|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001921128
007 cr mnu|||uuuuu
008 080118s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001882
Coole, David R.
Expansion and validation of the Political Skill Inventory (PSI ) :
b an examination of the link between charisma, political skill, and performance
h [electronic resource] /
by David R. Coole.
[Tampa, Fla] :
University of South Florida,
ABSTRACT: The present research was developed to reexamine the factor structure of the Political Skill Inventory (PSI), expand upon the political skill behavioral taxonomy to include charisma, and provide validity evidence for both the PSI and our new measure of charisma. In study one, using a large undergraduate student sample, confirmatory factor analysis provided evidence for a three factor structure of political skill. Charisma and networking ability were identified as unique factors of the political skill construct domain while PSI dimensions of social astuteness, interpersonal influence, and apparent sincerity collapsed to form a single dimension. Study One results also indicated a strong positive relationship between self-reports of political skill, charisma, and OCB.^ In Study Two, using a sample of public-sector triads consisting of professional level employees, their coworkers and their supervisors, mixed support was found for the convergent and divergent validity of the four PSI dimensions and charisma across reporting sources. As hypothesized, political skill predicted supervisor reports of overall job performance, task performance, and OCB. Charisma contributed to the prediction of supervisor ratings of overall performance and task performance after controlling for PSI total scores. At the dimensional level, social astuteness and charisma demonstrated the strongest predictive validity across all study criteria. Social astuteness and charisma also demonstrated a significant interaction when predicting supervisor ratings of overall performance and task performance. This interaction indicated that social astuteness plays more of a role in predicting job performance for employees low in charisma than for employees high in charisma.^ ^As an addition to the second study, the ability of the PSI and charisma to predict performance ratings was compared against an abridged version of a situational judgment test assessing practical intelligence, the Tacit Knowledge Inventory for Managers (TKIM; Wagner and Sternberg, 1991). After controlling for PSI total scores and charisma, the TKIM provided a modest contribution to the prediction of supervisor ratings of overall performance. Implications of these findings and directions for future research are provided.
Dissertation (Ph.D.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 101 pages.
Adviser: Walter C. Borman, Ph.D.
t USF Electronic Theses and Dissertations.
Expansion and Validation of the Political Skil l Inventory (PSI): An Examination of the Link Between Charisma, Political Skill, and Performance by David R. Coole A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Walter C. Borman, Ph.D. Michael T. Brannick, Ph.D. Marcie A. Finkelstein, Ph.D. Carnot E. Nelson, Ph.D. Walter R. Nord, Ph.D. Date of Defense: February 28, 2007 Keywords: influence, social effectiveness, OCB, organizational politics, leadership Copyright 2007, David R. Coole
Note to the Reader Note to the Reader: The original of this document contains color that is necessary for understanding the data. The or iginal dissertation is on file with the USF library in Tampa, FL.
Acknowledgements A portion of this research was conduc ted as part of a workplace study for the Civil Service Board of Hillsborough County Florida. Several people are deserving of appreciation for the roles th ey played in having this study come to fruition. First and foremost, I would like to thank Mr. Ronald Gardner, the Civil Service Board Director, for hi s approval and endorsement of this project. I would also like to thank the Hillsborough Count y Administrator (Ms. Pat Bean), the Clerk of the Circuit Cour t (Ms. Pat Frank), the Co unty Attorney (Ms. Rene Francis Lee), and the Director of the Environmental Protec tion Commission (Dr. Richard Garrity) for their support and cooperation during th e data collection process. I want to offer a special thanks to my mana ger, Kurt Wilkening, for his encouragement, guidance, and patience at a ll stages of this research. Finally, I would also like to extend my gratitude to Don Nott, my colleague and friend, for his assistance in data collection and hi s unwavering camaraderie throughout this demanding venture.
Political Skill, Charisma, & Performance i Table of Contents List of Tables iii List of Figures v Abstract vi Introduction 1 Background on Organizational Pol itics & Political Behavior 1 Political Behavior as Political Skill 4 Validation Evidence for the PSI 8 Political Skill & Tacit Knowledge 11 Political Skill & Charisma 14 Political Skill, Charisma & OCB 19 Method 27 Study 1 27 Sample and Procedure 27 Measures 27 Political Skill Inventory (PSI) 27 Charisma 28 Organizational Citizenship Behavior (OCB) 28 Self-Monitoring 28 Political Savvy 29 Social Desirability 29 Results 29 Item Analyses 29 Exploratory Factor Analysis 31 Confirmatory Fit Statistics and Alternative Models 33 Measure Reliabilities and Correlations 36 Regression Analyses 40 Study 2 43 Sample 43 Procedure 44 Measures 45 Political Skill 45 Charisma 45
Political Skill, Charisma, & Performance ii OCB 46 Task & Overall Performance 46 Tacit Knowledge 46 Results 48 Convergent and Discriminant Validity 48 Criterion-Related Validity 51 Regression Analyses 55 Exploratory Analyses 61 Discussion 69 Factor Structure of Political Skill 69 Multi-source Convergence of Political Skill 71 Evidence for Criterion-Related Validity 73 Limitations and Future Directions 76 References 79 Appendices 91 Appendix A: The Political Skill Inventory 92 Appendix B: Charisma Item Pool 93 Appendix C: Motowidlo & Van Scot ters (1994) 16-item scale of OCB 95 Appendix D: Snyders (1987) 18 It em Measure of Self-Monitoring 96 Appendix E: Chao et al.s (1994) Po litical Savvy Factor of Socialization 97 Appendix F: Strahan & Gerbasi s (1972) 10-item Measure of Social Desirability 98 Appendix G: Borman et al.s (1994) Job Performance BARS 99 About the Author End Page
Political Skill, Charisma, & Performance iii List of Tables Table 1 Coleman and Bormans ( 2000) Taxonomy of Citizenship Performance 22 Table 2 Exploratory Factor Analysis: Rotated Factor Loadings & Initial Eigenvalues 32 Table 3 Model Fit Statistics for 3 and 5 F actor Models of Political Skill & Charisma 36 Table 4 Correlations and Descriptive Statistics for Study 1 Measures 39 Table 5 Hierarchical Regression Anal yses Predicting Organizational Citizenship Behavior (OCB) from Charisma after Controlling for Political Skill 41 Table 6 Hierarchical Regression Anal yses Predicting Organizational Citizenship Behavior (OCB) from Political Skill after Controlling for Charisma 42 Table 7 Multi-Trait, Multi-Method Corre lation Matrix for Political Skill & Charisma 49 Table 8 Descriptive Statistics and Correlations of Study 2 Measures 53 Table 9 Correlations between Study 2 Predictors and Supervisor Ratings of Performance 54 Table 10 Hierarchical Regression Analys es Predicting Supervisor Reports of Performance from Study 2 Predictors 57 Table 11 Hierarchical Regression An alyses Predicting Self Reports of Performance from Study 2 Predictors 59 Table 12 Hierarchical Regression Anal yses Predicting Coworker Reports of Performance from Study 2 Predictors 61
Political Skill, Charisma, & Performance iv Table 13 Correlations between PSI Dime nsions, Charisma and Supervisor Ratings of Performance 62 Table 14 Hierarchical Regression Anal yses Predicting Supervisor Reports of Performance from Social Astute ness, Charisma, and TKIM Scores 63 Table 15 Hierarchical Regression An alyses Predicting Self-Reported Performance from Social Astutene ss, Charisma, and TKIM Scores 67
Political Skill, Charisma, & Performance v List of Figures Figure 1 The Interactive Effects of Ch arisma and Social Astuteness When Predicting Supervisor Ratings of Task Performance 65 Figure 2 The Interactive Effects of Ch arisma and Social Astuteness When Predicting Supervisor Ratings of Overall Performance 65
Political Skill, Charisma, & Performance vi Expansion and Validation of the Political Skill Inventory (PSI): An Examination of the Link Between Charisma, Political Skill, and Performance David R. Coole ABSTRACT The present research was developed to reexamine the factor structure of the Political Skill Inventory (PSI), expand upon the political skill behavioral taxonomy to include charisma, and provide validity evid ence for both the PSI and our new measure of charisma. In study one, using a large undergra duate student sample, confirmatory factor analysis provided evidence for a three factor structure of political sk ill. Charisma and networking ability were identifi ed as unique factors of the po litical skill construct domain while PSI dimensions of social astuteness, interpersonal infl uence, and apparent sincerity collapsed to form a single dimension. Study One results also indicated a strong positive relationship between self-repor ts of political skill, charisma, and OCB. In Study Two, using a sample of public-s ector triads consis ting of professional level employees, their coworkers and their supervisors, mixed support was found for the convergent and divergent validity of the four PSI dimensions and charisma across reporting sources. As hypothesized, political sk ill predicted supervisor reports of overall job performance, task performance, and OCB. Charisma contributed to the prediction of supervisor ratings of overall performance a nd task performance afte r controlling for PSI total scores. At the dimensi onal level, social astuteness a nd charisma demonstrated the
Political Skill, Charisma, & Performance vii strongest predictive validity across all study cr iteria. Social astuteness and charisma also demonstrated a significant interaction when predicting supervisor ratings of overall performance and task performance. This intera ction indicated that so cial astuteness plays more of a role in predicting job performan ce for employees low in charisma than for employees high in charisma. As an addition to the second study, the ability of the PSI and charisma to predict performance ratings was compared against an abridged version of a situational judgment test assessing pract ical intelligence, the Tacit Knowledge Inventory for Managers (TKIM; Wagner and Sternberg, 1991). After controlling for PSI total scores and charisma, th e TKIM provided a modest cont ribution to the prediction of supervisor ratings of overall performance. Implications of these findings and directions for future research are provided.
Political Skill, Charisma, & Performance 1 Introduction Background on Organizational Polit ics & Political Behavior Since the early 1970s, interest in or ganizational politics has been rapidly growing. The increasing presence of ambiguous and dynamic work environments has forced organizations and managers to a dopt new approaches to resolving business problems. Environmental uncertainties have necessitated a shift in the way business is conducted and how most organizations are structured and operated (Cascio, 1995). Advances in technology and a prevalence of industries focused on product improvement, specialization, information shar ing, and customer service ha ve often made traditional systems of business obsolete. Old-fashion mechanistic organizations, limited in their ability to cope with turbulent business conditi ons, are espousing more organic structures that place emphasis on the use of human and in tellectual capital in meeting organizational goals. These organizations have flatter hierar chies, are less formal, and are more flexible in addressing complex work problems with seemingly ambiguous resolutions (Daft, 2004). To cope with changes in organi zational environments and structures, organizational politics have become recognized as an important and necessary channel through which power is distri buted, decisions are made, and work goals are realized (Pfeffer 1981; 1992). Echoing the words of Pf effer, we feel that in many cases organizational politics are the best and only way to resolve work conflicts or make organizational decisions.
Political Skill, Charisma, & Performance 2 Salancik & Pfeffers (1977) interpretati on of the strategic-co ntingency model of power helped fuel the rising interest in how politics affect workers and organizations. These authors have suggested that the distri bution of organizational power is contingent upon the problems most consequential to th e organizations survival. Accordingly, managers can gain power by obtaining control of critical work ac tivities and exploiting ways of completing these activit ies through the use of social ca pital (i.e. other people). In response to this theory, concerns have been raised over the misuse or abuse of power within organizations and the use of manipula tive or deceptive political behaviors to gain power. Wary of exploitation, re searchers warned against the dark side of political behavior (Ferris & King, 1991). Accordingl y, political behavior began taking on a negative connotation and was perceived by most organizations or HR administrators as behavior to be discouraged. Apprehension over the deceptive and debi litating role of orga nizational political behavior led to disagreements regarding the definition of the phenomenon. Several of the literatures emerging definitions referred to political behavior as self-serving or testing the ethical or procedural boundaries of an organiza tion (Culbert & McDonough, 1980; Farrell & Paterson, 1982; Ferris, Fedor, & King, 1994; Mayes & Allen, 1977; Pfeffer, 1981). Only recently have theorists considered political behaviors as being motivated by desires for improved outcomes for the self or for the organizati on (Ferris, Perrewe, Anthony, & Gilmore, 2000). We take the pers pective that political behavior can be performed to achieve self, group, and organi zational level objectives and that these objectives are not always mutually exclusive. Furthermore, we contend that regardless of
Political Skill, Charisma, & Performance 3 the targeted outcomes, political behavior can be executed in a style that may or may not preserve ethical or procedural standards of an organization. Rather than politics being inherently manipulative or expedient, we ar gue that managers must make the choice whether or not to use politics for appropriate purposes and al so choose whether or not to execute them in an a manner that is percei ved as preserving the so cial or procedural norms of a context. A second clarification we would like to ma ke regarding the definition of political behavior concerns the audiences these behaviors are targeted to influence. We contend that political behavior ca n be exercised up, down, and la terally across the chain-ofcommand. Likewise, we believe these behaviors can be used to influence others within the organization to which an employee belongs or across organizati ons with which an employee interacts. In other words, we believe that employees can use political behaviors to influence supervis ors, subordinates, or latera l colleagues within their own organization, and they can also use these behaviors to influence workers spanning all levels of external organizations with which they routinely conduct business. We believe that political behaviors are a ppropriate for any situation where the influence of others, regardless of their relative rank or organizational membership, has the potential to result in desired outcomes. It should be noted, how ever, that the selection and expression of political behaviors may vary greatly dependi ng on the status and association of the individual an employee is attempting to influence.
Political Skill, Charisma, & Performance 4 Political Behavior as Political Skill Academics and practitioners, alike, are beginning to frame politics in a more positive light. A building block for this shift was laid nearly 50 years ago with Thompson and Tudens (1959) quadratic-categor ization of decision situations. These authors proposed that the way decisions ar e made is contingent upon the agreement over organizational goals and how these goals should be realized. According to their model, with exception to situations where there is full agreement over what to do and how to do it, attempts to influence or the use politics will always emerge when decisions are being made. Pfeffer (1981) expanded on the work of Thompson and Tuden with his theoretical modeling of the conditions producing the use of power and politics in organizations. Pfeffers model contends that the use of politics in organizations is the response to conflicts over important decisions when th ere is a dispersion of power across decision makers. He argues that conflict will arise wh en resources are scarce, organizational units are interdependent, or there ar e discrepancies in work goals across units or departments. In such situations, managers need to use pol itics to lobby for access to resources or for decision-making power. Pfeffer asserts that when the conditi ons of his model are met, the use of power is virtually inevitable and furthermore, it is the only way to arrive at a decision (pg. 70). Consequently, those ma nagers with the will and the skill to use politics are most likely to achieve their pe rsonal and/or organizational goals. It is difficult to imagine any organi zation where employees agree over all work decisions and where power is distributed from a single autocratic s ource. Attempting to
Political Skill, Charisma, & Performance 5 minimize major conflicts regarding the fairness of resource distributi on is a healthy goal for any organization; however, it is overly optimistic to be lieve organizations can avoid all conflicts that trigger the expression of political behavior. Accordingly, the inescapable presence of political environmen ts within organizations creates the demand to focus on how politics are executed rather than focusing only on how to minimize, condemn, or avoid them. A shift in how we conceptualize the use of politics must be supplemented with a shift in how we measur e, acknowledge, reward, and train political behavior. Answering this call, Ferris and his co lleagues (1999) have initiated a line of research treating the appropriate use of politic al behaviors as a skill-set indicative of good performance and successful outcomes rather than as actions detrimental to organizational functioning. These researchers define politic al skill as: The ability to effectively understand others at work, and to use such know ledge to influence othe rs to act in ways that enhance ones personal and/or or ganizational objectives (Ahearn, Ferris, Hochwarter, Douglas, & Ammeter, 2004: 311). Acknowledgement of political behavior as a skill-set and a viable business tool ha s opened many doors for HR interventions that were not explored while politics remained stig matized as an organizational ailment. By identifying and encouraging desired political behaviors, HR administrators can use the assessment of political skill for purposes of recruitment, selection, training, and managerial development. The Political Skill Inventory (PSI) was deve loped to target four key dimensions of desired political behaviors: so cial astuteness, interpersonal influence, networking ability,
Political Skill, Charisma, & Performance 6 and apparent sincerity (Ferris et al., 2005). The PSI is an expansion of a unidimensional, six-item measure of political skill introduced by Ferris et al. (1999) focusing primarily on the diagnosis of political a udiences and self-efficacy with building rapport. After stringent item-reduction procedures and c onfirmatory factor-analysis methodology, the PSI has emerged as a four-factor 18-item m easure providing a detailed assessment of the political skill construct domain. The first two dimensions of political skill, social astuteness and interpersonal influence, asse ss an individuals ability to read and understand social situations a nd select the most appropriate and influential behavioral strategies to suit those situations. These elements of political sk ill are similar to the characteristics of social inte lligence. Researchers studying the applications of social intelligence argue that effective leaders need to exercise social perceptiveness and behavioral flexibility when dealing with soci al interactions in the workplace (Zaccaro, Gilbert, Thor, & Mumford, 1991). In other wo rds, good managers need to be able to discriminate between social contexts and know how to monitor behavior depending on the demands of a specific context. Diagnosing situations and selecting suita ble behavior does well to describe the ability component of political skill. The final two dimensions of political skill, networking ability and apparent sincerity, assess how this ability is utilized to achieve positive outcomes for the individual or organiza tion. Politically skil led individuals are said to be masters of the quid pro quo accomplished in the ar t of negotiation, deal making, coalition building, and conflict resolution (Ferris et al., 2005). Pfeffer (1992) argues that successful managers strategi cally position themselves within the
Political Skill, Charisma, & Performance 7 communication network, develop powerful alli es, and build rapport with those who have access to resources. All of these behaviors are focused on the maintenance of networks and are geared toward increas ing resources through the shar ing of assets and cooperation of powerful individuals. Networking with supervisor s, coworkers, and outside constituents is the most overt political be havior observed in managers. In fact, networking ability corre lates more strongly with manageri al influence tactics including upward appeal (i.e. obtaining the support of i ndividuals higher up in the organizational hierarchy; r = .30), coalition building (i.e. obtaining the support of subordinates or coworkers to reinforce a position taken or a request for resources; r = .31), and assertiveness (i.e. demanding, ordering, setti ng deadlines, and checking up on others in order to exercise influence; r = .18) than the other dimensi ons of the PSI (Ferris et al., 2005). Politically skilled individuals enhan ce their ability to build connections, coalitions, and alliances by app earing to be sincere and genu ine in their intentions and aspirations. Apparent sin cerity could be coined the execution or delivery factor of political skill. Appropriate influence tactics or political behaviors will only be successful to the extent they are perceived as bei ng genuine and devoid of personal motives or hidden agendas. Followers and collaborators, alike, will be more likely to increase their commitment to an idea or be influenced by an individual when they feel they are not being manipulated or bullied. An employee perceived as being insincere will be less successful in political interactions regard less of how well he reads situations and understands what behavioral strategies are mo st effective across different contexts.
