Prepare an Analysis Report using data and information from studies on leadership styles and characteristics derived from empirical research, and explain how the information applies to successful leade

styles and effectiveness Mary B. Sarver Department of Sociology, Criminal Justice, and Women’s Studies, University of South Carolina Upstate, Spartanburg, South Carolina, USA, and Holly Miller College of Criminal Justice, Sam Houston State University, Huntsville, Texas, USA Abstract Purpose– The purpose of this paper is to examine the leadership styles of police chiefs and how these styles related to demographic, personality, and effectiveness.

Design/methodology/approach– Participants included 161 police chiefs in Texas who completed the Multifactor Leadership Questionnaire (MLQ 5X-Short) leader form, the NEO Five-Factor Inventory (NEO-FFI), and a background characteristics form.

Findings– Results indicate that the police chiefs were fairly evenly classified across leadership styles with the Transformational leaders rated as most effective. Transformational leaders are characterized as confident, energetic, and open-minded. Although few of the demographic variables predicted leadership styles, several of the personality characteristics were significant predictors.

Originality/value– Few previous studies have reported the relationship between police leadership style, personality, and effectiveness. This study adds to the body of knowledge regarding the relationship between these variables by specifically targeting police chiefs.

KeywordsLeadership, Effectiveness, Transformational, Passive/avoidant, Police chief, Transactional Paper typeResearch paper Introduction Traditionally, the predominant leadership style of law enforcement leaders has been an authoritarian style, where leaders do not include subordinates in the decision-making process (Beito, 1999). However, within recent years, there has been a transition toward more of a democratic and mutual/shared relationship style, also referred to as Transformational, where subordinates are encouraged to make decisions and set their own goals (Avolio and Bass, 2004; Bass and Avolio, 1997; Beito, 1999; Levasseur, 2005).

It is argued that these changes occurred because of the widespread adoption of community policing (Denhardt, 1993), because subordinates are no longer responding to an authoritarian style of leadership, and that the democratic and mutual/shared style is preferred and more effective (Andreescu and Vito, 2010; Krimmel and Lindenmuth, 2001; Silvestri, 2007).

According to the literature, police leader selection has occurred through many avenues, including cognitive testing, but identification and selection of good leaders is still a problem (Green, 2006). It is important to understand and determine the factors that have an effect on or predict leadership style in order to prepare leaders for the future more effectively, to teach them how to keep followers motivated to meet both organizational and community needs and goals, and to keep up with an ever changing environment. This understanding could aid in identifying, selecting, promoting, and training effective police leaders. It will also help researchers gain a better The current issue and full text archive of this journal is available at www.emeraldinsight.com/1363-951X.htm Received 16 March 2013 Revised 30 May 2013 31 May 2013 Accepted 31 May 2013 Policing: An International Journal of Police Strategies & Management Vol. 37 No. 1, 2014 pp. 126-143 rEmerald Group Publishing Limited 1363-951X DOI 10.1108/PIJPSM-03-2013-0028 126 PIJPSM 37,1 understanding of leadership styles in law enforcement, specifically those of police chiefs, and which characteristics chiefs possess that enable them to be effective leaders.

Literature review Leadership style definitions Research regarding leadership styles and effectiveness has revealed conflicting findings (Densten, 2003; Green, 2006; Morreale, 2003; Rasor, 1995), and indicates that police leaders utilize various styles, with no general consensus regarding which style is predominant (Beito, 1999; Caless, 2011; Engel, 2001; Kapla, 2005; Krimmel and Lindenmuth, 2001; Legault, 2005; Villarreal-Watkins, 2001), or what police leadership is exactly, how it should be measured, and if it simply consists of generic leadership styles seen in management positions outside of policing (Adlam and Villiers, 2003; Caless, 2011).

Caless (2011) examined 12 competencies for chief officers set forth by the Association of Chief Police Officers (ACPO) and argued that a large number of those criteria are not police specific, but are generic skills utilized in management positions outside of policing.

Many of these generic skills were also found by Goleman (2004) when he reviewed the competencies of large company leaders around the world. Generic skills needed include effective communication, working well with others, developing policy, creating a vision of the future, being open to change, and motivating others.

Through a review of the relevant literature, four overall leadership styles emerged:

transformational/emotionally intelligent, transactional/charismatic, authoritative, and passive/avoidant (Laissez-Faire). According to Avolio and Bass (2004), transformational leaders utilize proactive and innovative approaches to make effective changes within an organization. These leaders also influence their subordinates to make changes within themselves. Transformational leaders help workers create and achieve higher goals, perform above the standards, and discover what is important so they may maximize their potential (Antonakiset al., 2003; Avolio and Bass, 2004; Caless, 2011). Goleman (2004) asserted that an emotionally intelligent leader exists, similar to the transformational leader, in which intelligence goes beyond cognitive ability. Emotionally intelligent leaders should exude confidence, be committed to the organization, trustworthy, open to change, persuasive, and optimistic, and able to lead change effectively.

Antonakiset al.(2003) indicated that “Transactional leadership is an exchange process based on the fulfillment of contractual obligations and is typically represented as setting objectives and monitoring and controlling outcomes” (p. 265). Transactional leaders use a system of rewards and punishments as motivation, and do not attempt to implement change within the organization or their subordinates. Transactional leaders also do not influence their subordinates to go above and beyond the performance standards, but only expect workers to achieve goals that have already been established (Avolio and Bass, 2004; Bass, 1985, 1990; Caless, 2011; Levasseur, 2005). Caless (2011) identifies an extension of the transactional leader, known as the charismatic leader.

This individual can be characterized as having a fiery personality, being a calculated risk taker, concerned more with results than means, and able to motivate followers by examples and achievements (Caless, 2011; Goleman, 2004).

Authoritative leaders exhibit a more extreme type of transactional leadership.

Subordinates are not involved in the decision-making process, there is very little to no interaction between the leaders and workers, and fear from the threat of punishment is the chief motivator. These leaders dictate tasks to be completed without, at times, 127 Police chief leadership identifying the larger goals, and have little concern for the aspirations or needs of workers (Beito, 1999; Bruns and Shuman, 1988).

According to Avolio and Bass (2004), passive/avoidant leaders do not want to be involved in decision making, and avoid utilizing their authority or taking responsibility. Additionally, these leaders do not take action until a problem has already occurred and become serious (Antonakiset al., 2003; Avolio and Bass, 2004).

Leadership styles within law enforcement Studies have shown that police leadership styles are situational, in which they change as conditions change (Girodo, 1998), and that training can help supervisors identify the appropriate way to lead in various situations (Williams, 1993). It has also been found that styles vary between departments among leaders (Engel, 2001) and rank (Caless, 2011), and that either no dominant style may be exhibited (Kuykendall and Unsinger, 1982), or a mixed style can exist, where a combination of styles are displayed (Caless, 2011; Kapla, 2005).

