Trust and leader--member exchange: a closer look at relational vulnerability.

Trust in work relationships is an intriguing topic in leader--member exchange (LMX) research, however, previous LMX studies examined trust as unidimensional. Research on LMX has yet to explore the multidimensional nature of trust and how different dimensions may affect LMX quality. This study with 228 full-time employees explores the effect of different trust dimensions on LMX relationships. It is interesting that LMX quality was positively related to identification-based trust as well as calculus-based trust, which involves calculative exchanges. Results found support for a nonlinear association between calculus-based trust and LMX. Thus, contrary to expectations, trust appears to be vulnerable even in high-quality LMX relations.

Keywords: leader--member exchange; LMX; social exchange; trust

**********

More than four decades ago, leadership research began to challenge the existence of an average leadership style (Fleishman & Harris, 1962; Likert, 1967; Schriesheim & Stogdill, 1975; Taylor & Bowers, 1972). These studies indicated that leaders differentiate among followers rather than enacting one leadership style with all members (Dansereau, Graen, & Haga, 1975; Graen, 1976; Graen, Liden, & Hoel, 1982; Graen & Schiemann, 1978). When employees were asked to describe their relationship with their boss, different employees reported varying views of the same manager. Some employees reported high-quality exchanges characterized by a high degree of mutual trust, respect, and obligation (in-group). Others reported low-quality exchanges with their manager, where the employee fulfills job description requirements but contributes nothing extra; the relationship remains within the bounds of the employment contract (out-group) (Graen & Uhl-Bien, 1995).

These early studies led to the development of leader--member exchange (LMX) research, which posits that supervisor--subordinate relationships fall along a continuum ranging from low-quality, in which the relationship is based strictly on the transactional part of the employment contract, to high-quality relationships based on mutual liking, trust, obligation, and respect (Dienesch & Liden, 1986; Graen & Scandura, 1987; Graen & Uhl-Bien, 1995). More than 30 years of research repeatedly linked LMX quality to a wide range of work outcomes including promotion, turnover, organizational commitment, job satisfaction, citizenship behaviors, willingness to contribute, performance, and trust in supervisor (Dulebohn, Brouer, Bommer, Ferris, & Kato, 2008; Gerstner & Day, 1997; Ilies, Nahrgang, & Morgeson, 2007; Liden, Sparrowe, & Wayne, 1997).

Despite the fact that research on trust is essential to a deeper understanding of LMX relationships, previous research on LMX examined trust as a unidimensional construct, and research has yet to examine different trust dimensions in the context of LMX relationships. Essentially, researchers have suggested that high-quality LMX relations involve high levels of interpersonal trust, which carries the relationship beyond the formal employment contract (Bauer & Green, 1996; Dienesch & Liden, 1986; Uhl-Bien, Graen, & Scandura, 2000). Therefore, prior research primarily studied LMX as a social exchange relationship involving high levels of affective interpersonal trust. However, in social exchange relationships, social transactions may coexist with economic transactions (Cropanzano & Mitchell, 2005). Accordingly, high-LMX relationships may also involve a transactional (i.e., calculative) component, which may render trust to be vulnerable even in high-quality exchange relationships (Uhl-Bien, 2007).

Trust is a construct that has received broad attention in social sciences including psychology, sociology, political science, and economics (see Gambetta, 1988; Lewicki & Bunker, 1995; Lewicki, Tomlinson, & Gillespie, 2006, for reviews). In the context of organizations, research on trust has proliferated, which was reflected a decade ago in a 1998 special issue of Academy of Management Review on trust in work relationships. Furthermore, there are two recent metaanalyses on trust in organizations that found trust in manager to be positively related to job performance and organizational citizenship behavior and negatively related to counterproductive work outcomes, such as intention to quit the organization (Colquitt, Scott, & LePine, 2007; Dirks & Ferrin, 2002).

