Howell and Mendez’s models

Journal of Leadership & Organizational Studies 20(2) 185 –198 © Baker College 2012 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1548051812460099 http://jlo.sagepub.com

Trait views of leadership were proposed decades ago (for reviews, see Bass, 1990; Zaccaro, 2007), with early approaches focused on the main effects of leader character- istics (i.e., generally traits that individuals were either born with or not such as height, gender, etc.) on leadership effec- tiveness, with little consideration of the mechanisms that underlie such relationships. The renewed interest in trait- view research today has focused on multistage models that more precisely articulate the intermediate linkages between leader characteristics (e.g., individual differences) and lead- ership effectiveness, and have broadened their scope to include characteristics such as interpersonal attributes and task competencies (e.g., DeRue, Nahrgang, Wellman, & Humphrey, 2011; Van Iddekinge, Ferris, & Hefner, 2009; Zaccaro, 2007).

Ammeter, Douglas, Gardner, Hochwarter, and Ferris (2002) proposed a political theory of leadership in response to an appeal for such work by House and Aditya (1997), which theoretically specified the leader characteristics or traits, behaviors, and contextual factors contributing to leadership effectiveness. A central leader individual differ- ence variable proposed in this theory was political skill, which constitutes a synergistic set of social competencies argued to bring adaptability and situationally appropriate influence behavior to interactions with followers, inspiring trust and confidence (Ferris et al., 2007). As such, leader political skill is regarded as one of the most important traits/ characteristics of leaders that contribute to leadership effec- tiveness (DeRue et al., 2011).

It has been proposed that political skill should help both leaders and followers realize effectiveness (e.g., Ferris, Davidson, & Perrewé, 2005; Ferris, Treadway, Brouer, & Munyon, 2012; Ferris, Treadway, et al., 2005; Ferris et al., 2007), because political skill is a fundamen- tally interpersonal-behavior-oriented construct, and thus would be expected to demonstrate effects on more distal per- formance and effectiveness outcomes through effects on more proximal variables such as work relationships. Indeed, recent conceptualizations of leadership have supported such proximal and distal distinctions and positioned political skill-type traits/characteristics as influencing effectiveness outcome measures through particular types of leader behav- iors and processes such as relationship building (DeRue et al., 2011; Zaccaro, Kemp, & Bader, 2004).

In a small number of studies, leader political skill has demonstrated main effect relationships with various opera- tionalizations of leadership effectiveness (Ahearn, Ferris, Hochwarter, Douglas, & Ammeter, 2004; Douglas & Ammeter, 2004; Semadar, Robins, & Ferris, 2006; Treadway et al., 2004). Unfortunately, the intermediate linkages

460099 JLO20210.1177/1548051812460099Journal of Leadership & Organizational StudiesBrouer et al. © Baker College 2012

Reprints and permission: http://www. sagepub.com/journalsPermissions.nav

1State University of New York at Buffalo, Buffalo, NY, USA 2Florida State University, Tallahassee, FL, USA

Corresponding Author: Robyn L. Brouer, Department of Organization and Human Resources, School of Management, State University of New York at Buffalo, 266 Jacobs Management Center, Buffalo, NY 14260-4000, USA Email: [email protected]

Leader Political Skill, Relationship Quality, and Leadership Effectiveness: A Two-Study Model Test and Constructive Replication

Robyn L. Brouer1, Ceasar Douglas2, Darren C. Treadway1, and Gerald R. Ferris2

Abstract

Grounded in leader–member exchange, social exchange, political skill and influence theories, the present two-study investigation tests the model that leader political skill is related to both leader and follower effectiveness through leader– follower relationship quality. It is hypothesized that leader political skill is associated with leader effectiveness (Study 1) and follower effectiveness (Study 2) through relationship quality. The results support the hypotheses and were constructively replicated in Study 2, contributing to theory that leader political skill enhances relationship quality, which then ultimately affects leader and follower effectiveness relationships. Strengths, limitations, directions for future research, and implications for practice are discussed.

Keywords

political skill, leader–member exchange, contextual performance, leadership effectiveness

186 Journal of Leadership & Organizational Studies 20(2)

between, or mechanisms that underlie, the leader political skill–leadership effectiveness relationship have not been specified or empirically investigated to date, thus leaving theoretical gaps in our understanding of the leader political skill construct and how it works (i.e., Ahearn et al., 2004; Ammeter et al., 2002; Treadway et al., 2004).

Therefore, the purpose of this investigation is to examine the impact of leader political skill on leadership effective- ness, but to do so by explicating the intermediate linkages through which leader political skill achieves effectiveness outcomes. More specifically, the present two-study investi- gation expands political skill theory by arguing that leader political skill operates to achieve desirable organizational outcomes through the relationships these leaders have with their followers. The precise focus of this investigation is to test the portions of the DeRue et al. (2011) model suggest- ing that work relationship quality mediates the leader politi- cal skill–leadership effectiveness relationships. As such, this investigation builds on recent work on trait views, rela- tional approaches, and political perspectives on leadership (e.g., Ammeter et al., 2002; Carmeli, Ben-Hador, Waldman, & Rupp, 2009; DeRue et al., 2011; Uhl-Bien, 2006).

Leadership effectiveness is a broad term that represents the various ways we can construe leadership outcomes to reflect increased performance and effectiveness of both leaders and followers. Two of the ways to construe or oper- ationalize this broad effectiveness construct are leader per- formance and follower performance (DeRue et al., 2011). Indeed, we need to know much more about the antecedents of leader performance, as well as the leader determinants of follower performance. For example, given the importance of contextual performance in contributing to organizational effectiveness (e.g., Borman & Motowidlo, 1993), research on situational factors (e.g., leader characteristics, organiza- tional culture, etc.) that might encourage contextual perfor- mance is likely to be beneficial.

Specifically, for purposes of completeness in testing the proposed conceptualization that relationship quality between leaders and followers mediates the relationship between leader political skill and leadership effectiveness, both of these outcome manifestations of leadership effec- tiveness (i.e., leader effectiveness and follower effective- ness) are examined. Specifically, this investigation involves the development and testing of the hypothesized mediation model in Study 1 (i.e., using leader performance assessed from follower and superior perspectives as the criterion measure) and a subsequent test of the model in Study 2 (i.e., using follower contextual performance as the effec- tiveness measure) in an effort to provide constructive rep- lication and theoretical extension evidence through this multistudy research package (Eden, 2002; Hochwarter, Ferris, & Hanes, 2011; Lykken, 1968). The establishment of constructive replication in such research packages pro- vides for greater confidence in the validity of the obtained

results than single-study designs permit, which are always susceptible to questions of replicability of results.

Theoretical Foundations and Hypothesis Development Theoretical Foundations

Social exchange theories are central to the study of work relationships, in general, and leader–member exchange (LMX), in particular (e.g., Erdogan & Liden, 2002; Ferris et al., 2009). The basic tenet of social exchange theory is that relationships form and endure between individuals because of the mutual obligations and interdependence that materialize between them (e.g., see Cropanzano & Mitchell, 2005, for a review). Expectations of reciprocity typically develop between the participants in an exchange relation- ship over an extended and unspecified amount of time (see Liden, Sparrowe, & Wayne, 1997). The costs and benefits of such associations are the underpinnings of individual relational satisfaction within these relationships (Blau, 1964), and subsequent work has suggested that successful social exchanges foster trust and commitment in relation- ships (e.g., Cropanzano & Mitchell, 2005), which then can foster more positive social exchanges over time.