Other studies, however, indicate that one particular style can be exhibited most often in a police organization, but those studies do not demonstrate that one style is predominant in law enforcement as a whole. Some researchers have found that the authoritative leadership style is most used (Bruns and Shuman, 1988) and most favored (Engel, 2001; Legault, 2005), particularly among police managers in administration (Girodo, 1998), while others have found that the transformational leadership style is exhibited most often and most favored within law enforcement (Berringer, 2005; Kapla, 2005; Morreale, 2003; Singer and Singer, 1990). Additionally, Rasor (1995) asserted that the leaders in their study represented more of a transactional leadership style, yet Singer and Singer (1990) found that it was the least used and least favored. Morreale (2003) found, on the other hand, that the passive/avoidant style was the least used. Silvestri (2007) found that even though women police managers favor the transformational style and believe it to be more effective, the transactional style is still being used most often because of the difficulty of implementing a transformational leadership style in an organizational culture which puts so much emphasis on a rank structure.

Leadership outcomes, personality, and background characteristics Many studies have found that a transformational-type leadership style based on mutual participation can be more effective and have a significantly positive effect on organizational commitment, compliance, satisfaction, communication, influence, perceived fairness and extra effort (Adebayo, 2005; Beito, 1999; Morreale, 2003; Schwarzwaldet al., 2001; Singer and Singer, 1990). Additionally, Goleman (2004) indicated that a great amount of emotional intelligence would exist within an effective leader. Densten (1999), on the other hand, found that transformational leadership ratings were significantly lower than the norm, indicating that senior officers exerted a significantly lower amount of effort, believed their leaders to be significantly less effective than the norm, and significantly more satisfied (Densten, 1999) than the established norms of Bass and Avolio (1997). Densten (1999) also found higher levels of subordinate satisfaction with transactional leadership.

Kuykendall and Unsinger (1982) found that an authoritative-type style was more effective than a transformational-type, while Welsh (1988) indicated that effectiveness depended on the approach used toward subordinates. For instance, authoritative-type leaders were more effective when tasks were structured and had established goals, while the transformational-type leaders were more effective when orientation and personal contact were concerned. Additionally, researchers have found that the passive/avoidant 128 PIJPSM 37,1 leader is less effective than transformational and transactional leaders and has negative effects on the outcome measures of extra effort, effectiveness, and satisfaction (Antonakiset al., 2003; Avolio and Bass, 2004; Morreale, 2003). Results from Densten’s (2003) study demonstrated that for executives and superintendents, effectiveness and extra effort were negatively and positively predicted by Laissez-Faire, respectively.

These findings indicated that those leaders who did not delay decisions or avoid responsibility were viewed as being more effective, and that extra effort was expended when leaders were absent. Still others have found that either no significant differences exist between styles or no significant relationship exists with variables such as commitment, job satisfaction, and effectiveness (Berringer, 2005).

Additional variables such as age, gender, race, assignment, years of law enforcement experience, years in current position, education, rank, and agency type have also been examined to determine if a relationship exists with leadership styles and outcomes.

Berringer (2005) indicated that the transformational style was most favored, but no significant differences existed in regard to age or gender between a group of police field trainers who had attended a supervision training course and a group of police field trainers who had not completed the course. Girodo (1998), on the other hand, found that the oldest age group in their study utilized the transformational leadership style significantly more often than the youngest age group and that managers with more years of law enforcement experience utilized transformational leadership more often than those with fewer years of experience. Densten (2003) found that the number of years in the current position was not a significant predictor of either effectiveness or extra effort.

His findings, however, indicated that distinctive sets of leadership behaviors were held by senior officers at different ranks that increased the willingness of subordinates to exert extra effort and believe that their leaders were effective. These findings meant that rank did indeed matter. Villarreal-Watkins (2001) found that leadership was significantly positively correlated with rank and educational experience, which indicated that significantly greater differences existed between the ranks than within. Villarreal-Watkins (2001) stated that these findings could be due to the assertion that leadership can be learned, therefore, she proposed that police agencies reevaluate their training divisions and determine if an educational requirement should be established. Additionally, Andreescu and Vito (2010) found that, overall, police managers tended to prefer the transformational style and, more specifically, women, African Americans, and those given specific assignments, such as detective or special operations, preferred it as well.

They also found that those with fewer years in the current department were more likely to support a transactional style of leadership.

In regard to leadership and personality characteristics, Welsh (1988) found that an authoritative-type leader was the highest rated style by police sergeants, and scored higher on the personality attributes of aggression, dominance, heterosexuality, intraception, and achievement. Rasor (1995) also found that leadership styles were significantly predicted by personality in his study. Results demonstrated that the most exhibited Myers-Briggs Type Indicator (MBTI) personality attributes of the leaders was a combination of introversion, sensing, thinking, and judging. No significant differences were found between the personality types and agency type or rank.

Schafer (2009, 2010a, b) examined perceptions of traits which characterize both the effective and ineffective leader. Results indicated that ineffective leaders exhibited behaviors such as poor communication skills, were inconsistent in decision making, micromanaged, and were not open to others’ suggestions. Effective leaders, on the other hand, were characterized as being fair and honest, competent, open-minded, 129 Police chief leadership responsible, flexible, and having good decision-making and communication skills.

Schafer (2009) also found that leadership can be learned and enhanced by seeing other effective leaders in action, through education and training provided before and after promotion, and by mentoring new leaders as they apply what they have learned, allowing for an opportunity to not only gain experience, but also to get feedback.

Research has also found that training can help improve leadership skills by assessing their strengths and weaknesses prior to training through the California Psychological Inventory-260 (CPI-260) (Milleret al., 2009).

Current study Studies specifically examining police chief leadership have demonstrated varying styles (Kapla, 2005; Krimmel and Lindenmuth, 2001; Stamper, 1992). The traditional leadership style approach of a police chief has generally been seen as authoritative and militaristic, however, a more open-minded group of chiefs have changed their leadership styles dramatically (Denhardt, 1993). More recent research has found that the transformational leadership style was one of the most favored styles for police chiefs (Kapla, 2005; Morreale, 2003; Singer and Singer, 1990).

One of the most recent and important studies to note is Caless’s (2011) extension of Reiner’s (1991) work regarding the leadership style typologies of bobbies, bosses, and bureaucrats. Caless (2011) asked chief officers if they identified with one of Reiner’s (1991) typologies, or if they considered their leadership style to be something else, such as transactional, transformational, or mixed. Interviews indicated that many chief officers discard Reiner’s (1991) typologies and a majority of them use a mixture of styles, based on what may be fitting for the situation.

The leadership style of police chiefs can also be influenced by his/her attributes (Kapla, 2005). Research findings have indicated that educational level and whether or not chiefs were hired from inside or outside of the department were significantly correlated with leadership style. Thesestudies specifically found that the transformational leaders had higher levels of education, that chiefs with leadership and performance issues were more likely to have a high school degree or less, and that chiefs who were hired from within the department with more education scored higher on the performance and leadership indicators (Kapla, 2005; Krimmel and Lindenmuth, 2001). Additionally, Caless (2011) indicated that a large majority of chief officers have a college degree or postgraduate degree in the UK.