Although trust is critical to the success of professional relationships, little effort has been made to synthesize research on trust to advance LMX theory. A number of recent studies examined the association between trust and LMX and found support for a significant and positive association (Dulebohn et al., 2008; Gomez & Rosen, 2001; Wat & Shaffer, 2005). However, there is still a considerable lack of agreement among LMX researchers with regard to whether trust is an antecedent or a consequence of LMX quality. Some studies examined trust as an antecedent of LMX (Dulebohn et al., 2008; Gomez & Rosen, 2001), whereas others studied it as an outcome of LMX (Wat & Shaffer, 2005). Thus, research integrating trust and LMX literatures is timely and warranted.

LMX has been defined as a trust-building process that develops fairly quickly and remains stable over time (Bauer & Green, 1996; Graen & Cashman, 1975; Liden, Wayne, & Stilwell, 1993). This definition considers trust to be a unidimensional construct (Graen & Uhl-Bien, 1995). However, trust in work relationships is multidimensional and its meaning cannot be captured by a single, static definition (Brower, Schoorman, & Tan, 2000; Dirks & Ferrin, 2002; Lewicki & Bunker, 1996; Shapiro, Sheppard, & Cheraskin, 1992). Different forms of trust may have different influences on the development of work relationships (Shapiro et al., 1992). However, previous research on LMX has overlooked the multidimensional nature of trust.

In high-quality LMX relations, the leader engages in risk-taking behaviors such as delegation and empowerment (Bauer & Green, 1996; Schriesheim, Neider, & Scandura, 1998). According to Brower et al. (2000), the amount of risk taken is related to the level of trust the leader has in the employee. Despite the importance of trust in LMX, empirical research has only recently begun to integrate the two literatures (Mitchell & Uhl-Bien, 2004). In addition, previous LMX research studied trust as unidimensional, which has delayed a more in-depth study of the role of trust in exchange relations. However, to advance LMX theory, it is essential that research examines the effects of different trust forms on relationship quality. Thus, the primary goal of this study is to explore the relations of different trust dimensions with the quality of LMX relationships. An examination of trust as a multidimensional construct may help provide much-needed inquiry into the LMX literature concerning the role of trust in LMX relations.

First, we provide a brief introduction to the LMX literature. Next, we integrate the existing LMX models with the current literature on trust in business organizations. Finally, we test hypotheses derived from our model with a sample of 228 full-time employed professionals.

LMX Models

LMX theory articulates the development of a work relationship between the supervisor and the subordinate. LMX researchers proposed various models of this process (Dienesch & Liden, 1986; Graen & Scandura, 1987; Graen & Uhl-Bien, 1995). Graen and Scandura's (1987) role-making model described LMX as comprised of three phases: role-taking, role-making, and role-routinization. In the role-taking stage, the leader makes a request or assigns a task and evaluates the member's behavior and performance to assess the underlying motivation and potential. This phase corresponds to the first step in Dienesch and Liden's (1986) model in which demographic characteristics and personalities may influence the initial interaction between the two parties (due to limited information early in the relationship). Role-making is the continuation of the developmental process where the nature of the leader--member relationship becomes more defined. It is at this stage that managers and subordinates begin to cement the relationship (Bauer & Green, 1996). The leader provides an opportunity to the member by assigning an unstructured task. If the member accepts that opportunity, the relationship continues to develop into a high-quality exchange relationship (Liden et al., 1997). The third stage is role-routinization. At this point, leader and member develop a common understanding and clear mutual expectations. The behaviors of the leader and member become interlocked (Graen & Scandura, 1987) and the quality of exchange typically remains stable after this phase (Liden et al., 1997).

Although the role-making model has traditionally formed the basis of LMX research, more recent work has proposed the leadership-making model (Graen & Uhl-Bien, 1995). The LMX relationship in this framework begins with the stranger phase. The leader and member occupy interdependent organizational roles and begin interactions. They interact on a formal basis, which Graen and Uhl-Bien (1995) refer to as a "cash-and-carry" economic exchange. The model characterizes the stranger phase as low-quality LMX because of the absence of caring and commitment (Liden et al., 1997). In this phase, once an offer for an improved work relationship through career-oriented exchange has been accepted by both members, the dyad can then move to the acquaintance phase. In the acquaintance phase, leader and member begin to share greater information both on a personal level and work level. This is a critical stage because dyads that do not develop may revert to the stranger phase (Graen & Uhl-Bien, 1995). The next stage is referred to as mature partnership, in which the exchanges are not only behavioral but also emotional. The members count on each other for loyalty and support, and mutual trust, respect, and obligation also grow throughout this process. Leaders count on their followers to provide them with assistance whenever needed, and followers rely on their leaders for support, encouragement, and career investments. The leadership-making model refers to this stage as mature partnerships because it reflects high-quality LMX (Graen & Uhl-Bien, 1995).

