating a relationship between role-oriented attitudes and job performance, simply measuring one aspect of job performance is not sufficient. In order to address consistency, a broader range of police 56 behavior was measured compared to previous research. For example, to measure order maintenance activities, the number of issuing tickets for Basic Order Violations was obtained; for law enforcement duties, the number of arrests. was measured; and finally, for service provisions, time on service-provision was measured. The research was designed to measure attitude-consistent job performance in policing. However, measuring the dependent variables still appeared to have critical problems. This problem was illustrated particularly in the construction of scales. The univariate analysis of arrests indicated that half of the officers made less than three arrests during the past six months, resulting in little variation. In addition, measuring time spent on service-provisions was also problematic. A quarter of officers spent over 151 minutes during the day shift on service provisions. Comparing the mean time of 31-45 minutes, the officers’ time spent is quite difficult to interpret. It may be due to the measurement of scales or the overall difficulty in measuring the dependent variable of job performance. Another explanation of no-relationship for the effect of attitudes upon job performance may be attributed to the unique dynamics in the organization. Even if police on the street have broader discretion than other criminal justice officials, this may be restricted by increased organizational pressure. In Korea, the purpose of implementing Jigoo-Dae system was in large part, due to the deficiency of police force (KNPA, 2004). This resulted in the suspension of patrolling by police vehicles (KNPA, 2004). Regretfully, there has been little research that examines the perceptions of police officers’ concern with their environment. Several studies have examined the organizational influence on officers’ job performance. However, no firm conclusion has been reached among police scholars. For 57 example, Novak et al. (2002) found that assignment as a community policing officer has no direct effect on officers’ decision to arrest. In addition, Worden (1995) found that officers in more bureaucratic agencies use more force. Terrill and Mastrofski (2002) found that officers with more aggressive policies use more force. Although several studies have examined the organizational effects on arrest and coercion, there is little research for organizational effects on order maintenance or service provisions. Moreover, this author could not locate any research concerning a relationship between officers’ attitudes toward the organization and their job performance. Of particular interest, although not statistically significant, frustration from issuing few tickets or few arrests slightly increased the number of issued tickets or arrests, but decreased the time spent on service provisions. One possible explanation is that Korean police officers may be apprehensive regarding certain aspects of their job performance, which led them to perform their duties passively. Another explanation is that officers who were encouraged by frustration factors may not be likely to conduct their occupational duties proactively in circumstances where job performance is not monitored. Both explanations may be important for Korean police managers. One situational factor on officers’ job performance was assessed, since this study focused on attitudinal factors rather than other factors (and also due to cost restrictions). The most recent studies in the United States have utilized the observation method. For better observation techniques, Mastrofski et al. (1995) conducted the systematic social observation (SSO) methodology with trained researchers. Generally, previous research has found that situational factors affect arrest decision-making. Yet, regarding specific situational factors, prior research indicates differences. For example, a suspect’s race, sex, 58 and demeanor are respectively found as influential factors on officers’ job performance (Novak et al., 2002; Worden, 1995; Engel et al., 2000; Terrill and Mastrofski, 2002; Smith, 1986). The findings indicate that officers who perceived they encountered more suspects in their jurisdiction than in other jurisdictions made more arrests, while there was no significance for officers who perceived they encountered more individuals for issuing tickets. Although this finding was restricted by measurement flaws, it showed officers’ arrest decisions may be affected by situational factors. The result of the multivariate analysis including demographic factors displayed similar findings to the multivariate analysis of independent variables. Like the multivariate analysis of independent analysis, a high probability of encountering suspects increased actual arrests. However, several significant factors in the independent variables resulted in no statistically significant relationships. For example, motivation factors of making arrests had no statistically significant effect on time spent on service provisions. Likewise, the probability of encountering suspects for issuing tickets did not have a statistically significant effect on time spent on service provisions. Several demographic factors revealed a statistically significant relationship with officers’ job performance. Particularly, rank had a significant effect on officers’ job performance. These results of multivariate analysis clearly showed some demographic factors’ influence on officers’ job performance. 