Center for Army Leadership Technical Report 2011-3 Antecedents and consequences of toxic leadership in the U.S. Army: A two year review and recommended solutions John P. Steele Center for Army Leadership June 2011 The Center for Army Leadership An Organization of Leader Development and Education, U.S. Army Combined Arms Center Thomas P. Guthrie COL, IN Director ____________________________________________________ Leadership Research, Assessment and Doctrine Division Fort Leavenworth, Kansas 66027-2314 Jon J. Fallesen, Chief Distribution: Approved for public release; distribution is unlimited, June 2011. Form Approved OMB No. 0704-0188 REPORT DOCUMENTATION PAGE Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 222024302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY) 30-06-11 4. TITLE AND SUBTITLE 2. REPORT TYPE Technical Report 3. DATES COVERED (From - To) November 2009 – June 2011 5a. CONTRACT NUMBER W91QF-09-F-0113 Antecedents and consequences of toxic Leadership in the U. S. Army: A two year review and recommended solutions 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 5d. PROJECT NUMBER 6. AUTHOR(S) John P. Steele (CAL) 5e. TASK NUMBER 5f. WORK UNIT NUMBER 8. PERFORMING ORGANIZATION REPORT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Center for Army Leadership 9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) Center for Army Leadership Leadership Research, Assessment and Doctrine Division 10. SPONSOR/MONITOR’S ACRONYM(S) 290 Stimson Ave, Unit 4 Fort Leavenworth, KS 66027 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 2011-3 CAL 12. DISTRIBUTION / AVAILABILITY STATEMENT Distribution: Approved for public release; distribution is unlimited, June 2011. 13. SUPPLEMENTARY NOTES This report supplements CAL Technical Report 2011-1, 2010 Center for Army Leadership Annual Survey of Army Leadership (CASAL): Main Findings 14. ABSTRACT This report supplements the main survey findings CAL Technical Report 2011-1, and provides in-depth analysis on the issue of toxic leadership. This research is the first that is based on a representative sample and highlights the prevalence, severity, and correlates/consequences of toxic leadership in the U.S. Army. This report provides a common framework, explains the importance of the subject, presents potential solutions and highlights 2 years of CASAL data and other recently collected data in both the Army and open literature. Estimates from the data suggest that about 1 in 5 subordinates view their leader negatively, and most believe that they have interacted with toxic leaders, and that the problem is severe. Toxic behaviors include: micromanaging, being mean-spirited/aggressive, rigidity and poor decision-making, and having a poor attitude and setting a bad example. Toxic leadership was associated weakly with unit wasting time on unproductive tasks, expected unit performance, mission accomplishment, and career intention; moderately with penalizing honest mistakes, lack of frank discussions, unwillingness to implement good ideas from subordinates, discouraging creativity, solving problems at the surface level, and individual and unit morale; and strongly with lack of ethics, putting own needs ahead of unit, and lack of subordinate confidence to follow in life-or-death situations. 15. SUBJECT TERMS Leadership; Toxic Leadership; CASAL; Abusive Supervision; Ethics; Baseline; Data, Research; Negative leadership; Darkside; Command climate; Survey. 16. SECURITY CLASSIFICATION OF: a. REPORT Unlimited b. ABSTRACT Unclassified 17. LIMITATION OF ABSTRACT c. THIS PAGE Unclassified 18. NUMBER OF PAGES 19a. NAME OF RESPONSIBLE PERSON John P. Steele 19b. TELEPHONE NUMBER (include area Unlimited 42 code) 913-758-3240 Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. Z39.18 Table of Contents INTRODUCTION ............................................................................................................................................. 1 CASAL ........................................................................................................................................................ 1 Caveats ...................................................................................................................................................... 1 Importance ................................................................................................................................................ 2 TOXIC LEADERSHIP FRAMEWORK ................................................................................................................. 2 HOW THE FIELD SEES TOXIC LEADERS .......................................................................................................... 6 Data Sources and Pre-Screening ............................................................................................................... 6 Toxic Behaviors and Typology ................................................................................................................... 8 Does the Army Have Leaders Who Are Routinely Seen as Toxic? ........................................................... 10 Correlates and Effects of Toxic Leaders .................................................................................................. 16 Paradox of Performance & Perpetuation of Toxic Leadership ................................................................ 20 Severity of Toxic Leadership .................................................................................................................... 21 WHY DOES THE ARMY HAVE TOXIC LEADERS? ........................................................................................... 22 Leaders As a Contributing Factor ............................................................................................................ 22 Followers as a Contributing Factor ......................................................................................................... 24 Leadership Environment As a Contributing Factor ................................................................................. 24 System Causes of Toxic Leadership ......................................................................................................... 25 POTENTIAL SOLUTIONS ............................................................................................................................... 28 Systems-level Solutions ........................................................................................................................... 28 Leader-level Solutions ............................................................................................................................. 30 Follower-level Solutions .......................................................................................................................... 32 DISCUSSION................................................................................................................................................. 35 REFERENCES ................................................................................................................................................ 37 CASAL INTRODUCTION 1 The Center for Army Leadership Annual Survey of Army Leadership (CASAL) assesses and tracks trends (since 2005) in Army leader attitudes of leader development, the quality of leadership, and the contribution of leadership to mission accomplishment. CASAL provides research guidance for policy decisions and program development. CASAL is a reliable source because a rigorous scientific approach is used for survey development, data collection, and data analysis including a large random representative sample and high precision. Additionally, findings are calibrated with other Army research. This report will establish a framework, present qualitative and quantitative findings from the 2009 CASAL, quantitative findings from the 2010 CASAL, initial data from the 2011 Profession of Arms campaign (PoA) survey and senior leader survey, and open literature regarding toxic leadership. Caveats While what follows is a critical review of the state of Army leadership and toxic leadership, it is not the entire story. The reader is urged to first read either 2010 CASAL executive summary or the full 2010 CASAL main findings report (CAL Technical Report 2011-1, volumes 1 or 2, respectively), which highlights the complex integrative system that influences leadership and leader development within the Army. Each year, the CASAL uncovers far more positive than negative information regarding Army leadership. Most leaders are viewed quite positively and balance subordinate needs with meeting mission requirements, despite a demanding operational tempo. In addition, some behaviors that are considered toxic in one situation may appear necessary in others (Williams, 2005). That being said, the current estimated prevalence and severity of toxic leader behaviors is serious and warrants an in-depth examination. This examination has several important limitations. Limitations of this study include, but are not restricted to: the use of survey data and the fact that anonymity does not allow for direct study of identified toxic leaders. Although surveys are useful for collecting large amounts of data from a representative sample, they are limited in the amount of causality one can place on a given variable. For instance, we cannot say directly whether it is recruiting practices, promotion practices, or the recent decrease in priority of leader development in units that have led to perceived issue regarding toxic leadership (or if negative perceptions of leaders are spilling-over into these other assessments). There is no objective assessment of toxic leadership. What follows are perceptions; however, perceptions, even inaccurate perceptions, are important because they affect behavior, learning, and ultimately mission accomplishment. 1 This document is a summary of the toxic leadership and command climate from the 2009 and 2010 CASAL dataset. For a full description of CASAL see volume 2 main findings CAL Technical Report 2011-1. In addition to the main findings report, more in-depth analyses and recommendations are offered in topical reports on education (CAL Technical Report 2011-2) and Department of Army civilians (CAL Technical Report 2011-4). Focus groups are planned for this summer in order to provide greater context and disentangle some of the intricacies of this data. 1 Importance The Army needs to be concerned with toxic leadership because of the seriousness of consequences that are caused by leadership failure. Under worst case scenarios, toxic leadership in the Army can lead to mutiny and death, as well as a whole host of relatively less serious, but still troubling outcomes (Haslam, Reicher, & Platow, 2010). These outcomes include erosion of trust, reduced effectiveness, commitment and retention, break-downs in essential communication, and diminished follower well-being (Ashforth, 1997; Duffy, Ganster, & Pagon, 2002). Part of the solution to leadership failure and negative leadership is for individual leaders to establish a positive command climate that takes care of Soldiers. The positive command climate being suggested is one in which leaders routinely engage with subordinates, reduce the stigma of seeking help, and provide support while forbidding abuse, physical punishment, and ridicule. Recent CASAL data (Riley et al., 2011) indicate that only 59% of active component leaders believe that seeking help for stress-related problems is acceptable. Toxic leadership may be even more damaging in a military setting than in civilian corporations because the impact that toxic leaders have on their subordinates’ performance is greater for those who identify a strong sense of value and meaning in their jobs (Harris, Kacmar, Zivunska, & Shaw, 2007). In other words, the best Soldiers are the ones who are most likely to be affected by toxic leaders. In short, toxic leadership affects Soldier well-being, retention, and mission accomplishment. Civilian research has shown links with both important organizational-level and individual-level outcomes. Organizational outcomes include, but are not limited to: organizational functioning and finances (Ashforth, 1997; Keashly & Jagatic, 2000; Sutton, 2007), counterproductive work behavior (Duffy, et al., 2002) including interpersonal deviance (Tate, 2009), and work withdrawal behaviors including turnover intention (Tate). Outcomes for individuals include, but are not limited to: sexual harassment (Chan, Lam, Chow, & Cheung, 2008), decreased job satisfaction (Tate), and decreased psychological well-being (Cortina, Magley, Williams & Langhout, 2001). TOXIC LEADERSHIP FRAMEWORK This section provides a clear common framework. This framework can be readily applied to research, assessment, education, and policies. Stogdill (1974, p. 259) famously wrote nearly three decades ago that, “There are almost as many definitions of leadership as there are persons who have attempted to define the concept.” Toxic leadership, like leadership in general, is more easily described than defined (Reed, 2004). Although descriptions (e.g., assholes, abusive supervisors, bad leadership, bullies, corrosive leadership, dark leadership, destructive leadership, harassing leaders, health endangering leaders, intolerable bosses, jerks, tyrannical leaders, negative leaders, etc.) and definitions of toxic leadership vary, there are behavioral consistencies. Common behaviors that are repeated by toxic leaders include: avoiding subordinates, behaving aggressively toward others, denigrating subordinates, hoarding information, hoarding job tasks, blaming others for their own problems, overly critical of work that is done well, and intimidating others (Ashforth, 1994; Frost, 2004; Kellerman, 2004; Kile, 1990; Lipman-Blumen, 2005; Lombardo & McCall, 1984; Tepper, 2000). 