Within organizations, leader–member relationships are at the core of an employee’s work experience. LMX theory specifies that leaders develop and maintain differing rela- tionships with followers on the basis of social exchange theory (e.g., Dansereau, Graen, & Haga, 1975; Graen & Cashman, 1975). Some leaders and followers develop higher quality relationships based the norms of mutual obli- gation and reciprocity (Gerstner & Day, 1997). These rela- tionships go beyond the employment contract, with each party offering to do more than what is explicitly required by their job, and as a result tend to be more positive in nature. However, other leader–follower relationships consist of lower quality exchanges, with each party behaving merely in ways dictated by the employment contract (Gerstner & Day, 1997), which would be more transactional in nature.

Most researchers have examined the outcomes of LMX but not the antecedents or predictors (see Dulebohn, Bommer, Liden, Brouer, & Ferris, 2011), and few studies have examined the impact of leader characteristics (i.e., skills, personality, individual differences, etc.) on LMX (e.g., Dulebohn et al., 2011; Erdogan & Liden, 2002; Gerstner & Day, 1997). In this regard, theory and research in political skill and influence provides a complement to social exchange frameworks by explicating the individual social motives, situationally adaptive behavior, influence dynamics, and perceptual lenses that affect social exchanges and work relationship quality, which then influence both leader and follower effectiveness (e.g., Ferris et al., 2007; Ferris et al., 2012).

Brouer et al. 187

Leader Political Skill and the Effectiveness of Leaders and Followers

Nature of leader political skill. Political skill is best charac- terized as an individual difference variable that represents “a comprehensive pattern of social competencies, with cog- nitive, affective, and behavioral manifestations, which have both direct effects on outcomes, as well as moderating effects on predictor–outcome relationships” (Ferris et al., 2007, p. 291) and has been defined as “the ability to under- stand social interactions at work and to use this understand- ing to influence others to act in ways that enhance one’s personal or organizational goals” (Ferris, Treadway, et al., 2005, p. 127). Politically skilled individuals are socially astute, having an accurate perception of both their own and others behaviors as well as the social interaction in general, which allows them to be influential (Ferris, Davidson, et al., 2005; Ferris, Treadway, et al., 2005; Ferris et al., 2007). Those high in political skill also possess interpersonal influ- ence, which is reflected in a certain style that allows them to easily influence others, providing the adaptability to change their behaviors as the situation demands in order to achieve desired goals (Ferris et al., 2007).

Zaccaro (2002, p. 45) earlier suggested that “successful social influence by the leader requires the mastery of a range of skills and the ability to select and apply them to the appropriate situation,” and the political skill of leaders appears to function in this manner. Moreover, these indi- viduals are characterized by networking ability and are able to forge friendships and build coalitions, strong beneficial alliances, and networks, thus vesting the politically skilled with a great deal of information and social support that they can share with followers (e.g., Ferris et al., 2007). Last, politically skilled individuals are seen as possessing high degrees of integrity and authenticity, thus creating percep- tions of as genuineness and sincerity in their interactions (Ferris, Davidson, et al., 2005).

Leader political skill and leadership effectiveness. The accu- rate perception and reading of social relationships in organi- zations may well be what characterizes effective leadership. Political skill provides individuals with the ability to accu- rately read and understand social situations and themselves. These characteristics are central to arguments that leader political skill drives leadership outcomes (e.g., Ammeter et al., 2002; Treadway, Adams, Ranft, & Ferris, 2009). Although the empirical research on political skill in the leadership context is still relatively limited, some work has supported the positive influence of political skill on leader- ship effectiveness.

Among the earliest studies, Ahearn et al. (2004) found that teams with leaders high in political skill performed bet- ter than teams with leaders lower in political skill. Similarly, Douglas and Ammeter (2004) found leader political skill to significantly predict leader effectiveness, and Semadar et

al. (2006) reported leader political skill to significantly pre- dict leader performance beyond other the other individual difference predictors of leadership self-efficacy, emotional intelligence, and self-monitoring. Although impressive, this collection of studies lacks a clear and precise explanation of the mediating factors through which leader political skill enhances leadership effectiveness, that is, an understanding of the “black box,” if you will.

Relationship Quality as Mediator of Leader Political Skill–Leadership Effectiveness Linkage

The present two-study investigation seeks to expand the- ory and research in this area by demonstrating that the work relationship quality formed between leaders and followers mediates the relationship between leader politi- cal skill and measures of leadership effectiveness. More specifically, we hypothesize that the relationships between leader political skill and leader effectiveness (i.e., opera- tionalized as leader performance in Study 1) and follower effectiveness (i.e., operationalized as follower contextual performance in Study 2) are mediated by leader–follower relationship quality.

Leader political skill and relationship quality. Building and maintaining effective relationships at work is an important task that can serve as a source of increased effectiveness and competitive advantage (Uhl-Bien, 2003; Uhl-Bien, Graen, & Scandura, 2000). Because effective relationships are impor- tant to the success of leaders, and because high-quality work relationships provide positive outcomes for all involved (e.g., Dulebohn et al., 2011; Gerstner & Day, 1987), it stands to reason that leaders should strive to develop and maintain high-quality relationships with their followers.

Using social exchange theory, the political skill of lead- ers should be associated with follower perceptions of higher relationship quality. Politically skilled leaders should be more aware that high-quality relationships are necessary in the workplace. Additionally, politically skilled leaders pos- sess the ability to inspire trust; to use their understanding and adaptability to match their behaviors, rewards, and styles to each individual follower; and to use their vast net- work to access resources and share and invest such resources in follower relationship development. As such, social exchange process dynamics involving feelings of reciproc- ity between leaders and followers should operate in such work relationships. Therefore, leaders high in political skill should realize the critical need to develop and maintain effective relationships, which should direct such leaders to invest in high-quality work relationships, resulting in recip- rocation behavior in leader-initiated goal accomplishment directions by followers, through which leader and follower effectiveness is realized.

188 Journal of Leadership & Organizational Studies 20(2)

Not only will politically skilled leaders be able to recog- nize the need for high-quality work relationships but also such leaders will reflect a deeper understanding of social exchange processes because they are attuned to their envi- ronments and followers. These leaders possess keen social astuteness that allows them to understand the motivations of others (Ferris et al., 2007). This allows politically skilled leaders to understand when followers reflect issues and needs that need to be attended to and addressed by leaders and precisely what is the best solution for that particular follower. Additionally, politically skilled leaders will engage in behaviors (e.g., provision of resources and sup- port) that followers will wish to reciprocate.

Because political skill engenders a sense of calm and per- sonal security to those who possess it, leader political skill positively affects the perceptions of followers (Ferris et al., 2007). High-quality work relationships are built on an idea of mutual obligation (e.g., Graen & Uhl-Bien, 1995; Scandura & Pellegrini, 2008). Followers have to trust that when they go above and beyond their employment contract, their lead- ers will do so as well, therefore creating the feelings of reci- procity on which high-quality work relationships are built and maintained. In support of this, Treadway et al. (2004) found that politically skilled leaders were trusted more by followers, and followers felt they received more support, were more satisfied, and committed to those leaders.