A review of the literature in law enforcement has revealed little in regard to leadership style and its relationship with leadership outcomes, effectiveness, personality, and various individual characteristics. A vast majority of the literature has not statistically analyzed participant characteristics with leadership style and outcomes or personality. They have only described those characteristics in terms of frequencies, percentages, means, and standard deviations (Berringer, 2005; Densten, 1999; Schwarzwaldet al., 2001). Results of studies on self-perceptions like these could offer great insight into what factors affect, predict, or have a relationship with leadership styles and outcomes, especially effectiveness. Results could also aid in the selection and promotion of more effective police leaders, and in the improvement upon training programs that will help police leaders strengthen those characteristics of the most effective leaders.

The present study attempts to expand the knowledge of leadership styles in law enforcement and minimize the gap that exists within the literature by examining the correlates and predictors of leadership styles. This study attempts to determine if there 130 PIJPSM 37,1 are relationships among leadership styles, effectiveness, personality attributes, and various background characteristics of police leaders. The focus of this study is on police chiefs, as they are at the top of the hierarchy, hold a great deal of discretion, and have the power to influence subordinates to meet organizational and community needs and demands (Rainguet and Dodge, 2001).

Hypotheses . The transformational leadership style will be most prevalent among Texas police chiefs (Kapla, 2005; Morreale, 2003; Singer and Singer, 1990). . Transformational leaders will have higher mean scores on all Multifactor Leadership Questionnaire (MLQ) measures of extra effort, satisfaction, and effectiveness, and a summed measure of job satisfaction than transactional and passive/avoidant leaders, and passive/avoidant leaders will have higher mean scores for the number of formal and informal complaints received than transformational and transactional leaders (Morreale, 2003; Singer and Singer, 1990). . Length of time in law enforcement will be a significant positive predictor of the transformational leadership style (Girodo, 1998), the older police chiefs will utilize the transformational leadership style significantly more often than younger chiefs (Girodo, 1998), the type of jurisdiction characterized as suburban/ urban will be a positive significant predictor of the transactional leadership style (Kapla, 2005), and education will be a positive significant predictor of leadership style (Krimmel and Lindenmuth, 2001; Villarreal-Watkins, 2001). . Transformational leaders will have higher mean scores on extraversion and conscientiousness than transactional and passive/avoidant leaders, and passive/ avoidant leaders will have higher mean scores on neuroticism than transformational and transactional leaders (Rasor, 1995; Welsh, 1988).

Method Participants The participants in this study consisted of Texas police chiefs who were enrolled in the New Chiefs Development Program (NCDP) and the Texas Police Chief Leadership Series (TPCLS) program offered through the Bill Blackwood Law Enforcement Management Institute of Texas (LEMIT) at Sam Houston State University (SHSU) in Huntsville, Texas. At the time of this study, the population of acting Texas police chiefs was 1,016. The sample size of the study is 161, which was approximately 16 percent of the total population. In order to determine if the sample was representative of the population of Texas police chiefs, information for age, gender, and race were requested from LEMIT for all acting police chiefs.

In regard to the entire sample (see Table I), the average age of the participants was approximately 49 years (SD¼8.68), and an overwhelming majority of the entire sample (87 percent) was white. Of those who reported their gender, male participants accounted for almost 85 percent of the sample (n¼136). An examination of Table I also demonstrates that, in regards to age, gender, and race, the police chiefs included in the sample for the current study are similar to the population. The age and race statistics of the current study are comparable to both Krimmel and Lindenmuth (2001) and Stamper’s (1992) studies, where the average age for the four leadership typologies was between 48 and 53, and 90 percent of the sample was white. 131 Police chief leadership In order to determine whether the sample was statistically representative of the population,t-test andw 2tests were conducted. No significant differences were found between the sample and the population for age, gender, or race, which indicates that the sample is representative of the population in terms of those variables (Agresti and Finlay, 2008).

Measures MLQ (Avolio and Bass, 2004). Participants in the study completed the 45-item leader form (5X-Short) of the MLQ and were asked to indicate, on a scale from 0 (not at all) to 4 (frequently, if not always), how often they exhibit specific behaviors.

The transformational, transactional, and passive/avoidant leadership styles were measured through nine factors, and each factor was measured through four items, for a total of 36 items. Additionally, the MLQ measures the following three leadership outcomes through nine items: the amount of extra effort subordinates are willing to exert, the effectiveness of the supervisor, and subordinate satisfaction with their supervisor.

The transformational leadership style is measured by the five factors of idealized attributes, idealized behaviors, inspirational motivation, intellectual stimulation, and individual consideration. The transactional leadership style is measured by two scales, Contingent Reward and Management-by-Exception (active). Finally, the passive/ avoidant leadership style is assessed by the two scales of Management-by-Exception (passive) and Laissez-Faire (Antonakiset al., 2003; Avolio and Bass, 2004). Avolio and Bass (2004) reported that Cronbach’sareliability scores for the nine-factor model of the MLQ (self) are between 0.60 and 0.76.

To determine the leadership style of each police chief, MLQ items were first scored based on the directions provided within the scoring form. Second, initial Cronbach’s ascores were calculated for each of the nine factors and the items measuring each of the three leadership styles. Only two of the nineascores for the factors and oneascore for leadership style would have been acceptable at the 0.60 level (Hairet al., 2006).

Therefore, the scores for each item were transformed intoz-scores and factor analysis was conducted to determine which items would load highly together.

After examining the output for the factor analysis, it was determined that only 25 of the 36 leadership items would be utilized to establish the dominant style of each participant. Seventeen items were used for transformational leadership, two for transactional, and six for passive/avoidant. The decision to utilize only 25 items was Sample Population Characteristic (n¼161) (n¼1,016) Age: Mean (SD) 49.03 (8.68) 48.91 (8.76) Gender:n (percent) Male 136 (84.5) 944 (92.91) Female 9 (5.6) 37 (3.64) Race White 140 (87.0) 791 (77.85) Black/African American 5 (3.1) 40 (3.94) Hispanic/Latino/a 15 (9.3) 120 (11.81) Native American 1 (0.6) 8 (0.79) Note:Due to missing data some percentages will not add up to 100 percent Table I.

Comparison of age, gender, and race for the entire sample and the population of Texas police chiefs 132 PIJPSM 37,1 based on the loading of the items within the factor components and the Cronbach’s ascores of the new scales that were subsequently created. Six components were retained after examination of the eigenvalues, variance, and residuals. Factor loadings ranged from 0.469 to 0.781, and were all positive. The items utilized for each style were added and divided by the number of items in that category (leadership style) to calculate the style score. Cronbach’sawas again calculated for each style and it was found that that transformational, transactional, and passive/avoidant scales had reliability scores of 0.86, 0.64, and 0.66, respectively, which were acceptable.