Integrating LMX and Trust Literatures

The most commonly used definition of trust in the Academy of Management Review' s 1998 special issue on trust in organizations is "willingness to be vulnerable" (Mayer, Davis, & Schoorman, 1995, p. 712). Another commonly used definition proposed by McAllister (1995) defines trust as "an individual's belief in, and willingness to act on the basis of the words, actions and decisions of another" (p. 25).

Previous LMX research examined trust as unidimensional when characterizing high-quality exchanges by high degree of trust and low-quality exchanges by low trust (Dienesch & Liden, 1986; Graen & Scandura, 1987; Graen & Uhl-Bien, 1995). The pervasiveness of studying trust as unidimensional may be attributed to the fact that our understanding of trust in LMX relationships is still in its infancy (Dulebohn et al., 2008; Lewicki, McAllister, & Bies, 1998). Because trust evolves as parties interact (Mayer et al., 1995), one dimension of trust may be salient in low-quality exchanges, whereas other dimensions may be more salient in high-quality exchanges.

Shapiro et al. (1992) and Lewicki and Bunker (1995) presented three types of trust operating in work relationships: calculus-based trust, knowledge-based trust, and identification-based trust. Further elaboration of this model suggested that knowledge-based trust was a dimension of the relationship rather than a dimension of trust. Accordingly, more recent trust research suggests an examination of two fundamental forms of trust: calculus-based trust (CBT) and identification-based trust (IBT) (Lewicki, Wiethoff, & Tomlinson, 2005).

CBT is a market-oriented, transactional, economic calculation that involves weighing the outcomes from sustaining the relationship relative to the costs of severing it (Lewicki & Bunker, 1995). This form of trust is partial and fragile. In essence, CBT is akin to Graen and Uhl-Bien's (1995) "low-quality LMX" conceptualization. In relationships characterized with CBT, dyad members continually assess the rewards and costs associated with sustaining the relationship. In other words, they form a relationship because they believe it is in their interest to do so (Boyd & Taylor, 1998). Graen and Uhl-Bien (1995) characterize this phase as a cash-and-carry economic exchange. Lewicki and Bunker's (1995) definition of CBT is also conceptualized as a market-oriented, economic calculation.

Uhl-Bien et al. (2000) suggested that CBT would only occur at low levels of LMX. They argue that once LMX quality increases, CBT should dissipate. Liden et al. (1997) also argue that economic calculations depict low-quality LMX because they lack caring, loyalty, and commitment. Therefore, previous LMX research suggests a negative association between CBT and the quality of LMX relationship.

Social exchange theory argues that in exchange for providing personal and positional resources, supervisors may demand greater levels of assistance and may have higher performance expectations from subordinates in high-LMX relationships (Blau, 1964; Liden & Graen, 1980). Therefore, in high-LMX relationships, the demands of the supervisor may overwhelm the subordinate, which may result in increased stress for the follower (Harris & Kacmar, 2006). Accordingly, we suggest that the follower in an increasingly high-LMX relationship may start to reevaluate the costs and benefits of sustaining the relationship, despite the fact that it is a high-quality exchange relationship. Therefore, we expect that the subordinate's high-quality exchanges with the supervisor may no longer counteract the demands of increased workload and stress. Essentially, there may be a point of diminishing returns, where the increased support and communication no longer compensate for the increased obligations in the relationship. At this stage, the employee may begin to question whether the supervisor is taking advantage of the subordinate's commitment and loyalty. Therefore, we expect that high-LMX relationships (due to stress and increased workload) may still involve high levels of CBT, resulting in a curvilinear relationship.

Hypothesis 1: A U shape characterizes the relationship between CBT and LMX. Specifically, LMX is high when CBT is low, LMX decreases as CBT becomes moderate, and as LMX becomes high, CBT increases as well.