Moreover, politically skilled leaders can sense and understand the motivations of others, and this understand- ing gives them the information needed about what exactly their followers need and value. For instance, the politically skilled leader will know whether a particular follower needs or wants verbal praise, more interaction, less interaction, more informational resources to do their job, and so forth. Thus, this understanding coupled with the ability of the politically skilled to adapt (Ferris et al., 2007) will allow leaders to tailor their behaviors, rewards, and feedback to each specific follower. When followers feel they are getting what they need and want from their leaders, this will increase feelings of reciprocity toward the leader, resulting in follower perceptions of higher quality relationships and goal-directed behaviors that can increase effectiveness.

Last, House (1996) argued that leaders who are central in their networks are able to acquire more resources for their unit. Furthermore, positioning, coalition building, network- ing, and social capital creation and leveraging (e.g., Brass, 2001; House, 1995), all of which political skill facilitates, allow leaders to maximize resources, which contribute to positive follower reactions and effectiveness behaviors common ways leaders can leverage such network connec- tions on followers’ behalf, as well as through the increased reputational benefits these leaders acquire (Ammeter et al., 2002; Hall, Blass, Ferris, & Massengale, 2004).

Relationship quality and leadership effectiveness. The present investigation argues that the establishment of

high-quality work relationships with followers derived from politically skilled leaders will result in higher ratings of leader and follower effectiveness, because it is through such work relationships that leader and follower effective- ness is realized. This claim is supported by social exchange theory (e.g., Blau, 1964; Cropanzano & Mitchell, 2005), and the norm of reciprocity, which suggests that when an individual perceives favorable treatment from another individual, the individual receiving the favorable treat- ment will feel an obligation to repay the provider for such treatment (Gouldner, 1960) with behavior that favorably affects the provider in some ways. In fact, it is suggested that followers in high-quality work relationships work harder and exert more effort for the benefit of the organi- zation (Erdogan & Liden, 2002), which can only lead to greater job, unit, and/or organizational effectiveness.

The benefits of leader–political-skill-driven work rela- tionship quality extend beyond the employment contract, thus encouraging followers to do more than just what is required of them. Contextual performance is characterized as discretionary, individual behaviors that extend beyond the formal requirements of the job, helping to form the social context of all jobs, and thus promoting the effective func- tioning of jobs, work units, and organizations. Behaviors in this category include following rules and procedures, volun- teering, helping and cooperating with others, exhibiting effort, and supporting work and organization goals and objectives, and have been designated by similar terms of contextual performance, citizenship performance, and orga- nizational citizenship behavior (Borman & Motowidlo, 1993, 1997; Organ, 1997, 1988). One of the main ideas sug- gested about why individuals might choose to engage in contextual performance/OCBs involves the idea of reciproc- ity (e.g., Blau, 1964; Gouldner, 1960).

Specifically, individuals perform contextual perfor- mance because they are reciprocating what the organization (e.g., through the behaviors of leaders) has done for them. This would seem to indicate that more positive relation- ships with leaders will result in greater demonstration of contextual performance. In support of this, a recent meta- analysis found leader–follower relationship quality to be significantly and positively related to contextual perfor- mance-like behavior (Ilies, Nahrgang, & Morgeson, 2007). In light of the foregoing analysis, we argue that the relation- ships between leader political skill and leader and follower effectiveness measures operate through the quality of the work relationships between leaders and followers. DeRue et al. (2011) argued that leadership traits and characteristics (i.e., they included leader political skill as one of these) should affect leadership behaviors and attributions and rela- tional associations, which would then influence leadership effectiveness.

We follow DeRue et al.’s (2011) line of reasoning in sug- gesting that the political skill of leaders will encourage

Brouer et al. 189

them to actively make connections with followers, invest resources in such followers, and encourage feelings of reci- procity exemplified by high-quality work relationships. In turn, these leaders and their followers realize greater effec- tiveness through the attitudinal and behavioral reciproca- tion followers provide in return for the leaders’ support and resource provision that occurs in high-quality work rela- tionships. That is, the impact of leader political skill on leadership effectiveness will operate through the work relationship associations they form with followers, whereby leadership effectiveness (i.e., leader and follower) is realized.

Hypothesis: Follower perceptions of leader–follower relationship quality will mediate the relationships between leader political skill and leader effective- ness (i.e., operationalized as leader performance rated by both leaders and followers) and follower effectiveness (i.e., operationalized as follower contextual performance rated by both leaders and followers).

Plan of the Research The present research involves a two-study investigation designed to test the conceptualization indicating that leader political skill influences leader and follower effectiveness through leader–follower work relationship quality. Specifically, the mediation hypothesis is tested in Study 1, and then constructive replication and extension is sought in Study 2, testing a similar but slightly expanded model, which proposes that relationship quality mediates the rela- tionship between leader political skill and follower effec- tiveness, operationalized as contextual performance.

Research findings are difficult to accept as truly scien- tific observations until they have been repeated and subject to further scrutiny (Popper, 1959). Constructive replications (i.e., with changes in subjects, measures, rating sources, and/or sampling procedures) are particularly influential in helping establish confidence in the validity of results that have been reported initially in single-study designs (e.g., Eden, 2002; Lykken, 1968). Thus, the convergent findings reported in these multistudy research packages not only increase confidence in the validity of the obtained results but also can make meaningful contributions to theory in the field (Hochwarter et al., 2011).

Study 1: Method Procedure and Participants

Data for this study were collected in a school district in the Midwestern United States. Data for this study were collected from administrators, principals, program coordinators, and

staff personnel within one school district in the Midwestern United States. A total of 275 surveys were distributed through interdepartmental mail to staff personnel who were asked for information on their unit leaders. The leaders were either principals or program coordinators, and they provided data reflecting their personal leadership style. The follower response rate was 42% (N = 105), mostly female (75%), and with a majority (66%) having at least a 4-year college degree.

Of the 25 leaders, 60% were male (i.e., 15), and all reported having at least a master’s degree. This resulted in 105 matched dyads. A separate survey was also distributed to the program directors and superintendents (administra- tors) who had reporting authority over the principals and program coordinators (leaders) whose staff personnel responded to the initial survey. Administrators (e.g., the leaders’ superintendents) were asked to evaluate leader effectiveness. Each leader had two to five subordinates and the administrators had one to three leaders as direct reports. Participation was voluntary for all surveys.

Measures All measures were described on 7-point Likert-type scales, with 1 = Strongly disagree and 7 = Strongly agree as anchors. The items in each scale were summed and then averaged to arrive at an overall value for the scale, with higher scores indicating higher levels of each variable.

Leader political skill. Leaders self-reported their political skill using the 18-item Political Skill Inventory (PSI) devel- oped by Ferris, Treadway, et al. (2005). Examples of items include the following: “I know many important people and am well connected at work” and “I easily develop good rap- port with most people.” As suggested by Ferris et al. (2008), the four dimensions were summed into a single higher- order factor, as has been shown to be empirically valid (Ferris et al., 2008; Ferris, Treadway, et al., 2005). The coefficient alpha internal consistency reliability estimate for this scale is .96.