NEO Five-Factor Inventory (NEO-FFI; Costa and McCrae, 1992). The NEO-FFI was developed by Costa and McCrae (1992), and is a 60-item measure of the personality domains of neuroticism, extraversion, openness, agreeableness, and conscientiousness.

The NEO-FFI is a shortened version of the NEO Personality Inventory Revised (NEO PI-R). The NEO-FFI asks respondents to indicate on a five-point scale the degree to which they agree or disagree with each statement. The 12 items of each domain also created six specific facets, however, for this study, only the five domains were utilized in the data analysis. Costa and McCrae (1992) reported that Cronbach’sareliability scores for the five domains of the NEO-FFI range from 0.68 to 0.86.

The background characteristics form. Finally, participants completed a background characteristics form. Items in the form included age, gender, race, education, number of formal and informal complaints, number of sworn and civilian employees, promotion to chief from within and outside of the department, type of jurisdiction served, and the number of years in law enforcement, current police department, as supervisor, and as chief. Additionally, three satisfaction items were included from Dantzker’s (1994a, b) research. This information was employed to determine if there were any significant predictors of Texas police chief leadership styles. In addition, each of Dantzker’s (1994a, b) satisfaction variables (overall job satisfaction, changing police departments, and a job outside of policing) were assessed through one item called “job satisfaction.” The variable job satisfaction was created by simply summing the scores of the three satisfaction variables for each participant.

Results Descriptive statistics Sample characteristics are presented and compared across the leadership style categories (transformational, transactional, and passive/avoidant) in Table II.

Although males were fairly evenly split across leadership styles, the female participants were more likely to be a transformational leader. In regard to promotion, 84 (52.2 percent) participants had been promoted to the rank of chief from within the department, compared to 76 (47.2 percent) that were promoted from outside of the department. A slight majority of the transformational leaders were promoted from outside the department. The promotion statistics are comparable to Kapla’s (2005) study, where 54.7 percent of the participants were promoted to chief from within the department. It was also revealed that a majority of chiefs in the entire sample (n¼58) and in each of the three leadership style groups (n¼22,n¼19,n¼17) served in a jurisdiction characterized as rural.

Examining the number of formal and informal complaints, the average number of self-reported formal complaints the chiefs had received as a supervisor from both civilians and fellow officers for the entire sample was approximately six (SD¼28.77), while the average number of informal complaints was much higher at approximately 21 complaints (SD¼106.2). When broken down into leadership style categories, it 133 Police chief leadership appears that transformational leaders received fewer formal and informal complaints than both the transactional and passive/avoidant styles. The average number of civilian and sworn employees for the sample was approximately 11 (SD¼22.21) and 26 (SD¼63.0), respectively. An examination of these characteristics among the leadership style categories demonstrated that the transformational leaders managed more sworn (n¼43.48, SD¼96.28) and civilian (n¼15.72, SD¼30.82) employees than both transactional (n¼11.82, SD¼13.79;n¼8.87, SD¼18.57) and passive/avoidant (n¼22.80, SD¼44.51;n¼7.67, SD¼13.12) leaders.

Examining law enforcement experience, the entire sample of police chiefs had approximately 24 years of experience in law enforcement and had acted as a chief in their current department for an average of 5.5 years (SD¼5.8). This finding is consistent with Rainguet and Dodge (2001) study on police chief tenure where they found that chiefs served an average of 5.1 years. Additionally, the chiefs had at least 22 years of law enforcement experience across the leadership style groups.

Leadership style prevalence, satisfaction, and effectiveness Although it was hypothesized that the transformational leadership style would be most prevalent among this sample of police chiefs (Kapla, 2005; Morreale, 2003; Singer and Singer, 1990), results indicated an equal distribution of styles; of the 161 participants, 57 were categorized as transactional leaders (35.4 percent), followed by 54 transformational leaders (33.5 percent), and 50 passive/avoidant leaders (31.1 percent).

A MANOVA was conducted to examine leadership styles across the MLQ measures of extra effort, satisfaction, effectiveness and number of formal and informal complaints self-reported by the police chief (see Table III). MANOVA was the most appropriate analytical technique to use because it determines if statistical differences Total Transformational Transactional Passive/avoidant Variable (N¼161) (n¼54) (n¼57) (n¼50) Gender Male 136 (84.5) 41 (75.9) 49 (85.9) 46 (92.0) Female 9 (5.6) 6 (11.1) 1 (1.8) 2 (4.0) Selected to Chief Inside of dept. 84 (52.2) 30 (55.6) 29 (50.9) 25 (50.0) Outside of dept. 76 (47.2) 24 (44.4) 27 (47.4) 25 (50.0) Jurisdiction Urban 50 (31.1) 20 (37.0) 14 (24.6) 16 (32.0) Suburban 33 (20.5) 11 (20.4) 14 (24.6) 8 (16.0) Rural 58 (36.0) 22 (40.7) 19 (33.3) 17 (34.0) Other 6 (3.7) 1 (1.8) 3 (5.3) 2 (4.0) Complaints Formal 6 (28.8) 0.82 (1.5) 13.7 (47.8) 2.29 (7.3) Informal 21 (106.2) 1.94 (2.7) 50.63 (175.8) 8.29 (29.63) Years in Law Enforcement 23.78 (9.33) 24.15 (10.18) 24.3 (9.4) 22.80 (8.36) Years as a Supervisor 14.30 (8.54) 16.15 (9.09) 13.27 (8.48) 13.47 (7.82) Years in Current Police Department 9.47 (9.17) 10.0 (10.07) 6.75 (6.96) 11.98 (9.7) Years as Chief in Current Department 5.46 (5.8) 5.19 (6.05) 4.38 (4.82) 7.01 (6.30) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Note:Due to missing data some percentages will not add up to 100 percent Table II.

Sample characteristics 134 PIJPSM 37,1 exist in means between the three leadership styles on the variables identified above.

Results demonstrated statistically significant differences between extra effort, effectiveness, and satisfaction on the combined dependent variable:F(6, 314)¼4.132, p¼0.001, Pillai’s trace¼0.146,Z 2¼0.121. When the results for the dependent variables were considered separately, it was found that several differences reached statistical significance indicating that leadership style significantly affected extra effort (F(2, 158)¼7.919,p¼0.001,Z 2¼0.091), effectiveness (F(2, 158)¼10.616, po0.001,Z 2¼0.118), and satisfaction (F(2, 158)¼6.056,p¼0.003,Z 2¼0.071).