Political Skill, Charisma, & Performance 8 Using a four factor model, the PSI combines elements of several social effectiveness constructs (e.g. social inte lligence, self-monito ring, tacit knowledge, emotional intelligence, ego-resiliency, soci al self-efficacy, and self-monitoring) into a concise 18-item measurement of political behavior (Ferris, Perrewe, Anthony, & Gilmore, 2000). Politically skille d individuals exercise what Cu lbert (1996) refers to as a mind-set orientation in determining how to interact with those they wish to influence. Politically skilled managers build lasting re lationships with stakeh olders and influence audiences with diverse interests by analyzing political arenas, choosing strategies aligned with audience expectations and styles, and by demonstrating behavioral flexibility and genuineness in the executi on of these strategies. Validation Evidence for the PSI The first step in determining the utility of a new measure is to examine how well it converges with similar constr ucts and discriminates from different constructs, and to test how well it predicts organi zational outcomes. Given that political skill is a relatively new construct to the I/O and organizational be havior literature, th ere hasnt been an abundance of validation research conducted us ing the PSI. However, the research that does exist has consistently found positive resu lts regarding the divergence of the PSI from other measures of social skill or intell ectual abilities and its ability to predict job performance (Ahearn et al., 2004; Ferris et al., 2005; Semadar, R obins & Ferris, 2006). A major concern regarding the uniqueness of any social effectiveness measure is the extent to which it diverges from general mental ability (GMA). The meta-analytical work of Hunter and Hunter (1984) solidified general inte lligence as the highest order
Political Skill, Charisma, & Performance 9 predictor of job performance w ith reports of a corrected m ean correlation of .51 between the General Aptitude Test Battery (GATB) and overall performance for a cumulative sample of approximately 32,000 employees acro ss 512 job classifications. To satisfy critics of social effectiveness and practical intelligence measures (e.g. Schmidt & Hunter, 1993), these types of scales need to demons trate weak correlations with GMA. The divergence of political skill from GMA was fi rst evidenced by Ferris et al. (1999) when these researchers found a negative, nonsignifi cant correlation (r = .08) between the sixitem measure of political skill a nd a measure of GMA. To our knowledge, research using the 18-item PSI has yet to report how the sc ale correlates with GMA. Although it is beyond the scope of the current proposal, more research is needed to replicate the divergence of political skill from GMA that was demonstrated by the 6-item scale. Ferris et al. (2005) reported modest corre lations between PSI total scores and an array of variables falling under the umbrella of personality and social effectiveness including self-monitoring ( r = .39), conscientiousness ( r= .31) trait anxiet y (r = .31), and political savvy (r = .47). These authors also reported low to modest correlations between PSI total scores and three of Kipnis et al.s (1980) influence tactics including reports of upward appeal (r = .25), efforts to build coalitions (r = .21), and assertiveness (r = .09). While these correlations suggest some construct overlap with political skill, the relationships are sufficiently weak to dism iss concerns of construct redundancy. There have been relatively few criter ion-related validity st udies assessing the ability of the PSI to predict job performanc e. In a sample of public-sector casework teams, Ahearn et al. (2004) showed that le ader political skill, measured using the
Political Skill, Charisma, & Performance 10 unidimensional 6-item scale, was correlated si gnificantly with an objective measure of team performance (r = .19) and accounted for an additional 3% of the performance variance after controlling for several vari ables including team member experience and leader experience. Similarly, using the s hort scale, Higgins (2000) reported a positive link between political skill and recruitment in terviewer ratings and evaluations of job applicants. Since the expansion of the PSI to four dimensions and 18 items, researchers have reported stronger relationships between political skill and performance measures. Semadar, Robbins, and Ferris (2006) reporte d a correlation of .34 between PSI selfreports and supervisory ratings of job perfor mance in a sample of 400 managers from a large Australian automotive manufacturer. More importantly, this study demonstrated that PSI self-reports accounted for 85% of th e variance explained in performance when self-monitoring, emotional intelligence, and le adership self-efficacy were also included as predictors in the regressi on model. In a different study, Ferris et al. (2005) reported significant R2s when regressing effectiveness ratings and job performan ce ratings on PSI total scores using two samples spanning public and private sectors of industry. Although the criterion-related validity research for the PSI is limited due to the infancy of the construct, the existing research provides initial support for the PSIs ability to predict job performance. A goal of the current research wa s to further validate the predictive utility of the PSI by examining the link between political skill and multi-source reports of organizational citizenship behavior (OCB) and task performance.
Political Skill, Charisma, & Performance 11 Political Skill & Tacit Knowledge Indeed, given the limited number of exis ting studies, more validation research is needed to provide additional evidence of th e PSIs value as a sele ction or development tool. Existing research has shown the PSI to outperform several measures of social effectiveness including emo tional intelligence, self-mon itoring, and leadership selfefficacy in the prediction of job performance (Semadar, Robbins, & Ferris, 2006). However, researchers have yet to compare the predictive power of the PSI against a measure of tacit knowledge. In light of Ferris et als (2000) claim th at tacit knowledge is explained within the parameters of political skill, the PSI should converge with tacit knowledge as well as compliment its ability to predict job performance. To examine this hypothesis, our second study competitively tested the predictive power of the PSI against that of a well-validated measure of practical intelligence, the Tac it Knowledge Inventory for Managers (TKIM) (Wagner & Sternberg, 1991). Tacit knowledge is defined as practi cal know-how that rarely is expressed openly or taught directly (O xford University Dictionar y, 1933; taken from Wagner & Sternberg, 1991, p. 1). The constr uct, popularly referred to as street smarts, has been categorized into three distinct managerial dimensions: an employees ability to manage the self, manage tasks, and manage coworker s. The dimensions of self and coworker management draw similarities with the cons truct of political skill. Wagner and Sternberg (1990) argue that managers high on practical intelligence understand self motives and organizational strategies, and have an extensive know ledge of how to finesse subordinates, peers, and supervisors. Si milarly, politically sk illed individuals use
Political Skill, Charisma, & Performance 12 information about themselves, their coworkers, and work contexts to choose and execute effective political behaviors. The TKIM has been shown to predict supervisor ratings of performance (rs = .29-.56) in several studies using diverse academic and business samples (Wagner & Sternberg, 1991). In a sample of busi ness executives, Wagner and Sternberg (1990) reported that the TKIM was the single best predictor of performance on a managerial simulation, accounting for an additional 32% of the variance in performance after controlling for GMA. Tacit knowledge resear chers have consisten tly shown divergence of the TKIM from measures of general ment al ability or verbal reasoning (rs = .02-.30) using academic, managerial, and military sa mples (Hedlund, Forsyth, Horvath, Williams, Snook, & Sternberg, 2003; Wagner & Sternberg, 1990; Wagner & Sternberg, 1991). Tacit knowledge and political skill theorists a like have argued for the importance of experience or reputation in predicting th e acquisition and successful expression of the two constructs (Wagner & Sternberg, 1990; Ferris, Perrewe, & Douglas, 2002). However, only the TKIM has been shown to correlate significantly (r = .30) with managerial experience (Wagner & Sternberg, 1991). This may be because the mechanisms through which time and experien ce influence these variables are quite different. It could be ar gued that tacit knowledge is learned implicitly or informally as workers experience increasingly diverse contexts and are forced to make decisions across different situations. Political skill, on the other hand, may only develop over time if an individuals reputation as be ing resourceful and cooperative is established. Ferris, Perrewe, and Douglas (2002) argue that the progression of a managers reputation as an
Political Skill, Charisma, & Performance 13 important and fair player, maintaining rappor t with coworkers will ultimately result in more opportunities requiring the use of po litical skill. While tacit knowledge places emphasis on the role of work experiences, political skill focuses only on those experiences where employees are required to negotiate, collaborate, or influence others while protecting their image and reputation. Political skill wi ll only develop over time if a manager has the motivation or desire to manage impressions in efforts to be received favorably, and have the aspirati on to engage in political fo rums. Tacit knowledge, on the other hand, should grow with experience regard less of a managers st ylistic preferences for managing. Until a comparison of the TKIMs and the PSIs criterion-related validity was conducted, we didnt believe there were gr ounds for making hypotheses regarding which inventory would be the better predictor of performance. It should be noted however, finding comparable predictive power between the two measures would provide strong support for the predictive efficiency of the PSI, containing only 18 items as opposed to the TKIMs 90; not to mention the ease in scoring the PSI in comparison to the complex scoring methodology for the TKIM that is comm on among situational judgment tests. In any case, given the similarities between the two construct domains, we believe that PSI and TKIM total scores will be positively correlated. Hypothesis 1: PSI total scores will correlate positively and significantly with TKIM total scores.
Political Skill, Charisma, & Performance 14 Political Skill & Charisma Charisma was first introduced as an im portant contributor to leader success and organizational performance over 80 years ago with Webers (1925) development of the three dimensional typology of ideal author ity structures. According to Weber, charismatic authority utilizes a leaders creativity, character, consideration, and extraordinary qualities in efforts to make organizationa l changes, motivate workers, and achieve organizational goals. Organizationa l behaviorists and i ndustrial psychologists finally adopted the construct of charis ma in the 1970s and 80s by introducing and examining several leadership theories focu sing on charisma as a core antecedent to leadership success (e.g. Bass, 1985; Bu rns, 1978; Conger & Kanungo, 1987; House, 1977; Roberts, 1985). There has been disagreement regard ing the definition of charismatic and transformational leadership acros s these theories. However, they all refer to good leaders as demonstrating elements of individualized consideration, vision ar ticulation, intellectual stimulation, behavioral flexibi lity, and a capacity to challenge the status quo in search of improved methods of operation or decision ma king (Lowe, Kroeck & Sivasubramaniam, 1996). Because the current study is concerned with the initial deve lopment of a measure of employee charisma, we adopt a broad defini tion of charisma consistent with Burns (1978) conceptualization of tr ansformational leadership. We define the charismatic employee as an individual who engages with coworkers or subordi nates in such a way that all stakeholders involv ed achieve a higher level of motivation and commitment through mutual support in efforts to reach a common goal. We stipulate that the positive
Political Skill, Charisma, & Performance 15 outcomes of charisma transcend through the ar ticulation of a collec tive vision, flexibility in decision making, inspir ational communication, beha vioral role-modeling, and commitment to the proposed mission. Despite criticism over inconsistencies in the conceptualizations of charismatic and transformational leadership th eories (e.g. Yukl, 1989), charis ma has been consistently predictive of leader effectiveness across multiple criteria (e.g. performance ratings, subordinate satisfaction with the leader, & subordinate motiv ation). These findings have proven to be stable across alternative meas ures of charismatic and transformational leadership (Shamir & House, 1993; Bass & A volio, 1993). Charisma, a dimension of the five-factor model of transf ormational leadership, has been shown to account for the majority of the variance in studies asse ssing the measurement of transformational leadership (Bass & Avolio, 1993). Lowe, Kr oeck, and Sivasubramaniam (1996) reported a corrected mean correlation of .71 between the Multi-factor Lead ership Questionnaire subscale of charisma and ratings of leader e ffectiveness in a meta-analysis of 47 studies examining the relationship between tran sformational leadership and positive organizational outcomes. Although a vast amount of research promotes the positive effects of transformational attributes, like political skill, charisma can be detrimental to organizational outcomes if used in a manipula tive fashion for individual gains. Conger (1997: 215) warns of the darkside of leadership, suggesti ng that when a visionary leaders behaviors become exaggerated, lose touch with reality, or become vehicles for purely personal gain, they may harm the le ader and the organization. He argues that
Political Skill, Charisma, & Performance 16 successful and genuine leaders align their strategic visions with environmental resources and stakeholder needs, avoid fundamental attribution errors, and communicate with constituents openly and honestly. Consistent with this stream of logic, Yorges, Strikland, and Weiss (1999) found that leaders were more influential when making personal sacrifices in efforts to secure their visions This effect was augmented when followers perceived the leader as being charismatic and sharing collective interests. While selfsacrifice and personal accountability has been recognized as an important antecedent for the successful display of charisma (C onger & Kanungo, 1987), only re cently have these concepts been included in measures of char isma or transformationa l leadership (Conger & Kanungo, 1994; Strange & Mumford, 2002). In the current study, weve included several items evaluating self-s acrifice and accountability in the initial item pool of our employee charisma scale. It will be interes ting to see if these items differentiate from those items on the PSI assessing ap parent sincerity. Its reas onable to assume that those individuals who are perceived as being genuine and sincere would also be perceived as being accountable for their actions. Ferris, Davidson, and Perrewes recent book on political skill (2005) addresses the theoretical link between charisma and the su ccessful display of political skill. These authors frame charisma as the stylistic mechanism through which employees or managers convey political behavior. Rather than treating charisma and political skill as separate entities, they argue that political skill explains charisma in th at politically skilled leaders are effective because they astutely read contex ts, situationally adjust, adapt, and calibrate their behavior to create the desired image, leve rage their social capital to further reinforce
Political Skill, Charisma, & Performance 17 their image, and do all this in a sincere, au thentic and convincing wa y (pp. 167). Similar propositions linking charisma to social eff ectiveness have been made by Ashkanasy and Tse in their work on transformational leadersh ip and emotional intelligence. Emotional intelligence refers to the abil ity to recognize, access, gene rate, regulate, and understand emotions in social contexts to promote in tellectual and emotional growth (Mayer & Salovey, 1997). Under the premise that transf ormational leaders are successful in their display of charisma, Ashkanasy and Tse propos e that transformational leaders will be better equipped to engage followers emotionally, use emotional language and communication, effectively communicate thei r vision, understand and sympathize with followers needs, maintain closer relati onships with followers, appropriately use impression management techniques, and achieve higher levels of performance, follower satisfaction, and affective follower commitment. Although Ferris et al. (2005) argue that ch arisma is embedded in political skill, to our knowledge, there has not been an empi rical investigation te sting the covariance between charisma and the PSI dimensions. The first study proposed will reexamine the factor structure of the PSI and explore whether or not charisma provides a unique dimension that could be considered a fifth f acet of political skill. Though we believe that charisma will be correlated with each of the PSI dimensions, we also feel that the item content of the proposed charisma scale is supplemental to the be haviors assessed by the PSI. Hypothesis 2(a): Charisma will demonstrate a significant positive correlation with each of the four PSI dimensions.
Political Skill, Charisma, & Performance 18 Hypothesis 2(b): Charisma will be identified as a unique factor of political skill when factor analyzed with the four existing PSI dimensions. As mentioned above, an argument could be made for the overlap of the PSIs assessment of apparent sincerity and our conc eptualization of charisma. In particular, elements of charisma that demonstrate a w illingness to make self-sacrifices for goal attainment or an affinity for taking respons ibility for work outcomes should be linked to perceptions of genuineness. It is difficult to imagine an employee willing to forfeit personal gains for a collective vision while appear ing to be insincere in his or her actions. Though this proposal does not hypothesize a cau sal relationship between charisma and apparent sincerity, we feel that a reluctance to make self-sacrifices or take accountability for work outcomes would likely influence pe rceptions of an employees genuineness. Likewise, it is reasonable to assume that per ceptions of sincerity will be partly based on how well an employee can instill a vision a nd stimulate coworkers in an intellectual manner. It reasons that a charismatic manager sincerely communicates conviction, investment, and a desire for coworkers to share in the mutual gains of doing good work and achieving success. Accordingly, of th e four PSI dimensions, we believed that charisma would be most strongly related to appa rent sincerity. Hypothesis 2(c): Charisma will have a stronger positive correlation with apparent sincerity than with any of the other PSI dimensions. Researchers have established a positive link between political skill total scores and measures of self-monitoring (r = .33) and political savvy (r = .47) (Ferris et al., 2005). We included these two measures in the cu rrent research in an effort to assess their
Political Skill, Charisma, & Performance 19 convergence with charisma. Charismatic leaders have been acknowledged by some authors as more likely to utilize impressi on management techniques and more capable of identifying the environmental and emotional cues necessary for the successful execution of political behaviors or behaviors aimed at influencing or motivating others (Ashkanasy & Tse, 2000). Although self-monitoring measur es an individuals ability to manage impressions (Snyder, 1987), the political savvy f actor of socializati on (Chao et al., 1994) assesses an employees ability to identify the key players and political norms (i.e. environmental and social cues) within an or ganization. Consistent with the propositions of Ashkanasy and Tse, we believe that employees with higher levels of charisma will also demonstrate higher levels of self-mon itoring and political savvy. Hypothesis 3: Charisma will correlate significantly and positively with selfmonitoring and political savvy. Political Skill, Charisma, & OCB Since Organ (1988) introduced the construc t to the I/O literatu re nearly 20 years ago, Organizational Citizenship Behavior (OCB) has become widely acknowledged as an important topic of study in areas of sele ction, performance appraisal, and employee development. OCB, otherwise known as cont extual or citizenship performance, is defined as behaviors that shap e the organizational, social, and psychological context that serve as a catalyst for task activities and processes (B orman & Motowidlo, 1993: 71). A wealth of factor analysis research has congealed the factor structure of OCB as a parsimonious, three-factor construct assess ing employee personal support of coworkers, support for organizational norms and goals, a nd an employees displa y of conscientious
Political Skill, Charisma, & Performance 20 initiative (Coleman & Borman, 2000) (See Tabl e 1 for a complete taxonomy of the threefactor model of OCB). Not without debate, O CB researchers have agreed that contextual performance can be considered in-role work behaviors that contribu te to the successful functioning of an organization, and that their display may be recognized by organizational reward systems (Borman & Motowidlo, 1997; Organ, 1997; Organ & Paine, 1999). The positive influence of OCB on unit level performance and supervisor ratings of employee performance has been well establ ished and replicated across multiple studies using diverse samples and alternative criteria (MacKenzie, Podsakoff, & Ahearne, 1996; Podsakoff, Ahearne, & MacKenzie, 1997; Podsakoff & MacKenzie, 1994; Walz & Niehoff, 1996). In fact, Mo towidlo and Van Scotter (1994) found OCB to be just as important in predicting overal l performance ratings as employee task performance in a sample of 300 entry-level Air Force employ ees. Similar findings have been reported from several other studies, each supporting th e argument that OCB and task performance are commensurate in predicting an employ ees overall performance evaluation (Borman, White, & Dorsey, 1995; MacKenzie, Pods akoff & Fetter, 1991; Van Scotter & Motowidlo, 1996). Although it is evident that there is a link between OCB and both subjective and objective evaluations of performance, research has also provided support for the effects of OCB on the distribution of organizationa l rewards. Van Scotter, Motowidlo, and Cross (2000) found that citizenship performan ce was related to promotability ratings and the attainment of informal systemic reward s for two large military samples. Allen and
Political Skill, Charisma, & Performance 21 Rush (1998) found similar resu lts linking OCB to recommenda tions for salary increase, promotion, high profile projects, public r ecognition, and opportunities for professional development. In light of these findings, OCB has been shown to influence more than just organizational effectiveness; it also faci litates employees in the acquisition of organizational rewards and in effort s toward advancing ones career.