Leader–follower relationship quality. The 7-item Leader– Member Exchange (LMX) scale, developed by Scandura and Graen (1984), was used to measure the quality of the relationship between leader and follower (i.e., from the fol- lower’s perspective). Sample items are “My unit leader understands my problems and needs” and “I usually know where I stand with my unit leader.” The coefficient alpha reliability for this scale was .90.

We characterize political skill as an individual difference that will enhance the perceptions of LMX from the follow- er’s viewpoint. Thus, we do not see political skill as an average leader style that leads to the objective dyadic rela- tionship between the leader and follower. Like many others (e.g., Lewin, 1936), we argue that perceptions are the basis for individual actions in the workplace. Thus, rather than

190 Journal of Leadership & Organizational Studies 20(2)

examining actual dyadic relationships between the leaders and their followers, we examine the followers’ perceptions of the quality of their relationships with their leaders. This is because we believe that follower behaviors will stem directly from their perceptions and not necessarily reality (and these perceptions may or may not be accurate).

Leader effectiveness. Leader performance was used as the operationalization of leader effectiveness. Three items were used to measure leader performance, which focused on the leader’s effectiveness in representing the work unit to upper management, in meeting the job-related needs of work unit members, and in meeting the needs of the organization. Ratings on these items were taken from both followers and the leaders’ direct supervisors (i.e., the administrator), resulting in two outcome variables. The coefficient alpha reliability for the follower’s view was .86, and .92 for the administrator’s view of leader effectiveness.

Data Analyses Two methods of data analysis were used for this study. Because subordinates were nested within leaders, there may be concerns about leader effects, thus the pathway with the administrators’ view of leadership effectiveness was tested with hierarchical linear modeling (HLM). HLM is appropriate because it can account for the influence of leader factors (Raudenbush, Bryk, Cheong, Congdon, & Toit, 2004). For the model with follower ratings of effec- tiveness, Preacher and Hayes (2008) method of analyzing mediation using bootstrapping was used. This is appropri- ate because it allows a formal test of the indirect effect (Preacher & Hayes, 2008). Furthermore, it uses resampling strategies (i.e., bootstrapping), which are not subject to the assumption of normality of data (Preacher & Hayes, 2008; Shrout & Bolger, 2002). A SPSS macro was used to com- plete the mediation analyses.

Study 1: Results Descriptive Statistics and Intercorrelations of Variables

Means, standard deviations, intercorrelations, and coeffi- cient alpha reliabilities of all variables are presented in Table 1. As expected, these results reveal significant cor- relations between leader political skill, relationship quality, and leader effectiveness. Because followers each rated two variables in the model, we performed a Harmon one-factor test on leadership effectiveness and LMX. The one-factor solution did explain slightly more than half of the variance (60%). Therefore, this gives us some indication that com- mon method bias may be an issue with the data.

A confirmatory factor analysis (CFA) was run to deter- mine if these two variables were represented better as one

variable or two. Two models were compared to each other: a one-factor model, χ2(35, N = 105) = 129.51, and a two- factor model, χ2(34, N = 105) = 80.30. Next, the two models are compared to see if they are significantly different from each other using chi-square differences tests (∆χ2 = 49.21, ∆df = 1, p < .0001). The chi-square difference test indicated that the models were significantly different from each other, and because the two-factor model has the lowest chi-square, it fits the data better than the one-factor model. Thus, the leadership effectiveness and LMX variables are best repre- sented as separate variables.

Hierarchical Linear Modeling Results The HLM results are shown in Table 2. A null HLM equation (i.e., with no predictors) demonstrated that there is system- atic between-group variation in the administrator view of leader effectiveness ratings (χ2 = 63.57, df = 24, p < .00). The intraclass correlation coefficient (ICC; Hofmann, Griffin, & Gavin, 2000), calculated from the null model, indicates that 29% of the variance in ratings of administrators’ views of leadership effectiveness resides between work units.

The hypotheses were tested using “means as outcomes” analyses in HLM (e.g., Raudenbush et al., 2004). To test the proposed mediation in HLM, Baron and Kenny’s (1986) approach was adopted. As can be seen in Table 2, Step 1 of Baron and Kenny’s procedure was supported with leader political skill significantly predicting relationship quality (γ = .36, p < .0001). Next, in Step 2, supervisors’ view of leader effectiveness was regressed on leader political skill and was found to be significant (γ = .06, p < .0001). Finally, in Step 3, supervisors’ view of leader effectiveness was regressed on leader political skill and relationship quality. The impact of leader political skill dropped but remained significant (γ = .03, p < .0001), thus suggesting partial mediation.

Bootstrapping Results The results from the bootstrapping analysis indicate full mediation and are located in Table 3. Political skill is

Table 1. Descriptive Statistics, Correlations, and Reliabilities for All Variables (Study 1)

Variables M SD 1 2 3 4

1. Leader Political Skill 5.80 0.42 (.96) 2. Relationship Quality 5.44 1.10 .35** (.90) 3. Leader Performance

(follower) 5.62 1.12 .44** .73** (.86)

4. Leader Performance (administrator)

5.42 1.21 .40** .03 .16 (.92)

Note. N = 105. Reliability coefficients for the scales are on the diagonal. *p < .05. **p < .01.

Brouer et al. 191

positively related to LMX (β = .29, p < .001). LMX is positively related to leadership effectiveness (β = .04, p <05). Additionally, the indirect effect of political skill on fol- lower-rated leadership effectiveness is .03, with the bias corrected confidence intervals not containing zero, suggest- ing that political skill does operate through LMX to increase follower-rated leadership effectiveness.

Study 2: Method Procedure and Participants

Dyadic data were collected from 12 organizations to ensure generalizability. All organizations were located in the United States and included industries such as medical, technology, and manufacturing. In most organizations surveyed, we were able to invite all employees to partici- pate. In other organizations (especially larger ones), we were given an entire segment to survey. Surveys were mailed, emailed, or made available online, depending on the specific needs and requirements of the organizations (e.g., some organizations’ employees did not have regular access to computers). Leaders and followers were matched based on identification codes selected by the researchers. All participants were informed that their responses were confidential.

In all, 64 leaders and 474 followers were surveyed with 43 leader and 233 follower responses, resulting in 67% and 49% response rates, respectively. From the surveys received, there were 105 follower surveys and 37 leader surveys that could be matched, resulting in 105 dyads. The leaders held titles such as director, executive director, dentist, and man- ager. The leader respondents were 92% Caucasian, 2.7% African American, and 2.7% Asian, predominantly female (60%), averaged 51 years in age, and had 12.5 years in organizational tenure. The average number of subordinates they supervised was 24. The positions held by the subordi- nates included things such as administrative assistants, coordinators, nurses, dental assistants, and dental hygien- ists. The follower respondents were 81% Caucasian, 12.4% African American, 1.9%, Asian, and 3.8% Hispanic. The follower respondents were mostly female (81%), averaged 44 years in age, and 10 years in organizational tenure.

Measures Unless otherwise noted, all of the measures used a 5-point Likert-type scale ranging from 1 (Strongly disagree) to 5 (Strongly agree). The items in each scale were summed and then averaged to arrive at an overall value for the scale, with higher scores representing higher levels of each of the variables.