An inspection of the mean scores indicated that those with high scores of extra effort (M¼9.65, SD¼1.65), effectiveness (M¼13.48, SD¼1.63), and satisfaction (M¼6.78, SD¼0.84) were more likely a transformational leader than a passive avoidant leader with low scores of extra effort (M¼8.42, SD¼1.43), effectiveness (11.86, SD¼2.10), and satisfaction (M¼6.16, SD¼1.02). The findings also demonstrated that transactional leaders had higher mean scores on extra effort (M¼9.19, SD¼1.65) and effectiveness (M¼12.91, SD¼1.70) than passive/avoidant leaders.

MANOVA analyses also examined leadership style across formal and informal complaints received by the police chiefs (see Table III). There were statistically significant differences between the number of formal and informal complaints received on the combined dependent variable:F(4, 316)¼2.942,p¼0.021, Pillai’s trace¼0.072, Z 2¼0.036. When the results for the dependent variables were considered separately, the only difference to reach statistical significance was between transformational and transactional leaders for the number of formal complaints received (F(2, 158)¼4.314, p¼0.015,Z 2¼0.052). Because the number of formal complaints received was transformed using the inverse method, the results must be interpreted in reverse.

The mean scores demonstrated that those with the lowest mean of formal complaints (M¼0.57, SD¼0.35) were more likely a transactional leader than a transformational leader with a higher mean (M¼0.75, SD¼0.31). This finding indicates that transactional leaders have more formal complaints than transformational leaders.

These findings are important because they further support what has been found in previous studies regarding the MLQ (Adebayo, 2005; Densten, 2003; Morreale, 2003; Singer and Singer, 1990) and they support the assumption in the current study that transformational leaders are the most effective when effectiveness is defined by the MLQ measures of extra effort, effectiveness, and satisfaction, and the number of formal and informal complaints received.

Leadership style, personality, and predictors A MANOVA was conducted to determine if there were significant differences for personality between leadership styles (see Table IV). There were statistically SSdf MS F pZ 2 Extra Effort 39.83 2 19.92 7.919 0.001 0.091 Effectiveness 69.89 2 34.94 10.616o0.001 0.118 Satisfaction 10.24 2 5.12 6.056o0.001 0.071 Number of formal Complaints 0.955 2 0.477 4.314 0.015 0.052 Number of informal Complaints 0.760 2 0.380 2.906 0.058 0.035 Job satisfaction 8.30 2 4.15 0.933 0.396 0.012 Table III.

Summary table of MANOVA for leadership style prevalence, satisfaction, and effectiveness 135 Police chief leadership significant differences between neuroticism, extraversion, openness, agreeableness, and conscientiousness on the combined dependent variable:F(10, 310)¼3.861,po0.001, Pillai’s Trace¼0.222,Z 2¼0.111. When the results for the dependent variables were considered separately, it was found that several differences reached statistical significance.

Results indicated that leadership style significantly affected neuroticism (F(2, 158)¼4.494, p¼0.013,Z 2¼0.054), extraversion (F(2, 158)¼12.139,po0.001,Z 2¼0.133), openness (F(2, 158)¼6.330,p¼0.002,Z 2¼0.074), and conscientiousness(F(2, 158)¼8.30,po0.001, Z 2¼0.095). An inspection of the mean scores indicated that those with high scores of extraversion (M¼34.43, SD¼4.67), openness (M¼26.09, SD¼4.73), and conscientiousness (M¼37.11, SD¼5.50) were more likely a transformational leader than either a transactional leaderwith lower scores of extraversion (M¼32.11, SD¼5.12), or a passive/avoidant leader with low scores of extraversion (M¼29.62, SD¼5.12), openness (M¼22.12, SD¼6.86), and conscientiousness (M¼33.26, SD¼5.34). The mean scores demonstrated that those with high scores of neuroticism (M¼15.90, SD¼6.94) were more likely a passive/avoidant leader than a transformational leader with low scores of neuroticism (M¼11.91, SD¼6.79). The means scores also indicated that transactional leaders had higher mean scores on extraversion (M¼32.11, SD¼5.12) and conscientiousness (M¼36.84, SD¼5.25) than passive/avoidant leaders.

Binary logistic regression was conducted to examine whether experience, agency size, or background characteristics were significant predictors of leadership style (see Table V). Results of the transformational binary logistic regression model, as provided in Table V, indicated that the overall model adequately fit the data ( 2 log likelihood¼191.903,w 2(8)¼8.438,p¼0.392). Results also demonstrated that the model accounted for only 11.2 percent of the variance in the dependent variable and correctly classified 68.3 percent of the cases. Wald statistics indicated that length of time as a supervisor was a significant negative predictor of transformational leadership. Odds ratios for length of time as a supervisor (e B¼0.994) showed very little change in the likelihood of being classified as a transformational leader when compared to transactional and passive/avoidant leaders. This finding meant that those with experience as a supervisor were slightly less likely (e B¼0.994) to be classified as a transformational leader than a transactional or passive/avoidant leader.

Results of the transactional binary logistic regression model, as provided in Table V, indicated that the overall model adequately fit the data ( 2 log likelihood¼193.008, w 2(8)¼11.193,p¼0.191). Results also demonstrated that the model accounted for 13.2 percent of the variance in the dependent variable and correctly classified 68.9 percent of the cases. Wald statistics indicated that the number of sworn employees was a significant predictor of transactional leadership. Odds ratios for the number of sworn employees (e B¼3.745) showed great change in the likelihood of being classified as a transactional leader. This finding meant that those with sworn employees were almost four times as likely (e B¼3.745) to be classified as a transactional leader.

SS df MS F pZ 2 Neuroticism 418.78 2 209.39 4.494 0.013 0.054 Extraversion 599.64 2 299.82 12.139o0.001 0.133 Openness 413.63 2 206.81 6.330 0.002 0.074 Conscientiousness 477.24 2 238.62 8.30o0.001 0.095 Agreeableness 158.10 2 79.05 2.617 0.076 0.032 Table IV.

Summary table of MANOVA results for leadership style and personality 136 PIJPSM 37,1 Results of the Passive/avoidant binary logistic regression model, as provided in Table V, indicated that the overall model adequately fit the data ( 2log likelihood¼185.361,w 2(8)¼14.548,p¼0.069). However, none of the independent variables made a significant difference in the likelihood of being classified as a passive/ avoidant leader when compared to transformational and transactional leaders. In addition, it was found that the model accounted for 11.8 percent of the variance in the dependent variable and correctly classified 70.2 percent of the cases.

In order to examine whether any of the background characteristics of the police chiefs were predictors of leadership style, binary logistic regressions were computed with each leadership style as a dependent variable. Results of the transformational binary logistic regression model, as provided in Table VI, indicated that the overall model adequately fit the data ( 2 log likelihood¼190.527,w 2(8)¼6.371,p¼0.606). However, none of the independent variables made a significant difference in the likelihood of being classified as a transformational leader when compared to transactional and passive/avoidant leaders. In addition, it was found that the model accounted for 7.5 percent of the variance in the dependent variable and correctly classified 66.5 percent of the cases.