Political Skill, Charisma, & Performance 22 Table 1 Coleman and Bormans (2000) Taxonomy of Citizenship Performance (i.e. OCB) Personal Support Helping others by offering suggestions, t eaching them useful knowledge or skills, directly performing some of their tasks, and providing emotional support for their personal problems. Cooperating with othe rs by accepting suggestions, informing them of events they should know about, and putting team objectives ahead of personal interests. Showing considerati on, courtesy, and tact in relati ons with others as well as motivating and showing confidence in them. Subdimensions: Helping Cooperating Courtesy Organizational Support Representing the organization favorably by defending and promoting it, as well as expressing satisfaction and showing loyalty by staying with the organization despite temporary hardships. Supporting the organi zations mission and objectives, complying with organizational rules and proced ures, and suggesting improvements. Subdimensions: Representing Loyalty Compliance Conscientious Initiative Persisting with extra effort despite difficult conditions. Taking the initiative to do all that is necessary to accomplish objectives even if not normally a part of own duties, and finding additional productive work to pe rform when own duties are completed. Developing own knowledge and skills by taki ng advantage of opport unities within the organization and outside the organization using own time and resources. Subdimensions: Persistence Initiative Self-Development
Political Skill, Charisma, & Performance 23 Given the important role OCB has in predicting an employees overall performance or career success, it falls in line that any thorough investigation of a measures criterion-related valid ity should include an assessme nt of its relationship with OCB. Validity research for the PSI has ye t to formally investigate the correlation between the political skill and the OCB constr uct domain. In fact only one study testing the predictive power of the PSI has utilized a comprehensive, 28 category assessment of job performance (Ferris et al., 2005). Regrettably, thes e researchers did not report dimensional level analyses inspecting the relationships among the many facets of job performance and political skill. The PSI validity research is also lacking a full 360 degree assessment of the scales convergence across self, peer, and supervisor reports. Semadar, Robins, and Ferris (2006) have pr ovided initial evidence of the convergence between PSI self and supervisory reports (r. = .36) and the superior ability of PSI self reports to predict job performance (r = .34) over supervisor reports (r = .26). However, a study examining the predictive power of the PSI has yet to include peer reports or perform analyses looking at the correlations between polit ical skill and the different dimensions of job performance. The sec ond study of this proposal examined the relationships among self, peer, and supervisor reports of political skill, OCB, task performance, overall performance, and charisma. As noted earlier, there has been a popular shift in contemporary work environments away from classic top-dow n organizational structures plagued by interdepartmental barriers, in efficiencies in decision maki ng, and a mechanistic inability to effectively manage cross-functional work units. In response to an increasingly
Political Skill, Charisma, & Performance 24 dynamic business environment, many organizations have adopted organic structures characterized by interdependent departments, broadly defined jobs, flexibility in work tasks and responsibilities, subj ective reward systems, a constant challenging of the statusquo, and a focus on interpersonal relationships among coworkers (Daft, 2004). Due to their reliance on social intera ctions, organic organizations demand the careful expression of OCB from their workers. To be successf ul, employees need to show support for their peers, share in organizational pride and camaraderie, and enthusiastically manage ambiguous work projects. Ferris, Davidson, & Perrewe (2005) addre ss the connection betw een political skill and contextual performance. They argue th at the interpersonal na ture of OCB requires employees to be socially astute, flexible, and adaptable. In other words, they postulate that politically skilled employees are more likely to display OCB than workers who are inept in social or political relations. Consistent with th is reasoning, we believe that political skill will not only predict task or overall performance as it has in the past (e.g. Semadar, Robins, & Ferris, 2006), but it will al so predict multi-source reports of OCB. Furthermore, because the dimension of pers onal support represents several politically driven behaviors such as interpersonal cons ideration, cooperation, and collaboration, we proposed that political skill would be most effec tive in predicting this dimension of OCB. Hypothesis 4: PSI total scores will correlate si gnificantly and positively with self, coworker, and supervisory ratings of OCB, task performance, and overall job performance.
Political Skill, Charisma, & Performance 25 Borman & Motowidlo (1993) argue that di fferences in the conceptualizations of task performance and OCB constitute the need to consider performance antecedents in reference to what type of performance is being examined. Although GMA has been identified as be the best predictor of ove rall job performance (e.g. Hunter & Hunter, 1984), Borman and Motowidlo contend that an individuals dis position or social effectiveness should predict OCB better than GMA. Their claims have been supported through numerous research efforts demons trating positive relationships between OCB and personality measures incl uding conscientiousne ss, agreeableness, positive affectivity, locus of control, and prosocial personal ity (Borman, Penner, Allen, & Motowidlo, 2001; Organ & Ryan, 1995). Political skill, like O CB, has a dispositional component and has been shown to correlate signifi cantly with conscientiousness (r = .31) (Ferris et al., 2005). Due to the constructs social and dispositional nature, we expected political skill to do a better job of predicting OCB th an task performance. Hypothesis 5: The positive correlation be tween PSI total scores and coworker/supervisor OCB ratings will be stronger than the correlation between PSI total scores and coworker/superv isor task performance ratings. The transformational and charismatic lead ership literature has yet to empirically test the link between employee charisma and OC B. However, there has been a wealth of research reporting positive outcomes for l eaders exhibiting high charisma such as increased leadership effectiveness ratings and reports of increased subordinate satisfaction (e.g. Lowe, Kroeck, & Sivas ubramaniam, 1996). We theorize that charismatic behaviors will also be positively correlated with the expression of OCB.
Political Skill, Charisma, & Performance 26 Employees who wish to articulate a vision, in spire excellence, and gain the commitment of fellow coworkers need to show support for those they wish to influence, promote ideas globally through organization-wide channels and model performance ideals by going beyond general expectations, alwa ys being dependable, and ma intaining a constant focus on self-development. Consistent with our be lief that charisma is at least partially exclusive from the current PSI dimensions, we felt that the addition of charisma to the measurement of political skill will improve the PSIs ability to predict OCB. Although we did not make a hypothesis regarding ch arismas ability to account for additional variance in task or overall performance ratings beyond polit ical skill, exploratory analyses were be performed to test the unique contribution of charis ma to the prediction of both task and overall performance. Hypothesis 6: Charisma will demonstrate significant positive prediction of self, coworker, and supervisor reports of OCB ratings after controlling for the other PSI dimensions.
Political Skill, Charisma, & Performance 27 Method Study 1 Sample and Procedure A total of 1,445 undergraduate psychology studen ts at a large southeastern university completed online surveys. All participants who completed surveys were compensated with course credit. Complete data were obtained for 1,094 participants. The average age of the respondents in this sample wa s 20.63 (SD = 3.82), 74% were female, and 64% were part-time workers. 45% of the participan ts were in their junior or senior year of college. The 1094 participants were randomly split into two samples of 547 participants each. In an effort to preserve the statisti cal assumptions of fact or analysis, the first sample was used for exploratory analyses the second for confirmatory analyses. Measures Political skill. Ferris et al.s (2005) Political Skill Inventory (PSI) was used to assess political skill and its dimensions. Speci fically, the scale contains four dimensions including social astuteness (5 items), interper sonal influence (4 items), networking ability (6 items), and apparent sincerity (3 items). Respondents indicated the extent to which they agreed with each statement about th emselves using a 7 point Likert scale (1=Strongly Disagree, 7=Strongly Agree). The coefficient alphas for each of the four PSI dimensions were .85 for social astuteness, .88 for interpersonal influence, .87 for
Political Skill, Charisma, & Performance 28 networking ability, and .87 for a pparent sincerity. See Appendi x A for a complete list of the PSI items. Charisma. A total of 28 items were generate d to assess the cons truct of employee charisma. Several of these items were modified from already existing measures of charisma (Conger & Kanungo, 1994; Strange & Mu mford, 2002). Charisma reflects the ability to communicate high expectations, inst ill confidence, inspire others to reach high goals, communicate a sense of mission, and c onvey a powerful presence (Kudisch et al., 1995). Using the same rating format as th e PSI, respondents indicated the extent to which they agreed with each statement about themselves using a 7 point Likert scale (1=Strongly Disagree, 7=Str ongly Agree). The coefficient alpha for the original charisma scale was .96. A copy of the 28 items generated to assess charisma is included in Appendix B. Organizational citizensh ip behavior (OCB). OCB was measured using Motowidlo and Van Scotters (1994) 16-item s cale (See Appendix C). The scale contains three dimensions including conscientious initia tive (6 items), persona l support (5 items), and organizational support (5 items). Scale instructions were modified to solicit selfreports of OCB rather than supervisor ratings of subordinate perfor mance. To maintain consistency across study measures, individuals indicated the extent to which they agree with each statement using a 7 point Like rt scale (1=Strongly Disagree, 7=Strongly Agree). The coefficient alpha for OCB was .94. Self-monitoring. Self-monitoring was measured using Snyders (1987) 18-item scale. Self-monitoring refers to the extent to which individuals m onitor and control how
Political Skill, Charisma, & Performance 29 they present themselves in social situations. The coefficient alpha for the scale was .59. A copy of the measure can be found in Appendix D. Political savvy. Political savvy was measured usi ng Chao et al.s (1994) six-item instrument (See Appendix E). This scale a ssesses an individual s understanding of the existence and workings of po litics within an organization. The coefficient alpha for political savvy was .76. Social desirability. The Strahan and Gerbasi (1972) 10-item measure was used to assess social desirability. Social desirability refers to extent to which individuals desire to be perceived positively by those with whom they interact. The coefficient alpha for the scale was .55. See Appendix F for a complete list of the social desi rability items. Results Item analyses. Because our interest was to further develop the PSI to include an additional dimension of charisma, we wa nted only the most representative and parsimonious set of items assessing charisma to be included in the final scale. Accordingly, a three-step item reduction pro cedure was utilized. Consistent with the methodology employed by Ferris et al. (2005), charisma items were first eliminated if they failed to express sufficient item-tot al correlations. Foll owing the recommended cutoffs of Nunnally (1978), only those items with item-to-total correlations of .40 or greater were retained for factor analysis and cross validation. This resulted in the elimination of one item (i.e. Item 1 in the appendix). Second, charisma items were eliminated if they correlated higher than .10 with social-desirability total scores. Researchers have become increasingly critical of self-
Political Skill, Charisma, & Performance 30 report measures of bio-data, personality, or social skill that are vulnerable to socially desirable responses, especially when thes e measures are being used for employee selection (e.g. Smith, Hanges & Dickson, 2001; Snell & McDaniel, 1998; Stokes, Hogan & Snell, 1993). The goal in this study was to create items of charisma that have a low susceptibility to participant faking. Though we originally hope d to eliminate all charisma items demonstrating a significan t correlation with social-desirability, the large sample utilized for this study caused correlations greater than .083 to become significant when using a two-tail test of significance ( = .05). Accordingly, we set a cutoff of r = .10 as the decision rule for eliminati ng items based on their relationshi p with social-desirability. This resulted in the eliminat ion of an additional 4 items (i.e. Items 16, 21, 26, and 28 in the appendix). Third, a principal axis fa ctor analysis using Varimax rotation was conducted to identify and eliminate charisma items demons trating high cross-loadi ngs with the already existing political skill dimensions. Accordi ngly, items with loadings on charisma lower than .60, and items with loadi ngs higher than .45 on factors other than charisma were eliminated from confirmatory analyses a nd cross validation. Th is resulted in the elimination of 16 items (i.e. Items 2, 3, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 22, 23, 24, and 25 in the appendix). Ultimately, the three-st ep item reduction procedure yielded a set of 7 items that met the criteria for inclusion in confirmatory analyses. The seven-item charisma scale had a coefficient alpha of .89, and item-total correlations ranging from .72 to .81. T-tests were computed on charisma tota l scores to assess if there were significant mean differences between the ratings provided by males and females. The t -test results
Political Skill, Charisma, & Performance 31 indicated no significant mean differences. Additional t -tests also indicated no statistically significant mean differences among any of the study variables in regard to gender. Exploratory factor analysis. The principal axis factor analysis that was performed as part of the charisma item-reduction proce dure also served as our exploratory factor analysis in regard to the political skill and charisma. In our initial analysis, we followed the Kaiser-Guttman criterion of retaining factors by only extracti ng and retaining those factors with eigenvalues gr eater than 1.0. Though this me thodology has been criticized for over-identifying reliable factors (Zwick & Velicer, 1986) our first analysis only extracted 3 factors with eigenvalues greater than 1.0. Each of these factors had coefficient alphas ranging from .87 to .94. According to Cliff (1988), retaining only factors with eigenvalues greater than 1 is a credible appro ach if the factor components demonstrate strong reliability coefficients. A total of four principal axis factor analyses were performed to eliminate charisma items. Each of the analyses yielde d a three-factor soluti on. The final analysis included the 18-item PSI scale and a 7-item char isma scale. As shown in Table 2, factor eigenvalues ranged from 1.33 to 12.58, with 63% of the total vari ance explained. The pattern of factor loadings indicate that the PSI dimens ions of social astuteness, interpersonal influence, and networking abi lity collapsed to form the first factor, explaining 50.31% of the variance in the mode l. Factor 2, charisma, explained 7.70% of the variance. The PSI dimension of networ king produced the third factor and explained an additional 5.33% of the variance.
Political Skill, Charisma, & Performance 32 Table 2 Exploratory Factor Analysis : Rotated Factor Loadings & Initial Eigenvalues Items Intended Dimension Factor 1 (SA) + (II) + (AS) Factor 2 Charisma Factor 3 Networking Ability 1. I always seem to instinctively know the right thing to say or do to influence others. Social Astuteness .44 .21 .49 2. I have a good intuition or savvy about how to present myself to others.Social Astuteness .66 .25 .39 3. I am particularly good at sensing the motivations and hidden agendas of others. Social Astuteness .50 .32 .28 4. I pay close attention to peoples facial expressions. Social Astuteness .76 .28 .14 5. I understand people very well. Social Astuteness .70 .24 .25 6. It is easy for me to develop good rapport with most people. Interpersonal Influence .77 .23 .28 7. I am able to make most people feel comfortable and at ease around me. Interpersonal Influence .73 .16 .30 8. I am able to communicate easily and effectively with others. Interpersonal Influence .71 .26 .31 9. I am good at getting people to like me. Interpersonal Influence .70 .18 .34 10. I spend a lot of time and effort at work networking with others. Networking Ability .16 .16 .77 11. At work, I know a lot of important people and am well connected. Networking Ability .27 .20 .67 12. I am good at using my connections and networks to make things happen. Networking Ability .29 .23 .73 13. I have developed a large network of colleagues and associates at work who I can call on for support when I really need to get things done. Networking Ability .25 .39 .60 14. I spend a lot of time at work developing connections with others. Networking Ability .22 .34 .63 15. I am good at building relationships with influential people at work. Networking Ability .52 .33 54 16. It is important that people believe I am sincere in what I say and do. Apparent Sincerity .79 .20 .15 17. I try to show a genuine interest in other people. Apparent Sincerity .78 .27 .19
Political Skill, Charisma, & Performance 33 Table 2 (Continued) Items Intended Dimension Factor 1 (SA) + (II) + (AS) Factor 2 Charisma Factor 3 Networking Ability 18. When communicating with others, I try to be genuine in what I say and do. Apparent Sincerity .80 .24 .12 19. I have vision and often bring up ideas about possibilities for the future. Charisma .28 .70 .25 20. I provide inspiring strategic and organizational goals. Charisma .29 .70 .29 21. I consistently generate new ideas for the future of the organization. Charisma .10 .80 .26 22. I take into account the needs of the organization when making my work decisions. Charisma .44 .62 .21 23. I try to positively reward or reinforce coworkers for performing in line with my goals. Charisma .37 .66 .19 24. I delegate authority to my coworkers regarding work tasks in line with my goals/vision. Charisma .14 .72 .18 25. I demonstrate to my coworkers how committed I am to my ideas. Charisma .45 .66 .23 Initial Eigenvalue 12.17 1.92 1.33 Percentage of Variance Explained 50.31 7.69 5.33 Coefficient Alpha .94 .90 .87 Note. N = 547; SA = Social Astuteness; II = Interpersonal Influence; AS = Apparent Sincerity Extraction Method: Principa l Axis Factor Analysis Rotation Method: Varimax with Kaiser Normalization Confirmatory fit statis tics and alternative models. The 25 items used in the final exploratory analysis (18 PSI items; 7 Charis ma items) were included in a confirmatory factor analysis using the principal axis met hod and oblique, direct oblimin factor rotation. We first tested the fit of the three-factor representation of political skill and charisma that was extracted during our exploratory anal yses. Upon the recommendation of Hair,
Political Skill, Charisma, & Performance 34 Anderson, and Tatham (1997) oblique factor rotation was used as an alternative to orthogonal rotation because of the fewer constraints it imposes early in scale development. We used structural equati on modeling software (Lisrel 8; Joreskog & Sorbom, 1993) to perform the 3 factor confir matory analysis. The same software was used to test the plausibility of a five-factor model treating the four PSI dimensions and charisma as separate factors. All analys es were performed using covariance matrices extracted from SPSS data worksheets (SPSS 11.5, 2003). Several recommended measures of model f it were used including the Normed Fit Index (NFI), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Non-Normed Fit Index (NNFI), Goodness of Fit Index (GFI), the Adjusted Goodness of Fit Index (AGFI), the root mean squared error of approxima tion (RMSEA), the standardized root mean squared residual (SRMR), and the ratio of chisquare relative to th e degrees of freedom ( /df) (Hu & Bentler, 1999; La Du & Tana ka, 1989; Tucker & Lewis, 1973; Wheaton, Muthen, Alwin, & Summers, 1977). It is suggested that the CFI, NFI, NNFI and GFI should be higher than .90 for the tested model to have demonstrated acceptabl e fit (Hatcher, 1994; Medsker, Williams, & Holohan, 1994; Mulaik, James, Van Alstine, Benn et, Lind, & Stillwell, 1989). It is also recommended that an AGFI higher than .80 (Gefen, Straub, & Boudreau, 2000), an RMSEA lower than .06, an SRMR lower than .08 (Hu & Bentler, 1999), and values less than five for the /df ratio should be obtained in order to infer acceptable levels of model fit (Wheaton et al., 1977).