2 Army War College students operationalized toxic leaders as those whom are: “focused on visible short-term mission accomplishment. They provide superiors with impressive, articulate presentations and enthusiastic responses to missions. But, they are unconcerned about, or oblivious to, staff or troop morale, and or climate. They are seen by the majority of subordinates as arrogant, self-serving, inflexible, and petty” (as quoted in Bird 2010, p.6). Some (e.g., Howell & Avolio, 1992) disregard toxic leadership as an oxymoron, and maintain that leadership is, by definition, positive. However, Kellerman (2004) aptly argued that denying bad leadership in leadership discussion, curricula, and training is comparable to medical schools teaching only health, while ignoring disease. Research (Ashforth, 1994; Tepper, 2000) has shown that toxic leadership is different from the absence of effective leadership behavior, and leader incompetence. FM 6-22 (1-1) defines an Army leader as: “Anyone who by virtue of assumed role or assigned responsibility inspires and influences people to accomplish organizational goals. Army leaders motivate people both inside and outside the chain of command to pursue actions, focus thinking, and shape decisions for the greater good of the organization.” In other words, Army leaders motivate individuals for the greater good of the organization. So, what would be a toxic Army leader? Toxic leaders work to promote themselves at the expense of their subordinates, and usually do so without considering long-term ramifications to their subordinates, their unit, and the Army profession. It is not that these Army leaders don’t engage in leadership as defined by FM 6-22, or that they aim to harm their units; rather, it is their specific behaviors, especially their treatment and outlook regarding subordinates that make them toxic. This description sets the condition that such leaders do not necessarily set out to be toxic; however, there are also some who are just “assholes” (Sutton, 2007). Figure 1 and the foundational approach emphasize the dynamic of balancing subordinate needs and organizational needs. This is a starting point and was used in the 2010 CASAL survey to allow respondents to classify their leaders’ behavior more objectively than by directly asking respondents to make a toxic attribution. This classification scheme comes from over 50 years of leadership research and is similar to the initiating structure and individual consideration factors from The Ohio State Leadership Studies and The Leadership Grid (e.g., Blake & Mouton, 1964; Fleishman, 1953). These prior classifications are a strong start, but as most military scholars and actual leaders would tell you are incomplete and too situationally-driven. As a result, this initial framework and Einarsen, Aasland, and Skogstad’s (2007) initial grid was expanded to identify those leaders who act in a way that is dominated by self promotion (see Figure 1). 3 Figure 1. Leader Classification. Constructive Leadership Derailed Leadership Supportivedisloyal Leadership This figure places toxic leadership in the block of a leader who is high on pro-organizational behaviors and self-interests, and high on anti-subordinate behaviors. Essentially, these leaders are often setting out to accomplish organizational objectives (or to appear that they are doing so) in the short-term at the expense of their subordinates, and usually without considering long-term ramifications. These leaders generally behave in accordance with the goals, tasks, mission and strategies of the organization, but they typically obtain results not through, but at the cost of subordinates (Asforth, 1994; Tepper, 2000). Subordinate and unit performance may actually increase as a result of this intense focus and hard approach. Ma, Karri, and Chittipeddi (2004) called this the paradox of managerial tyranny explaining that toxic leadership may lead to extraordinary performance, even when subordinates suffer. This problem is exacerbated when promotion decisions are focused on the leader obtaining results because now the toxic leader is reinforced, and is in a position to negatively affect even more individuals. A similar, but separate leadership category and issue is derailment. Note that derailed leaders share toxic leaders’ hallmark of low consideration for subordinates, but differ by also having a low concern for themselves and their organization. Data from the 2009 CASAL (Hilgenkamp & Steele, 2010) indicated that the field believes that much of derailment is caused by personal issues and not living up to Army values. Additional indicators included reckless behaviors of alcohol/substance abuse and adultery/fraternization, or personality conflicts and politicking. These indicators and accompanying behaviors are different from the toxic leadership findings that will be presented later. As stated before, toxic leaders are usually not incompetent or ineffective leaders in terms of accomplishing explicit mission objectives. Many times they are strong leaders who have ‘the right stuff’, but just in the wrong intensity, and with the wrong desired end-state, namely selfpromotion above all else (Williams, 2005). 4 According to Reed (2004), one specific behavior is not necessarily toxic leadership, rather it is “the cumulative effect of demotivational behavior on unit morale and climate over time that tells the tale” (p.67). Reed specified that toxic leaders routinely: •See their subordinates as disposable instruments, rather than as people. •Have a destructive personality or interpersonal skills that have deleterious effects on climate. •Appear motivated primarily by self-interest. Bird (2010) summed up the literature aptly saying that the process for destructive leaders involves dominance, coercion, and manipulation, as opposed to constructive leaders who use influence, persuasion, and commitment. 5 HOW THE FIELD SEES TOXIC LEADERS Data Sources and Pre-Screening In November-December 2009 2, quantitative and qualitative CASAL data were obtained from a random and representative sample of Army uniformed leaders with ranks from Sergeant – Colonel (N = 1186), and Department of Army (DA) civilians (N = 923). The sampling error was approximately +/- 3%, which means that obtained percentages (of perceptions) are accurate to within plus or minus 3 percentage points. In November-December 2010, additional quantitative items were asked of nearly 12,000 active component Army leaders, a little over 10,000 reserve component Army leaders, and nearly 5,900 civilians with a sampling error of approximately +/0.6%. The respondents also completed additional items that are not presented here, but are accessible in the CASAL main findings report (CAL Technical Report 2011-1). In both the 2009 and 2010 CASAL the toxic label and similarly biased language did not appear in the questions or instructions to the participants. The toxic leadership label was only applied after analyzing the responses (see verbatim examples on next page). In short, the survey respondents created and applied their own definition of ‘toxic leadership’ within a few parameters. Specifically, 2009 CASAL respondents were primed to think of a leader who was over-controlling, inhibited innovative thinking, or generally created a negative work environment. Within these parameters, the respondents were then tasked to describe behaviors that were exhibit that had a negative effect on the work environment. It is important to point out that this only served as a primer, and the actual nature and range of responses were far more severe than anticipated. A cognitive psychology researcher and a Field Grade researcher (neither of which is the author) were asked to independently examine the severity of the statements. Overall, they had 100% agreement that 14% of the statements were extreme, 50% were moderate, and 36% were still negative, but not concerning. As a result of the breadth and severity of the comments, the term toxic is used to describe these behaviors and individuals. 2 Respondents in the 2009 sample were randomly assigned to recall a successful influence attempt or an unsuccessful influence attempt. Specifically, participants were asked to recall a recent attempt at influencing someone whom they had no legitimate authority or power over. The difference between the two conditions was in priming the participants to recall a particularly salient personal example of a successful influence situation, or a failed influence situation. Results indicated that the assigned condition did not affect participant’s attitudes toward toxic leadership. Frequency of toxic leaders encountered, F (1, 2411) = 2.29, p = .13 in the unsuccessful influence condition (Mean = 2.06) were about the same as those in the successful condition (Mean = 2.16). A similar nonsignificant effect was obtained for ratings of seriousness of toxic leadership as a problem; F (1, 2403) = 3.04, p = .08, with unsuccessful influence condition reporting nearly the same level of seriousness (Mean = 4.56) compared to those who were in the successful condition (Mean = 4.51). Therefore, the results that follow are unlikely biased by mood or question order. 6 The following are verbatim descriptions that are considered extreme: “Serving as primary staff officer in a maneuver battalion where all decisions were very centralized, primary staff officers would have to wait two week for an appointment with the battalion commander. Initiatives, suggestions, opinions not vetted by the Executive Officer were crushed with prejudice. Honest mistakes resulted in comments like "CPT if you do that again I will put a gun to your head and pull the trigger," collective "reamings" of the staff and attributional behavior. One of the officers was also sleeping with subordinates wives which divided the battalion as well.” “Lack of concern; self before service; he was out for #1 - made the statement "I do not have to develop you, just work you til you drop; when you drop, I will just replace you with another officer." “Domineering, distrust of other, and uncompromising behaviors led to a work environment of paranoia and leadership by fear and intimidation.” “The superior was intimidating and derisive of employees. He required certain workers to remove family pictures from their work areas and had different behavior standards for various employees based on their national heritage.” The following are verbatim descriptions that are considered moderate: “Punative, argumentative, overbearing, always right, didn't listen to other opinions, quick to react.” “The individual was drunk with power. his integrity was lacking as he wrote checks he couldnt cash in order to get people to align with him. he surrounded himself with yes men, and acted like a tyrant. his actions caused his staff to function independently and not as a collective Staff. The whole time period was painful at best.” “Constantly putting team members down as incompetent. Conducting daily nonproductive meetings, playing silly games. Displayed no knowledge of mission requirements nor concern for the mission nof the organization or welfare of the organization's military, civilian, and contract members.” The following are verbatim descriptions that are negative, but not concerning: “Information flow did not go freely and leader resisted input from others.” “Micromanagement. Assigned same task to multiple people.” 