The second type of trust is IBT. At this level, the members effectively understand and appreciate each other's needs. They serve as each other's agent and substitute for the other in interpersonal transactions. IBT allows the members to be confident that their interests will be fully protected and that no surveillance of the other member is necessary (Lewicki & Bunker, 1996). This conception is in line with McAllister's (1995) affective trust, which highlights the emotional bond between the parties. We expect that this level of trust will exist in mature partnerships (high-quality LMX) characterized by a high degree of respect, trust, and obligation (Graen & Uhl-Bien, 1995). In this stage, the members fully count on one another for loyalty and support. Leaders continue to count on their followers to provide them with partnership assistance, and followers confidently rely on their leaders for support, encouragement, and career investments. Therefore, we expect to find a positive association between IBT and high-quality LMX.

Hypothesis 2: There is a positive linear relationship between IBT and LMX.

Method

Participants

The sample included 228 full-time employed professionals enrolled in an Executive MBA program at a large Southeastern university. Self-administered surveys were distributed during the last week of classes to be returned to the professor within a week, either personally or by mail or e-mail. Participation was voluntary and anonymous (no individual identification was collected on the questionnaire). Over four semesters, 242 questionnaires were distributed and 228 were returned, resulting in a response rate of 94%.

Slightly more than half of the participants were female (53%). The majority of the participants worked in the service sector (30%), followed by manufacturing (22%), health care (15%), and education (6%). The average age of the participants was 31 years (SD = 7.8 years), with an average tenure in their current organization of 3.0 years (SD = 2.6 years).

Measures

LMX was measured using the seven-item scale from Scandura and Graen (1984). This scale uses a 4-point response format with higher scores representing higher exchange quality. A sample item from this scale is, "How would you characterize your working relationship with your manager?" Anchors corresponding to this question ranged from 1 (extremely ineffective) to 4 (extremely effective). Coefficient alpha for the scale scores was .89.

Trust was measured with Lewicki, Bunker, and Stevenson's (1997) 11-item scale. All survey items had a 5-point response format with 1 (strongly disagree) to 5 (strongly agree) as anchors. Participants were provided with instructions that read, "Please answer the following questions about your relationship with your manager." The CBT scale included five items with a Cronbach alpha of .88. An example item from this scale is, "This person knows that the benefits of maintaining the trust are higher than the costs of destroying it." The IBT scale included six items with a Cronbach alpha of .91. An example item from this scale is, "This person and I share the same basic values."

Discriminant validity was examined through a confirmatory factor analysis. As shown in Table 1, model fit was significantly improved with the three-factor model (CBT, IBT, and LMX) as compared with the two-factor model (LMX and a unidimensional trust construct) ([DELTA][[chi square].sub.2] = 280.16, p < .001). Thus, empirical results support the distinctiveness of LMX and two trust dimensions as independent constructs.

Results

Table 2 shows the means, standard deviations, and intercorrelations among the study variables. Consistent with LMX theory, relationship duration was not related to either LMX or trust (Liden et al., 1993).

Calculus-Based Trust

A hierarchical regression analysis was conducted to examine whether the association between CBT and LMX was nonlinear. First, we mean-centered CBT scores to prevent the introduction of multicollinearity into the model (Cohen, Cohen, West, & Aiken, 2003). Next, we controlled for the influence of IBT. In Step 2, the linear CBT term was entered. In Step 3, the quadratic CBT term was entered, and in Step 4, the cubic CBT term was entered. If linearity is the best representation of the CBT-LMX relationship, CBT alone should explain a significant amount of the variance. However, if the inclusion of the squared CBT term explains a significant amount of variance beyond that due to the linear term, then a curvilinear form (U shape or an inverted U shape) would better represent the relationship. Finally, if the cubed term explains a significant amount of variance beyond that accounted for by the linear and squared terms, then the relationship would have two bends in the curve (an S shape).

As shown in Table 3, neither the linear ([R.sup.2] = .08, p > .05) nor the squared ([R.sup.2] = .07, p > .05) CBT terms were significant. However the cubic term was significantly related to LMX ([R.sup.2] = .09, p < .05). Therefore, in contrast to the hypothesized U-shaped association, the results suggest a third-order polynomial relationship (i.e., S shaped) between CBT and LMX (see Figure 1).