59 Research Implicgtions This study improves upon past research in several ways. Most studies examining the relationship between officers’ attitudes and behavior have focused on one aspect of job performance as a behavior. To increase the scope, this study attempted to measure diverse dimensions of police performance. Moreover, this is consistent with the suggestion of social psychologists who assert the consistency of the relationship between attitudes and behavior is an important factor for Operationalization. For consistency, service provisions were measured, corresponding to attitudes toward service provisions as a police function; and order maintenance was measured, corresponding to attitudes toward order maintenance as a police function. Additionally, this also enables a comparison of each job performance per the independent variables. This research also suggests that more elaborate items concerning attitudes are needed. South Korean police officers often interpret several items differently than US. police officers. For example, in US. research, the item of “how often do you think that patrol officers should respond to nuisance violations that be problematic for community residents?” is intended to measure an attitude toward community policing. For the South Korean police, however, this item was interpreted as an order maintenance function. Moreover, South Korean police identified the role of service provision with activities of serving citizens. Through factor analysis, this study suggests that South Korean police officers identified their role as three types: 1) law enforcement, 2) order maintenance through resolving citizens’ disputes or expected problems, and 3) service provisions by serving citizens. These role conceptions are slightly different than Wilson’s typology. Therefore, additional research is needed to identify “what police do” from officers’ 60 perspectives. Despite its strengths, this study has several limitations which must be expressed. First, there is a concern regarding the constructed scales of items relating to job performance. Although a pre-test for the scale of job performance was administered, the result of this study raises a question that several items incorporated into the scale may be inappropriate for proper measurement. For example, on average, officers made fewer than four arrests during the past six months, which corresponds with the range of two to eight. This finding confounded the analysis because the variance was very small. In addition, for a more precise measurement of officers’ job performance, the observation technique is needed. Second, although the survey instrument captured diverse attitudinal dimensions, this study could not measure several situational factors. Noted from previous research which found a relationship between situational factors and officers’ job performance, the observation technique, such as the systematic social observation (SSO) method, is a useful and important consideration for the research design. Policy Implications One policy implication is the importance of training. The findings suggest that officers’ job performance regarding both the issuance of tickets and service provisions was affected by officers’ training experience. Many scholars in the United States have considered training as an important tool in the process of facilitating change within police organizations. Yet, one finding of interest is that training for arrest did not indicate a significant influence on officers’ job performance. That is, only training concerning order 61 maintenance increased service provisions, while training concerning law enforcement duties did not demonstrate a significant influence on any job performance. On the basis of the findings, for more effective training, police managers are suggested to focus on order maintenance training rather than law enforcement based training. Yet, as these findings came from the responses to two questions asking whether or not officers experienced training and supervisors’ special directives, it is unclear what factors of training affect officers’ job performance. Therefore, additional research is needed to identify both factors and mechanisms. 62 Conclusion This research investigated three relationships between: 1) officers’ attitudes and their job performance, 2) organizational factors and officers’ job performance, and 3) situational factors and job performance. Following Wilson’s (1968) typology of police, officers’ job performance and their attitudes toward their role were categorized into three areas: law enforcement duties, order maintenance, and service provisions. The findings of this study do not provide any evidence for attitudinal influence on job performance. Therefore, the first hypotheses were not supported in regard to law enforcement, order maintenance, and service provisions. However, there is some evidence for organizational factors’ influence on job performance with respect to order maintenance and service provisions. Yet, the findings do not provide support for the influence of organizational factors on job performance. Moreover, the second hypothesis was partially supported with regard to order maintenance and service provision activities. These results do provide support for situational factors’ influence on job performance. However, this was found in regard to law enforcement duties. The third hypothesis was also partially supported for law enforcement. The above results should be approached with caution, since statistical significance was reported only in light of specific areas of job performance. The findings only suggest that many factors exist which influence officers’ job performance although unexplained. In addition, the R-squared values in the multivariate models were quite low. This suggests that there are other important variables that affect officers’ job performance. Therefore, the hypotheses tests and results reveal ambiguity in job performance for the influence of attitudinal, organizational, and situational factors. 