Table 2. HLM Results for Leader–Follower Relationship Quality as Mediator of the Leader Political Skill–Leader Performance Relationship (Study 1)

Equation Dependent Independent γ Estimate SE t p

1 Relationship Quality Leader Political Skill .36 .09 3.86 .00 2 Leader Performance Leader Political Skill .06 .01 4.90 .00 3 Leader Performance Leader Political Skill .03 .01 2.88 .00 Relationship Quality .09 .01 9.29 .00

Note. N = 105.

Table 3. Results of Preacher and Hayes Mediation Analysis

Predictor β SE t p

IV to Mediator Leader Political Skill 0.31 0.13 2.31 0.02 Mediator to DV LMX 0.10 0.01 10.20 0.00

Boot Indirect Effect Boot SE CI Lower End CI Upper End

Indirect effect of IV on DV through proposed mediator Total 0.03 0.03 0.01 .06

LMX 0.03 0.03 0.01 .06

Note. N = 105. IV = independent variable; DV = dependent variable; CI = confidence interval. Unstandardized regression coefficients are reported. Boot- strap sample size = 1,000. Values represent selected output provided by the Preacher, Rucker, and Hayes (2007) macro.

192 Journal of Leadership & Organizational Studies 20(2)

Leader political skill. Leaders self-reported perceptions of their political skill using the Ferris, Treadway et al. (2005) PSI (α =.91), also employed in Study 1.

Leader–follower relationship quality. Relationship quality was measured using the 7-item unidimensional scale pro- posed by Graen and Uhl-Bien (1995), assessed by follow- ers. A sample item is “My leader understands my job problems and needs” (α = .91).

Follower effectiveness. Follower contextual performance was measured from both the leader’s perspective (α =.85; sample item: “This subordinate helps others who have been absent”) and the follower’s perspective (α =.84; sample item: “I help others who have been absent”) using Williams and Anderson’s (1991) 14-item scale.

Data Analyses The hypothesized model was tested using the same meth- ods as Study 1.

Study 2: Results Descriptive Statistics and Intercorrelations of Variables

Table 4 provides descriptive statistics and intercorrela- tions of the variables in the study. As expected, leader political skill was significantly related to follower ratings of relationship quality (r = .27, p < .01), and member rat- ings of relationship quality were significantly related to both follower ratings of contextual performance (r = .34, p < .01) and leader ratings of contextual performance (r = .25, p < .01). Interestingly, follower ratings of contextual performance were not significantly correlated with leader ratings of contextual performance (r = .07, ns). Additionally, because leaders and followers each rated two variables each in the model, we performed a Harmon one-factor test for each set of respondents. In each case, the one-factor solution did not explain the most variance (follower, 29%; leader,

39%). Therefore, this gives us some indication that com- mon method bias may not be that strong of an issue in this data.

Hierarchical Linear Modeling Results Because subordinates were nested within leaders, there may be concerns about leader effects; thus, HLM was employed. A null HLM equation (i.e., with no predictors) demonstrated that there was systematic between-leader variation in leader-rated contextual performance (χ2 = 165.81, df = 36, p < .0001). The ICC coefficient (Hofmann et al., 2000), calculated from the null model, indicated that 62% of the variance in leaders’ ratings of contextual per- formance reside between leaders, which indicates that the use of HLM is appropriate.

The HLM results are shown in Table 5. The hypotheses were tested using “means as outcomes” analyses in HLM (e.g., Raudenbush et al., 2004). To test the proposed mediation of relationship quality in the leader political skill–leadership effectiveness relationships in HLM, Baron and Kenny’s (1986) approach was adopted. As can be seen in Table 6, Step 1 of Baron and Kenny’s procedure was sup- ported, with leader political skill significantly predicting relationship quality (γ = .40, p < .10). Next, in Step 2, leader-rated contextual performance was regressed on leader political skill and was found to be significant (γ = .28, p < .05). Finally, in Step 3, leader-rated contextual per- formance was regressed on leader political skill and rela- tionship quality. The impact of leader political skill dropped to nonsignificance (γ = .22, p < .10), thus providing support for mediation and the hypothesis.

Bootstrapping Results The results from the bootstrapping analysis indicate full mediation and are located in Table 3. Political skill is positively related to LMX (β = .46, p < .01). LMX is posi- tively related to leadership effectiveness (β = .17, p < .001). The model summary for the dependent variable is significant. Additionally, the indirect effect of political skill on follower-rated contextual performance is .08, with the bias corrected confidence intervals not containing zero, suggesting that political skill does operate through LMX to increase follower-rated contextual performance.

Discussion Contributions to Theory and Research

The results of this two-study investigation indicate that leader political skill has a significant indirect effect on leader effectiveness through leader–follower work relationship

Table 4. Means, Standard Deviation, and Intercorrelations for the All Variables (Study 2)

Variable M SD 1 2 3 4

1. Leader Political Skill 4.16 0.44 .28 2. Relationship Quality 4.03 0.77 .04 .79** 3. Leader-rated Contextual

Performance 4.18 0.50 −.03 .19* .25*

4. Follower-rated Contextual Performance

4.31 0.38 .01 .35** .34** .07

Note. N = 105. *p < .05. **p < .01.

Brouer et al. 193

quality, thus contributing significantly to theory in this area by helping explain how leader political skill operates to pre- dict both measures of leadership effectiveness. Generally, this research responded to calls by leadership scholars for increased work on models that more precisely articulate the intermediate linkages between leader traits/characteristics and leadership effectiveness (e.g., DeRue et al., 2011; Van Iddekinge et al., 2009; Zaccaro, 2007; Zaccaro et al., 2004). More specifically, the present investigation addresses appeals by organizational scientists for empirical research to investi- gate the mediating conditions between leader political skill and both leader and follower effectiveness (e.g., DeRue et al., 2011; Ferris et al., 2007; Ferris et al., 2012).

Although researchers have examined leader political skill in the past, most investigated its direct impact on fol- lower attitudes or performance outcomes (e.g., Ahearn et al., 2004; Treadway et al., 2004). The finding that leader politi- cal skill affects leader–follower work relationship quality, which ultimately affects leader and follower effectiveness, represents an important contribution to theory in the social exchange, LMX, and political skill areas, thus providing more specificity into how leader political skill operates to achieve positive performance outcomes. Furthermore, these results were constructively replicated and extended in a sec- ond study, which achieved comparable strength of results in terms of bootstrapping results and coefficients, and thus

provide further confidence in the validity of the obtained results (e.g., Eden, 2002; Hochwarter et al., 2011; Lykken, 1968).

These results are important because they help illuminate the “black box” with respect to leader political skill effects, suggesting that it partially operates through the relation- ships leaders develop with their followers, which then drive the effectiveness of both leaders and followers. Although several studies have examined the impact of leader political skill on leadership effectiveness, group effectiveness, and follower attitudes, there was little understanding of how political skill operates to achieve these effects (Ahearn et al., 2004; Douglas & Ammeter, 2004). Because there was a significant indirect relationship, and evidence of partial mediation replicated using bootstrapping and HLM, some of the “how it operates” has been explained, thus contribut- ing more empirical specification to theoretical arguments on political skill (Ferris et al., 2007; Ferris et al., 2012; Ferris, Treadway, et al., 2005; Treadway et al., 2004).