Political Skill, Charisma, & Performance 35 The fit statistics for the three-factor and the hypothesized five-factor model are provided in Table 3. The thre e-factor solution demonstrat ed reasonable fit by meeting the recommended fit cutoffs fo r 6 of the 8 fit indices. Only the GFI (.84) and the RMSEA (.08) values failed to meet the r ecommended criteria; however, the values for both of these indices approached the desired cutoffs. Based on these findings, we believe the three-factor model achieves reasonable fit and should be considered a plausible representation of the pol itical skill and charisma construc t domain. Since charisma was identified and confirmed as the second fact or in the model, we supported hypothesis 2(b), which made the prediction that charisma would be extracted as an additional or unique factor in the model beyond th e four PSI dimensions. When testing the 5 factor model as an a lternative to the 3 factor model, we found remarkably similar results in regard to fit statistics (See Table 3). The 5 factor model also met the recommended cutoffs for 6 of the 8 fit indices, again, only falling short in regard to RMSEA and the GFI. Based on th ese findings, the 5 factor model should also be considered as a reasonable representati on of the political skill/charisma construct domain. Differences in fit statistics between the two competing models were marginal. The 5 factor model did demonstrate a modestly favorable ratio of Ch i-Square to degrees of freedom, while the 3 factor model had slig htly superior GFI and AGFI indices. These modest differences in fit statistics ma ke it difficult to select one model as better fitting over the other. Nonetheless, the 3-factor pr ovides a less restrictive representation of the data and was also extracted as the predicte d model during exploratory analyses. Until additional research allows for a firm judgment of model superiority, we tend to favor the
Political Skill, Charisma, & Performance 36 less restrictive representation of the constr uct. Based the present studys results, however, both models should be given credence as possible conceptualizations of the political skill and charisma construct domain. Table 3 Model Fit Statistics for 3 and 5 Factor Models of Political Skill & Charisma Fit Indices 3-Factor Model 5-Factor Model Comparative Fit Index (CFI) .97 .98 Normed Fit Index (NFI) .97 .96 Non-Normed Fit Index (NNFI) (or Tucker-Lewis Index) .97 .97 Goodness of Fit Index (GFI) .84 .82 Adjusted Goodness of Fit Index (AGFI) .81 .79 Root Mean Square Error of the Approximation (RMSEA).08 .08 Standardized Root Mean Squared Residual (SRMR) .06 .06 Ratio of Chi-Square to Degrees of Freedom ( 2/ df ) 4.27 2.41 Measure reliabilities and correlations. Internal consistency was estimated for all study variables using Chronbach s reliability esti mate. With the exception of selfmonitoring ( = .57) and social desirability ( = .54), all reliability estimates exceeded the .70 level recommended by Nunnally (1978) w ith coefficient alphas ranging from .76 for political savvy to .95 for PSI total scor es. Pearson product-moment correlations between all study variables were computed and are presented in Table 4. The four PSI dimensions demonstrated strong convergence with each other as evidenced by intercorrelations ranging from .57 ( < .001) to .81 ( < .001). Similarly, as predicted by hypothesis 2(a), charisma also demonstrated a positive and significant correlation with each of the four PSI dimensions ranging from .56 ( < .001) with apparent sincerity to .70
Political Skill, Charisma, & Performance 37 with networking ability. A ccording to its operational definition, charisma involves gaining the commitment of others to a proposed vision. Since gaining the support of others relies heavily on the establishment of trust and others belief in the purpose of the vision, charisma was hypothesized to be more st rongly related to appa rent sincerity than the other three PSI dimensions (See Hypothesis 2c). Howe ver, we did not find support for this prediction. Nonetheless, the pattern of positive correlations between charisma and all of the PSI dimensions indicate a ra ther strong convergence be tween the constructs of political skill and charisma. It should be noted, however, the convergence of the PSI dimensions with PSI total scores (rs = .84 to .91, < .001) was more pronounced than the convergence of charisma w ith PSI total scores (r = .72, < .001). Although there does seem to be overlap between the four f actor measure of political skill and our7 new measure of charisma, tests of discriminant validity need to be evaluated before a conclusion can be made regard ing construct redundancy. Hypothesis 3 predicted that charisma w ould be positively correlated with both political savvy and self-monitoring. We f ound partial support for this prediction. Although charisma correlated significan tly with political savvy (r = .58, < .001), no relationship was found between char isma and self-monitoring (r = -.03, = .317). PSI total scores correlated positively and signi ficantly with both political savvy (r = .74, < .001) and self-monitoring (r = .06, < .05). However, given the magnitude of the correlation between the PSI and self-mon itoring, the relationshi p between the two variables could be considered marginal at be st. As expected, we found initial support for our prediction (Hypothesis 4) that PSI total scores would be positively correlated with
Political Skill, Charisma, & Performance 38 self-reports of OCB (r = .46, < .001). Similarly, each of the four PSI dimensions as well as charisma were found to be significan tly and positively relate d to self-reported OCB, with correlations ranging from .39 ( < .001) for networ king ability to .49 ( < .001) for charisma.
Political Skill, Charisma, & Performance 39 Table 4 Correlations and Descriptive Stat istics for Study 1 Measures Measures M SD 1 2 3 4 5 6 7 8 9 10 1. Political Skill Total 5.27 1.01 (.95) 2. Social Astuteness 5.28 1.07 .91 (.85) 3. Interpersonal Influence 5.54 1.21 .92 .81 (.88) 4. Networking Ability 4.83 1.13 .86 .67 .68 (.87) 5. Apparent Sincerity 5.75 1.24 .84 .74 .79 .57 (.87) 6. Charisma 4.88 1.06 .72 .63 .59 .70 .56 (.89) 7. Political Savvy 5.04 .98 .74 .70 .64 .62 .64 .58 (.76) 8. Self Monitoring 1.55 .17 .06* .11**.06 .04 .00 -.03 .05 (.57) 9. Social Desirability 1.48 .20 .04 .02 .05 .04 .06 .09* .05 -.28 (.54) 10. OCB 5.53 .95 .46 .40 .40 .39 .44 .49 .38 -.05 .18 (.94) Note. Due to missing values, sample sizes range from 1,072 to 1,093. The values in parentheses re present the coefficient alphas for each of the measures. p < .05; ** p < .01; p < .001 for all correlations in bold
Political Skill, Charisma, & Performance 40 Regression analyses. To test our hypothesis that charisma would add to the prediction of self-reported OCB after contro lling for the four PSI dimensions (See Hypothesis 6), we performed a series of hier archical regression an alyses. Specifically, we regressed OCB total scores on PSI total sc ores, the four PSI dimensions, and charisma in 5 separate, two-step regression analyses. In the first step of each analysis, either PSI total scores or a single PSI dimension was en tered into the regression equation; charisma was then entered in the second step of th e analysis. Following the procedures of Pedhazur (1997), F ratios were computed to test for significant increases in R2 between the two regression models. Before conducting dimensional analyses, we first performed a preliminary outlier analysis regressing se lf-reported OCB onto PSI total scores and charisma. Researchers have recommended trea ting data cases with studentized deleted residuals higher than 2 standard deviations as statistical outliers (Pedhazur & Schmelkin, 1991). Taking a conservative approach to el iminating cases, we only removed those cases with studentized dele ted residuals grea ter than 2.5 standard deviations. Accordingly, we identified and removed 18 out liers (1.6% of total sample) from Study 1 regression analyses. As shown in Table 5, charisma predicted a significant portion of unique variance in self-reported OCB after cont rolling for individual PSI dime nsions and PSI total scores. More specifically, charisma explained an additional 6% of the variance in OCB selfreports beyond PSI total scores and an additional 10 to 11% of the variance when entered after individual PSI dimensions. These fi ndings provide initial support for our hypothesis
Political Skill, Charisma, & Performance 41 that charisma would significantly add to the prediction of self-reported OCB beyond the four-factor model of political skill. Table 5 Hierarchical Regression Analyses Predicting Organizational Citizenship Behavior (OCB) from Charisma after C ontrolling for Political Skill Step Predictors B F R2 R2 Dependent Variable: Self-re ports of OCB (N = 1,058) 1 Political Skill Total Score.487 .514 379.38 .264* 2 Charisma .285 .324 84.96 .319* .055* 1 Apparent Sincerity .366 .473 305.04 .224* 2 Charisma .335 .381 165.48 .329* .105* 1 Interpersonal Influence .361 .454 274.62 .206* 2 Charisma .346 .394 164.52 .313* .107* 1 Social Astuteness .404 .456 277.20 .208* 2 Charisma .345 .392 149.43 .306* .098* 1 Networking Ability .345 .420 225.99 .176* 2 Charisma .388 .442 157.88 .284* .107* Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to second step of the hi erarchical regression. p < .001 Although we did not make a hypothesis regard ing the predictive efficiency of the PSI or our newly developed charisma scale, we did perform a second series of regression analyses in order to compare each measure s ability to predict unique variance in selfreported OCB. In this second set of explorat ory analyses, we entered charisma into the regression equation in the first st ep of the analyses and PSI total scores and dimensions in
Political Skill, Charisma, & Performance 42 the second step. As shown in Table 6, PSI to tal scores only explai ned an additional 4% of the variance in OCB self-reports after we controlled for charisma. Similarly, increases in R2 ranged between .008 (networking ability) a nd .054 (apparent sincerity) when each of the four PSI dimensions were entered into the regression analyses after controlling for charisma. These findings, when compared agai nst the results of the first set of regression analyses, indicate that charisma consisten tly explains more unique variance in selfreported OCB than PSI total scores or individua l PSI dimensions. This is a particularly impressive finding when considering that char isma was measured with only 7 items while the PSI contains 18 items. In other words, charisma did a better job of predicting selfreported OCB even though the scale is less than half the length of the PSI. Table 6 Hierarchical Regression Analyses Predicting Organizational Citizenship Behavior (OCB) from Political Skill af ter Controlling for Charisma Step Predictors B F R2 R2 Dependent Variable: Self-re ports of OCB (N = 1,058) 1 Charisma .461 .525 401.11 .275* 2 Political Skill Total Score.275 .290 67.95 .319* .044* 2 Apparent Sincerity .211 .273 85.01 .329* .054* 2 Interpersonal Influence .187 .235 58.65 .312* .038* 2 Social Astuteness .195 .220 47.00 .306* .031* 2 Networking Ability .101 .123 12.11 .284* .008* Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to second step of the hi erarchical regression. p < .001
Political Skill, Charisma, & Performance 43 Study 2 Sample Data were collected as part of a wor kplace study for a large southeastern county government spanning 23 agencies and over 11,000 employees. A total of 495 respondents participated in the study. Th e respondents consis ted of 193 upper-level managers (supervisors), 169 subordinates (t arget workers), and 133 employees that worked closely with the target workers (c oworkers). Of the upper-level managers providing demographic data, 61% were male (N = 116), 75% were white (N = 143), 12% were black (N = 23), and 10% were of Hispanic decent (N = 19). The average organizational tenure for the supervisor sample was 17.16 years (SD = 8.49) and the mean tenure for their current position was 11.71 (SD = 6.15). On average, these respondents supervised target work ers for 5.18 years (SD = 3.97). Target workers spanned 20 functional ar eas of the county government with the strongest representation in areas including administration (13%), accounting and finance (11%), environmental protection (10%), e ngineering (9%), social services (9%), management (8%), and technological services (5%). Of the target workers providing demographic data, 46% were male (N = 76), 68% were white (N = 112), 18% were black (N = 30), and 7% were of Hispanic decent (N = 12). The average organizational tenure for the target workers was 14.54 years (SD = 8.70) and their mean job tenure was 8.79 years (SD = 5.36). 66% of the target worker s had direct supervis ion (i.e. managerial duties) of one or more employees (N = 111). The median salary for target workers was between $45,000 and $49,999 and 54% of the sample had a Bachelors degree or higher.
Political Skill, Charisma, & Performance 44 With regard to coworkers, 47% were male (N = 61), 72% were white (N = 93), 16% were black (N = 21), and 9% were of Hispanic decent (N = 12). The average organizational tenure for coworkers was 13.98 years (SD = 9.33) and their mean job tenure was 8.42 years (SD = 6.33). On average, coworkers had worked with the target ratee for 6.65 years (SD = 4.53). All coworker participants included in the study had worked with the target worker a minimum of six months. Procedure Cover letters and instruction sheets e xplaining the purpose of the study were sent via interoffice mail to 600 upper-level managers /supervisors working in their current job for a minimum of one year. S upervisors were responsible for selecting the target worker (subordinate) and coworker partic ipating in the study. They were instructed to choose a subordinate that reported di rectly to them and for whom they provided formal organizational performance evaluations. All subordinates chosen for the study had to meet the criteria of being in a managerial position or earning a minimum salary of $35,000. Supervisors were asked to distribute a target worker instruction sheet to the subordinate of they chose to participate in the st udy. They were also instructed to distribute a coworker instruction sheet to an employee that worked closely with their subordinate. Supervisor, target, and coworker instruction sheets ex plained the purpose of the study and directed participants to a web address hosting three separate links for online surveys. Participants accessed the online surveys by following the appropriate survey link and entering a participation code provided on their instruction sheet.
Political Skill, Charisma, & Performance 45 Participants were informed that the purpose for completing surveys was to identify important training objectives and to develop future training courses geared toward managerial development. They were asked to be candid in their responses and were assured of their anonymity. Complete data were collected from 193 supervisors (response rate of 32%), 169 target workers (response rate of 28%), and 133 coworkers (response rate of 22%). In all, complete data were co llected for 100 full participant triads. Measures Political skill. Political skill was measured usi ng the Ferris et al.s (2005) 18 item PSI that was used in Study 1. Using a seve n-point Likert response format, participants indicated the extent to which they agreed with each item (1 = Strongly Agree, 7 = Strongly Disagree). All items were modified from the first person to the third person to elicit appropriate responses from target worker s, supervisors, and coworkers. The same response format was employed to assess all st udy variables with th e exception of tacit knowledge. The coefficient alphas for each of the PSIs four dimensions ranged from .77 (social astuteness, target repor ts) to .93 (interpersonal influe nce, coworker reports) across target worker, supervisor, and coworker repo rts. The coefficient alphas for PSI total scores ranged from .92 for target repo rts to .96 for coworker reports. Charisma. Charisma was measured using the seven charisma items from Study 1 that were retained for factor analyses. Th e coefficient alphas for charisma were .85 for target workers, .89 for supervisors, and .91 for coworkers.
Political Skill, Charisma, & Performance 46 OCB. OCB was measured using five items from Borman, Ackerman, & Kubisiaks (1994) behaviorally-anchored ra ting scale (BARS) of job performance (See Appendix G). Target, supervisor, and cowork er ratings of initiative, adaptability, dependability, cooperation, and integrity were co mbined to create an OCB total score. Study results are reported at bot h the OCB composite level and at the dimensional level. The coefficient alphas for OCB total scor es were .68 for target workers, .81 for supervisors, and .85 for coworkers. Task & overall performance. Task performance was measured using 6 items from Borman et al.s (1994) job perfor mance BARS (See Appendix G). Target, supervisor, and coworker ratings of jo b knowledge, task proficiency, productivity, problem solving, and oral/wr itten communication were comb ined to create a task performance total score. Ratings of overall performance were obtained using a single item administered after all other performance dimensions had been rated. Study results are reported at both the task performance comp osite level and at the dimensional level. The coefficient alphas for task performance to tal scores were .68 for target workers, .79 for supervisors, and .77 for coworkers. Tacit knowledge. Tacit knowledge was measured using five of the nine situational stems from Wagne r and Sternbergs (1991) Taci t Knowledge Inventory for Managers (TKIM). The TKIM is a 90-item situational judgment test (SJT) asking respondents to rate the appr opriateness of action-items relating to nine separate situational vignettes using a 7point Likert scale. The TKIM was developed to assess the experience-based knowledge or prac tical intelligence of civilian managers. In the current
Political Skill, Charisma, & Performance 47 study, the five TKIM scenarios (51 items) involving the use of influence tactics or interpersonal interaction were administered to target workers. Due to the length of the TKIM and limited access to participant work time, we were only able to include the situation stems having the most conceptual overlap with the othe r key study variables (i.e. political skill and charisma). Only targ et workers completed the abridged version of the measure. TKIM total scores were computed by comparing individual target worker responses to consensus reference patterns on each item. Legree (1995) argues that using consensus reference patterns to score situationa l judgment tests is particularly appropriate when it is difficult to identify experts on the c onstruct of interest. He supports his claim with research demonstrating correlations rang ing from .72 to .95 between mean ratings of experts and nonexperts on a situational judgm ent test assessing tacit knowledge in a military sample (Legree, 1994). In the case of tacit knowledge, it would be extremely difficult to isolate subject matter experts with unique expertise in practical intelligence. Accordingly, using consensus reference patter ns to score the TKIM would be consistent with Legrees recommendations. In order to calculate TKIM total scores z-score transformations were computed for all target worker responses. Z-score tr ansformations control for response bias and do not punish participants for using response anch ors that are different from the consensus reference pattern (Legree, 1995; Legree, Mart in & Psotka, 2000). TKIM total scores were computed by summing across the absolute values of item z-scores for each target worker participant.
Political Skill, Charisma, & Performance 48 Results Convergent and discriminant validity. Table 7 presents the correlations among self, supervisor, and coworker reports of PSI total scores, the four political skill dimensions, and charisma within a multitr ait-multimethod matrix. Inspection of the validity diagonals in Table 7 indicates that se lf and supervisor ratings of overall political skill were significantly correlated (r = .33, p < .01). Likewise, the correlation between self and coworker ratings of overall polit ical skill (r = .28, p < .01) as well as the correlation between coworker and supervisor ratings of political skill (r = .32, p < .01) were also significant. This same general pattern holds true for each of the four PSI dimensions. That is, there were significant correlations between reports of each of the four PSI dimensions regardless of the reporting source. The magnitude of these correlations ranged from .18 (p < .05) for the relationship between self and supervisor ratings of apparent sincerity, to .40 (p < .01) for the relationship between self and supervisor ratings of social astuteness. These findings provide evidence for the convergence of political skill acro ss self, supervisor, and coworker reports. It should be noted, however, that this finding was less pr onounced when comparing self-ratings to coworker ratings. Further inspection of the va lidity diagonals in Table 7 indicates that self and supervisor ratings of charisma were also significantly correla ted (r = .37, p < .01). However, the positive correlation between self and coworker ratings of charisma (r = .13, p = .15), and the positive correlation between coworker and supervisor ratings of charisma (r = .13, p = .18) di d not reach significance.