Results of the transactional binary logistic regression model, as provided in Table VI, indicated that the overall model adequately fit the data ( 2 log likelihood¼192.584, w 2(8)¼5.952,p¼0.653). Results also demonstrated that the model accounted for only 6.8 percent of the variance in the dependent variable and correctly classified 65.2 percent of the cases. Wald statistics indicated that race was the only significant predictor of transactional leadership. Odds ratios for race (e B¼3.944) showed great change in the BSEWaldpExp(B) Transformational Time in law enforcement 0.004 0.003 2.464 0.116 1.004 Time as a supervisor 0.006 0.003 4.785 0.029 0.994 Time as chief in current PD 0.732 0.484 2.288 0.130 2.079 Civilian employees 0.232 0.540 0.185 0.667 1.261 Sworn employees 0.935 0.605 2.390 0.122 0.393 Constant 0.551 0.679 0.659 0.417 1.736 NagelkerkeR 2¼0.112 Transactional Time in law enforcement 0.004 0.002 3.420 0.064 0.996 Time as a supervisor 0.003 0.003 1.158 0.282 1.003 Time as chief in Current PD 0.018 0.511 0.001 0.971 0.982 Civilian employees 0.784 0.522 2.261 0.133 0.456 Sworn employees 1.320 0.597 4.893 0.027 3.745 Constant 0.138 0.684 0.041 0.840 0.871 NagelkerkeR 2¼0.132 Passive/avoidant Time in law enforcement 0.001 0.002 0.204 0.652 1.001 Time as a supervisor 0.003 0.003 1.461 0.227 1.003 Time as chief in current PD 0.905 0.516 3.071 0.080 0.405 Civilian employees 0.781 0.575 1.845 0.174 2.185 Sworn employees 0.557 0.642 0.752 0.386 0.573 Constant 1.970 0.761 6.697 0.010 7.168 NagelkerkeR 2¼0.118 Note:df¼1 Table V.

Logistic regression coefficients for the three leadership styles, experience, and agency size 137 Police chief leadership likelihood of being classified as a transactional leader when compared to transformational and passive/avoidant leaders. This finding meant that non-whites were almost four times (e B¼3.944) as likely to be classified as a transactional leader than a transformational or passive/avoidant leader.

Results of the passive/avoidant binary logistic regression model, as provided in Table VI, indicated that the overall model adequately fit the data ( 2log likelihood¼185.284,w 2(8)¼9.126,p¼0.332). However, none of the independent variables made a significant difference in the likelihood of being classified as a passive/avoidant leader when compared to transformational and transactional leaders. In addition, it was demonstrated that the model accounted for only 5.8 percent of the variance in the dependent variable and correctly classified 68.4 percent of the cases.

Discussion As demonstrated in the literature review for the current study, there is no consensus regarding the predominant style of police leaders, including police chiefs, and which of those styles have been most effective in the past. Few studies have been conducted examining the correlates of leadership style, and results of those studies have offered BSEWaldpExp(B) Transformational Age 0.015 0.022 0.427 0.513 0.985 Education 0.056 0.071 0.610 0.435 0.946 Gender (1) 0.865 0.463 3.493 0.062 2.376 Race (1) 0.983 0.673 2.136 0.144 0.374 Employees 0.068 0.312 0.048 0.827 1.070 Promotion (1) 0.084 0.375 0.050 0.823 0.920 Jurisdiction (1) 0.278 0.397 0.490 0.484 1.320 Constant 2.181 1.599 1.862 0.172 8.856 NagelkerkeR 2¼0.075 Transactional Age 0.001 0.023 0.003 0.955 1.001 Education 0.009 0.069 0.016 0.899 1.009 Gender (1) 0.074 0.487 0.023 0.879 0.929 Race (1) 1.372 0.529 6.731 0.009 3.944 Employees 0.215 0.307 0.489 0.484 1.240 Promotion (1) 0.004 0.372 0.000 0.991 1.004 Jurisdiction (1) 0.070 0.396 0.032 0.859 0.932 Constant 0.764 1.543 0.245 0.620 0.466 NagelkerkeR 2¼0.068 Passive/avoidant Age 0.015 0.023 0.420 0.517 1.015 Education 0.047 0.070 0.450 0.502 1.048 Gender (1) 1.039 0.592 3.078 0.079 0.354 Race (1) 0.783 0.609 1.654 0.198 0.457 Employees 0.306 0.315 0.940 0.332 0.737 Promotion (1) 0.076 0.379 0.040 0.841 1.079 Jurisdiction (1) 0.215 0.407 0.278 0.598 0.807 Constant 1.264 1.614 0.613 0.434 3.539 NagelkerkeR 2¼0.058 Note:df¼1 Table VI.

Logistic regression coefficients for the three leadership styles and demographic predictors 138 PIJPSM 37,1 little in regard to what characteristics the most effective leaders possess. In addition, few studies have focussed on the leadership styles and effectiveness of police chiefs.

It is important to determine which style is most effective so that police chiefs can strengthen those characteristics in order to become more effective in meeting both organizational and community goals and needs.

The police chiefs in the current study were fairly evenly split across the leadership styles of transformational, transactional, and passive/avoidant, but this finding does not suggest that no dominant leadership style exists, as the MLQ does not allow for this.

Additionally, the MLQ does not measure the authoritative style; however, the transactional leadership is most comparable. This finding is contrary to what has been found in previous studies, where the police leaders have been classified more as transformational leaders (Berringer, 2005; Morreale, 2003; Singer and Singer, 1990). Identifying leadership styles utilized by police chiefs will enable departments to create and adjustment training programs to help leaders develop skills needed to become more effective.

MANOVA was utilized to answer the hypotheses regarding mean scores on the MLQ measures of extra effort, satisfaction, and effectiveness, the number of formal and informal complaints, and the summed measure of job satisfaction. Results indicated that those categorized as transformational leaders were more likely to have higher scores on the MLQ measures of extra effort, effectiveness, and satisfaction than passive/avoidant leaders. It was demonstrated that transactional leaders had higher mean scores on extra effort and effectiveness than passive/avoidant leaders.

Results also showed that those categorized as transactional leaders were more likely to have lower mean scores on the number of formal complaints than transformational leaders.

Based on the chiefs’ evaluation of what they think their subordinates would report on the MLQ items, these findings suggest that transformational leaders are more successful in motivating their subordinates to exert extra effort on the job, are more effective, and are more successful in satisfying their subordinates with the way they lead than passive/avoidant leaders. These findings also suggest that transactional leaders are more successful in motivating and satisfying their subordinates than passive/avoidant leaders. In addition, these findings indicate that transactional leaders have received more formal complaints than transformational leaders. These findings are important because they further support what has been found in previous studies (Adebayo, 2005; Densten, 2003; Morreale, 2003; Singer and Singer, 1990) and they support the assumption in the current study that transformational leaders are the most effective when effectiveness is defined by the MLQ measures of extra effort, effectiveness, and satisfaction, and the number of formal and informal complaints received. These findings, however, should be replicated and validated with data other than self-report information for effectiveness, satisfaction, and extra effort with a better measure of leadership. If the results of this study are validated, training programs could utilize this information to alter their programs in order to focus on strengthening those characteristics found to correspond with the most effective leadership style.