7 Before analyzing the results, the data were assessed to see if they could validly be combined and reported at the aggregate level. Analysis of variance (ANOVA) was used to identify significant differences on toxic leadership perceptions between genders, among racial groups, and among Army components. Analysis of gender revealed no significant differences on perceived seriousness of a toxic leadership problem, F (1, 2339) = 0.46, p = .50; however, there was a weak, but still statistically significant difference regarding the number of leaders observed, F (1, 2347) = 6.14, p = .01, with women reporting slightly less toxic leader occurrences (Mean = 1.98) than men (Mean = 2.16). Race revealed a nearly identical pattern with no significant differences on seriousness of a problem, F (1, 2364) = 0.09, p = .76; but, with non-whites reporting a slightly lower number of toxic leaders (Mean = 1.91) than whites (Mean = 2.16). In other words, ‘minorities’ perceive toxic leadership as less of an issue, compared to non-minorities, but the difference is small and not meaningful. Active Duty, National Guards, and Reserve respondents were not significantly different in terms of seriousness of a problem, F (2, 1131) = 0.77, p = .46; and number of leaders observed, F (2, 1133) = 0.46, p = .63. In short, this preliminary analysis suggested aggregating respondent groups and focusing on the overall findings 3. Note that some analyses still focus on specific groups or cohorts in order to emphasize or clarify a finding. Toxic Behaviors and Typology Toxic leadership behavioral typologies were derived from 2009 CASAL open-ended comments, some of which were quoted on the previous page. The original typology was made up of 9 categories of behaviors (micromanaging, mean-spirited/aggressive, rigid/poor decision-making, poor attitude, narcissistic, inconsistent, interfering with work or processes, unreachable, and incompetent), which were later condensed down to 5, by retaining only those that were mentioned in at least 5% of the comments (see Figure 2). Micromanaging refers to overly controlling behaviors such as restricting pertinent information and results in under-utilization of resources. Mean-spirited/aggressive includes inexcusable behaviors that are often illegal including intimidation, physical abuse, and sexual harassment. Lastly, rigid/poor decisionmaking describes a closed-minded leader who ignores important information, sticks to a failing plan, and is incapable of adapting. Figure 2 shows that both Army civilian and uniformed leaders produced the same typology and the same weighted average of responses making up each toxic leader category or type, with the greatest (but still not statistically significant) difference appearing between comments describing micromanaging behaviors (39% civilian and 45% Army). This typology was constructed independently, though is remarkably similar to one observed from 707 survey takers, who generated the same 4 types of destructive leadership behaviors (Shaw, Erickson, & Harvey, in press). 3 Note that just because no meaningful differences were detected, this does not rule out complex relationships. For example, these data were analyzed without knowing the demographics of the target [‘toxic leaders’], only the respondents. 8 Figure 2. Comparison of Categorization of Toxic Behaviors for Civilian and Army Leaders. 12% Poor Attitude/Example 13% 15% Rigid/Closed-Minded 17% Army Civilian 19% Mean Spirited-Aggressive 23% 45% MicroManaging 39% 0% 20% 40% 60% 9 80% 100% Does the Army Have Leaders Who Are Routinely Seen as Toxic? Reed and Bullis (2009) noted that toxic statistics obtained from both corporate and military settings are biased because they under-report the negatives since the data obtained are only from the “survivors”. Conceivably, the worst leaders have driven away personnel, and those that have stayed have become more calloused to it; however, the damage is also limited because the most serious offenders have likely been relieved as well. Recently, LTG(R) Ulmer (personal communication 21 April, 2010) estimated that roughly 8% 12% of Army officers at the rank of colonel and higher are so toxic that they need to be removed from command. The 2011 Profession of Arms campaign (PoA) senior leader survey revealed that only 27% responded that the Army is effective (47% ineffective) in identifying ineffective or negative leaders, and even less thought the Army was effective (17% effective; 57% ineffective) in rehabilitating or removing such leaders. Based on several 2010 CASAL leader-level and unit-level data points, it is estimated that roughly 1 in 5 uniformed leaders are viewed negatively. A set of questions helped classify leaders according to their subordinates’ responses (using the classification given in Figure 1). The survey items were framed without reference to toxic leadership: • does things and behaves in a way that is positive for the organization, themselves, and subordinates (75%; i.e., constructive-authentic classification) • does things and behaves in a way that is negative for the organization and themselves, but positive for subordinates (2%; i.e., supportive-disloyal classification) • does things and behaves in a way that is negative for the organization, themselves, and subordinates (5% uniformed and 6% civilian; i.e., derailed classification). • does things and behaves in a way that is positive for the organization and themselves, but negative for subordinates (18% uniformed and 19% civilian; i.e., toxic classification) Other items on the survey provide insight as to the number of leaders who rated: • their superior “a real jerk” sometimes, often or definitely (25% uniformed; 10% civilian) • that their superior puts own needs ahead of their unit (22% uniformed 4; 11% civilian) • that the unit holds honest mistakes against them (21% uniformed; 17% civilian) • that the unit promotes a zero-defect mentality (30% uniformed; 34% civilian) It is also important to note that Hannah et al. (2010) observed an average of 2.46 (SD = 1.1; N = 1415) on Tepper’s abusive supervision scale (2000) for Army participants. This is meaningful because 28 published studies using the same measure indicated an average of only 1.63 on a 5point scale (S. Hannah, personal communication, December 29, 2010). In fact, Hannah et al. only found a single study (Rafferty, Restuborg, & Jimmieson, 2010) that reported a higher average (M = 2.92, SD = 1.64; N = 210). However, that sample was relatively small and limited to MBA students in Philippine universities. This indicates that Army service members rate abusive supervision higher than other occupational samples. 4 The 2011 PoA survey reported 28% agreement with, “My leader puts his/her career ahead of the mission and his/her people). 10 Furthermore, Reed and Bullis (2009) provided data of 172 participants from the class of 2008 at a military senior service college, and found that 100% of the 171 leaders they assessed had experienced toxic leadership during their careers. The Army needs to be concerned with toxic leadership because of both the prevalence and seriousness of consequences caused by leadership failure. Like all CASAL items, perceptions of toxic leadership were more negative in lower rank cohorts; however, they were still quite negative even at the field grade level (see Table 1). Table 1 Number of Army Toxic Leaders Encountered by Cohort for 2010 CASAL Cohort N Average # of Toxic Leaders 2 0 *Change 1-2 Toxic For 0 Toxic Leaders TL Leaders 24% +9% 49% Field Grade 3034 Company Grade 2 19% +7% 49% 2898 Warrant Officers 2 21% -3% 49% 2103 SR NCOs 2 19% +9% 48% 4018 JR NCOs 2 14% +5% 43% 4840 *Change compared to the same question in the 2009 CASAL. 3-4 Toxic Leaders 15% 5+ *Change Toxic For 5+ Leaders TL 12% -2% 18% 14% +2% 17% 13% -5% 19% 14% -1% 21% 22% -2% In 2009 and 2010 CASAL data, the vast majority (83%) indicated directly encountering leaders who were over-controlling, inhibitive of innovative thinking or generally created a negative working environment in the last year. In fact, over a third (35% in 2009; 46% in 2010) indicated that they had firsthand experience with 3 or more leaders demonstrating these behaviors, indicating significant prevalence. The 2009 CASAL asked, “How much of a problem are these negative types of behaviors (1- ‘Not a problem at all’ to 6- ‘A serious problem’)”. Although only the extreme anchors were provided, it is safe to conclude that at least over half (61% responded 5 or 6) thought that these negative leader behaviors were a problem. In 2010, the scale was changed to a 1 to 7 and 39% selected the two highest response anchors. In both 2009 and 2010, only 13% responded with a 1 or 2, indicating that very few indicated that negative leadership is not a problem. 11 Command Climate Another way to examine toxic leadership is to examine the command climate. The 2011 PoA senior survey results are concerning in this regard. Less than half (45%; 34% disagreed) agreed that the Army allows people to offer candid opinions without fear of repercussions and 25% believe that making an honest mistake in the Army can ruin someone’s career. Lastly, 10% of military leaders and 15% of Army civilian leaders are believed to conceal problems from their superiors. Table 2 presents the favorability and correlations (i.e., strength of relationship between command climate and toxic and constructive leader characteristics) for the 2010 CASAL data. Table 2 shows that while most (77%) think that their unit can effectively accomplish their mission and (70%) consider their unit to be high performing, relatively few (27%) believe that their organization allows the frank and free-flow of ideas, or that their organization does NOT hold honest mistakes against them (38%). In short, approximately 20% of 2010 CASAL respondents saw their superior as toxic and unethical. 12 Table 2 Favorability (agreement with positive items and disagreement with negative items) and Zero-Order Correlations Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1. Superior is a real jerk 75% 2. *# of toxic leaders observed .24 17% 3. Severity of toxic leadership problem .17 .45 21% 4. Superior puts unit needs ahead of self -.52 -.21 -.14 78% 5. Superior is a real star -.57 -.22 -.17 .68 80% 6. *# of constructive leaders observed -.20 .01 -.06 .25 .26 97% 7. Superior enforces ethical standard -.53 -.24 -.16 .68 .66 .26 83% 8. Confidence follow life-or-death -.55 -.21 -.17 .68 .81 .27 .66 68% 9. Leader accomplishes their goals -.03 -.09 -.15 .10 .08 .05 .11 .10 65% 10. Leader’s promotion potential -.41 -.16 -.11 .65 .69 .22 .59 .66 .11 74% 11. Emulate immediate superior -.57 -.24 -.18 .69 .82 .25 .65 .78 .06 .70 56% 12. Unit Morale -.36 -.36 -.28 .36 .42 .26 .42 .42 .14 .34 .44 41% 13. Individual Morale -.35 -.32 -.23 .39 .40 .28 .40 .42 .10 .32 .41 .69 52% 14. *Career Intentions -.14 -.16 -.10 .14 .15 .12 .16 .15 .06 .10 .14 .24 .29 82% N = 8780 to 9580. All correlations > |.02| are significant at the .01 level (two-tailed). Diagonal values are favorability meaning agreement with positive items, effectiveness, and disagreement with negative items. For favorability calculations, variable 2 was calculated by identifying the percentage selecting ‘none’; variable 6 was calculated by identifying the percentage who did not select ‘none’; variable 14 was calculated by those who planned to stay in the Army beyond their obligation. 