[FIGURE 1 OMITTED]

Identification-Based Trust

Once again, we conducted a hierarchical regression analysis to test Hypothesis 2. As shown in Table 4, the linear term was significantly and positively related to LMX ([R.sup.2] = .08, p < .001), however, neither the squared ([R.sup.2] = .07, p > .05) nor the cubic ([R.sup.2] = .07, p > .05) terms were statistically significant. Accordingly, the results suggest a linear relationship between IBT and LMX, providing support for Hypothesis 2 (see Figure 2).

[FIGURE 2 OMITTED]

Discussion

Prior LMX research examined trust as unidimensional. In this study, through an examination of a multidimensional trust model, we were able to address important, yet unexplored, assumptions in the LMX model. More specifically, LMX research suggests that high-quality relationships are characterized by social transactions involving high levels of affective trust in contrast to calculative and economic transactions (Graen & Uhl-Bien, 1995). However, these findings challenge this assumption and demonstrate that trust may be fragile even in high-quality exchanges.

Theoretical Implications

Earlier research suggested a negative relationship between CBT and LMX. Uhl-Bien et al. (2000) stated that CBT would only occur at low levels of LMX, and once LMX quality increases, CBT should dissipate. However, our results suggest that CBT may operate across levels of LMX. Specifically, as prior research suggests, CBT starts low in high-LMX relations; however, as CBT reaches mid-level, LMX may start to increase. LMX quality decreases only after CBT reaches its highest levels. Recently, Schoorman, Mayer, and Davis (2007) stated that unlike LMX, trust is not necessarily mutual and reciprocal. One of the implications of this argument is that the leader may create a high-quality LMX environment via sharing personal and positional resources, however, subordinates' trust in the leader may still be vulnerable.

These findings are consistent with Blau's (1964) depiction that exchange relationships involve both socio-emotional and economic components. Recently, Cropanzano and Mitchell (2005) suggested that social exchange relations (e.g., LMX) may involve economic as well as social transactions. Therefore, although these results may challenge previous LMX theory, the findings are consistent with more recent work on LMX (Cropanzano & Mitchell, 2005) as well as Blau's original depiction of exchange relationships involving social and economic components.

The curvilinear association between LMX and CBT is an important finding because it suggests that economic exchange considerations where members continually consider the rewards and costs associated with sustaining the relationship may increase together with LMX quality. CBT may increase with LMX because the demands of the supervisor in high-LMX relations may overwhelm the subordinate, which may result in increased workload and added stress for the follower (Harris & Kacmar, 2006). Therefore, the follower may continue to reevaluate the costs and benefits of sustaining the relationship, despite the fact that it is an increasingly higher quality exchange relationship.

Furthermore, perhaps high-LMX relations in today's dynamic work environment are not as stable as suggested by research from more than a decade ago. It may be that in today's organizations, individuals maintain some focus on themselves even in high-quality exchange relationships (Uhl-Bien & Maslyn, 2003). From a practitioner standpoint, this finding suggests that both members and leaders in a high-quality exchange relationship should be conscious of the other party's continual weighing of costs and benefits toward sustaining the relationship. These findings suggest that high-quality exchange should not be taken for granted as a stable process. On the contrary, it may still involve a CBT component that makes the relationship vulnerable at all times.

Research on LMX has more recently focused on the extent to which subordinates' relationships in organizational networks mirror their relations with their managers (i.e., LMX). Sparrowe and Liden (2005) suggested that as the leader's sponsorship of the member increases, the member's position as a reliable and trustworthy exchange partner approaches that of the leader. In other words, members borrow social capital from their formal leaders. Sparrowe and Liden found that the relation between LMX and the member's influence in the organization is moderated by the leader's access to resources (i.e., leader's network centrality), which may be independent of the leader's positional resources because it may flow through informal networks (Brass, 1984). Given that the network centrality of the leader is attained through the leader's informal relationships, it is not surprising that high-quality LMX relations still involve high levels of CBT, especially because sharing trust ties with a leader who is low in centrality may be detrimental to the member's influence in the organization (Sparrowe & Liden, 2005). This line of research makes these findings even more relevant by extending the significance of this study's results beyond the boundaries of the immediate work group to the organizational level.