MANOVA was conducted to answer the hypotheses regarding personality characteristics and results indicated that those categorized as transformational leaders were more likely to score higher on extraversion, openness, and conscientiousness than either the transactional or passive/avoidant leaders. Results also demonstrated that passive/avoidant leaders were more likely to score higher on neuroticism than transformational leaders, and that transactional leaders had higher mean scores on extraversion and conscientiousness than passive/avoidant leaders. In regard to the 139 Police chief leadership domain of extraversion, these findings mean that transformational leaders can be characterized as confident, dynamic, energetic, positive, optimistic, and outgoing.

Transformational leaders who scored high on openness can be described as curious, open-minded, considerate of feelings, and they do not judge. In addition, those transformational leaders who scored high on conscientiousness can be characterized as strong-minded, focussed, persistent, dependable, and meticulous. These findings also suggest that those passive/avoidant leaders who scored high on neuroticism can be described as having illogical thoughts, not as capable as others in controlling their impulses, and dealing with stress in poor ways. These findings are important because, again, they further support previous findings in the literature, and they demonstrate which characteristics comprise the most effective leaders (Rasor, 1995; Welsh, 1988).

With regards to policy, based on these findings (and after replication and validation), training facilities could use this information to modify the content of their training programs and focus on helping police leaders strengthen those characteristics found to correspond with the transformational leadership style. Adding material that focusses on such things as sensitivity and stress management training to these programs could prove beneficial, by increasing motivation and reducing both officer and citizen complaints. Strengthening these characteristics could enable the police chiefs to become more effective in meeting both organizational and community needs.

With the constant change in police organizations and their respective communities (i.e. diversity in the community, a younger generation of officers) it is imperative for law enforcement leaders to develop the necessary skills to adapt to these changing environments. In addition, knowing which characteristics correspond with the most effective leaders could aid in developing promotion criteria, enabling the selection of the most appropriate police chief candidate.

Binary logistic regression analysis was conducted to determine if any of the background characteristics of the police chiefs were predictors of leadership style.

Results indicated that length of time as a supervisor was a significant negative predictor of transformational leadership, that the number of sworn employees and race were significant predictors of transactional leadership, and that none of the independent variables made a significant difference in the likelihood of being classified as a passive/avoidant leader. These findings mean that those with experience as a supervisor were slightly less likely to be classified as a transformational leader than a transactional or passive/avoidant leader, that those with sworn employees were almost four times as likely to be classified as atransactional leader, and that transactional leaders were almost four times as likely to be non-white as transformational or passive/ avoidant leaders. These findings are important because they are in conflict with the literature, where neither of these variables has been found to be significantly correlated with leadership style. Future research would be needed to determine if these same findings would result.

The results of the binary logistic regression analyses reveal very little about the demographic and background correlates of police chiefs in predicting leadership styles. These findings are important, however, because they conflict with previous findings (Girodo, 1998; Kapla, 2005; Krimmel and Lindenmuth, 2001; Villarreal- Watkins, 2001). They demonstrate an unexpected finding in regard to race, and illustrate the need for future research on the correlates of leadership style. Although little significance was found, it is encouraging to find that qualities of effective leadership may generally not be based on ascribed characteristics such as age, gender, and race. 140 PIJPSM 37,1 Limitations The biggest limitation of this study was the MLQ instrument; the MLQ demonstrated weak psychometric properties and does not assess the possibility that individuals have no dominant leadership style. An additional limitation was the large number of variables utilized in analyses in relation to the sample size. Variables had to be broken into groups to ensure that no more than eight variables were included in any one analysis.

The sample size could have impacted the results and generalizability of the findings.

A final limitation to this study was the utilization of only self-report responses to the MLQ items. Without subordinate responses, it is more difficult to be certain that the police chiefs are leading the way they report they are leading. Three studies utilized only self-report responses to the MLQ, resulting in conflicted findings, which further supports the use of rater responses (Berringer, 2005; Densten, 1999, 2003).

Suggestions for future research The sample size of the current study included only 161 Texas police chiefs, therefore, it is suggested that this study be replicated with a larger sample size to ensure generalizability to the population of police chiefs. It is suggested that the findings of the current study be replicated and validated with data other than self-report measures on effectiveness, satisfaction, and extra effort with a better measure of leadership.

Because of the lack of rater responses for the current study, future research should focus on comparing both self and rater reports to ensure that the police chief is indeed leading the way he/she states they are leading. The comparison of leader and rater responses would also enable researchers to determine if the subordinates believe their leader is being effective in meeting organizational and individual goals and needs.

References Adebayo, D.O. (2005), “Perceived workplace fairness, transformational leadership and motivation in the Nigeria police: implications for change”,International Journal of Police Science and Management, Vol. 7 No. 2, pp. 110-122.

Adlam, R. and Villiers, P. (Eds) (2003),Police Leadership in the 21st Century: Philosophy, Doctrine, and Development, Waterside Press, Winchester.

Agresti, A. and Finlay, B. (2008),Statistical Methods for the Social Sciences, 4th ed., Prentice Hall, Upper Saddle River, NJ.

Andreescu, V. and Vito, G.F. (2010), “An exploratory study on ideal leadership behaviour:

the opinions of American police managers”,International Journal of Police Science & Management, Vol. 12 No. 4, pp. 567-583.

Antonakis, J., Avolio, B. J. and Sivasubramaniam, N. (2003), “Context and leadership: an examination of the nine full-range leadership theory using the multifactor leadership questionnaire”,Leadership Quarterly, Vol. 14 No. 3, pp. 261-295.

Avolio, B. J. and Bass, B.M. (2004),Multifactor Leadership Questionnaire: Third Edition Manual and Sampler Set, Mind Garden Inc, Menlo Park, CA.

Bass, B.M. (1985), “Leadership: good, better, best”,Organizational Dynamics, Vol. 13 No. 3, pp. 26-40.

Bass, B.M. (1990), “From transactional to transformational leadership: learning to share the vision”,Organizational Dynamics, Vol. 18 No. 3, pp. 19-31.

Bass, B.M. and Avolio, B.J. (1997),Full Range Leadership Development, Manual for Multifactor Leadership Questionnaire, Mind Garden Inc, San Francisco, CA.

Beito, L.R. (1999),Leadership Effectiveness in Community Policing, Wyndham Hall Press, Bristol, IN. 141 Police chief leadership Berringer, R.D. (2005), “The effect of formal leadership training on the leadership styles of police field trainers”,Dissertation Abstracts International, Vol. 43 No. 2, pp. 393-498.

Bruns, G.H. and Shuman, I.G. (1988), “Police managers’ perception of organizational leadership styles”,Public Personnel Management, Vol. 17 No. 2, pp. 145-157.