Political Skill, Charisma, & Performance 49 TABLE 7 Multi-Trait, Multi-Method Correlation Matrix for Political Skill & Charisma Measures M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Self Ratings, N = 169 1. Political Skill Total 5.54 0.73 -2. Social Astuteness 5.44 0.80 .86 -3. Interpersonal Influence 5.87 0.82 .80 .67 -4. Networking Ability 4.98 1.05 .87 .61 .51 -5. Apparent Sincerity 6.39 0.78 .66 .50 .54 .41 -6. Charisma 5.55 0.84 .73 .65 .51 .64 .58 -Supervisor Ratings, N = 193 7. Political Skill Total 5.36 0.91 .33 .35 .28 .24 .21 .28 -8. Social Astuteness 5.15 1.06 .37 .40 .34 .25 .23 .30 .90 -9. Interpersonal Influence 5.80 1.12 .29 .33 .33 .14 .23 .21 .88 .74 -10. Networking Ability 4.90 1.05 .23 .22 .13 .23 .09 .24 .84 .71 .55 -11. Apparent Sincerity 6.04 1.00 .22 .23 .16 .15 .18 .19 .76 .57 .79 .43 -12. Charisma 5.35 0.97 .32 .30 .22 .30 .17 .37 .79 .75 .64 .67 .61 -Coworker Ratings, N = 133 13. Political Skill Total 5.36 1.05 .28 .27 .27 .17 .29 .10 .32 .26 .29 .30 .24 .18 -14. Social Astuteness 5.17 1.10 .31 .32 .30 .19 .30 .15 .28 .27 .25 .25 .17 .17 .94 -15. Interpersonal Influence 5.64 1.32 .18 .18 .22 .08 .20 -.03 .32 .25 .35 .24 .29 .12 .91 .83 -16. Networking Ability 5.12 1.15 .29 .25 .21 .24 .28 .18 .30 .24 .18 .36 .17 .25 .86 .75 .64 -17. Apparent Sincerity 5.78 1.22 .20 .21 .22 .07 .28 .03 .22 .16 .29 .13 .25 .05 .87 .80 .86 .59 -18. Charisma 5.42 1.06 .16 .20 .16 .07 .16 .13 .17 .13 .15 .15 .17 .13 .82 .79 .71 .70 .77 -Note. Due to missing values, sample sizes ranged from 121 to 193. Bold-underlined numbers represent monotrait-heteromethod correlat ions. Roman numbers represent heterotrait-heteromethod correlations. Bold-italic num bers represent heterotrait-monomethod correlations. r .18, p < .05 (two-tail). r .23, p < .01 (two-tail).
Political Skill, Charisma, & Performance 50 These findings provide evidence for the convergence of charisma between self and supervisor reports. On the other hand, th ese data suggest that coworker reports of charisma fail to converge with ei ther self or supervisor rating s of the construct. Also of interest is the fact that th e magnitude of the heterotrait-monomethod correlations for selfratings of all four PSI dimensions and charis ma appear to be somewhat smaller than the same correlations among coworker and supervis or ratings. This pa ttern of findings may indicate a stronger halo effect associated with supervisor and coworker ratings. Similar findings have been found in past research involving multi-source reports of OCB (Allen et al., 2000). Interestingly, this pattern of correlations was also observed among Study 2 criteria (OCB, task performance, & overall performance). These data are presented in Table 8. In order to assess the divergence of the four PSI dimensions and charisma across reporting sources, we compared the monotrait, hetero-method correlations with corresponding hetero-trait, hetero-method correlations provide d in Table 7. In regard to self and supervisor reports, charisma a nd social astuteness demonstrated the best divergence from the other PSI dimensions. Fo r both of these dimensions, there were no reversals between mono-trait, hetero-method correlations a nd hetero-trait, hetero-method correlations. In other words, the mono-trait, hetero-method correlations for both of these dimensions were higher than all eight of their correspondi ng hetero-trait, hetero-method correlations. Interpersonal influence showed only one reversal while networking ability and apparent sincerity each yielded three. The pattern of these correlations provides
Political Skill, Charisma, & Performance 51 mixed support for the divergence of these five dimensions when comparing ratings from target workers and their supervisors. A similar pattern of correlations is found for the four PSI dimensions when comparing coworker and supervisor reports For these two data sources, social astuteness, interpersonal influence, and ne tworking ability yielded no reversals when comparing mono-trait, hetero-method corre lations with corresponding hetero-trait, hetero-method correlations. This finding supports the divergence of these three dimensions. On the other hand, apparent si ncerity demonstrated two reversals while charisma yielded a total of five, sugge sting a strong overlap between these two dimensions and the other three facets of the PSI. When considering self and coworker reports, only social astuteness demonstrated ze ro reversals. Interp ersonal influence and apparent sincerity each demonstrated one re versal, networking ability yielded two, and charisma yielded five. Taken together, ther e is mixed support for the divergence of the four PSI dimensions and charisma. The ex tent to which each of these dimensions discriminate from one another is influenced by the reporting source. In general, these five dimensions tend to be most divergent when comparing ratings from target workers and their supervisors, or when comparing ra tings from supervisors and coworkers. Criterion-related validity. Table 8 provides Pearson product-moment correlations among Study 2 predictors and criteria. Hypothe sis 4 argued that self -reported PSI total scores would be positively and significan tly correlated with self, coworker, and supervisor ratings of OCB, task performan ce, and overall job performance. With the exception of coworker ratings of OCB and ta sk performance, we found that self-reported
Political Skill, Charisma, & Performance 52 political skill correlated positively and signifi cantly with all Study 2 criteria irregardless of reporting source. These corr elations appear to be highe r for self-reported criteria, ranging in magnitude from .35 (p < .01) for ove rall performance to .53 (p < .01) for OCB, than they were for supervisor-reported crit eria with correlations ranging from .19 (p < .05) for OCB to .25 (p < .01) for task perfor mance. In regard to coworker reported criteria, political skill dem onstrated a significant correla tion with overall performance (r = .20, p < .05). These findings provided support for 7 of the 9 predictions proposed by hypothesis 4. A closer inspection of the co rrelations between supervisor ratings of individual performance dimensions and poli tical skill (See Table 9) indicated that PSI scores significantly correlated with 7 of the 10 performance dimensions (rs = .19 to .26, ps < .01), with the strongest correlation bei ng with problem solving (r = .26, p < .01). Taken together, these results indicate a fairly consistent linkage between political skill and performance across different reporting sources and diverse performance dimensions. Hypothesis 5 proposed that self-reported political skill would correlate more strongly with coworker and supervisor repor ts of OCB than with reports of task performance from the same sources. We did not find support for this prediction. In fact, PSI total scores had higher co rrelations with supervisor ra tings of task performance (r = .25, p < .01) than with supervisor ratings of OCB (r = .19, p <.05). Neither coworkerreported OCB (r = .15, p = .13) nor coworker-r eported task performance (r = .15, p = .12) were significantly correlated with self-reported political skill.
Political Skill, Charisma, & Performance 53 TABLE 8 Descriptive Statistics and Corre lations of Study 2 Measures Measures M SD 1 2 3 4 5 6 7 8 9 10 11 12 Self Ratings 1. PSI Total Score 5.54 0.73 -2. Charisma 5.55 0.84 .73 -3. TKIM Total 29.07 6.80 .01 .01 -4. OCB 31.38 2.80 .53 .51 -.06 -5. Task Performance 29.51 3.25 .43 .38 -.05 .57 -6. Overall Performance 6.07 0.67 .35 .34 -.10 .46 .59 Supervisor Ratings 7. OCB 30.16 4.04 .19 .22 .16 .23 .29 .21 -8. Task Performance 29.31 4.03 .25 .31 .15 .13 .36 .29 .67 -9. Overall Performance 6.09 0.92 .23 .29 .20 .15 .33 .30 .78 .82 -Coworker Ratings 10. OCB 29.41 4.69 .15 .06 -.07 .15 .23 .18 .45 .27 .41 -11. Task Performance 28.98 4.07 .15 .17 .09 .14 .26 .20 .37 .38 .50 .76 -12. Overall Performance 5.93 0.96 .20 .15 .05 .19 .28 .27 .34 .42 .51 .83 .81 -Note. Due to missing values, sample sizes ranged from 108 to 168. Bold-underlined numbers represent monotrait-heteromethod correlations. Roman numbers in columns 4 thru 9 represent heterotrait-heteromethod correlations. Bold-italic numbers represen t heterotrait-monomet hod correlations. r .19, p < .05 (two-tail). r .25, p < .01 (two-tail).
Political Skill, Charisma, & Performance 54 Table 9 Correlations between Study 2 Predictors and Supervisor Ratings of Performance Performance Dimension PSI Total Scores Charisma TKIM Job Knowledge .07 .16 -.05 Task Proficiency .20 .24 .04 Productivity .20 .24 .26 Problem Solving .26 .26 .16 Communication .19 .23 .19 Initiative .22 .29 .16 Adaptability .23 .22 .16 Dependability -.03 .04 .14 Cooperation .20 .15 .12 Integrity .11 .14 -.01 Overall .23 .29 .20 Note. N = 144-147. r > .16, p < .05. r > .22, p < .01. Interestingly, self ratings of OCB demonstrat ed higher correlations w ith self ratings of political skill (r = .53, p < .001) than did self -ratings of task performance (r = .43, p < .001). However, the difference between these two correlations did not reach statistical significance (t (143) = 1.53, p > .05) when conducting a significance test for the difference between dependant correlati on coefficients (e.g. Cohen & Cohen, 1983). Charisma also demonstrated several positive correlations with Study 2 criteria across two of the three reporting sources. As shown in Table 9, charisma was positively and significantly correlated with self reports (rs = .34 to .51) and s upervisor reports (rs = .22 to .31) of OCB, task performance, and overall performance. Similar to political skill, charisma also correlated with 7 of the 10 dimensional performance ratings provided by supervisors (rs = .16 to .29), the highest co rrelation being with initiative (r = .29, p < .01). In regard to TKIM scores, the abridged SJT failed to correlate with self ratings (rs
Political Skill, Charisma, & Performance 55 = -.05 to -.10) or coworker ratings (rs = -.07 to .09) of overall performance, task performance, or OCB. Intere sting, all the correl ations between TKIM scores and selfreported criteria were negative, though none of these correlati ons was significant. TKIM scores only demonstrated a si gnificant relationship with supe rvisor ratings of overall job performance (r = .20, p < .05) among all of the Study 2 criteria. However, when looking at individual performance dimensions, ta cit knowledge did correlate positively and significantly with five of the ten performa nce dimensions rated by supervisors, the highest correlations bein g with employee productivity (r = .26, p < .01) and communication (r = .19, p < .05). Finally, contra ry to our expectations (See Hypothesis 1), the TKIM did not correlate with the eith er of the other Study 2 predictors, political skill (r = .01, p = .93) or charisma (r = .01, p = .87). This is surprising considering the conceptual overlap shared by political skill, charisma, and tacit knowledge. We address possible explanations for this null fi nding in the discussion. Regression analyses. To investigate the extent to which Study 2 predictors differentially explained unique variance in multi-source performance ratings, we conducted several hierarchical regression analys es. In the first series of analyses we regressed multi-source reports of overall performance, task performance, and OCB on political skill, charisma, and TKIM scores We initially controlled for employee organizational tenure, salary, education and age, but found that these control variables failed to predict a statistically significant proportion of the vari ance in any of the performance criteria regardless of reporting source (R2s = .013 to .061, M = .030, ps > .05). Accordingly, we removed the control va riables from each regression analysis and
Political Skill, Charisma, & Performance 56 only reported variance explained by each of the Study 1 predictors. Since we were interested in the extent to which charis ma and tacit knowledge explained variance in performance criteria beyond the PSI, we ente red PSI scores in the first step of each analysis, charisma in the second, and TKIM sc ores in the third and final step. Before reporting our final results, we performed pre liminary outlier analyses and removed cases with studentized deleted residua ls greater than 2 standard deviations. Accordingly, we removed between 5 and 10 cases for each regression analysis depending on the reporting source of the criteria and the performance dimensions being regressed (e.g. overall performance, task performance, or OCB). Table 10 shows the results for the anal yses treating supervisor performance reports as the dependent variable. The addi tion of PSI total scores to the regression analyses resulted in the explanation of a signi ficant portion of the variance in supervisor ratings of overall performance ( R2 = .062, p < .01), task performance ( R2 = .068, p < .01), and OCB ( R2 = .039, p < .05). At the second step of the analysis, charisma significantly increased the pred iction of supervisory reports of overall performance ( R2 = .029, p < .05) and task performance ( R2 = .036, p < .05) when being entered after PSI scores. However, charisma failed to account for additional va riance in supervisor ratings of OCB ( R2 = .021, p = .08). Based on these findings, we found partial support for hypothesis 6. Charisma significantly contribut ed to the prediction of two of the three supervisor performance ratings after controlling for political skill. The addition of TKIM scores at the thir d step of analyses significantly increased the prediction of supervisor ratings of overall performance by 4%. However, the TKIM
Political Skill, Charisma, & Performance 57 did not significantly add to the prediction of supervisory ratings of task performance ( R2 = .000, p = .94) or OCB ( R2 = .003, p = .55). As such, the TKIM was only successful in contributing to the prediction of one of the three supervisor performance ratings after controlling for political skill and charisma. Taken together, the Study 2 predictors explained a significant porti on of the variance in superv isor ratings of overall performance (13%), task perf ormance (10%) and OCB (6%). Table 10 Hierarchical Regression Analyses Predicti ng Supervisor Reports of Performance from Study 2 Predictors Step Predictors Included B F R2 R2 Dependent Variable: Supervisor -reported Overall Performance 1 Political Skill Total Score .267 .249 8.83 .062* -2 Charisma .230 .245 4.31 .091* .029+ 3 TKIM Total Score .021 .187 5.26 .126* .035+ Dependent Variable: Supervisor -reported Task Performance 1 Political Skill Total Score 1.19 .261 9.76 .068* -2 Charisma 1.12 .279 5.37 .104* .036+ 3 TKIM Total Score 0.00 .007 0.01 .104* .000 Dependent Variable: Su pervisor-reported OCB 1 Political Skill Total Score .867 .198 5.53 .039+ -2 Charisma .816 .213 3.03 .061+ .021 3 TKIM Total Score .026 .051 0.37 .063+ .003 Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to the hierarchical regression. + p < .05; p < .01; N = 135-136 Table 11 provides results from anal yses regressing self-reported overall performance, task performance, and O CB on each of the Study 2 predictors. The
Political Skill, Charisma, & Performance 58 inclusion of PSI total scores at the first st ep of the regression an alyses resulted in the explanation of a significant por tion of the variance in self ratings of overall performance ( R2 = .176, p < .01), task performance ( R2 = .236, p < .01), and OCB ( R2 = .291, p < .05). At the second step of the analysis, char isma significantly incr eased the prediction of self-reports of ta sk performance ( R2 = .024, p < .05) and OCB ( R2 = .045, p < .01) when being entered after PSI scores. Ho wever, charisma failed to account for a significant portion of the variance in self ratings of overall performance ( R2 = .015, p = .09). Based on these findings, we again found partial support for Hypothesis 6. Charisma significantly contributed to the prediction of two of the three self-rated performance criteria after controlling for political skill.
Political Skill, Charisma, & Performance 59 Table 11 Hierarchical Regression Analyses Predicti ng Self Reports of Performance from Study 2 Predictors Step Predictors Included B F R2 R2 Dependent Variable: Self-re ported Overall Performance 1 Political Skill Total Score .328 .420 32.79 .176* -2 Charisma .126 .188 2.90 .192* .015 3 TKIM Total Score -.018 -.218 9.48 .240* .048* Dependent Variable: Self-re ported Task Performance 1 Political Skill Total Score 1.79 .486 47.67 .236* -2 Charisma 0.74 .231 5.07 .261* .024+ 3 TKIM Total Score 0.04 .105 2.29 .272* .011 Dependent Variable: Self-reported OCB 1 Political Skill Total Score 1.93 .539 63.10 .291* -2 Charisma 0.98 .315 10.45 .336* .045* 3 TKIM Total Score -0.03 -.087 1.74 .343* .007 Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to the hierarchical regression. + p < .05; p < .01; N = 154-155 The addition of TKIM scores at the thir d step of analysis significantly increased the prediction of self ratings of overall performance by 5%. In terestingly, th e direction of this relationship was negative. In other word s, individuals with hi gher TKIM scores were found to rate themselves lower on overall perf ormance than individuals with lower TKIM scores. This is contrary to our expectati ons, especially because there was a significant positive relationship identified between TKIM total scores and supervisor reports of overall performance. In regard to the othe r two self-reported perf ormance criteria, the TKIM did not significantly add to the predic tion of self ratings of task performance ( R2 = .011, p = .13) or OCB ( R2 = .007, p = .19). As such, the TKIM was only successful in
Political Skill, Charisma, & Performance 60 contributing to the prediction of one of the three self-rated perfor mance criteria after controlling for political skill and charisma, and this prediction was negative in direction. Nonetheless, when all three Study 2 predictors were included in analyses they explained a significant portion of the vari ance in self ratings of overa ll performance (24%), task performance (27%) and OCB (34%). Table 12 provides results from analyses regressing coworker-reported overall performance, task performance, and OCB on eac h of the Study 2 predictors. As shown in the table, only PSI total scores predicted a significant portion of variance in any of the coworker-reported performance criteria. Specifically, political skill predicted a significant portion of the variance in co worker ratings of overall performance ( R2 = .079, p < .01) when included in the first step of the analysis. Charis ma failed to predict additional variance in any of the criteria when added at the second step. Likewise, TKIM total scores were unsuccessful in explaining variance in coworker-reported criteria when entered at the third step of each analysis. Based on these findings, we did not support our hypothesis that charisma would explain addi tional variance in coworker performance ratings after controlling for political skill.