Practical Implications

Our results have some significant practical implications. First, previous theory indicates that calculative, transactional exchanges occur only in low-quality LMX relationships (Graen & Scandura, 1987; Graen & Uhl-Bien, 1995). However, our results indicate that high-LMX relationships may also involve economic transactions. Furthermore, prior research suggested that LMX as a trust-building process develops quickly and remains stable over time (Bauer & Green, 1996). However, these results suggest that LMX may be vulnerable to trust violations even in high-quality relationships. Thus, this study demonstrates that LMX relationships may not be as stable as suggested by prior research and, therefore, effective managers need to not only gain the trust of their subordinates but also learn to sustain their trust.

Leaders and followers are partners in an exchange relationship. If either member of the dyad experiences lack of trust, it will be difficult to maximize the potential positive outcomes evolving from this relation (Brower, Lester, Korsgaard, & Dineen, 2008). Therefore, to ensure that high-LMX relationships are successfully sustained, organizations should inform their managers of the potential negative results of increased obligations and stress in high-LMX relationships (Harris & Kacmar, 2006). Managers need to be aware that LMX relationships are susceptible to trust violations at any level.

Strengths and Limitations

Notwithstanding its contributions to the extant research on LMX, our study had some limitations. First, the data were cross-sectional, which limits any causal conclusions that can be made about the relations between different trust components and LMX. Also, we measured LMX only from the subordinate's perspective. We suggest that further research measure LMX from both the leader's and the member's perspective to examine whether measurement perspective may act as a moderator of the relationship between different types of trust and LMX. In addition, the data were obtained from the same source, which may raise concerns about common-rater bias. However, inspection of the correlation coefficients does not indicate inflated correlations as would be expected if such a bias were having an effect on the data (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003; Spector, 2006). Furthermore, we followed Podsakoff and Organ's (1986) procedure and performed a principal components analysis on all items. If common method variance is present, the factor analysis should reveal either a single factor or a dominant general factor that accounts for a majority of the variance in individual responses. The results suggested that no method factor was apparent.

Despite these limitations, this study offers an important contribution to the advancement of LMX theory and practice. These results suggest that before we make definitive conclusions about LMX relationships, we need further understanding of the underlying trust characteristics prevalent in these exchanges. Undoubtedly, more research is needed to understand the complexity of exchange relationships between leaders and followers. However, we trust that these findings will shed some light on the trust dynamics in high-quality LMX relationships.

Directions for Further Research

We suggest a number of fruitful areas for future research studies. First, research is needed that examines whether individual differences (e.g., propensity to trust) and social context (e.g., organizational culture) may moderate the curvilinear relation found between CBT and LMX. Furthermore, it would be enlightening to study LMX as a multidimensional construct. In this study, we followed Graen and Scandura's (1987) unidimensional LMX model, however, Liden and Maslyn (1998) have empirically shown that LMX is multidimensional and made up of four dimensions: contribution, loyalty, professional respect, and affect. Further research should examine trust using a multidimensional LMX model and see what additional insights can be acquired into the curvilinear effect found in this study.

Another area that warrants attention is cross-cultural research on trust in LMX relations. There is still a dearth of research that examines how trust is different across cultures (Schoorman et al., 2007). International LMX research may find divergent results concerning the effectiveness of certain management functions, such as fair treatment (i.e., organizational perceptions) and participative decision making (i.e., empowerment) on trust formation. For example, Pellegrini and Scandura (2006) found that employees in the Middle East may be disinterested in delegation. They argue that employees in high-power-distance cultures may expect the leader to take charge and give orders rather than delegate decision-making authority to the subordinate. There is also some evidence in the culture literature that initial trust of strangers varies across cultures (Schoorman et al., 2007). The majority of published research on trust and LMX has been conducted with Western samples, however, globalization increasingly challenges today's managers to become more cross-culturally adept. Thus, international research that examines trust in dyadic work relationships is a promising direction for further research.