63 The nature of ambiguity involved in the existence and explanation of the relationships is meaningfully compared with the result of Engel and Worden’s (2003) research on attitudes, behavior, and supervisors’ influence. They primarily did not find a relationship between attitudes and behavior, and supervisors’ priorities modestly, but negatively affect officers’ job performance. The authors attributed this failure of finding significance for the attitude-behavior relationship to the variation in how officers perform their jobs as being not congruent with their occupational attitudes. They also stated that situational pressures or factors inherent to the police organization may have also accounted for their findings. In this study, officers’ job performance, similar to findings in the United States, was scarcely affected by officers’ attitudes. Therefore, Engel and Worden’s assumption can be applied to Korea in regard to this relationship. In conclusion, these issues raise important questions for further direction of future research. In some areas of social psychology, conducting research with newly developed models and measurements has provided support for the attitude-behavior relationship. Therefore, if future researchers develop more sophisticated models, they may find different results. While the present study found some organizational or situational factors’ influence on job performance, these factors may or may not directly affect job performance, or may also have an interactive effect. 64 APPENDICES 65 Appendix A. Factor Analyses of Independent Variables Variables Survey Items (range of each item = 1-4) :23: 82:56:11” Law 12. The primary function of police is to handle crime related 64 Enforcement problems. - ' 1 24 19. Enforcing the law in a strict way is a patrol officer’s most 86 ' important responsibility. ' Order 21. How often do you think that patrol officers should respond 83 Maintenance to neighbor disputes? ' 22. How often do you think that patrol officers should respond 85 to domestic disputes? ' 1.96 23. How often do you think that patrol officers should respond to nuisance businesses that may cause problems for community .71 residents (e.g. abandoned cars and houses)? Service 14. The primary function of police is to provide community 80 Provision service. ' 16. The primary purpose of police is to serve the citizens. .85 2.01 17. The primary function of police is to safeguard citizens’ 73 rights ' Motivation 45. Making many arrests will help you to achieve your 89 Factors promotion. ' (Arrest) 46. Making many arrests helps you receive a good evaluation .93 2.37 47. Making many arrests helps to gain recognition for your good 85 work from your supervisor ' Frustration 48. Making few arrests leads to consequences from your 75 Factors supervisor ' (Arrest) 49. Making few arrests leads to a transfer to an unexpected 87 1 99 workplace or department ' ' 50. Making few arrests leads to undermining your competency 83 among fellow police officers ' Motivation 55. Issuing many police tickets helps you in achieving your 83 Factors promotion. ' (Police 56. Issuing many police tickets helps you in receiving a good 91 2 12 Ticket) evaluation ' ' 57. Issuing many police tickets helps you in receiving 74 recognition from your supervisor ' Frustration 58. Issuing few police tickets leads to consequences from your 72 Factors supervisor ' (Police 59. Issuing few police tickets leads a transfer to an unexpected 81 1 90 Ticket) workplace or department ' ' 60. Issuing few police tickets leads to undermining my 81 competency amogg fellow police officers/officials. 66 Appendix B. Questionnaires SOUTH KOREAN OFFICERS’ JOB ATTITUDES AND JOB PERFORMANCE Dear Colleagues, This scientific research study is intended to identify relationships between officers’ attitudes and job performance. That is, the questionnaire is designed to assess the effects of your perceptions of police work, supervisors, and the organization upon job performance. Please fill out the questionnaire as completely as possible. The survey administrator does not know the names of the respondents who choose to participate in the study nor do the surveys have any identification marks. All responses will remain confidential to the maximum extent allowable by law and will not be used in anyway that may identify the participant. Please remember that your participation in this study is completely voluntary, and you may refuse to answer any questions or stop participation at any time without penalty. You indicate your voluntary agreement to participate by completing and returning this questionnaire. This survey should take approximately 20 minutes to complete. Should you have any questions regarding this research, please feel free to contact us. If you have questions or concerns regarding your right as a study participant, or are dissatisfied at any time with any aspect of this study, you may contact - anonymously, if you wish - Peter Vasilenko, Chair of the University Committee on Research Involving Human Subjects by email: ucrihs@msu.edu, or regular mail: 202 Olds Hall, East Lansing, IVII 48824, USA. Many thanks and much appreciation for your time and cooperation. Jongbo Kim Email: l_