This research reinforces Carmeli et al.’s (2009) find- ings that leader relational behaviors affect performance through the cultivation of effective work connections and relationships. Also, the present results support Uhl-Bien’s (2003) suggestion that leader relational skills affect dyadic work relationship quality. Furthermore, these findings expand the work relationships literature, generally, and the

Table 5. HLM Results for Leader–Follower Relationship Quality as Mediator of the Leader Political Skill–Leader-Rated Contextual Performance Relationship (Study 2)

Equation Dependent Independent γ Estimate SE t p

1 Relationship Quality Leader Political Skill 0.40 0.23 1.73 .09 2 Contextual Performance Leader Political Skill 0.28 0.13 2.27 .03 3 Contextual Performance Leader Political Skill 0.22 0.13 1.73 .09 Relationship Quality 0.14 0.05 2.65 .01

Note. N = 105.

Table 6. Results of Preacher and Hayes Mediation Analysis

Predictor β SE t p

IV to Mediator Leader Political Skill 0.46 0.17 2.81 .01 Mediator to DV LMX 0.17 0.05 3.64 .01

Boot Indirect Effect Boot SE CI Lower End CI Upper End

Conditional indirect effect of IV on DV through proposed mediator Total 0.08 0.08 0.03 .14 LMX 0.08 0.08 0.03 .14

Note. N = 105. IV = independent variable; DV = dependent variable; CI = confidence interval. Unstandardized regression coefficients are reported. Boot- strap sample size = 1,000. Values represent selected output provided by the Preacher, Rucker, and Hayes (2007) macro.

194 Journal of Leadership & Organizational Studies 20(2)

LMX literature, specifically. Finally, the results of this investigation shed light on potential work context (e.g., leader characteristics, etc.) effects on employee demonstra- tion of contextual performance.

Strengths and Limitations This research has a number of strengths and limitations that deserve mention, beginning with several methodological strengths that should be noted. First, the two studies together provide constructive replication, which contributes to greater confidence in the validity of the obtained findings (e.g., Eden, 2002; Hochwarter et al., 2011; Lykken, 1968). Second, rather than relying solely on self-report data, the research was conducted using matched dyadic data from followers and their leaders and followers and the leader’s administrator. Data from multiple viewpoints alleviates concerns of response bias. Third, between the two studies, we were able to capture two different aspects of leader effectiveness. DeRue et al. (2011) and Yukl (2006) have suggested that when investigating leadership effectiveness, it is important to examine a variety of perspectives. We attempted to do this by examining multiple perspectives regarding some of the different variables (i.e., leaders, the leaders’ administrators, and followers).

To fully understand the results of this investigation, it is necessary to recognize some limitations. First, it needs to be recognized that the use of a newly developed subjective outcome measure (i.e., perceptions of leader performance) is not ideal and, therefore, represents a limitation of this research. Objective measures probably would provide more robust results. It is acknowledged in the research literature that those in higher quality work relationships with their leaders tend to be rated higher in performance regardless of their actual performance (e.g., Gerstner & Day, 1997). Additionally, one of the measures of leadership effective- ness was collected from the same viewpoint as the LMX relationship. The CFA results indicated that they were best represented by two factors. Additionally, the results of the Harmon one-factor test indicated a modest amount of com- mon method bias between the followers’ viewpoints of LMX and leadership effectiveness.

However, we did use multiple viewpoints of effective- ness as suggested by researchers (e.g., DeRue et al., 2011; Erdogan & Liden, 2002; Yukl, 2006), and we found simi- lar relationships across two studies and three different viewpoints of leadership effectiveness. This increases confidence that the results are not just artifactual of empir- ical confound or common method bias, because the other relationships were from different viewpoints and did not suffer from common method bias. Thus, we have three dif- ferent relationships that confirm that the one relationship is not merely because of empirical artifacts. Additionally, finding support for our hypotheses with different studies

(i.e., constructive replication) minimizes some of these concerns (e.g., Hochwarter et al., 2011).

Furthermore, the sample sizes were relatively small. We were able to conduct the analyses on only 105 dyads for both studies. However, although this is a small sample size, it still has more than 15 cases per variable, as suggested by Stevens (1999). Furthermore, our measures of relationship quality and follower view of leader effectiveness in Study 1, and relationship quality and follower-rated contextual performance in Study 2, were based on single-point-in-time- target (i.e., follower) reports. Although target reports pro- vide multiple views of leader behavior, such views have the potential to be biased by recent events. If possible, future research should attempt to collect each measure at different points in time.

Directions for Future Research These two studies examine the impact of leader political skill on work relationship quality and leadership effective- ness. Investigating the impact of follower political skill on the development and maintenance of high-quality leader– follower relationships is another promising area for future research. Dienesch and Liden (1986) argued that both leader and follower characteristics could affect relationship quality. These authors also maintained that followers’ uses of influencing behaviors could affect relationship develop- ment through their impact on leaders’ attributions about follower performance. It might be the case that followers’ political skill also would affect the development and main- tenance of leader–follower relationships through its effect on follower influence behaviors. Brouer, Duke, Treadway, and Ferris (2009) investigated both follower political skill and LMX. However, rather than examining the impact of follower political skill on relationship development, these authors examined the moderating effect of political skill on the racial dissimilarity—LMX relationship.

Additionally, recent work has begun to explore the rela- tionship development of the leader beyond that of dyadic exchanges (Carmeli et al., 2009; Uhl-Bien, 2006). It is argued that leaders can go beyond developing relationships just between themselves and followers, and they can be a key driver of relationship development among followers (e.g., Carmeli et al., 2009). Because political skill enables individuals to build alliances, networks, and coalitions, it stands to reason that politically skilled leaders should be able to help their group members develop positive, strong relationships.

Thus, a logical extension would be to see if the leaders’ political skill goes beyond affecting just the dyadic relation- ship between the leader and follower. For instance, it may be that politically skilled leaders serve as role models for how to interact with others in the workplace because of their ability to set others at ease and easily influence them.

Brouer et al. 195

Therefore, the entire work group might recognize these behaviors and try to model themselves after the leader. Thus, we might expect leader political skill to affect the quality of relationships formed among team members (i.e., referred to as “team-member exchange,” Seers, 1989).

Recent research on the nature of dyadic work relation- ships has argued for a focus on the underlying dimensions of such relationships, such as trust, respect, time, distance, and affect (Ferris et al., 2009). Thus, future research might seek to extend the present results to investigate which par- ticular dimensions of work relationships between leaders and followers are driven by leader political skill. Because politically skilled leaders are believed to reflect a calm sense of self-confidence, which inspires trust and confi- dence, it might be hypothesized that the major factors con- tributing to high-quality work relationships with politically skilled leaders are the trust and confidence followers develop in them, as well as the support they feel the leader will provide them.