Conclusion

These results challenge the assumption that work relationships (such as LMX) follow the same patterns as relational trust. Although one may expect CBT to be associated with low-quality relationships, we found that high-quality LMX may also involve CBT. LMX may stabilize in a short period of time and remain stable over the course of the relationship (Liden et al., 1993), however, keeping promises and commitments and following through are important to maintain the relationship. Our results highlight the vulnerability of high-quality relationships to trust violations at any level and the continued need for the leader to attend to the expectations of followers.

References

Bauer, T. N., & Green, S. G. (1996). The development of leader--member exchange: A longitudinal test. Academy of Management Journal, 39, 1538-1567.

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

Boyd, N. G., & Taylor, R. R. (1998). A developmental approach to the examination of friendship in leader-follower relationships. Leadership Quarterly, 9, 1-25.

Brass, D. J. (1984). Being in the right place: A structural analysis of individual influence in an organization. Administrative Science Quarterly, 29, 518-539.

Brower, H. H., Lester, S. W., Korsgaard, M. A., & Dineen, B. R. (2008). A closer look at trust between managers and subordinates: Understanding the effects of both trusting and being trusted on subordinate outcomes. Journal of Management. Prepublished February 5, 2008. DOI: 10.1177/0149206307312511.

Brower, H. H., Schoorman, E D., & Tan, H. H. (2000). A model of relational leadership: The integration of trust and leader--member exchange. Leadership Quarterly, 11, 227-250.

Cohen, J., Cohen, E, West, S., & Aiken, L. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.

Colquitt, J. A., Scott, B. A., & LePine, J. A. (2007). Trust, trustworthiness, and trust propensity: A meta-analytic test of their unique relationships with risk taking and job performance. Journal of Applied Psychology, 92, 902-927.

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

Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad approach to leadership within formal organizations. Organizational Behavior and Human Performance, 13, 46-78.

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.

Dirks, K. T., & Ferrin, D. L. (2002). Trust in leadership: Meta-analytic findings and implications for research and practice. Journal of Applied Psychology, 87, 611-628.

Dulebohn, J. H., Brouer, R. L., Bommer, W. H., Fenis, G. R., & Kato, K. (2008). A meta-analysis of the antecedents of leader--member exchange: Integrating the past with an eye toward the future [Working paper]. East Lansing: Michigan State University.

Fleishman, E. A., & Harris, E. F. (1962). Patterns of leadership behavior related to employee grievances and turnover. Personnel Psychology, 15, 43-56.

Gambetta, D. (1988). Trust: Making and breaking cooperative relations. Oxford, UK: Basil Blackwell.

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.

Gomez, C., & Rosen, B. (2001). The leader-member exchange as a link between managerial trust and employee empowerment. Group & Organization Management, 26, 53-69.

Graen, G. B. (1976). Role-making process within complex organizations. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp. 1201-1245). Chicago: Rand McNally.

Graen, G. B., & Cashman, J. (1975). A role-making model of leadership in formal organizations: A developmental approach. In J. Hunt & L. L. Larson (Eds.), Leadership frontiers (pp. 143-166). Kent, OH: Kent State University Press.

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

Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of dyadic organizing. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (Vol. 9, pp. 175-208). Greenwich, CT: JAI Press.

Graen, G. B., & Schiemann, W. (1978). Leader--member agreement: A vertical dyad linkage approach. Journal of Applied Psychology, 63, 206-212.

Graen, G. B., & 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.

Harris, K. J., & Kacmar, K. M. (2006). Too much of a good thing: The curvilinear effect of leader-member exchange on stress. Journal of Social Psychology, 146(1), 65-84.

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.

Lewicki, R. J., & Bunker, B. B. (1995). Trust in relationships: A model of trust development and decline. In B. B. Bunker & J. Z. Rubin (Eds.), Conflict, cooperation, and justice (pp. 133-173). San Francisco, CA: Jossey-Bass.

Lewicki, R. J., & Bunker, B. B. (1996). Developing and maintaining trust in work relationships. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 114-139). Thousand Oaks, CA: Sage.

Lewicki, R. J., Bunker, B. B., & Stevenson, M. A. (1997). The three components of interpersonal trust: Instrument development and differences across relationships. Paper presented at the meeting of the International Association of Conflict Management, Germany.