13 The correlation between a superior seen as being a “real jerk” and enforcing an ethical standard is strong, (r = -.56), as is the relationship between seeing superior as a “real jerk” and being confident to follow that leader in a life or death situation (r = -.55). Table 2 also shows that toxic leaders are associated with toxic leadership environmental characteristics including:      Penalizing honest mistakes; Stifling frank communication, good ideas, and creativity; Solving problems at the surface level; Poorer unit performance including time wasting; Reduced individual and unit morale. Interestingly, having a zero-defect command climate was unrelated to toxic leadership, and more weakly-related to the handling of honest mistakes than what was expected. To more fully explore the relationship between command climate and how leaders are seen, constructive leadership was also assessed. The command climate and performance indicators for constructive/authentic leaders are the opposite of toxic leaders (see Table 2 vs. Table 3). Constructive leaders: encourage frank and free-flowing idea discussion, implement good ideas, encourage creativity and innovation, and solve problems by examining the root-cause. Interestingly, favoring results over methods, and having a zero-defect climate was not indicative of either toxic or constructive leaders. Seeing one’s superior as a real star strongly correlated with that superior enforcing ethical standards (r = .66), which was virtually the same correlation between superior putting unit needs of their own needs, and superior enforcing ethical standards (r = .68). Said differently, there was a strong relationship between constructive leadership and behaving ethically. Ethical behavior had a similar pattern of correlations with all the variables previously listed that constructive leadership is related to (see Table 3). 14 Table 3 AC Favorability (agreement with positive items and disagreement with negative items) and Zero-Order Correlations Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1. Superior is a real star 80% 2. *# of exceptional leaders observed .26 97% 3. Superior puts unit needs ahead of self .68 .25 78% 4. Superior enforces ethical standard .66 .26 .68 83% 5. Results favored over methods -.08 -.08 -.06 -.09 30% 6. Honest mistakes are held against -.21 -.20 -.27 .28 .21 50% 7. Zero-defect climate -.01 -.04 .01 .03 .17 .29 38% 8. Frank Free-flowing idea discussion .42 .25 .39 .42 -.09 -.46 .02 27% 9. Good ideas implemented .41 .23 .38 .41 -.09 -.43 .05 .80 48% 10. Creativity/Innovation Encouraged .45 .28 .40 .43 -.09 -.46 -.01 .71 .69 87% 11. Root-cause Problem-solving .39 .25 .35 .38 -.06 -.34 .09 .58 .60 .57 49% 12. Waste time on unproductive tasks -.31 -.20 -.26 -.26 .17 .38 .05 -.43 -.44 -.44 -.46 29% 13. Relative unit performance .35 .24 .33 .35 -.04 -.28 .11 .48 .50 .48 .61 -.32 70% 14. Mission Accomplishment Efficacy .36 .25 .36 .39 -.05 -.33 .10 .52 .53 .51 .60 -.34 .73 77% N = 8,700 to 11,400. All correlations > |.02| are significant at the .01 level (two-tailed). Diagonal values are favorability meaning agreement with positive items, effectiveness, and disagreement with negative items. For favorability calculations, variable 2 was calculated by identifying the percentage that did not select ‘none’. 15 Correlates and Effects of Toxic Leaders It is important to reflect that like a disease, toxic leadership can be contagious (Kelly & Barsade, 2001). One toxic leader can negatively affect dozens or hundreds of Soldiers, and abused subordinates will, in turn, negatively affect even more personnel. Although, they did not specifically study toxic leadership, Hannah et al. (2010) recently found that at the platoon-level, ‘abusive supervisors’ affected social cohesion, task cohesion, and was the best predictor of average disciplinary actions, and average reprimands received. As previously stated, under worst case scenarios, toxic leadership can lead to mutiny and death, as well as a whole host of relatively less serious, but still troubling outcomes (Haslam et al., 2010). These outcomes include erosion of trust, reduced effectiveness, commitment and retention, and break-downs in essential communication (Ashforth, 1997; CASAL 2009 and 2010 data; Duffy et al., 2002). Table 4 categorizes the direct effects that toxic leadership has on important variables based on 2010 CASAL data. Table 4 2010 CASAL data indicate that toxic leadership is related to subordinate perceptions of superior variables: *Weakly with Unit wasting time on unproductive tasks Expected unit performance **Moderately with† Penalizing honest mistakes ***Strongly with† Lack of ethics Lack of frank discussions Mission accomplishment Lack of implementing good ideas from subordinates Discouraging creativity and innovation Solving problems at the surface level instead of rootcause Unit Morale Individual Morale Putting own needs ahead of unit Lack of confidence to follow in life-or-death-situations Lack creating a positive environment Lack demonstrating interpersonal tact Career intention†† *Weak refers to an effect size that around .2, meaning that these variables are 20% higher than what’s expected by chance (and that toxic leadership DIRECTLY explains around 4% of the variation in these outcomes); **moderate refers to an effect size around .35, meaning that the relationship holds about 68% of the time, as opposed to the 33% of the time expected by chance (and that toxic leadership DIRECTLY explains around 12% of the variation in these outcomes); ***strongly refers to around .55, meaning that this relationship holds about 78% of the time, which is 55% higher than the 23% expected by chance (and that toxic leadership DIRECTLY explains around 30% of the variation in these outcomes). †All competencies and attributes of the Leadership Requirements model moderately or strongly correlated with toxic leadership, but only the two strongest are listed. ††These are only direct effects; subsequent analyses presented later shows that toxic leaders have an indirect effect on career intention of affected subordinates. 16 Toxic leadership (assessed as my superior is a real jerk) is well-explained by the Leadership Requirements Model (FM, 6-22). When combined, effectiveness ratings on the competencies and attributes account for 53.8% of the variability in toxic leadership. Specifically, the Creates a Positive Environment competency itself accounts for 46.5% of the variability in toxic leadership and the addition of the Interpersonal Tact attribute adds an additional 4%. Repeating this analysis with the dichotomy that the superior does things and behaves in a way positive for organization, self, but negative for subordinates (i.e., classified as toxic leader versus classified as anything else) is less explained by the Leadership Requirements Model. Analysis revealed Creates a Positive Environment again as the highest weighted predictor accounting for 17.7% of the variability, and the other requirements adding only an additional 3.4% 5. For constructive leadership, the Leadership Requirements Model accounts for 70.4%, with 60.5% of the variance in superior is a real star coming from the Leads Others competency. Unsurprisingly, toxic leadership is associated with negative outcomes, and constructive leadership is associated with positive outcomes, and the magnitudes of most of the effects were similar (see Table 5). The key differences are in promotion potential and emulation of superior. With both variables, the effects of constructive leadership are considerably stronger than the effects of toxic leadership. Respondent perception that their superior is a real jerk correlates with perceptions that this superior puts their own needs ahead of their unit. This underscores selfinterest as a key element of toxic leadership. In addition to direct effects, toxic leadership has indirect effects on important outcomes. For example, the 2009 CASAL shows that toxic leadership impacts turnover intention via reduced morale. In statistical terms, morale totally mediates the relationship between toxic leadership and intention to remain in the Army. 6 In support of the aforementioned finding, Reed and Bullis (2009) also found that 57% of their sample agreed that they seriously considered leaving military service because of how they were treated by their superior. 5 Less variance accounted for should be expected because the outcome was dichotomous instead of on a more continuous Likert-type scale. 6 Baron and Kenny (1986) proposed specific criteria that if met indicated the presence of a totally mediated relationship. For morale to function as a mediator it must fulfill the following. First, toxic leadership must significantly be related to both the outcome of career intention and the mediator of morale. The frequency of toxic leaders is significantly associated with career intention, ß = -.34, p < .001, and morale, ß = -.13, p < .01 Second, morale must explain a significant portion of variance in career intention after removing the variance attributable to toxic leadership; which it did, ß = .23, p < .001. Third, the effect of toxic leadership on career intention after controlling for morale should be non-significant; which it is, ß = - .05, p = .11. 17 As part of the larger 2009 CASAL survey, data were also collected regarding the priority of leader development that occurs within the unit, interpersonal trust in superior, and talent management. At a descriptive level, the data revealed favorability (endorsement of positive anchors e.g., “agree” and “strongly agree” for positive items, and endorsement of negative anchors e.g., “not a serious problem” for negative items) of: 28% for the accuracy of existing personnel policies and procedures. 32% for the fairness of existing personnel policies and procedures. 51% for the priority of leader development that occurs within units. 22% for strong interpersonal reliance on superior (i.e., high superior trust). The relationships between these variables and toxic leadership were also evaluated. These findings provide greater insight into Army toxic leadership. The data indicate moderate inverse relationships between the level of priority that one’s unit places on leader development and toxic leadership (frequency r = -.34; severity r = -.26). In other words, units that make leader development a higher priority also tend to report fewer toxic leaders, and consider toxic leadership less of a problem. Superior trust showed a similar, but weaker pattern, (frequency r = -.22; severity r = -.15), meaning reliance on a superior was related to perceptions of toxic leadership. Perceptions regarding the accuracy of the personnel system was also inversely related with the toxic leadership (frequency r = -.34; severity r = -.28), as was the fairness of the selection and promotion systems (frequency r = -.29; severity r = -.24). These analyses were cross-validated with the other half of the 2009 CASAL sample and produced nearly identical results. These analyses indicate that toxic leadership perceptions and other important organizational outcomes are related. Subordinate Behavioral Responses to Toxic Leadership. In the 2009 CASAL survey, respondents who encountered toxic leadership were asked to indicate what their responses were. Overall, the most common behavioral responses were staying mission focused (34%), followed closely by direct confrontation and communication with the toxic leader (28%), and avoiding the toxic leader (22%). This is one of the few areas where the overall sample differed depending on the group, with uniformed Army leaders being much less likely to confront and being more likely to stay mission focused compared to Army civilians. In fact, only 15% of uniformed leaders said they directly confront and only 3% said that they discuss with others when they are faced with a toxic leader; and 45% said that they keep their head down and focus on the mission. Reed (2004) cited frequent assignment changes as a reason for abused subordinates to attempt to wait-out toxic leaders, rather than to act toward any real improvement, which may be the reason for this difference in subordinate behavioral reactions. 