Political Skill, Charisma, & Performance 61 Table 12 Hierarchical Regression Analyses Predicti ng Coworker Reports of Performance from Study 2 Predictors Step Predictors Included B F R2 R2 Dependent Variable: Coworker-r eported Overall Performance 1 Political Skill Total Score .322 .282 8.81 .079* -2 Charisma .103 .099 0.54 .084+ .005 3 TKIM Total Score .010 .088 0.85 .092+ .008 Dependent Variable: Coworker -reported Task Performance 1 Political Skill Total Score .772 .157 2.56 .025 -2 Charisma 1.013 .225 2.67 .050 .025 3 TKIM Total Score .059 .117 1.47 .063 .014 Dependent Variable: Coworker-reported OCB 1 Political Skill Total Score .490 .096 0.95 .009 -2 Charisma .516 .109 0.60 .015 .006 3 TKIM Total Score -.012 -.023 0.06 .016 .001 Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to the hierarchical regression. + p < .05; p < .01; N = 103-104 Exploratory analyses In order to competitively assess the predictive efficiency of the four PSI dimensions and charisma we first investigated the strength of the correlations between each of the dimensions a nd supervisor ratings of performance (See Table 13). Consistent with the findings of Ferris et al (2005), social astuteness demonstrated the strongest positive relations hip with each of the performance ratings, with correlations ranging from .24 with OCB to .33 with overall performance. However, while social astuteness was the only PSI di mension to significantly correlate with performance in the Ferris et al. study, we f ound significant positive correlations for four of the five dimensions of interest. In re gard to the magnitude of these correlations,
Political Skill, Charisma, & Performance 62 charisma was a close second to social astute ness with correlations ranging from .22 with OCB to .29 with overall performa nce. Apparent sincerity and interpersonal influence demonstrated modest relationships with perf ormance as indicated by correlations ranging between .18 and .22 across performance ratings Networking abilit y, on the other hand, was the only dimension that failed to correlate significantly with any of the supervisor ratings of performance (rs = .08 .13, ps > .05). Table 13 Correlations between PSI Dimensions, Charis ma and Supervisor Ratings of Performance Predictors by Dimension Mean SD Overall Performance Task Performance OCB Social Astuteness 5.45 0.81.33** .34** .24** Interpersonal Influence 5.89 0.82.21* .20* .18* Networking Ability 4.96 1.07.09 .13 .08 Apparent Sincerity 6.39 0.78.22** .21* .20* Charisma 5.55 0.84.29** .31** .22** Note. N = 144-147. p < .05; ** p < .01 Since apparent sincerity and charisma de monstrated the strongest positive trend in predicting performance criteria among the 5 dimensions of interest, we included only these two dimensions and TKIM total scores as predictors in our exploratory regression analyses. Consistent with the first round of analyses, the control va riables (organizational tenure, salary, education, and age) failed to predict a significant portion of variance in any of the performance ratings. This findi ng was observed across all reporting sources (R2s = .013 to .068, M = .031, ps >.05). Acco rdingly, these variables were removed from analyses. We also conducted preliminary outlier analyses for each of our exploratory regressions. This resulted in the identification of 6 to 10 outliers (Ns = 134
Political Skill, Charisma, & Performance 63 to 138) that were removed from correspondi ng analyses before final results were reported. Table 14 shows the results for the exploratory anal yses treating supervisor performance ratings as dependent variables in each regression equation. The addition of social astuteness in the first step of the regr ession analyses resulted in the explanation of a significant portion of the variance in supe rvisor ratings of overall performance ( R2 = .153, p < .01), task performance ( R2 = .144, p < .01), and OCB ( R2 = .055, p < .01). Table 14 Hierarchical Regression Analyses Predicti ng Supervisor Reports of Performance from Social Astuteness, Charisma, and TKIM Scores Step Predictors Included B F R2 R2 Dependent Variable: Supervisor -reported Overall Performance 1 Social Astuteness .336 0.39 24.04 .153* -2 Charisma .194 0.23 4.97 .184* .031+ 3 Soc. Astuteness x Charisma -.164 -1.85 6.31 .221* .038+ 4 TKIM Scores .000 0.00 0.00 .221* .000 Dependent Variable: Supervisor -reported Task Performance 1 Social Astuteness 1.58 0.38 22.49 .144* -2 Charisma 0.72 0.18 2.86 .162* .018 3 Soc. Astuteness x Charisma -1.40 -3.26 21.02 .277* .115* 4 TKIM Scores -0.01 -0.01 0.02 .277* .000 Dependent Variable: Su pervisor-reported OCB 1 Social Astuteness .945 0.24 7.97 .055* -2 Charisma .596 0.15 1.94 .069* .013 3 Soc. Astuteness x Charisma -.629 -1.51 3.76 .094* .025 4 TKIM Scores .013 0.02 0.08 .095* .001 Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to second step of the hier archical regression. + p < .05; p < .01; N = 134-137
Political Skill, Charisma, & Performance 64 Adding charisma at the second step of the anal yses resulted in a si gnificant increase in the prediction of supervisory reports of overa ll performance ( R2 = .031, p < .05). However, charisma did not enhance the prediction of task performance ( R2 = .018, p = .09) or OCB ( R2 = .013, p = .17) after controlling for social astuteness. To examine the possible inte ractive effects of charisma and social astuteness, we created an interaction term for these two predictors and entered it in the third step of the regression analyses. This interaction wa s found to be significant for both task performance ( R2 = .038, p < .05) and overall performance ( R2 = .115, p < .01). As depicted in Figures 1 and 2, so cial astuteness had more infl uence in predicting task and overall performance for employees with low charisma scores than for employees with high charisma scores. In other words, employ ees with low charisma were more likely to receive high task and overall performance ratings if they ca refully adhered to social and contextual cues while interacting with othe rs at work. Taken t ogether, these findings indicate that employees with strong charisma are not as reli ant on astuteness in order to make a good impression on their supervisors. In terestingly, this effect was three times as strong when predicting supervisor ratings of task performance, than when predicting supervisor ratings of overall performance. When taking a clos er look at how this interaction predicted individua l task performance dimensions, we found the interaction to be statistically significant for all five task performance di mensions with the strongest interaction being observed when pred icting dimensions of job knowledge ( R2 = .091, p < .01) and communication ( R2 = .099, p < .01).
Political Skill, Charisma, & Performance 65 high lowCharisma 3.004.005.006.007.00Social Astuteness 20.00 25.00 30.00 35.00 40.00S u p e r v i s o r R a t i n g s o f T a s k P e r f o r m a n c e high lowCharisma 3.004.005.006.007.00Social Astuteness 5.00 6.00 7.00S u p e r v i s o r R a t i n g s o f O v e r a l l P e r f o r m a n c e Figure 1. The interactive effects of char isma and social astuteness Figure 2. The interactive effects of ch arisma and social astuteness when predicting supervisor ra tings of task performance. when predicting supervisor ratings of overall performance. Note. Note. High Charisma represents cases with charisma Z-scores greater than 1 SD High Charisma represents cases with charisma Z-scores g reater than 1 SD Y = 3.89 + 5.27(Social Astuteness); R2 = 0.06 Y = 4.54 + 0.33(Social Astuteness); R2 = 0.07 Low Charisma represents cases with charisma Z-scores less than -1 SD Low Charisma re presents cases with charisma Z-scores less than -1 SD Y = 23.47 + 1.20(Social Astuteness); R2 = 0.77 Y = 2.75 + 0.65(Social Astuteness); R2 = 0.36
Political Skill, Charisma, & Performance 66 When entered in the final step of each an alysis, TKIM scores failed to account for additional variance across supe rvisor performance ratings. Accordingly, the unique variance explained in supervisor performan ce ratings demonstrated by TKIM scores in the first round of regression analyses was not replicated. Instead, th e data suggest that social astuteness, charisma and their inte raction are collectivel y exhaustive of the variance explained in supervisor reports of performance when considering these two predictors along with TKIM scor es in the regression equation. In fact, when considering only social astuteness and charisma, these tw o predictors and thei r interaction accounted for a total of 22% of the vari ance explained in supervisor ra tings of overall performance, 28% of the variance explained in task performance, and 9% of the variance explained in OCB. These findings are impressive given th e fact that social astuteness and charisma are measured with only 12 items. Table 15 shows the results for the explor atory analyses treating each self-reported performance rating as the dependent variable in each regression equa tion. The addition of social astuteness in the first step of the regression analyses result ed in the explanation of a significant portion of the variance in supervisor ratings of overall performance ( R2 = .151, p < .01), task performance ( R2 = .233, p < .01), and OCB ( R2 = .241, p < .01). Adding charisma at the second step of the anal yses resulted in a si gnificant increase in the prediction of self-repo rts of overall performance ( R2 = .034, p < .05), task performance ( R2 = .037, p < .05), and OCB ( R2 = .135, p < .05).
Political Skill, Charisma, & Performance 67 Table 15 Hierarchical Regression Analyses Predic ting Self-Reported Performance from Social Astuteness, Charisma, and TKIM Scores Step Predictors Included B F R2 R2 Dependent Variable: Self-re ported Overall Performance 1 Social Astuteness .278.388 27.29 .151* -2 Charisma .166.245 6.32 .184* .034+ 3 Soc. Astuteness x Charisma .028.384 0.29 .186* .002 4 TKIM Scores -.019-.228 10.24 .237* .052* Dependent Variable: Self-re ported Task Performance 1 Social Astuteness 1.61.482 46.65 .233* -2 Charisma 0.80.250 7.67 .269* .037* 3 Soc. Astuteness x Charisma 0.22.652 0.93 .274* .004 4 TKIM Scores -0.05-.127 3.42 .290* .016 Dependent Variable: Self-reported OCB 1 Social Astuteness 1.51.491 49.16 .241* -2 Charisma 1.38.472 33.29 .376* .135* 3 Soc. Astuteness x Charisma -0.17-.525 0.71 .379* .003 4 TKIM Scores -0.03-.079 1.55 .385* .006 Note. B indicates unstandar dized beta weights; indicates standardized beta weights; F indicates results from incremental F tests; R2 indicates the amount of variance explained in the dependent variable; R2 indicates the increase in R2 when adding a variable to second step of the hier archical regression. + p < .05; p < .01; N = 134-137 Interestingly, the interaction between soci al astuteness and charisma did not explain additional variance in any of the self-reported criteria. However, entering TKIM scores in the final step of analyses did explain an additional 5% of the va riance in self-reported overall performance. Still, consistent with the first round of regression analyses, this relationship was found to be negative. In regard to coworker ratings of perfor mance, the predictors in the exploratory analyses failed to predict a si gnificant portion of the variance in these criteria. The only
Political Skill, Charisma, & Performance 68 exception was the ability of social astute ness to modestly pred ict coworker-reported overall performance ( R2 = .072, p < .01). In regard to the social astuteness/charisma interaction, the interac tion term did not explain unique va riance in coworker reports of overall performance, task performance, or OCB ( R2s = .000 .012, ps > .05). Our results indicate that the interaction between social astuteness and charisma is only observed when predicting supervisor -reports of job performance.
Political Skill, Charisma, & Performance 69 Discussion Factor Structure of Political Skill One major goal of the current research was to examine and expand upon Ferris et als (2005) four-factor structure of political skill. In particul ar, we were interested in the stability of the original four-factor model, and the extent to which our measure of charisma provided a unique contribution to the political skill constr uct domain. Study 1 factor analyses identified charisma as a unique addition to the existing political skill behavioral taxonomy. On the other hand, expl oratory analyses did not provide support for the differentiation of the PSIs four f actors of political sk ill. Specifically, PSI dimensions of social astuteness, interpersona l influence, and apparent sincerity converged into a single dimension, producing a three fact or representation of political skill when combined with networking ability and charisma. One possible explanation for this finding could be the result of using a student sample. Although the majority of the partic ipants in Study 1 were part or full-time workers, an argument could be made that co llege students have ye t to experience work scenarios requiring them to differentiate between behaviors of social astuteness, interpersonal influence, and apparent sincerit y. This is a reasonabl e contention given that these three factors of political skill, though conceptually different, share a commonality in regard to reading social cu es. By definition, socially astu te workers have the ability to read and understand social situ ations to determine the appr opriate course of action in
Political Skill, Charisma, & Performance 70 response to a given scenario. Interpersonal influen ce requires the selection of effective persuasion techniques that are dependent upon the information provided by the social context in which the worker wishes to exer cise influence. Similarly, the successful display of apparent sincerity is also reliant on the envir onment in which a message or idea is delivered. For instance, audience ch aracteristics and/or the mode of delivery should have some impact on the content or ex pression of a message which is intended to be perceived as genui ne or sincere. Based on their limited exposure to divers e work scenarios requiring the use of politics or influence, undergraduate students may collapse their interpretation of social astuteness, interpersonal influence, and appa rent sincerity into a single, higher-order factor broadly focused on the ability to mold behavior to fit social or environmental contexts. Professional level employees, on the other hand, may be more akin to differentiate between these thre e factors as a result of work experiences requiring skill in unique contexts such as assessing social /environmental cues, using discretion when choosing influence tactics, and conveying ideas genuinely through the expression of emotions such as modesty, passion, commitme nt, or accountability. Unfortunately, the current research was unable to test for diffe rences in political skill conceptualizations between undergraduate students and professi onals using exploratory or confirmatory factor analyses; Study 2 did not provide a su fficient professional sample to conduct such analyses. Nonetheless, taken together with the work of Ferris et al., the present findings suggest that there are incons istencies in how individuals distinguish between political behaviors depending on their leve l of professional experience.
Political Skill, Charisma, & Performance 71 Multi-source Convergence of Political Skill Another important goal of the present re search was to test the convergence of political skill and charisma across multiple reporting sources (i.e. self-reports, coworkerreports, and supervisor-reports). By demons trating some consistency in agreement of employee political skill across sources, we would have greater confidence that the measure was able to assess meaningful facet s of social effectiveness. Although Study 2 provided mixed results for the convergence of the PSI dimensions and charisma across self, coworker, and supervisor reports, a posit ive trend did emerge. Specifically, there was consistent evidence for the convergence of political skill and charisma between self and supervisor reports with significant converg ent validity coefficients ranging from .18 for apparent sincerity to .40 for social astuteness. More importantly, 3 of the 5 dimensions of interest (social astuteness, interpersonal influence, and charisma) demonstrated coefficients higher than .30 be tween self and superv isor ratings. Taken together, these findings provide evidence that there is some level of agreement regarding employee political skill between self and supervisor report s. Evidence of agreement across sources lends support to the assertion th at self-reported political skill and charisma tap important elements of interpersonal effectiveness that may be indicative of future performance ratings. Interestingly, we found that the conver gent validity coefficients between selfreports and reports from others to be comparable to the validity coefficients between coworker and supervisor reports for the four political skill dimensions. These data are inconsistent with past re search attempting to dem onstrate the convergence of
Political Skill, Charisma, & Performance 72 performance ratings across sources. For example, research efforts testing the multisource convergence of both OCB and overall performance ratings tend to yield higher validity coefficients for ratings between sources external to the target (e.g. supervisorcoworker) than for coefficients of self-oth er reports (e.g. Allen et al., 2000; Becker & Vance, 1993; Harris & Schaubroeck, 1988). Pa st findings suggest that individuals external to an employee have higher agr eement regarding employee performance than they do with the employees pe rception of their own performa nce. Although, this trend of results was found for the convergence of OCB, task performance, and overall performance ratings in Study Two of the curr ent research, the PSI dimensions failed to demonstrate the same pattern of convergence. One possible explanation for why this tr end failed to emerge for ratings of political skill could be linked to limitations in our coworker sample and a tendency for political skill levels to be audience specific. Our samp ling instructions requested supervisors to select coworker participants that worked closely with the target employee. Consistent with these instruc tions, and feedback we received from supervisors, coworkers were selected on a basis of how often they interacted with the target worker. This produced a coworker sample consisting of pa rticipants that were either lateral or subordinate to the target. Wh ile we contend that political behavior can be exercised up, down, and across the chain-of-c ommand, we also believe the appropriate selection and expression of these behaviors vary based on the status or level of the individual an employee intends to influence. We also believe that an employees motivation or will to exercise political skill may vary as a function of the status or power of the individual or
Political Skill, Charisma, & Performance 73 audience being addressed. Accordingly, a disc repancy in the rank or organizational level of a coworker participant, in comparison to the target employee, may result in varied perspectives of the targets political skill and charisma, thus causing coworker-supervisor agreement to be deflated. Evidence for Criterion-Related Validity A third focus of the present research wa s to examine the criterion-related validity of self-reported political skill and charisma. As expected, Study 2 provided evidence for the ability of political skill and charisma to predict supervisor reports of task performance (R2 = .104) and overall performance (R2 = .091). However, contra ry to our expectations, charisma and political skill only accounted for a marginal portion of the variance in supervisor ratings of OCB (R2 = .061). On a positive note, when inspecting the individual OCB dimensions tapped by the Borm an et al.s (1994) behaviorally anchored rating scale (i.e. initiative, adaptability, dependability, coope ration, and integrity), only dimensions of dependability and integrity fa iled to yield significant correlations with charisma and/or PSI total scores. Based on these findings, an argument could be made that the extent to which political skill and ch arisma predict supervisor ratings of OCB is dependent upon the type of citizenship behavior being assessed. Our results indicate that political skill and charisma predict supervisor ratings of initiative, adaptability, and cooperation just as well they predict ove rall performance and task performance dimensions including job knowledge, task pr oficiency, productivit y, problem solving, and oral/written communication.