Such research could then expand on previous work reporting that leader political skill leads to follower per- ceived support from leader and trust in leader (i.e., Treadway et al., 2004). Furthermore, Smith, Plowman, Duchon, and Quinn (2008) found that managers conveyed political skill through affability and a sense of humility, and through creating accountability and the development of trust, all of which are important qualities involved in effective work relationships (Ferris et al., 2009). This lends itself to another area of future research that stems directly from an interesting nonsignificant finding in our results. Specifically, our results indicated the administrator and follower ratings of leader effectiveness were unrelated (r = .16, p > .05). It would be interesting to compare the politi- cally skilled with the low politically skilled in this instance. It may be that politically skilled leaders would be more likely to be seen as uniformly effective by their followers and their leaders, whereas with those low in political skill, there may be disagreement between the two sets of evaluators.

Implications for Practice The results of this investigation indicate that leader political skill represents a particularly important set of social com- petencies for organizational leaders. Although political skill is thought to be partly dispositional, it is believed to reflect trainable competencies (e.g., Ferris, Davidson, et al., 2005), thus providing potential development opportunities on how to become more socially astute, use influence effectively, and how to build and manage networks. Therefore, organi- zations could use the PSI as a management selection tool, or in managerial training programs.

This article also confirmed that leader–follower quality is related to numerous positive outcomes, such as both

leader and follower effectiveness (i.e., including follower contextual performance). High-quality relationships are effective and offer a host of positive outcomes for organiza- tions (e.g., Carmeli et al., 2009; Gerstner & Day, 1997). Therefore, it could be helpful for organizations to train lead- ers on how to develop and maintain such high-quality rela- tionships with their followers. Much of this training can be integrated with the training in political skill. However, in addition to training leaders on how to be good listeners, organizations can train leaders to spend more time with their followers, discussing their needs and expectations for them (Graen, Liden, & Hoel, 1982).

Conclusion Leadership is vital for organizational success. This two- study investigation found that leaders high in political skill were able to develop high-quality relationships with their followers, which facilitated both leader and follower effectiveness. Therefore, this investigation highlights the critically important mediating role of leader–follower relationship quality in the linkages between leader political skill and leadership effectiveness measures. Hopefully, these results will stimulate further investigation into the roles of leader characteristics and the nature of relationship quality as they affect important work outcomes.

Acknowledgments

The authors wish to express their gratitude to Abraham Carmeli, David Day, K. Michele Kacmar, Lei Wang, and Kenneth J. Harris for their helpful comments on an earlier version of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, author- ship, and/or publication of this article.

References

Ahearn, K. K., Ferris, G. R., Hochwarter, W. A., Douglas, C., & Ammeter, A. P. (2004). Leader political skill and team perfor- mance. Journal of Management, 30, 309-327.

Ammeter, A. P., Douglas, C., Gardner, W. L., Hochwarter, W., & Ferris, G. R. (2002). Toward a political theory of leadership. Leadership Quarterly, 13, 751-796.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Concep- tual, strategic, and statistical considerations. Journal of Per- sonality and Social Psychology, 51, 1173-1182.

Bass, B. (1990). Bass & Stodgill’s handbook of leadership: The- ory, research, and managerial applications (3rd ed.). New York, NY: Free Press.

196 Journal of Leadership & Organizational Studies 20(2)

Blau, P. M. (1964). Exchange and power in social life. New York, NY: John Wiley.

Borman, W. C., & Motowidlo, S. J. (1993). Expanding the crite- rion domain to include elements of contextual performance. In N. Schmitt & W. C. Borman (Eds.), Personnel selection on organizations (pp. 71-98). San Francisco, CA: 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.

Brass, D. J. (2001). Social capital and organizational leader- ship. In S. J. Zaccaro & R. J. Klimoski (Eds.), The nature of organizational leadership (pp. 132-152). San Francisco, CA: Jossey-Bass.

Brouer, R. L., Duke, A., Treadway, D. C., & Ferris, G. R. (2009). Moderating effect of political skill on the demographic dissim- ilarity: Leader-member exchange quality relationship. Leader- ship Quarterly, 20, 61-69.

Carmeli, A., Ben-Hador, B., Waldman, D. A., & Rupp, D. E. (2009). How leaders cultivate social capital and nurture employee vigor: Implications for job performance. Journal of Applied Psychology, 94, 1553-1561.

Cropanzano, R., & Mitchell, M. S. (2005). Social exchange the- ory: An interdisciplinary review. Journal of Management, 31, 874-900.

Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach to leadership within formal organizations: A longitudinal investigation of the role-making process. Organi- zational Behavior and Human Performance, 13, 46-78.

DeRue, D. S., Nahrgang, J. D., Wellman, N., & Humphrey, S. E. (2011). Trait and behavioral theories of leadership: An integra- tion and meta-analytic test of their relative validity. Personnel Psychology, 64, 7-52.

Dienesch, R. M., & Liden, R. C. (1986). Leader-member exchange model of leadership: A critique and further development. Academy of Management Review, 11, 618-634.

Douglas, C., & Ammeter, A. P. (2004). An examination of leader political skill and its effect on ratings of leader effectiveness. Leadership Quarterly, 15, 537-550.

Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. (2011). A meta-analysis of antecedents and con- sequences of leader-member exchange: Integrating the past with an eye toward the future. Journal of Management. doi: 10.1177/0149206311415280

Eden, D. (2002). From the editors. Academy of Management Review, 11, 618-634.

Erdogan, B., & Liden, R. C. (2002). Social exchanges in the work- place: A review of recent developments and future research directions in leader-member exchange theory. In L. Neider & C. A. Schriesheim (Eds.), Leadership (pp. 65-114). Green- wich, CT: Information Age.

Ferris, G. R., Blickle, G., Schneider, P. B., Kramer, J., Zettler, I., Solga, J., & . . . Meurs, J. A. (2008). Political skill construct and criterion-related validation: A two study investigation. Journal of Managerial Psychology, 23, 744-771.

Ferris, G. R., Davidson, S. L., & Perrewé, P. L. (2005). Political skill at work. Palo Alto, CA: Davies-Black.

Ferris, G. R., Liden, R. C., Munyon, T. P., Summers, J. K., Basik, K., & Buckley, M. R. (2009). Relationships at work: Toward a multidimensional conceptualization of dyadic work relation- ships Journal of Management, 35, 1379-1403.

Ferris, G. R., Treadway, D. C., Brouer, R. L., & Munyon, T. P. (2012). Political skill in the organizational sciences. In G. R. Ferris & D. C. Treadway (Eds.), Politics in organizations: Theory and research considerations (pp. 487-528). New York, NY: Routledge.

Ferris, G. R., Treadway, D. C., Kolodinsky, R. W., Hochwarter, W. A., Kacmar, C. J., Douglas, C., & Frink, D. D. (2005). Develop- ment and validation of the political skill inventory. Journal of Management, 31, 126-152.

Ferris, G. R., Treadway, D. C., Perrewé, P., Brouer, R. L., Douglas, C., & Lux, S. P. (2007). Political skill in organizations. Jour- nal of Management, 33, 290-320.

Gerstner, C. R., & Day, D. V. (1997). Meta-analytic review of leader-member exchange theory: Correlates and construct issues. Journal of Applied Psychology, 82, 827-844.

Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161-177.

Graen, G., & Cashman, J. (1975). A role-making model of lead- ership in formal organization: A development approach. In J. G. Hunt & L. L. Larson (Eds.), Leadership frontiers (pp. 143-165). Kent, OH: Kent State University Press.