18 Figure 3. Uniformed and Civilian Subordinate Behavioral Responses to Toxic Leadership. Positive, 6% Discussed w/ Others, 3% Ignored, 7% Mission Focused, 34% Avoided, 22% Confront, 28% What’s most troubling about these results is that only about a third of the comments, namely the sum of discussed with others (3%), and direct confrontation (28%) indicated that someone was made aware that these behaviors were having a harmful effect on personnel or the organization. In other words, over two-thirds of the time this negative behavior was never directly questioned nor reported. Having gained an understanding of what made a leader be considered toxic, and how subordinates responded to toxic leaders, the next question was, is there a relationship between the two (i.e., are certain toxic behaviors associated with particular behavioral responses)? Phicoefficients, which can be interpreted like correlations were calculated between type of toxic leader and subordinate behavioral reactions. The coefficients were relatively weak, with the strongest effect between directly confronting micro-managing leaders (r = .21) and for avoiding mean-spirited and aggressive leaders (r = .23). This means that the most common response type for micro-managers is to directly confront them and to avoid mean-spirited aggressive leaders; however, the relationship is not strong, indicating considerable variation in subordinate strategy. After studying the types of toxic leaders and their behaviors, and subordinates’ behavioral reactions, the next point of analysis was to see if either the type of toxic leader or the behavioral response resulted in improving the situation. Phi-coefficients indicated very small effect-sizes with the largest association (r = .12) between direct confrontation and effective improvement. In other words, no particular types of toxic leaders or subordinate strategies were much more effective than others, which suggests a training need, the influence of the situation, or the dominance of toxic leaders. 19 Paradox of Performance & Perpetuation of Toxic Leadership As previously mentioned, Ma et al. (2004) described the paradox of managerial tyranny arguing that toxic leadership may lead to extraordinary performance, even when subordinates suffer. This problem is exacerbated when promotion decisions are focused on the leader obtaining results because now negative behaviors are reinforced, and the toxic leader is in a position to negatively affect even more individuals. 2010 CASAL data indicates no direct linear relationship between the degree to which one’s superior was seen as ‘a real jerk’ and that superior leader accomplishing their goals (r = -.03), or the degree to which one’s superior was seen as ‘a real star’ and their goal accomplishment (r = .07). To further assess the performance tyranny paradox each leader classification from Figure 1 (i.e., toxic, constructive, derailed, and supportive-disloyal) was compared with corresponding ratings of the extent to which that leader accomplishes their missions and goals, their perceived promotion potential, and subordinate desire to emulate that leadership approach. Results confirm that leaders who were classified as toxic got their intended results more than any other leadership type; however, the difference between leader classification and goal accomplishment is very small at only 2% (see Table 6). That being said, promotion potential and emulation showed large differences between the constructive classification, and all other classifications. In fact, the majority of constructive leaders is expected to receive more leadership responsibility and is emulated. Unfortunately, 50% of toxic leaders are also expected to achieve a higher level of leadership responsibility, and are still emulated by 18% of their subordinates. Table 6 Comparing Leader Classification and Perpetuation of Behaviors Leader Type Toxic (18%) Constructive (75%) Derailed (5%) Supportive-Disloyal (2%) Leader Goal Accomplishment Effectiveness 66% 64% 51% 59% 20 Promotion Potential Agreement Emulation Agreement 50% 81% 19% 28% 18% 70% 3% 10% Severity of Toxic Leadership On a 1-7 scale of how much of a problem these negative types of behaviors are, only 11% selected 1 or 2 (13% in 2009) indicating few see this as not much of a problem at all, and 42% selected a 6 or 7 (57% in 2009) indicating that many perceive this as a serious problem. Examination by rank groups revealed that all ranks groups considered it a problem. Those in the 2010 CASAL who answered a ‘5’, ‘6’, or ‘7’ included: • • • • • 55% of field grade officers 61% of company grade officers 60% of warrant officers 60% of senior NCOs 66% of junior NCOs In sum, not only is toxic leadership prevalent, but the majority of leaders consider it a problem, regardless of their rank. The overall severity of toxic leadership across the 2010 uniformed leader respondents is graphed in Figure 4. Figure 4. Overall Severity of Toxic Leadership Problem. 2 7% 1 Not a Problem at All 4% 3 10% 7 A Serious Problem 27% 4 16% 6 15% 5 21% *Severity was rated on a 1-7 scale, with 7 = Maximum. 21 WHY DOES THE ARMY HAVE TOXIC LEADERS? Figure 5 depicts research (Padilla, Hogan, and Kaiser (2007) explaining that all leadership, including toxic leadership, is dependent upon three factors: the leaders themselves, their followers, and the environment in which they lead. Figure 5. Factors that Influence Leadership. Leaders As a Contributing Factor Colonel Williams (2005) argued that toxic leaders are not necessarily all-bad, and the Army, as a system, may unsurprisingly be producing these individuals. She questioned if toxic leaders are the extreme result of the Army’s focus of leaders who are confident, decisive, and demonstrate control. Taken to the extreme, such leaders can be self-serving and arrogant, rigid and unwilling to admit mistakes, and unwilling to develop others, and micromanage instead. Keltner, Gruenfeld, and Anderson (2003) described a scenario in which those with power become more socially aggressive as consequences of their actions are less likely to be punished. Kipnis’ (1978) metaphoric model of power states that repeated exercising of power will cause a person to become more arrogant and subsequently start to denigrate and avoid subordinates. In other words, the more one engages in toxic leadership and is not punished the worse they will become. 22 Thomas (1999) examined over 2,000 ROTC cadets and found that ambition, extraversion, hedonism, and power (all factors that in their extreme form are associated with toxic behaviors) were all significantly higher in the ROTC sample compared to a normative general population sample. He also found that conscientiousness and affiliation (factors associated with more inspirational leadership behaviors) were all significantly lower. In addition, ambition predicted cadet evaluation scores. Campbell (1995) reported executive officers indicated inflexibility and a lack of concern for the human element. In the California Psychological Inventory, the strongest traits among general officers were dominance, self-acceptance, inflexibility, and achievement via conformity. It should also be noted that such findings are not limited to the U.S. For instance, Bradley & Charbonneau (2004) found that for Canadian Forces cadets, extraversion and dominance personality factors predicted cadets’ overall course performance. Ideology refers the leader’s vision compromised of beliefs, goals, and resulting actions. Part of a leader’s ideology has already been described in Figure 1, which includes pro-subordinate, personal, and organizational orientation in the context of a constructive leadership, or antisubordinate, but pro-personal and organizational ideology for toxic leaders. Toxic leadership is closely linked with ethics. In fact, as previously stated, CASAL 2010 data show that the correlation between the two is .56, which means over 30% overlap between the two constructs. For case studies, read the Army Times’ (March 6, 2011) recent description of a U.S. Army brigade commander’s activities to hide his true command climate from his superior officer or Bourke’s (2005) description of UK executive officers’ attempts to dismiss Soldier brutalization, prisoner abuse, and bullying. In the Canadian Force (Bradley & Charbonneau 2004), only 32% disagreed with the statement, “Leaders in the Army above my unit’s chain of command are prepared to do unethical or immoral things to further their career.” Hannah et al.’s (2010) ethics study with U.S. Army platoon leaders revealed that 78% self-reported that they would never confront a direct report, and 79% would never confront a superior who is engaging in ethical misconduct. They concluded (p.21) that: “Abusive leaders are toxic to units. Not only do they create a negative culture and climate in their unit, but our results showed that they increase ethical transgressions. The Army should develop leaders who understand the line between being firm yet caring, and being abusive; and identify and separate those found to be abusive.” Recent data collected as part of the U.S. Army Profession of Arms Campaign (2011) revealed that 12% of respondents agreed that they had been pressured to cover up issues or act unethically, and 18% agreed that it would be hazardous to their career to speak up about ethical violations. In addition, only 50% of subordinates agreed that leaders are good role models for what the subordinates want to be like in the future. 23 Followers as a Contributing Factor Similar follower aspects also affect leaders and their behaviors. Padilla et al.’s (2007) review found that followers who are low in maturity, and have poorly developed ethics and values, or who hold a similar Machiavellian view reinforce toxic leadership. As a result of their positive organizational intentions (and either negative or lack of consideration of their subordinates), some toxic leaders may not even be aware of their negative behaviors and harmful effects (Reed, 2004). Reed (p. 69) provided a quote from Harvard Business review authors (Goleman, Boyatzis, & McKee, 2002): “It’s not that leaders don’t care how they are perceived; most do. But they incorrectly assume that they can decipher this information themselves. Worse, they think that if they are having a negative effect, someone will correct them if they’re wrong.” Such corrections, rarely exist. Recall Hannah et al.’s (2010) recent findings, which indicated subordinates’ unwillingness to report serious ethical violations and problems stemming from a superior. CASAL results (Keller-Glaze et al., 2010 and Riley, Hatfield, Nicely, Keller-Glaze, & Steele, 2011) demonstrate that most learning occurs through informal paths (e.g., job experience, informal mentoring, etc.). 2010 CASAL data indicate that subordinate leaders who see senior leaders model a behavior are inclined to emulate that behavior, and even though they do so with much greater frequency for constructive behaviors, some also do so for toxic behaviors. Leadership Environment As a Contributing Factor A common speculation (e.g., Kusy & Holloway, 2009; Lipman-Bluemen, 2005; Sutton, 2007) is that followers and organizations not only tolerate toxic leaders, but actually produce toxic leaders. Said differently, toxic people can only thrive in toxic systems. In fact, Harvard professor Robert I. Sutton has gone so far as to publish his philosophy summarized as the “No Asshole Rule” and argued that the absence of such a rule is almost a guarantee of assholes in the workplace. Direct and indirect reinforcement are only one piece, albeit the largest, in the environmental factor. Padilla et al. (2007) also suggested that some environments are simply more likely to facilitate toxic leadership including unstable environments with many perceived threats, organizations with low or poor cultural values, organizations that have a lack of checks and balances, and organizations that are perceived as struggling and appear ineffective and inefficient. The reader can decide for themselves which of Padilla et al.’s environmental conditions, if any, are present in the U.S. Army, and should be aware that 48% of 2010 CASAL respondents agree with the statement, “Members of my unit/organization waste time and energy on unproductive tasks.” In conclusion, even if organizational leaders do not intend to create environments conducive to toxic personalities, their lack of attention and ignorance of the problem enables toxic behaviors, and can reinforce toxic leadership. 