Political Skill, Charisma, & Performance 74 In regard to coworker-reported criteria, we found very little support for the ability of political skill or charisma to predict performance ratings. The same issues we believe are responsible for the weak convergence of supervisor and coworker ratings of the political skill dimensions may also be at play in this context. Since our coworker sample consisted of individuals in positions that were either lateral or subor dinate to the target participant, the organizationa l level of the coworker may have influenced the target employees selection and expression of po litical behavior. Its possible that an employees level of political skill may vary when attempting to influence audiences spanning different ranks in th e chain-of-command. In addi tion, employees, regardless of their level of political skill, may fluctuate in their motivation to ex ercise appropriate and effective political behaviors depending on the st atus or power of the individual they are attempting to influence. Fluctuations in ta rget workers skill level or motivation as a function of the organizational level of their coworkers co uld have resulted in lower correlations between ratings of performance and either political skill or charisma. When examining the criterion-related validity of political skill and charisma at the dimensional level, we observed an interac tion between charisma and social astuteness when predicting supervisor reports of task and overall performance. Ferris et al. projected political skill as a potentially impor tant moderator that should facilitate the effectiveness of influence tac tics on performance (pp. 148, 2005) Consistent with this hypothesis, recent research has provided eviden ce for the moderating effects of political skill when regressing performance ratings on several impression management tactics (Harris, Kacmar, Zivnuska, & Shaw, 2007). Harr is et al. found that politically skilled
Political Skill, Charisma, & Performance 75 individuals using impression management t actics including supplic ation, intimidation, ingratiation, self-promotion, a nd exemplification achieved higher performance ratings than individuals who used the same tactics but were not politically skilled. In the present research, however, we found the interaction betw een charisma and social astuteness to be counteractive rather than complimentary. Our results in dicate that social astuteness has more influence in predicting task and overall performance for employees low in charisma than for employees high in charisma. Rather than facilitating the effect of charisma on performance ratings, our findings su ggest that social astuteness serves as an alternative or substitute to charismatic be havior. We believe this fi nding provides additional support for the argument that charisma is a unique element of political skill, not simply an influence tactic or stylistic mechanism alrea dy captured within Ferri s et al.s four-factor taxonomy. As an addition to Study 2, we also test ed the criterion-rela ted validity of the TKIM. Our objective in doing so was to comp etitively test the predictive validity of the TKIM, the PSI, and our measure of charisma. Contrary to past research examining the predictive validity of the TKIM (e.g. Wagner & Sternberg, 1991), our results failed to indicate a positive relationship between TKIM scores and performance ratings, regardless of reporting source. The only exception to this pattern was a modest co rrelation (r = .20) between TKIM scores and supervisor reports of overall performance. One possible explanation for these findings could be the result of using an abridged version of the measure. In an effort to reduce the length of study materials, we only used five of the nine situational stems found in the complete ve rsion of the TKIM. If all nine vignettes
Political Skill, Charisma, & Performance 76 had been administered to target participants validity coefficients between TKIM scores and study criteria may have been enhanced. Al so of note, despite the conceptual overlap of tacit knowledge and political skill, the TKIM did not correlate signi ficantly with either PSI total scores or charisma Although tacit knowledge is theoretically similar to constructs of political skill and charisma, our results indicate that the TKIM does not assess the same facets of social effectiveness as the PSI or our measure of charisma. Whats more, our findings suggest that bot h political skill and charisma are more predictive of performance ratings than tacit knowledge. Limitations and Future Directions Like all empirical studies, the present research was not without methodological limitations. As previously mentioned, Study 1 findings are limited in their generalizability due to the use of a student sample. The second studys coworker sample produced unforeseen variance in regard to coworker organizational level in comparison to the rank of target workers. Also in St udy 2, estimates of the TKIMs criterion-related validity are limited in their generalizability due to the use of an abridged version of the measure. Despite these shortcomings, th e current research provided a wealth of validation evidence for both the PSI and our new measure of charisma. However, in order to increase our confidence in the st ability of the current findings, additional research is needed to replicate these results across diverse populations spanning organizations in both the pub lic and private sectors. Pfeffer (1981) was the first to argue the no tion of political beha vior being driven by elements of both skill and will Following this stream of thought, we believe that
Political Skill, Charisma, & Performance 77 political behaviors demand competence in read ing political cues a nd selecting effective strategies, as well as the motivation to successf ully execute strategies in order to achieve desired outcomes. Political skill enthusiasts (e.g. Ferris et al., 2002; 2005) have advanced the study of political behavior by operationalizing the skill and perceptual components of the construct. As of yet, however, research ers have not explored the motivational factors soliciting the expression of political behavi ors measured by the PSI. We contend that there may be variance in political skill when considering the rank, st atus, or power held by the individual one intends to influence. Fu ture research efforts need to examine the audience characteristics that so licit the effective and ineffectiv e use of political skill (e.g. organizational rank, access to fiscal resour ces, access to human capital, expertise, leadership style, etc.). Likewise, future research also needs assess the specific organizational or situational factors (e.g. value of outcome, team membership, decisionmaking process, etc.) and employee traits (e.g. charisma, need for achievement, positive affectivity, etc.) that facilitate efforts to exercise political skill within and across organizational levels. Beyond efforts to explore the antecedents of political skill and provide additional validity evidence for the PSI, researchers and pr actitioners need to explore new ways of measuring political skill for purposes of em ployee selection and promotion. Riggio and Riggios (2001) chapter on interpersonal sensitivity highlights the usefulness of measuring self-reported social effectivene ss constructs for purposes of assessment and development. In the realm of employee deve lopment, these types of assessments may be valid if employees are motivated to provide honest responses in an effort to obtain
Political Skill, Charisma, & Performance 78 accurate performance feedback. It should be noted, however, Likert-based self assessments are vulnerable to rating inflati on. In the scope of employee selection or promotion, candidates will likely inflate Likert-b ased, self-reports of social effectiveness in order to be perceived as putting the best foot forward To counteract socially desirable responses, self-report assessmen ts of political skill need to be expanded to include formats that are less susceptible to faking such as SJTs, assessment center exercises, or behavioral-based interv iew questions. Finally, as discussed by Ferris, Perrewe, and Douglas (2002), the job performance literature has become saturated with an a bundance of social eff ectiveness constructs, most of which are hypothesized to enha nce job performance or organizational effectiveness. These constructs include social intelligence, emotional intelligence, political skill, and prosocial work behavior, ju st to name a few. Clearly, research is needed to supplement ongoing efforts to understand the commonalities of these social constructs and to identify whic h these constructs represent unique elements of social effectiveness.
Political Skill, Charisma, & Performance 79 References Ahearn, K. K., Ferris, G. R., Hochwarter, W. A., Douglas, C. D., & Ammeter, A. P. (2004). Leader political sk ill and team performance. Journal of Management, 30(3), 309-327. Allen, T. D., Barnard, S., Rush, M. C., & Ru ssell, J. (2000). Ratings of organizational citizenship behavior: Does the source make a difference? Human Resource Management Review, 10(1), 97-114. Allen, T. D., & Rush, M. C. (1998). The eff ects of organizational citizenship behavior on performance judgments: A field study and a laboratory experiment. Journal of Applied Psychology, 83(2), 247-260. Ammeter, A. P., Douglas, C., Gardner, W. L ., Hochwarter, W. A., & Ferris, G. R. (2002). Toward a political th eory of leadership. Leadership Quarterly, 13 751-796. Ashkanasy, N. M., & Tse, B. (2000). Transf ormational leadership as management of emotion: A conceptual review. In Ashka nasy, N. M., & Hartel, C. E. (Eds.), Emotions in the Workplace: Research, theory, and practice. (pp. 221-235). Westport, CT: Quorum Books/G reenwood Publishing Group, Inc. Bass, B. M. (1985). Leadership and Performance Beyond Expectations, New York, NY, Free Press.
Political Skill, Charisma, & Performance 80 Bass, B. M., & Avolio, B. J. (1993). Transfor mational leadership: A re sponse to critiques. M. M. Chermers & R. Ayman (Eds.), Leadership theory and research: Perspectives and directions (pp. 49-80). San Diego, CA: Academic Press. Becker, T. E., & Vance, R. J. (1993). Constr uct validity for three types of organizational citizenship behavior: An illustration of the direct product model with refinements. Journal of Management, 19, 663-682. Borman, W. C., Ackerman, L. D., & Kubisiak, U. C. (1994). Development of a performance rating program in support of Department of Labor test validation research (Contract Nos. 93-2 and 93-3), S acramento, CA: Cooperative Personnel Services. Borman, W. C., & Motowidlo, S. J. (1993). Expanding the criterion domain to include elements of contextual performance. In N. Schmitt, W. C. Borman, & Associates (Eds.), Personnel Selection in Organizations (pp. 71-98). San Francisco: Jossey-Bass. Borman, W. C., & Motowidlo, S. J. ( 1997). Task performance and contextual performance: The meaning for personnel selection research. Human Performance, 10, 99-109. Borman, W. C., Penner, L. A., Allen, T. D., & Motowidlo, S. J. (2001). Personality predictors of citizenship performance. International Journal of Selection and Assessment, 9, (1-2), 52-69.
Political Skill, Charisma, & Performance 81 Borman, W. C., White, L. A., & Dorsey, D. W. (1995). Effects of ratee task performance and interpersonal factors on superv isor and peer performance ratings. Journal of Applied Psychology, 80, 168-177. Burns, J. M. (1978). Leadership. New York: Harper & Row. Cascio, W. F. (1995). Whither industrial and organizational psychology in a changing world of work. American Psychologist, 50, 928-939. Chao, G. T., OLeary-Kelly, A. M. Wolf, S ., Klein, H. J., & Gar dner, P. D. (1994). Organizational socialization: Its content and consequences. Journal of Applied Psychology, 79, 730-743. Cliff, N. (1988). The eigenvalues-greater-thanone rule and the reliab ility of components. Psychological Bulletin, 103(2), 276-279. Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. Coleman, V. I., & Borman, W. C. (2000). I nvestigating the underl ying structure of the citizenship performance domain. Human Resource Management Review, 10, 2544. Conger, J. A., & Kanungo, R. N. (1987). Towards a behavioral theory of charismatic leadership in organizational setting, Academy of Management Review, 12 637647. Conger, J. A. (1997). The dark side of leadership. In Vecchio, R. P. (Ed.), Leadership: Understanding the dynamics of power and influence in organizations. (pp. 215 232). Notre Dame, IN: University of Notre Dame Press.
Political Skill, Charisma, & Performance 82 Conger, J. A., & Kanungo, R. N. (1994). Char ismatic leadership in organizations: Perceived behavioral attri butes and their measurement. Journal of Organizational Behavior, 15, 439-452. Culbert, S. A. (1996) Mind-set management. New York, NY: Oxford University Press. Culbert, S. A., & McDonough, J. J. (1980). The invisible war: Pursuing self-interests at work. New York: John Wiley. Daft, R. L. (2004). Organisation Theory and Design (8th ed.). New York, NY: West Publishing Company. Farrell, D., & Patterson, J. C. (1982). Pattern s of political behavi or in organizations. Academy of Management Review, 7, 403-412. Ferris, G. R., Davidson, S. L., & Perrewe, P.L. (2005). Political Skill at Work: Impact on work effectiveness Mountain View, CA: Davies-Black Publishing. Ferris, G. R., Hochwarter, W. A., Douglas, C., Blass, F. R., Kolodinsky, R. W., & Treadway, D. C. (1999). Development and initial validati on of the political skill inventory. Paper presented at the Academy of Management, 59th Annual National Meeting, Chicago. Ferris, G. R., & King, T. R. (1991). Politics in human resource decisions: A walk on the dark side. Organizational Dynamics 20, 59-71. Ferris, G. R., Perrew, P. L., & Douglas, C. ( 2002). Social effectiven ess in organizations: Construct validity and research directions. Journal of Leadership and Organizational Studies, 9, 49-63.
Political Skill, Charisma, & Performance 83 Ferris, G. R., Perrew, P. L., Anthony, W. P ., & Gilmore, D. C. (2000). Political skill at work. Organizational Dynamics, 28(4), 25-37. Ferris, G. R., Russ, G. S., & Fandt, P. M. (1989). Politics in organizations. In R. A. Giacalone, & P. Rosenfeld (Eds.), Impression management in the organization (pp. 143-170). Hillsdale, NJ: Lawrence Erlbaum. Ferris, G. R., Treadway, D. C., Kolodinsky, R. W., Hochwarter, W. A., Kacmar, C. J., Douglas, C., & Frink, D. D. (2005). Deve lopment and validation of the political skill inventory. Journal of Management, 30(1), 126-152. Gefen, D., Straub, D., & Boudreau, M. C. (2000). Structural e quation modeling and regression: Guidelines for research practice. Communications of the Association for Information Systems, 4, 1-78. Hair, J., Anderson, R. O., & Tatham, R. 1987. Multidimensional data analysis. New York: Macmillan. Harris, K. J., Kacmar, K. M., Zivnuska, S., & Shaw, J. D. (2007). The impact of political skill on impression management effectiveness. Journal of Applied Psychology, 92(1), 278-285. Harris, M. M., & Straubroeck, J. (1988). A me ta-analysis of self-supervisor, self-peer, and peer-supervisor ratings. Personnel Psychology, 41, 43-62. Hatcher, L. (1994). A step-by-step approach to using th e SAS system for factor analysis and structural equation modeling. Cary, NC: SAS Institute, Inc.
Political Skill, Charisma, & Performance 84 Hedlund, J., Forsythe, G. B., Horvath, J. A., Williams, W. M., Snook, S., & Sternberg, R. J. (2003). Identifying and assessing t acit knowledge: understanding the practical intelligence of military leaders. The Leadership Quarterly, 14, 117-140. House, R. J. (1977). A 1976 theory of charismatic leadership. In: Hunt, J. G. & Larson, L. L. (Eds), Leadership: The Cutting Edge Southern Illinois University Press, Carbondale, IL, pp. 189-207. Hu, L., & Bentler, P. M. (1999). Cutoff crite ria for fit indexes in covariance structure analysis: Conventional criter ia versus new alternatives. Structural Equation Modeling, 6, 1-55. Hunter, J. E., & Hunter, R. F. (1984). Validity and utility of alternat ive predictors of job performance. Psychological Bulletin, 96, 72-98. Joreskog, K., & Sorboen, D. (1993). LISREL 8 Users Guide. Uppsala, Sweden: University of Uppsala. Kipnis, D., Schmidt, S. M., & Wilkinson, I. ( 1980). Intraorganizationa l influence tactics: Explorations in getting ones way. Journal of Applied Psychology, 65, 440-452. Kudisch, J. D., Poteet, M. L., Dobbins, G. H., & Rush, M. C. (1995). Expert power, referent power, and charisma: Toward the resolution of a theoretical debate. Journal of Business & Psychology, 10(2), 177-195. La Du, T. J., & Tanaka, J. S. (1989). Infl uence of sample size, estimation method, and model specification on goodness of fit asse ssment in structural equation models. Journal of Applied Psychology, 74, 625-635.
Political Skill, Charisma, & Performance 85 Legree, P. J. (1994). The effect of response format on reliability estimates for Tacit Knowledge Scales (Research Note No. 94-25). Alexandria, VA: U.S Army Research Institute for the Behavioral and Social Sciences. Legree, P. J. (1995). Evidence for an oblique so cial intelligence factor established with a Likert-based testing procedure. Intelligence, 21(3), 247-266. Legree, P. J., Martin, D. E., & Psotka, J. (2000). Measuring cognitive aptitude using unobtrusive knowledge tests: A new survey technology. Intelligence, 28(4), 291308. Lowe, K. B., Kroeck, K .G., & Sivasubramania m, N. (1996). Effectiv eness correlates of transformational and transactional leader ship: A meta-analytic review of the MLQ literature. Leadership Quarterly, 7(3), 385-425. MacKenzie, S. B., Podsakoff, P. M., & Ah earne, M. (1996). Unpublished data analysis. Indiana University, Bloomington. MacKenzie, S. B., Podsakoff, P. M., & Fe tter, R. (1991). Organi zational citizenship behavior and objective productivity as de terminants of managerial evaluations of salespersons performance. Organizational Behavior and Human Decision Processes, 50, 123-150. Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence. In: Sluyter, D. J. (Ed), Emotional development and emoti onal intelligence: Educational implications, Basic Books Inc., New York, NY, pp. 3-34. Mayes, B. T., & Allen, R. W. (1977). Towa rd a definition of organization politics. Academy of Management Review, 2, 672-678.
Political Skill, Charisma, & Performance 86 Medsker, G. J., Williams, L. J., & Holohan, P. J. (1994). A review of current practices for evaluating causal models in organi zational behavior and human resources management research. Journal of Management, 20, 439-464. Motowidlo, S. J., & Van Scotter, J. R. (1994) Evidence that task performance should be distinguished from c ontextual performance. Journal of Applied Psychology, 79, 475-480. Mulaik, S. A., James, L. R., Van Alstine, J ., Bennet, N., Lind, S., & Stillwell, C. D. (1989). An evaluation of goodness-of-fit in dices for structural equation models. Psychological Bulletin, 105, 430-445. Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill. Organ, D. W. (1997). Organizatio nal citizenship behavior: Its construct clean-up time. Human Performance, 10, 85-97. Organ, D. W. (1988). Organizational citizenship behav ior: The good soldier syndrome. Lexington, MA: Lexington. Organ, D. W., & Paine, J. B. (1999). A new kind of performance for industrial and organizational psychology: Recent contri butions to the study of organizational citizenship behavior. International Review of Industrial and Organizational Psychology, 14, 337-368. Organ, D. W., & Ryan, K. (1995). A meta-analy tic review of attitudi nal and dispositional predictors of organizatio nal citizenship behavior. Personnel Psychology, 48, 775802.
Political Skill, Charisma, & Performance 87 Pedhazur, E. J. (1997). Multiple regression in behavio ral research: Explanation and prediction (3rd ed.). Fort Worth, TX: Harcourt Brace & Company. Pedhazur, E. J., & Schmelkin (1991). Measurement, Design, and Analysis: An integrated approach. Hillsdale, NJ: Lawrence Erlbaum Associates Inc. Pfeffer, J. (1981). Power in organizations. Marshfield, MA: Pitman Publishing Inc. Pfeffer, J. (1992). Managing with power. Boston, MA: Harvard Business School Press. Podsakoff, P. M., Ahearne, M., & MacKenzie, S. B. (1997). Organizational citizenship behavior and the quantity and qua lity of work group performance. Journal of Applied Psychology, 82(2), 262-270. Podsakoff, P. M., & MacKenzie, S. B. ( 1994). Organizational citi zenship behavior and sales unit effectiveness. Journal of Marketing Research, 31, 351-363. Riggio, R. E., & Riggio, H. R. (2001). Se lf-report measurement of interpersonal sensitivity. In J. A. Hall & F. J. Bernieri (Eds.), Interpersonal sensitivity: Theory and measurement (pp. 127-142). Mahwah, NJ: Lawrence Erlbaum. Roberts, N. C. (1985). Transforming l eadership: a process of collective action, Human Relations, 38, 1023-1046. Salancik, G. R., & Pfeffer, J. (1977). Who gets power a nd how they hold on to it: A strategic-contingency model of power. Organizational Dynamics ?, 3-21. Schmidt, F. L., & Hunter, J. E. (1993). T acit knowledge, practical intelligence, general mental ability, and job knowledge. Current Directions in Psychological Science, 2, 8-9.
Political Skill, Charisma, & Performance 88 Semadar, A., Robins, G., & Ferris, G. R. (2006). Comparing the validity of multiple social effectiveness constructs in the prediction of managerial job performance. Journal of Organizational Behavior, 27(4), 443-461. Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effect of charismatic leadership: A self-concept based theory. Organizational Science, 4, 577-594. Smith, D., Hanges, P. J., & Dickson, M. W. (2001). Personnel selection and the five factor model: Reexamining the effects of applicants frame of reference. Journal of Applied Psychology, 86(2), 304-315. Snell, A. F., & McDaniel, M. A. (1998). Faking: Getting data to answer the right questions. Paper presented at the 13th annual conference of the Society for Industrial and Organizati onal Psychology. Dallas, TX. Snyder, M. (1987). Public appearances, private reali ties: The psychology of selfmonitoring. New York: W.H. Freeman. SPSS 2003. Statistical Package for Social Science Vr. 11.5 Palo Alto, CA: Lead Technologies Inc. Stokes, G. S., Hogan, J. B., & Snell, A. F. (1993). Comparability of incumbent and applicant samples for the development of biodata keys: The influence of social desirability. Personnel Psychology, 46(4), 739-762. Strahan, R., & Gerbasi, K. C. (1972). Shor t, homogenous version of the Marlow-Crowne social desirability scale. Journal of Clinical Psychology, 28, 191-193. Strange, J. M., & Mumford, M. D. (2002). Ch arismatic versus ideological leadership. Leadership Quarterly, 13(4), 343-377.