Graen, G., Liden, R. C., & Hoel, W. (1982). Role of leadership in the employee withdrawal process. Journal of Applied Psychol- ogy, 67, 868-872.

Graen, G., & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi- level multi-domain perspective. Leadership Quarterly, 6, 219- 247.

Hall, A. T., Blass, F. R., Ferris, G. R., & Massengale, R. (2004). Leader reputation and accountability in organizations: Impli- cations for dysfunctional leader behavior. Leadership Quar- terly, 15, 515-536.

Hochwarter, W. A., Ferris, G. R., & Hanes, T. J. (2011). Multi-study packages in organizational science research. In D. J. Ketchen & D. D. Bergh (Eds.), Building methodological bridges (Research Methodology in Strategy and Management, Vol. 6, pp. 163-199). Bingley, England: Emerald Group.

Hofmann, D. A., Griffin, M., & Gavin, M. B. (2000). The applica- tion of hierarchical linear modeling to organizational research. In K. J. Klien & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations (pp. 467-511). San Francisco, CA: Jossey-Bass.

House, R. J. (1995). Leadership in the twenty-first century. In A. Howard (Ed.), The changing nature of work (pp. 411-450). San Francisco, CA: Jossey-Bass.

House, R. J. (1996). Path-goal theory of leadership: Lessons, legacy, and a reformulated theory. Leadership Quarterly, 7, 323-352.

Brouer et al. 197

House, R. J., & Aditya, R. N. (1997). The social scientific study of leadership: Quo vadis? Journal of Management, 23, 409-473.

Ilies, R., Nahrgang, J. D., & Morgeson, F. P. (2007). Leader- member exchange and citizenship behaviors: A meta-analysis. Journal of Applied Psychology, 92(1), 269-277.

Lewin, K. (1936). Principles of topological psychology. New York, NY: McGraw-Hill.

Liden, R. C., Sparrowe, R. T., & Wayne, S. J. (1997). Leader- member exchange theory: The past and potential for the future. In G. R. Ferris (Ed.), Research in personnel and human resources management (Vol. 15, pp. 47-119). Green- wich, CT: JAI Press.

Lykken, D. T. (1968). Statistical significance in psychological research. Psychological Bulletin, 70, 151-159.

Organ, D. W. (1997). Organizational citizenship behavior: It’s construct clean-up time. Human Performance, 10, 85-97.

Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Lexington, MA: Lexington Books.

Popper, K. (1959). The logic of scientific discovery. London, England: Hutchison.

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resam- pling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879-891.

Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation hypotheses: Theory, methods, and pre- scriptions. Multivariate Behavioral Research, 42, 185-227.

Raudenbush, S., Bryk, A., Cheong, Y. F., Congdon, R., & Toit, M. (2004). HLM6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International.

Scandura, T. A., & Graen, G. (1984). Moderating effects of initial leader-member exchange status on the effects of a leadership intervention. Journal of Applied Psychology, 69, 428-436.

Scandura, T. A., & Pellegrini, E. K. (2008). Trust and leader-member exchange: A closer look at relational vulnerability. Journal of Leadership & Organizational Studies, 15, 101-110.

Seers, A. (1989). Team-member exchange quality: A new con- struct for role-making research. Organizational Behavior and Human Decision Processes, 43, 118-135.

Semadar, A., Robins, G., & Ferris, G. R. (2006). Comparing the validity of multiple social effectiveness constructs in the pre- diction of managerial job performance. Journal of Organiza- tional Behavior, 27, 443-461.

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommenda- tions. Psychological Methods, 7, 422-445.

Smith, A. D., Plowman, D., Duchon, D., & Quinn, A. (2009). A qualitative study of high-reputation plant managers: Political skill and successful outcomes. Journal of Operations Manage- ment, 27, 428-443.

Stevens, J. (1999). Applied multivariate statistics for the social sciences. Mahwah, NJ: Lawrence Erlbaum.

Treadway, D. C., Adams, G. L., Ranft, A. L., & Ferris, G. R. (2009). A meso-level conceptualization of CEO celebrity effectiveness. Leadership Quarterly, 20, 554-570.

Treadway, D. C., Hochwarter, W., Ferris, G. R., Kacmar, C. J., Douglas, C., & Ammeter, A. P. (2004). Leader political skill and employee reactions. Leadership Quarterly, 15, 493-513.

Uhl-Bien, M. (2003). Relationship development as a key ingredient for leadership development. In S. E. Murphy & R. E. Riggio (Eds.), The future of leadership development (pp. 129-145). Mahwah, NJ: Lawrence Erlbaum.

Uhl-Bien, M. (2006). Relational leadership theory: Exploring the social processes of leadership and organizing. Leadership Quarterly, 17, 654-676.

Uhl-Bien, M., Graen, G., & Scandura, T. A. (2000). Implica- tions of leader-member exchange (LMX) for strategic human resource management systems: Relationships as social capital for competitive advantage. In G. R. Ferris (Ed.), Research in personnel and human resources manage- ment (Vol. 18, pp. 137-185). Greenwich, CT: JAI Press.

Van Iddekinge, C. H., Ferris, G. R., & Hefner, T. S. (2009). Test of a multi-stage model of distal and proximal predictors of leader performance. Personnel Psychology, 62, 461-493.

Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 17, 601-617.

Yukl, G. (2006). Leadership in organizations (6th ed.). Upper Saddle River, NJ: Pearson-Prentice Hall.

Zaccaro, S. J. (2002). Organizational leadership and social intelli- gence. In R. E. Riggio, S. E. Murphy, & F. J. Pirozzolo (Eds.), Multiple intelligences and leadership (pp. 29-54). Mahwah, NJ: Lawrence Erlbaum.

Zaccaro, S. J. (2007). Trait-based perspectives of leadership. American Psychologist, 62, 6-16.

Zaccaro, S. J., Kemp, C., & Bader, P. (2004). Leader traits and attributes. In J. Antonakis, A. T. Cianciolo, & R. J. Sternberg (Eds.), The nature of leadership (pp. 101-124). Thousand Oaks, CA: Sage.

Bios

Robyn L. Brouer, PhD, is an assistant professor of organization and human resources at the State University of New York at Buffalo and has interests in social effectiveness, adaptive leader- ship, and leader–follower relationships. She has published articles in journals such as Journal of Management, Leadership Quarterly, and Journal of Organizational Behavior.

Ceasar Douglas is an associate professor of management at Florida State University. He received a PhD in management from the University of Mississippi. His research interests are in the areas of work team development, leadership, leader political skill, and temporary work-force issues.

198 Journal of Leadership & Organizational Studies 20(2)

Darren C. Treadway is an associate professor of organization and human resources at the State University of New York at Buffalo. He received a PhD from Florida State University and an MBA from Virginia Tech. His research interests include political will and skill, leadership, and social influence processes in organizations.

Gerald R. Ferris is the Francis Eppes Professor of Management and Professor of Psychology at Florida State University. He received a PhD in business administration from the University of Illinois at Urbana-Champaign. His research interests include the areas of social influence and effectiveness processes, reputation, and work relationships in organizations.