24 System Causes of Toxic Leadership The data continually indicate that toxic leadership is a problem. Toxic leaders are either recruited or made. In academic terms, toxic leadership in the Army could be explained through the attraction, selection, attrition (ASA) Theory. The ASA Theory (Schneider, Goldstein, & Smith, 1995) states that: Attraction-people are attracted to organizations that are similar to their own interests and personality. Selection-organizations choose people who share common personal attributes (although not necessarily common competencies). Selection will be applied more broadly here to indicate systematic reinforcement. Attrition--those who do not fit the environment will leave, and those who remain make-up an even more homogenous group than those who were initially attracted to the organization. ASA Theory implies that if toxic leadership is a problem in the Army then either the Army attracts such individuals who have toxic tendencies in the first place, or the Army fosters toxic leadership directly or indirectly in existing personnel through reinforcement (selection). The attrition component suggests that those who are different from the group (nontoxic whistleblowers in the case of a toxic environment, or toxic leaders in the case of a constructive environment) will either be systematically removed from the organization, or be uncomfortable enough that they just leave. In the case of attraction, the questions are: does the Army recruit individuals who have a proclivity for behaving toxically, or are such individuals attracted to the Army? Certainly, the Army is a powerful, hierarchical, and ideological organization whose missions frequently include involvement in dangerous conflicts; however, at the same time the Army employs rigorous screening and has strict rules, laws, and policies. Frankly, there is little direct assessment available to answer such a question, but there are indirect indicators. As previously reported, Thomas’s (1999) study led to the finding that ambition, extraversion, hedonism and power 7 were all significantly higher in ROTC cadets than the general U.S. population, as well as the finding that conscientiousness and affiliation were significantly lower in ROTC cadets. Similarly, Lall and colleagues (Lall, Holmes, Brinkmeyer, Johnson, & Yatko, 1999) evaluated over 500 third-year midshipmen at the US Naval Academy and found that compared to the general population, these future Naval officers were higher on personality factors of ambition and extraversion. Taken together, this provides some (albeit weak) support for the attraction element. 7 Narcissism is closely linked with toxic leadership (Hogan, Raskin, & Fazzini, 1990) and research (Lee & Ashton, 2005) shows that extraversion and narcissism are positively correlated. 25 While young recruits may be more extraverted and power-hungry the same could be said for a number of professionals such as sales professionals and executives. So, the real question becomes: what is being rewarded or reinforced? Thomas (1999) reported ambition predicted cadet evaluation scores (evidence for selection). According to Lall et al., not only were these future Naval officers higher on extraversion than the US general population (evidence for attraction), but the Midshipmen with the highest-class rankings were also less likely to be empathetic and to experience guilt (evidence for selection). Dorn, Graves, Ulmer, Collins, & Jacobs (2000, p.70) also reported that the United States military selection system rewards behavior aimed at, “Short-term, career enhancing accomplishments at the expense of long-term institutional needs.” Research on general officers in Germany, the US, and Canada found that the rate of promotions for high-level officers depended on “political skills in working the system rather than any other personal competency” (Cotton, 1997). Cotton described one of these systems as “Open to abuse by the dark side of leadership and narcissistic personalities.” Bondy (2005, p.10) remarked that the Canadian, United Kingdom, New Zealand, and United States militaries have selected for “dominance, inflexibility, and conformity” resulting in “strong, covert, and non-adaptive [military] cultures.” Recent data from the 2011 PoA U.S. Army survey found that less than 2/3 (64%) agreed that most leaders in the Army get there as a result of hard work and dedication to duty. As stated earlier, preliminary results of the Senior Leader Survey from the Profession of Arms Campaign indicate that only 28% of AC senior leaders (CSMs, CW5s, and COL-GEN) agreed that the Army effectively identifies ineffective or negative leaders, and the full survey showed even less agreement (22%). Correlation analyses show that the best indicators of these perceptions are: • • • • Army effectiveness in weeding out members who don't meet performance standards (r = .61) Army effectiveness in ensuring only good leaders are promoted (r = .55) Army effectiveness in certifying the skill level of those considered for promotion (r = .43) Army effectiveness in promoting high performing members (r = .38) This means there is a strong relationship between perceptions regarding the Army’s talent management/personnel system and the effective identification of ineffective or negative leaders. Unfortunately, 2009 CASAL findings showed little trust and low regard for the Army’s talent management system: • • 28% favorability for the accuracy of existing personnel policies and procedures. 32% favorability for the fairness of existing personnel policies and procedures. 26 Again, the attrition element means that if the Army is facilitating a toxic environment then those that stay will likely conform to a standard and those that are contra-toxic would leave. Of course, the opposite would also be true if the Army is facilitating a constructive environment. The attrition component is the most difficult to assess. Direct evidence is sparse. Reed and Bullis’ (2009) statistics of 57% considering leaving military service because of how they were treated by their boss is useful, but they did not analyze how many actually left, and if those that stayed conformed. In addition, preliminary results of the Senior Leader Survey from the PoA Campaign indicate that only 19% (18% for the full PoA survey) thought that Army effectively rehabilitates or removes ineffective or negative leaders. 27 POTENTIAL SOLUTIONS In 2003, the Secretary of the Army, Thomas E. White asked the U.S. Army War College to assess and detect leaders with “Destructive leadership styles”. Four major recommendations came from Secretary White’s tasking including: 1. Augmenting the Army’s supervisor-centric leader evaluation system with peer and subordinate input. 2. Pursuing both evaluative and developmental approaches to prevent toxic leaders. 3. Modifying unit climate assessments so that they focus on components useful to commanders. 4. Focusing on long-term success by recognizing legitimate concerns about subordinate input, applying a top-down approach, reinforcing chain of command responsibilities of providing feedback instead of relying on centralized selection boards, and minimizing the administrative load by leveraging web-based technology. These recommendations are still valid. Additional recommendations are offered here to address toxic leadership at various levels. Systems-level Solutions The bottom line is that a personnel systems approach requires an acknowledgement of the presence and detriment of toxic leadership in the Army. The Navy does this so openly that it prompted a recent editorial in the Army Times (June 25, 2011) questioning why the Army and other branches cannot fire their bad commanders openly like the U.S. Navy does. It also requires accurate and consistent assessment, input from subordinates, and a focus beyond what gets done in the short-term, toward a focus of how things get done, and the long-term effects associated with constructive leadership. LTG(R) Ulmer and his team (Ulmer, Shaler, Bullis, DiClemente, & Jacobs, 2004) asked division commanders what could be done about toxic leaders. The consensus of the respondents was that toxic leadership is rooted in personality and not amenable to change, and that this is more of an identification and selection challenge than a developmental change. The study team, did not completely accept toxic leaders as permanent, but did agree that, “de-toxicification should in any case begin early in an officer’s career” (p.32). They went on to advocate that Army systems should be designed to ensure that commanders meet comprehensive leadership standards. These standards must be aligned and integrated with selection, evaluation, education, and assignment systems. The idea that toxic leadership can and should be addressed systemically is consistent with many others (e.g., Kusy & Holloway, 2009; Lipman-Bluemen, 2005; Sutton, 2007) who advocate changing personnel policies and procedures, and initially screening out toxic individuals to begin with. The insights obtained by Hannah et al. (2010) and Ulmer et al. (2004) are similar to the 4 main recommendations made to Secretary White (Reed, 2004). 28 The most salient point from all (Hannah et al.; Reed; Ulmer et al.) is that toxic leadership identifies an issue in assessment and selection (selection here is broadly used to encompass initial screening, placement, and promotion). Currently, the U.S. Army is considering how to update or improve current evaluations (i.e., modify OER and NCOER). 2009 CASAL survey data led to the recommendation (Riley et al., 2009) to reinforce what’s important by modifying the OER and NCOER to reflect Army leadership doctrine (p. XII): “At a minimum, Part IV of Form 67-9 of the OER should be updated to align with the Army leadership competencies and attributes of Army Leadership, FM 6-22. Leader development and leadership development are most effective when systems and processes are aligned. The objective of adopting a competency model for leadership in FM 6-22 was to set a consistent, enduring model of leader development. This model that includes creating a positive environment should be extended to full implementation and practice.” Riley et al. (2009) went on to recommend expanding the company commander climate survey requirement to the battalion level, arguing that (p. XII), “Nested climate surveys from battalion, brigade, and division would allow consistent checks on climate and give an opportunity for command initiative on climate.” While following equal employment opportunity and discrimination laws is important, they do not guarantee a positive command climate. Riley et al. also advised evaluating and promoting leaders based, in part, on their responsibility to foster and maintain a positive command climate. The appeal of these recommendations is that: 1) they stem from relevant data, 2) are based on established Army leadership doctrine, 3) can be readily implemented as part of a change that is already being examined, 4) focus on subordinate assessment and needs being met, and 5) allow for greater integration of Army systems. A further benefit of these recommendations is that at the same time that the Army addresses toxic leadership it could finally address the lowest rated leader competency of Developing Others, and getting away from focusing on the strongest competency of Gets Results (Riley et al., 2011). Riley et al. (2009) recommended establishing more accountability and including results of unit level assessment and action plans in the Quarterly Training Brief (QTB). This suggestion stemmed from work by Aude, Keller-Glaze, Riley, and Fallesen (2007) who advocated establishing leader accountability by assessment and evaluation and suggesting that installation and units be provided with “organization/unit level assessment toolkits” for the purposes of assessing leader development. The assessment results would become part of the QTB and also provide a systematic method of processing assessment findings and creating an action plan to sustain and improve leader development practices, systems, and programs. This approach is consistent with Schein (1985) who emphasized the importance of senior leadership in shaping culture of their organizations and transmitting the importance of an issue by paying attention to it via a variety of means. 