Caless, B. (2011),Policing at the Top: The Roles, Values and Attitudes of Chief Police Officers, The Policy Press, Portland, OR.

Costa, P.T. and McCrae, R.R. (1992),Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO FFI): Professional Manual, Psychological Assessment Resources Inc, Lutz, FL.

Dantzker, M.L. (1994a), “Identifying determinants of job satisfaction among police officers”, Journal of Police and Criminal Psychology, Vol. 10 No. 1, pp. 47-56.

Dantzker, M.L. (1994b), “Measuring job satisfaction in police departments and policy implications: an examination of a mid-size, southern police department”,American Journal of Police, Vol. 13 No. 2, pp. 77-101.

Denhardt, R.B. (1993),The Pursuit of Significance: Strategies for Managerial Success in Public Organizations, Waveland Press Inc, Long Grove, IL.

Densten, I.L. (1999), “Senior Australian law enforcement leadership under examination”,Policing:

An International Journal of Police Strategies & Management, Vol. 22 No. 1, pp. 45-57.

Densten, I.L. (2003), “Senior police leadership: does rank matter?”,Policing: An International Journal of Police Strategies & Management, Vol. 26 No. 3, pp. 400-418.

Engel, R.S. (2001), “Supervisory styles of patrol sergeants and lieutenants”,Journal of Criminal Justice, Vol. 29 No. 4, pp. 341-355.

Girodo, M. (1998), “Machiavellian, bureaucratic, and transformational leadership styles in police managers: preliminary findings of interpersonal ethics”,Perceptual & Motor Skills, Vol. 86 No. 2, pp. 419-427.

Goleman, D. (2004), “What makes a leader?”,Harvard Business Review, Vol. 82 No. 1, pp. 82-91.

Green, J.L. (2006), “Executive leadership in American law enforcement: the role of personality in leader effectiveness”,Dissertation Abstracts International,Vol.66No.8, pp. 4527-4663.

Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006),Multivariate Data Analysis, 6th ed., Pearson Prentice Hall, Upper Saddle River, NJ.

Kapla, D.P. (2005), “Personalizing leadership: institutional and individual factors affecting the leadership styles and policy choices of Wisconsin police chiefs”,Dissertation Abstracts International, Vol. 66 No. 4, pp. 1494-1708.

Krimmel, J.T. and Lindenmuth, P. (2001), “Police chief performance and leadership styles”,Police Quarterly, Vol. 4 No. 4, pp. 469-483.

Kuykendall, J. and Unsinger, P.C. (1982), “The leadership styles of police managers”,Journal of Criminal Justice, Vol. 10 No. 4, pp. 311-321.

Legault, M. (2005), “Preferred police leadership styles in northern British Columbia”,Dissertation Abstracts International, Vol. 43 No. 2, pp. 455-533.

Levasseur, R.E. (2005), “The impact of transforming leadership style on follower performance and satisfaction: a meta-analysis”,Dissertation Abstracts International, Vol. 65 No. 7, pp. 2673-2866.

Miller, H.A., Watkins, R.J. and Webb, D. (2009), “The use of psychological testing to evaluate law enforcement leadership competencies and development”,Police Practice and Research, Vol. 10 No. 1, pp. 49-60.

Morreale, S.A. (2003), “Analysis of perceived leader behaviors in law enforcement agencies”, Dissertation Abstracts International, Vol. 63 No. 12, pp. 4385-4681. 142 PIJPSM 37,1 Rainguet, F. and Dodge, M. (2001), “The problems of police chiefs: an examination of the issues in tenure and turnover”,Police Quarterly, Vol. 4 No. 3, pp. 268-288.

Rasor, C.E. (1995), “An analysis of the relationship between personality preference traits of executive level and mid-level law enforcement/corrections leaders and exemplary leadership practices”,Dissertation Abstracts International, Vol. 56 No. 4, pp. 1436-1714.

Reiner, R. (1991),Chief Constables: Bobbies, Bosses, or Bureaucrats?, Oxford University Press, New York, NY.

Schafer, J.A. (2009), “Developing effective leadership in policing: perils, pitfalls, and paths forward”,Policing: An International Journal of Police Strategies and Management, Vol. 32 No. 2, pp. 238-260.

Schafer, J.A. (2010a), “Effective leaders and leadership in policing: traits, assessment, development, and expansion”,Policing: An International Journal of Police Strategies and Management, Vol. 33 No. 4, pp. 644-663.

Schafer, J.A. (2010b), “The ineffective police leader: acts of commission and omission”,Journal of Criminal Justice, Vol. 38 No. 4, pp. 737-746.

Schwarzwald, J., Koslowsky, M. and Agassi, V. (2001), “Captain’s leadership type and police officers’ compliance to power bases”,European Journal of Work and Organizational Psychology, Vol. 10 No. 3, pp. 273-290.

Silvestri, M. (2007), “Doing police leadership: enter the new smart macho”,Policing & Society, Vol. 17 No. 1, pp. 38-58.

Singer, M.S. and Singer, A.E. (1990), “Situational constraints on transformational versus transactional leadership behavior, subordinate’s leadership preference, and satisfaction”, Journal of Social Psychology, Vol. 130 No. 3, pp. 385-396.

Stamper, N.H. (1992),Removing Managerial Barriers to Effective Police Leadership: A Study of Executive Leadership and Executive Management in Big-City Police Departments, Police Executive Research Forum, Washington, DC.

Villarreal-Watkins, R.J. (2001), “Law enforcement leadership style as a function of educational experience and professional development”,Dissertation Abstracts International, Vol. 61 No. 8, pp. 3027-3136.

Welsh, J.S. (1988), “A two-study multivariate analysis of the relationships among personal variables, leadership style predispositions, and performance effectiveness”,Dissertation Abstracts International, Vol. 49 No. 6, pp. 2410-2605.

Williams, A.J. (1993), “The effects of police officer supervisory training on self-esteem and leadership style adaptability considering the impact of life experiences”,Dissertation Abstracts International, Vol. 53 No. 12, pp. 4482-4640.

About the authors Dr Mary B. Sarver is an Assistant Professor in the Department of Sociology, Criminal Justice, and Women’s Studies at the University of South Carolina Upstate. She earned her PhD in criminal justice from the Sam Houston State University in 2008. Her research interests include law enforcement training and leadership, and media and crime. Dr Mary B. Sarver is the corresponding author and can be contacted at: [email protected] Dr Holly Miller is an Associate Dean of Academic Programs and a Professor in the College of Criminal Justice at the Sam Houston State University. Her teaching, research, and consulting interests include psychological assessment, malingered mental illness, psychopathy, offender risk/need assessment, and the assessment and treatment of sexual offenders.

To purchase reprints of this article please e-mail:[email protected] Or visit our web site for further details:www.emeraldinsight.com/reprints 143 Police chief leadership