Political Skill, Charisma, & Performance 89 Tucker, L. R., & Lewis, C. (1973). A reli ability coefficient for maximum likelihood factor analysis. Psychometrika, 38, 1-10. Van Scotter, J. R., & Motowidlo, S. J. (1996) Interpersonal facilita tion and job dedication as separate facets of contextual performance. Journal of Applied Psychology, 81, 525-531. Van Scotter, J. R., Motowidlo, S. J., & Cro ss T. C. (2000). Effects of task performance and contextual performance on systemic rewards. Journal of Applied Psychology, 85(4), 526-535. Wagner, R. K., & Sternberg, R. J. (1990). Stre et smarts. In K. E. Clark, & M. B. Clark (Eds.), Measures of leadership (pp. 493-504). West Orange, NJ: Leadership Library of America. Wagner, R. K., & Sternberg, R. J. (1991). The Common Sense Manager: Tacit knowledge inventory for managers. San Antonio, TX: Harcourt Brace & Company. Walz, S. M., & Niehoff, B. P. (1996). Orga nizational citizenship behaviors and their effect on organizational eff ectiveness in limited-menu restaurants. In J. B. Keys & L. N. Dosier (Eds.), Academy of Management be st papers proceedings (pp. 307-311). Briarcliff Manor, NY: Academy of Management. Wheaton, B., Muthen, B., Alwin, D., & Summe rs, G. (1977). Assessing reliability and stability in panel models. In D. Heise (Ed.), Sociological methodology. (pp. 84136). San Francisco, CA: Jossey-Bass. Yerkes, S. L., Weiss, H. M., & Strickland, O. J. (1999). The effect of leader outcomes on
Political Skill, Charisma, & Performance 90 influence, attributions, a nd perceptions of charisma. Journal of Applied Psychology, 84(3), 428-436. Yukl, G. A. (1989). Leadership in organizations. Englewood Cliffs, NJ: Prentice-Hall. Zaccaro, S. J., Gilbert, J. A., Thor, K. K ., & Mumford, M. D. (1991). Leadership and social intelligence: Linking social pers pectiveness and behavioral flexibility to leader effectiveness. Leadership Quarterly, 2(4), 317-342.
Political Skill, Charisma, & Performance 91 Appendices
Political Skill, Charisma, & Performance 92 Appendix A. Ferris et al. (2005) Political Skill Inventory (PSI) Using the following 7-point scale, please indicate how much you agree with each statement about yourself. Strongly Disagree Strongly Agree Social Astuteness: 1. I always seem to instinctively know the right thing 1 2 3 4 5 6 7 to say or do to influence others. 2. I have good intuition or savvy about how to 1 2 3 4 5 6 7 present myself to others. 3. I am particularly good at sensing the motivations 1 2 3 4 5 6 7 and hidden agendas of others. 4. I pay close attention to peoples facial expressions. 1 2 3 4 5 6 7 5. I understand people very well. 1 2 3 4 5 6 7 Interpersonal Influence: 6. It is easy for me to develop good rapport with 1 2 3 4 5 6 7 most people. 7. I am able to make most people feel comfortable 1 2 3 4 5 6 7 and at ease around me. 8. I am able to communicate easily and effectively 1 2 3 4 5 6 7 with others. 9. I am good at getting people to like me. 1 2 3 4 5 6 7 Networking Ability: 10. I spend a lot of time and effort at work networking 1 2 3 4 5 6 7 with others. 11. At work, I know a lot of important people and am 1 2 3 4 5 6 7 well connected. 12. I am good at using my connections and networks to 1 2 3 4 5 6 7 make things happen at work. 13. I have developed a large network of colleagues and 1 2 3 4 5 6 7 associates at work who I can call on for support when I really need to get things done. 14. I spend a lot of time at work developing connections 1 2 3 4 5 6 7 with others. 15. I am good at building relationships with influential 1 2 3 4 5 6 7 people at work. Apparent Sincerity: 16. It is important that people believe I am sincere in 1 2 3 4 5 6 7 what I say and do. 17. I try to show a genuine interest in other people. 1 2 3 4 5 6 7 18. When communicating with others, I try to be genuine 1 2 3 4 5 6 7 in what I say and do.
Political Skill, Charisma, & Performance 93 Appendix B. Charisma Item Pool Using the following 7-point scale, please indicate how much you agree with each statement about yourself. Strongly Disagree Strongly Agree Conger & Kanungo (1994) C-K Vision and Articulation 1. Im an exciting public speaker 1 2 3 4 5 6 7 2. Im a skillful performer when presenting to a group 1 2 3 4 5 6 7 3. Im inspirational and able to motivate by articulating 1 2 3 4 5 6 7 effectively the importance of what organizational members are doing. 4. I have vision and often bring up ideas about 1 2 3 4 5 6 7 possibilities for the future. 5. I provide inspiring strategic and organizational goals. 1 2 3 4 5 6 7 6. I consistently generate new ideas for the future 1 2 3 4 5 6 7 of the organization Strange & Mumford (2002) Examples of Charisma (Modified) 7. I act according to a certain vision that specifies 1 2 3 4 5 6 7 a better future state. 8. I strive toward distal rather than proximate goals. 1 2 3 4 5 6 7 9. I communicate messages that contain reference 1 2 3 4 5 6 7 to my overall vision. 10. I personally model the values implied by the vision 1 2 3 4 5 6 7 I set forth. 11. I express high performance expectations to those 1 2 3 4 5 6 7 I work with. 12. I express confidence that my coworkers have the 1 2 3 4 5 6 7 ability to perform at high levels. 13. I will sacrifice my time, resources, or reputation at the 1 2 3 4 5 6 7 expense of my work vision. 14. I back up my requests with justification based on the 1 2 3 4 5 6 7 goodness of my vision. 15. I care about my image and will play to the desires of 1 2 3 4 5 6 7 influential coworkers. 16. I have a genuine interest in the preferences of 1 2 3 4 5 6 7 my coworkers. 17. I am motivated and rewarded when my work vision 1 2 3 4 5 6 7 is realized. 18. I take into account the needs of the organization when 1 2 3 4 5 6 7 making my work decisions. 19. I try to positively reward or reinforce coworkers for 1 2 3 4 5 6 7 performing in line with my goals.
Political Skill, Charisma, & Performance 94 Appendix B: (Continued) Strongly Disagree Strongly Agree 20. I delegate authority to my coworkers regarding work 1 2 3 4 5 6 7 tasks in line with my goals/vision. 21. I am flexible in changing my work goals to meet the 1 2 3 4 5 6 7 needs of my coworkers and organization. 22. At work, I exude confidence and a sense of purpose. 1 2 3 4 5 6 7 23. I interact closely with my coworkers when giving 1 2 3 4 5 6 7 direction or attempting to influence them. Additional Charisma Items 24. I am expressive with my face and hands when 1 2 3 4 5 6 7 supporting my ideas. 25. I encourage those I work with to share ownership 1 2 3 4 5 6 7 of my ideas. 26. I formulate my vision based on critical organizational 1 2 3 4 5 6 7 goals. 27. I demonstrate to my coworkers how committed I am 1 2 3 4 5 6 7 to my ideas. 28. Im willing to take accountability for both good and 1 2 3 4 5 6 7 bad outcomes that result from my ideas.
Political Skill, Charisma, & Performance 95 Appendix C. Motowidlo and Van Sco tters (1994) 16-item scale of OCB While performing his or her job, how likely is it that this person would Not at all likely Extremely likely 1. Comply with instructions even when supervisors 1 2 3 4 5 6 7 are not present. 2. Cooperate with others in the team. 1 2 3 4 5 6 7 3. Persist in overcoming obstacles to complete 1 2 3 4 5 6 7 a task. 4. Display proper company appearance and manner. 1 2 3 4 5 6 7 5. Volunteer for additional responsibilities. 1 2 3 4 5 6 7 6. Follow standard operating procedures and avoid 1 2 3 4 5 6 7 unauthorized shortcuts. 7. Look for challenging assignments. 1 2 3 4 5 6 7 8. Offer to help others accomplish their work. 1 2 3 4 5 6 7 9. Pay close attention to important details. 1 2 3 4 5 6 7 10. Defend the supervisors decisions. 1 2 3 4 5 6 7 11. Render proper business courtesy. 1 2 3 4 5 6 7 12. Support and encourage a coworker with 1 2 3 4 5 6 7 a problem. 13. Take the initiative to solve a work task. 1 2 3 4 5 6 7 14. Exercise personal discipline and self-control. 1 2 3 4 5 6 7 15. Tackle a difficult work assignment 1 2 3 4 5 6 7 enthusiastically 16. Voluntarily do more than the job requires to help 1 2 3 4 5 6 7 others or contribute to company effectiveness.
Political Skill, Charisma, & Performance 96 Appendix D. Snyders (1987) 18 It em Measure of Self-Monitoring Indicate whether each of the following statemen ts about you are primar ily true or false. True False 1. I find it hard to imitate the behavior of other people. (F) T F 2. At parties and social gatherings, I do not attempt to do or say things T F that others will like. (F) 3. I can only argue for ideas which I already believe. (F) T F 4. I can make impromptu speeches even on topics about which I have T F almost no information. (T) 5. I guess I put on a show to impress or entertain others. (T) T F 6. I would probably make a good actor. (T) T F 7. In a group of people, I am rarely the center of attention. (T) T F 8. In different situations and with different people, I often act like very T F different persons. (T) 9. I am not particularly good at making other people like me. (F) T F 10. Im not always the person I appear to be. (T) T F 11. I would not change my opinions (or the way I do things) in order to T F please someone or win their favor. (F) 12. I have considered being an entertainer. (T) T F 13. I have never been good at games like charades or improvisational acting. (F) T F 14. I have trouble changing my behavior to suit different people and T F different situations. (F) 15. At a party I let others keep the jokes and stories going. (F) T F 16. I feel a bit awkward in company and do not show up quite as well T F as I should. (F) 17. I can look anyone in the eye and tell a lie with a straight face T F (if for a right end). (T) 18. I may deceive people by being friendly when I really dislike them. (T) T F
Political Skill, Charisma, & Performance 97 Appendix E. Chao et al.s (1994) Polit ical Savvy Factor of Socialization Using the following 7-point scale, please indicate how much you agree with each statement about yourself at work. Strongly Disagree Strongly Agree 1. I have learned how things really work on the inside 1 2 3 4 5 6 7 of this organization. 2. I know who the most influential people are in my 1 2 3 4 5 6 7 organization. 3. I do not have a good understanding of the politics in 1 2 3 4 5 6 7 my organization. 4. I am not always sure what needs to be done to get the 1 2 3 4 5 6 7 most desirable work assignments in my area. 5. I have a good understanding of the motives behind the 1 2 3 4 5 6 7 actions of other people in the organization. 6. I can identify the people in this organization who are 1 2 3 4 5 6 7 most important to getting the work done.
Political Skill, Charisma, & Performance 98 Appendix F. Strahan & Gerbasis (1972) 10item Measure of Social Desirability True False 1. Im always willing to admit it when I make a mistake. (T) T F 2. I like to gossip at times. (F) T F 3. I never resent being asked to return a favor. (T) T F 4. There have been occasions when I took advantage of someone. (F) T F 5. I sometimes try to get even rather than forgive and forget. (F) T F 6. I have never deliberately said something that hurt someones T F feelings. (T) 7. At times I have really insisted on having things my own way. (F) T F 8. I have never been irked when people expressed ideas very T F different from my own. (T) 9. I always try to practice what I preach. (T) T F 10. There have been occasions when I felt like smashing things. (F) T F
Political Skill, Charisma, & Performance 99 Appendix G. Job Performance BARS (Borman, Ackerman & Kubisiak, 1994) Task Performance Job Knowledge: | Does not know many aspects of own job; is not | Knows ow n job reasonably well; is knowledgeab le about | Knows own job insid e and out; is very knowledgeable about | knowledgeable about methods, procedures, equipment | methods, procedures, equipment, etc., regarding own job, | methods, pro cedures, equipment, etc., as appropriate for | etc., related to own job. | but is not consid ered an expert. | successful job performance. 1 2 3 4 5 6 7 Task Proficiency: | Displays poor technical proficiency; is inaccurate in own | Performs most technical tasks with reasonable competence; | Di splays considerable mastery of all work tasks; is very | work, often makes mistakes or errors, a nd work products | is generally accurate in own work, typically avoids mistakes | accu rate in own work, consistently avoids mistakes or errors, | may lack quality. | or errors, and produces s ound products. | and produces very high quality products. 1 2 3 4 5 6 7 Effort and Productivity: | Is often late in completing work; may put forth little effort, | Is typically on time in completing tasks; usually works ha rd, | Completes all work tasks effi ciently and in a timely manner; | display poor work habits, or allow even minor obstacles, | but slacks off at times; produces av erage quantity of work; | puts forth considerable effort to complete a high quantity of | distractions, etc., to interfere with task completion; produ ces | for the most part, overcomes obstacles, distractions, etc., to | work; overcomes obstacles, dist ractions, etc., to complete the | a low quantity of work. | complete the work. | work. 1 2 3 4 5 6 7 Judgment and Problem Solving: | Tends to make poor decisions when confronted with a | For the most part, makes good decisions toward solving | Consistent ly makes good decisions toward solving even | problem; is often inaccurate at sizing up situations or | pr oblems; accurately assesses most situations or problems | diffi cult, complex problems; is always accurate at assessing | problems and ineffective at choosing a cour se of action. | and usually determines an effective course of action. | situations or problems and consistently determines an effective | course of action (e.g. ma y show excellent judgment in | interpreting work rules, procedures, etc.). 1 2 3 4 5 6 7
Political Skill, Charisma, & Performance 100 Appendix G. (Continued) Oral and Written Communication: | Does not explain things well, orally or in writing, so that | Generally speaks and writes satis factorily to the standards of | Communicates very effectively, both orally and in written | supervisors, coworkers, etc., are often confused or do not | the job; expresses self clearly enough to be understood most | form, as appropriate fo r job; expresses self very clearly so that | understand what is being communicated; has trouble with | of the time; shows reasonably good listening skills. | he/she is a lways understandable; consistently demonstrates | listening skills. | excellent listening skills. 1 2 3 4 5 6 7 Citizenship Performance Initiative: | Shows little or no interest in new/additi onal job assignments | Is willing to take on new/additional job assignments and | C onsistently seeks new/additional job assignments, responsibilities, | and responsibilities; never volunteers suggestions for | respons ibilities, but does not actively seek them out; | and chall enges; is definitely a self-s tarter; often gets involved to | improvements, new ways to acco mplish tasks, etc. | sometimes gets involved to make suggestions for | make good suggestions f or improvements, new ways to accomplish | improvements, new ways to accomplish tasks, etc. | tasks, etc. 1 2 3 4 5 6 7 Adaptability: | Has considerable trouble adapting to any organizational | Is reasonably flexible in adapting to changes in technology, | Ver y effectively adapts to changes in technology, supervision, the | changes; may be inflexible about change or otherwise react | supervision, the job or organization, etc.; in most situations, | organization, etc.; always responds well and reacts constructively to | poorly to stress, setbacks, frustrations, etc., brought on by | responds well to stress, setbacks, or frustrations related to | stress, setbacks, or frustrations related to change. | changes. | change. 1 2 3 4 5 6 7 Dependability: | Often arrives late for work, appointments, etc., and may fail | For the most part, follows attendance rules and complies with | Always conforms to organiza tion attendance rules, and follows | to follow important organization rules and procedures; has | regulations, procedures etc., is reasonably responsible and | regulations, procedures, etc.; can always be counted on to show a | discipline problems on the job and does not work very | well disc iplined at work, especially when supervisor is | high degree of responsibility and personal discipline and to work | reliably even when supervisor is present. | pres ent. | reliably with minimal or no supervision. 1 2 3 4 5 6 7
Political Skill, Charisma, & Performance 101 Appendix G. (Continued) Cooperation: | Has trouble working and interacting with s upervisors and/or | Works reasonably smoothl y with supervisors and coworkers, | Wor ks very smoothly and cooperative ly with both supervisors and | coworkers; may upset coworkers with unnecessary | for the most part, is a good team player, but works better with | coworker s; is a very good team player, avoids unnecessary conflict, | confrontations, show disrespect to superv isors, etc.; may be | some types of people than others; usually demonstrates good | and works well with all types of people; demonstrates excellent | selfish, uncooperative or otherwise sow poor service | service orientation toward internal/ext ernal customers. | service ori entation toward internal organizational customers and | orientation toward internal/external customer s. | (as appropriate ) external customers. 1 2 3 4 5 6 7 Integrity and Professionalism: | On some occasions behaves unethically on the job; tends to | Behaves ethically and, for the most part, honestly with | Alwa ys behaves ethically and is hone st and open with supervisors | blame others for own mistakes; may even steal money or | s upervisors/coworkers; does not blam e others for own work| and co workers; can be counted on to admit work-related mistakes | property from fellow employees or the or ganization. | related mistakes, but may not be very open to admitting them; | and not blame others; can be truste d beyond a doubt with money, | is basically trustworthy regardi ng money, organization | organization property, sens itive information, etc. | property, sensitive information, etc. 1 2 3 4 5 6 7 Overall Performance Overall Performance: | Considering all the factors already rated, and only these | C onsidering all the factors already rated, and only these | Con sidering all the factors already rated, and only these | factors, overall performance is usually inferior and seldom | factors, overall performance is adequate and generally meets | factors, overall performance is superior and always exceeds | meets performance standards. | performa nce standards. | performance standards. 1 2 3 4 5 6 7
About the Author David Coole received his Bachelor s Degree in Psychology from Central Michigan University in 2000 and an M.A. in Industrial/Organizational Psychology from the University of South Florida in 2004. Following his M.A., Mr. Coole entered the Ph.D. program at the Univers ity of South Florida. As a doctoral candidate, Mr. Coole worked with the Louis de la Parte Florida Mental Health Institute, Flor idas Department of Education, and the Hillsborough County Civil Service Board (HCCSB), and also cons ulted on several contra cts spanning both the public and private sectors. Throughout his candidacy, he remained active in the I/O research community and presented multiple posters and symposiums at national conventions of the Society of Industrial/Org anizational Psychology and the Academy of Management. Mr. Coole is currently em ployed by HCCSB as a Personnel Research Specialist and works in areas of selection, assessment, and employee development.