29 In addition to what has already been suggested, senior leaders could take a greater role and convey more importance by personally teaching and coaching incoming commanders. Given overall CASAL data 8 and previous evaluations of Army assessment and selection systems, such an approach advocated by Riley et al. (2009, 2011) is important. Leader-level Solutions A simplistic indicator that leaders can look at to see if they are viewed negatively is to observe if their subordinates emulate their behaviors and leadership approach. The data indicated that constructive leaders will notice a strong majority of their subordinates emulating them, but toxic leaders should see much lower subordinate emulation. A leader may be unaware that they are not addressing subordinate needs, or that their short-term pro-self/organization agenda is having deleterious effects. Leaders need to honestly self-assess and receive forthright feedback from others regarding their behaviors, motivation, and be aware of the extreme-side and downside of positive attributes. A Center for Army Leadership (CAL) response to facilitate the constructive leadership process was to create a handbook titled Developing Leadership During Unit Training Exercises. This handbook was developed over the past year to provide trainer-to-counterpart observation and feedback techniques focused on constructive leadership. The handbook contains information on creating a climate for development (i.e., avoiding a zero-defect results only environment), observing and understanding leadership behaviors (consistent with FM 6-22), and delivering feedback to make an impact and facilitate learning. As previously discussed, command climate surveys are a potential valuable source of information, and could become even more valuable if the requirement was extended to the battalion level (Riley et al., 2009). Having multi-level assessment in the same organization provides ownership of climate to each commander and allows for senior leaders to see where breakdowns occur. Unfortunately, such an approach is limited to full buy-in by an entire organization. However, there is an efficient and effective alternative, namely 360-degree feedback. The Army’s Multi-Source Assessment and Feedback (MSAF) program is a developmental tool that incorporates insights from one’s superiors, subordinates, and peers and compares with self-ratings. The goal is to provide unbiased feedback from multiple perspectives so that the leader can gain the personal insight needed to maintain leadership strengths and address leadership developmental needs. 8 (i.e., the moderate inverse relationships between toxic leadership and priority of leader development and accuracy and fairness of selection system; low favorability of the talent management system and moderate favorability of priority of leader development; the fact that Gets Results remains the highest rated competency, while Develops Others remains the lowest). 30 The main tenet behind the utility of the 360-degree approach is that all leaders have blind spots, or unnoticed skill deficiencies and strengths (Karrasch, Halpin, & Keene, 1997), and feedback from multiple sources can facilitate leader development (London & Smither, 1995). It is also much easier for the ratee to consider that a single rater is biased or inaccurate, but it is much more difficult to be dismissive when there is recurring information provided by multiple raters from each source (Steele & Garven, 2009). The additional benefit is that followers are given a voice, and an opportunity to discuss difficult subjects in a safe, anonymous, and productive way. This is especially important given the low rate of those whom will confront, and the high prevalence of leaders perceived as being toxic. Certainly, some toxic leaders (particularly many in the mean-spirited/aggressive category) may be uninterested in developmental feedback, but others may, for the first time, learn that their positive intentions or zealous actions are actually having counterproductive effects on their subordinates. The MSAF program also offers an additional benefit of producing a unit or organization roll-up report, which increases organizational leaders’ chances of identifying command problems. CREW is a new method that was introduced at the U.S. Army Veterans Hospital Administration, which aims to enhance civility respect and engagement in the workplace. In short, the purpose is to make all organizational members act courteous and considerate towards each other. This is similar to the previous consideration of others training (CO2) in goals, but is different in terms of focus and process. CO2 training included very different issues such as alcohol and drug abuse, weather-related injuries, HIV and safe sex, racism and sexism, and religious accommodation, but shares CREW’s goal of building unit cohesion. CO2’s philosophy is based on awareness and sensitivity, whereas CREW’s philosophy is based on skill improvement, identification of maladaptive behavioral patterns, and individual ownership. Unlike CO2, the CREW process builds on research (e.g., Olson-Buchanan & Boswell, 2008), which recognizes the importance of providing workers a voice and ownership of the process. At the management level, this means providing clear guidance for collegial behavior and moving away from an independent decisionmaking and micromanaging leadership style to more participative. The CREW approach has 5 defining principles (Osatuke et al., 2009): 1) Building civility requires an accurate assessment and reflection on interactions; 2) Practicing new ways to interact; 3) Facilitation from experts who help break maladaptive behavioral patterns; 4) Support from management is essential for success; 5) Employees must own the process for it to be successful. It should be noted that both CO2 and CREW use a methodology similar to operational mission analysis and execution. CREW pushes adaptability in individuals and also incorporates it into the program itself (e.g., few set rules as to how an intervention should work and customization is expected for each work group according to its own standard). Early results from over 1,400 work units that have used CREW are positive. CREW has been linked to less burnout among employees and higher rates of organizational commitment. The intervention also seems to have some lasting effects, with one year follow-up studies showing significant increases over baseline rates of civility (Osatuke et al., 2009). 31 The start of the CREW process involves an initial assessment, which is used by facilitators to understand the specific issues of the organization, as well as the strengths. For six months (the program is flexible on the timeline), employees frequently meet with one another to work on effective interpersonal interactions at work. These workers are assisted by trained facilitators. Unit members also complete structured exercises designed to move them out of their comfort zone and to learn alternatives. For example, workers may be coached on identifying disrespectful behavior and effectively responding to it with a variety of options, which highlight potential responses in others. During this entire process, leadership is expected to demonstrate support for a civil workplace by: making public statements regarding its importance and how it aligns with the organization’s values, writing articles in organizational publications underscoring the importance of the issue, leading by example, and displaying a signed commitment from a senior leader. At a midpoint and endpoint, refresher training is offered, as well as additional assessment to track progress. The refresher training teaches group facilitation, establishing a culture of trust, forming group rules, and active listening as well as identifying common challenges such as those who are reluctant to participate. While not all aspects of CREW may be readily transferred to an Army unit, the focus of improving civility and interrupting dysfunctional processes should be considered part of setting and improving a positive command climate. Consequently, leaders may wish to enhance the climate assessment and their own behaviors relative to the CREW process and FM 6-22. Follower-level Solutions Followers should not emulate their toxic leaders. Not only does this send positive feedback to their superior to continue these negative behaviors, but the unit now has to deal with several toxic leaders. Followers must also take a proactive role and realize that inaction is tantamount to supporting a toxic leader’s approach. Followers should examine environmental factors and what kind of a climate and expectations that they are actively or inactively creating. Specific actions follow, but in short followers can: 1) Take advantage of Army programs (MSAF Army-360, Comprehensive Soldier Fitness, etc.) 2) Use ingratiation tactics to make themselves less of a target. 3) Engage in cognitive reframing. Comprehensive Soldier Fitness (CSF) is an Army specific program, which seeks to enhance Soldier and Army civilian physical, emotional, social, family, and spiritual strength. Over 800,000 Soldiers have participated in some aspect of the program. This is both a systematic approach because it is implemented and supported by the institutional Army, like MSAF Army360, but also a follower approach because individuals can participate without permission, and the results of the program are kept confidential. This $50 million program focuses on developing resilience through fortifying a Soldier’s mental toughness, maximizing their potential, and sidestepping the pitfalls of deployment stress and anxiety. 32 The program includes the Global Assessment Tool (GAT), a self-appraisal tool that was designed to boost personal growth, strengthen relationships, and give individuals better coping skills for dealing with potentially traumatic events. There is no “right” or “wrong”, and GAT results are not shared, but self-reported problematic areas are shown in results. The user is then given access to training modules, resources, and videos that were developed by the world’s leading resiliency experts. In addition to the GAT and accompanying self-development modules, the other pillars of the CSF include institutional military resilience training that are taught in TRADOC schools and by Master Resilience Trainers (MRTs). The idea is Soldiers helping Soldiers at battalion, brigade, and installation levels. MRTs teach individuals: • To avoid thinking traps (e.g., jumping to conclusions, reflecting on one’s own actions, etc); • To detect icebergs, which are the deep beliefs and core values, not the overriding emotions; • energy management (regulating emotion to think clearly and respond with control) • problem-solving; • gaining perspective by identifying the worst, best, and most likely outcomes of a given situation. Note that resiliency training and CSF was designed to be used proactively. Thus, it was not designed that if one experiences trauma or is bullied by a toxic leader that they go “do” this. Rather it was designed so that individuals would have the knowledge, skills, abilities, and reserve to deal with negative circumstances, both at the time of the event, and after, in a way that would be healthy and constructive. Building on the idea of resilience is the concept of positive affectivity (PA). PA is the tendency to have a positive reflection of one’s own well-being, emotions, and level of engagement of both interpersonal relations and achievement (Baron, 1996). Harvey, Stoner, Hochwarter, and Kacmar (2007) demonstrated that those who possessed high levels of PA reported less negative effects from abusive supervision. A related behavior is ingratiation, which is social influence strategy using flattery, opinion-conformity, or performing favors for others. A common goal of ingratiation and similar tactics is to gain control over others (Castro, Douglas, Hochwarter, Ferris, & Fink, 2003; Harrell-Cook, Ferris, & Dulebohn, 1999). This is important because those who are proactive in exerting influence when handling workplace stressors experience less distress than those who do not (May, 1972). Ingratiation facilitates the accrual of social support in the workplace (Wayne, Liden & Sparrowe, 1994), which is useful to individuals coping with workplace stress (Aspinwall & Taylor, 1997). Research (Harvey et al., 2007) showed that those who used ingratiation and had high levels of PA reported the least negative effects from abusive supervision. The authors speculated that it may be that ingratiation allows employees to come across as likable and non-threatening, which sends a message of “pick on someone who is more problematic” to the toxic leader. Such advice 33 and conclusions are similar to the 2009 CASAL respondents’ most common reaction of trying to keep their head down and remain mission focused. The researchers also showed that those with the low levels of PA and who did not engage in ingratiation experienced the most distress. Those with low levels of PA who also engaged in ingratiation exacerbated the effects of abusive supervision. Therefore, ingratiation should be used strategically, and mainly reserved with those who feel well about themselves, and their engagement in important life activities. For those with low PA, avoidance may be useful for dealing with overly negative and aggressive leaders (Steele, 2009). Harvey et al. (2007) concluded that ingratiation is a moderately effective coping mechanism, but it is much better for employees to be able to report and be comfortable with reporting toxic leaders to those with the power to stop the mistreatment. It should be noted that this body of research is restricted to those leaders who behaved aggressively and were meanspirited, and not necessarily the other types of toxic leadership behaviors (e.g., micromanaging). The effectiveness of other influence strategies such as assertiveness, rationality, sanctions, exchange, upward appeals, blocking, and coalitions remains to be seen and should be more thoroughly researched. A more flexible approach that works regardless of toxic leader behaviors is the use of reframing. Reframing has been the focus of not only cognitive-behavioral stress inoculation training (Novaco, 1975), but also resilience training (Reivich & Shatte, 2002). Research (e.g., Liberman, Samuels, & Ross, 2004) shows that something as simple as phrasing a game such as the “community game” versus the “Wall Street” game can prime individuals to behave cooperatively or competitively. This is an important point, because phrasing conflict in cooperative or competitive terms has real implications for individual and organizational outcomes (Steele, 2009). Sutton (2007) concluded reframing to be a viable alternative for handling workplace bullies, and suggested banding together with nicer individuals. 34 DISCUSSION This research is the first representative sample confirmation of toxic leadership in the U.S. Army. In addition, the data uncovered several important relationships with other factors (e.g., ethics, command climate, turnover intention, etc.). CASAL data indicated that the vast majority of U.S. Army leaders observed a toxic leader in the last year, and over a third indicated that they had firsthand experience with 3 or more toxic leaders, indicating significant prevalence. The consistency among 2009 CASAL data, 2010 CASAL data, PoA data, Reed and Bullis’ (2009), and open literature (e.g., Shaw et al., in press) findings adds credibility to the argument that toxic leadership in the Army is an important issue, and there is little support for the contention that the situation can just fix itself. The present data are useful not just in terms of highlighting this issue, but also by validating a toxic leadership framework. This framework allows for the separation of toxic leaders from derailed leaders, and shows that perceptions of toxic leadership are related to important individual, unit, and institutional variables. Selfishness and narcissism were recurring themes. These themes were expected because almost all toxic and related leadership literature has mentioned these elements. These themes are also part of the theoretical framework (Einarsen et al., 2007). Toxic leaders, by definition, focus on themselves and their organizations, rather than on subordinates. Thus, selfishness is probably due to a direct observation of these leaders behaving in a way to advance their own agenda (or artificially make their organizations, and in turn, themselves look good), and a halo effect, in which the leader is viewed as so toxic, that everything must be negative and self-centered. This occurred regardless of cohort, but was most prevalent amongst JR NCOs. Some of the reported toxic behaviors are positive and effective when used under the right circumstances and in moderation. For instance, micromanagement can be effective when a subordinate is incompetent or wants tight guidance, and being rigid is sometimes necessary--at some point, a leader needs to just make a decision. It could also be argued that being critical is the hallmark of having high standards, which is also important. Likewise, caring and working hard to advance one’s career is desirable (as long as it is not at the expense of others). On the other hand, the mean-spirited/aggressive category offers no upside. It was uncommon to read descriptions that said the leader sometimes gets carried away, or holds onto an effective and constructive strategy even when it is not appropriate. Commonly, the descriptions indicated that the toxic leader used an approach that is not appropriate in hardly any scenario. The descriptions also indicated that such approaches were considered quite damaging to individuals, and often also to the host and other organizations, and that the toxic leader either does not care, or does not recognize their harmful effect. In short, the toxic leader is viewed as having no redeeming qualities. This is likely a bias (i.e., halo effect), rather than an accurate perception, but either way it is a real issue. 35 As previously stated, a halo effect is likely occurring in which, due to very poor behaviors in one area (e.g., interpersonal relations), a leader is believed to be poor in several other areas (e.g., decision-making, ethics). It is unlikely that toxic leaders have no redeeming characteristics (they can improve productivity, and they reached positions of authority by having some desirable knowledge, skills, and abilities). Rather, it is that their behavioral patterns are so toxic to the individuals affected that they simply cannot look past the caustic actions. In Sutton’s (2007) language, ‘who cares what they do; if they are an asshole then I do not want to work with them’. While the data do not allow for such a test, it would be useful to know if intention and roles were considered by followers when they formed their judgments (e.g., a constructive leader who is carrying out a negative course of action that was directly ordered by a superior toxic leader versus a toxic leader that is purposely creating a zero-defect climate of fear and punishment). As expected, the inverse relationship between perceptions and respondent rank was confirmed. Five years of survey data (Riley et al., 2011) have consistently shown that, regardless of content area, junior leaders are less satisfied and hold more negative perceptions than senior leaders. As expected, JR NCOs reported the highest frequency of toxic leadership and warrant and field grade officers reported the least. Reed and Bullis (2009) also confirmed this was the case in their study of toxic leadership in the Army. While this is expected, that does not mean it should be ignored. It could be that the most harmful leaders would be ones with the least experience (i.e., lack the requisite KSAOs to effectively deal with stressors and problematic subordinates) and least amount of culling and challenges by the Army system (i.e., less opportunity to be caught). Regardless of whether or not this is actually the true state, it is important because distorted perceptions have just as much power and affect behaviors and outcomes the same as accurate perceptions. As a result, it is important for the Army to identify the source of these negative perceptions, and either correct the problems, or adjust strategic communications to challenge these misunderstandings. As previously described, most toxic leaders believe that if they are negatively affecting their subordinates, then their subordinates will discuss the issue with them, or they will receive some indication that there is a problem (Goleman et al., 2002). Unfortunately CASAL research shows that this is not the case for toxic leaders, Hannah et al. (2010) showed that this is not the case for ethical violations, and PoA data showed that uniformed Army personnel believe that the Army is unable to effectively identify such negative leaders, let alone rehabilitate them or reduce their harmful effects. Worse still, almost a fifth of 2010 CASAL participants said that they emulate these toxic leaders. Several potential solutions have been presented. It is important to clarify that these changes will be largely ineffective if adopted individually, or slowly. Given that perceptions regarding toxic leadership also relate to unit and institutional perceptions, this problem must be attacked simultaneously at several levels. This includes examining all elements of the attractionselection-attrition model and senior leadership deciding if 1 out of 5 leaders seen as routinely exhibiting toxic behaviors are acceptable at a time when only 38% of Army leaders agree that the Army is headed in the right direction to prepare for future challenges. 36 REFERENCES Army Times. (2011, June 25). Fire commanders openly [editorial]. Army Times. Retrieved June 29, 2011, from: http://www.armytimes.com/community/opinion/army-fire-commandersopenly-editorial/ Ashforth, B. E. (1994). Petty tyranny in organizations. Human Relations, 47, 755-778. Ashforth, B. E. (1997). Petty tyranny in organizations: A preliminary examination of antecedents and consequences. Canadian Journal of Administrative Sciences, 14, 126-140. Aspinwall, L. G., & Taylor, S. E. (1997). A stitch in time: Self-regulation and proactive coping. Psychological Bulletin, 121, 417-436. Aude, S., Keller-Glaze, H., Riley, R., & Fallesen, J. (2007). Commander’s handbook for unit leader development. Fort Leavenworth, KS: Center for Army Leadership. Baron, R. A. (1996). Essentials of Psychology. Boston: Allyn & Bacon. Baron, R. M., & Kenney, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Bird, J. M. (2010). Destructive leadership in the army; Prescriptions for change. United States Army War College. Blake, R., & Mouton, J. S. (1964). The managerial grid. Houston: Gulf. Bondy, H. (2005). Personality type and military culture in the Anglo-West. Australian Defence Force Journal, 169, 4-14. Bourke, J. (2005, February 26). From Surrey to Basra, abuse is a fact of British Army life: Officers who blame “a few bad apples” ignore a culture of brutalization. The Guardian. http://www.guardian.co.uk/uk/2005/feb/26/military.iraq Bradley, P., & Charbonneau, D. (2004). Transformational leadership: Something new, something old. Canadian Military Journal, 5(1), 7-14. Campbell, D. P. (1995). The psychological test profiles of brigadier generals: Warmongers or decisive warriors? In D. Lubinski & R. Davis (Eds.), Assessing individual differences in human behavior: New concepts, methods, and findings (pp 145-175) Palo Alto, CA: Davies-Black. Castro, S.L., Douglas, C., Hochwarter, W. A., Ferris, G. R., & Frink, D. D. (2003). The effect of positive affect and gender on the influence tactics – job performance relationship. Journal of Leadership & Organizational Studies, 10, 1-13. 37 Center for the Army Profession and Ethic. (2011). Profession of Arms Campaign interim report: U.S. Army Training and Doctrine Command. Center for the Army Profession and Ethic. (2011). Profession of Arms Campaign, se...