Lewicki, R. J., McAllister, D. J., & Bies, R. J. (1998). Trust and distrust: New relationships and realities. Academy of Management Review, 23, 438-458.

Lewicki, R. J., Tomlinson, E. C., & Gillespie, N. (2006). Models of interpersonal trust development: Theoretical approaches, empirical evidence, and future directions. Journal of Management, 32, 991-1022.

Lewicki, R. J., Wiethoff, C., & Tomlinson, E. (2005). What is the role of trust in organizational justice? In J. Greenberg & J. Colquitt (Eds.), Handbook of organizational justice: Fundamental questions about fairness in the workplace (pp. 247-270). Mahwah, NJ: Lawrence Erlbaum.

Liden, R. C., & Graen, G. (1980). Generalizability of the vertical dyad linkage model of leadership. Academy of Management Journal, 23, 451-465.

Liden, R. C., Sparrowe, R. T., &Wayne, S. J. (1997). Leader-member exchange theory. The past and potential for the future. Research in Personnel and Human Resources Management, 15, 47-119.

Liden, R. C., Wayne, S. J., & Stilwell, D. (1993). A longitudinal study on the early development of leader-member exchange. Journal of Applied Psychology, 78, 662-674.

Likert, R. (1967). The human organization: Its management and value. New York: McGraw-Hill.

Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational mast. Academy of Management Review, 20, 709-734.

McAllister, D. J. (1995). Affective and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal, 38, 24-59.

Mitchell, M. S., & Uhl-Bien, M. (2004). Exploring the relationship of trust and leader-member exchange: A social exchange perspective. Presented at the annual meeting of the Academy of Management, New Orleans, LA.

Pellegrini, E. K., & Scandura, T. A. (2006). Leader--member exchange (LMX), paternalism and delegation in the Turkish business culture: An empirical investigation. Journal of International Business Studies, 37(2), 264-279.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. E (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903.

Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12, 531-544.

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, 71, 579-584.

Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative model of organizational trust: Past, present, and future. Academy of Management Review, 32, 344-354.

Schriesheim, C. A., Neider, L. L., & Scandura, T. A. (1998). Delegation and leader-member exchange: Main effects, moderators, and measurement issues. Academy of Management Journal, 41, 298-318.

Schriesheim, C. A., & Stogdill, R. (1975). Differences in factor structure across three versions of the Ohio State Leadership Scales. Personnel Psychology, 28, 189-206.

Shapiro, D. L., Sheppard, B. H., & Cheraskin, L. (1992). Business on a handshake. Negotiation Journal, 8, 365-377.

Sparrowe, R. T., & Liden, R. C. (2005). Two routes to influence: Integrating leader--member exchange and social network perspectives. Administrative Science Quarterly, 50, 505-535.

Spector, P. E. (2006). Method variance in organizational research: Truth or urban legend? Organizational Research Methods, 9, 221-232.

Taylor, J., & Bowers, D. (1972). The survey of organizations: A machine-scored standardized questionnaire instrument. Ann Arbor: Institute for Social Research, University of Michigan.

Uhl-Bien, M. (2007). The Achilles' heel of LMX: Reexamining the measurement-theory link. Panelist, the Society for Industrial and Organizational Psychology meetings, New York, NY.

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

Uhl-Bien, M., & Maslyn, J. M. (2003). Reciprocity in manager--subordinate relationships: Components, configurations, and outcomes. Journal of Management, 29, 511-532.

Wat, D., & Shaffer, M. A. (2005). Equity and relationship quality influences on organizational citizenship behaviors: The mediating role of trust in the supervisor and empowerment. Personnel Review, 34, 406-422.

Terri A. Scandura

University of Miami, Coral Gables, Florida

Ekin K. Pellegrini

University of Missouri--St. Louis

Terri A. Scandura, PhD, is the dean of the graduate school and a professor of management at the University of Miami. Her research focuses on leadership, mentoring, and applied research methods.

Ekin K. Pellegrini, PhD, is an assistant professor in the Department of Management at the University of Missouri--St. Louis. Her research focuses on leadership and mentoring. Table 1 Confirmatory Factor Analysis Results for Leader-Member Exchange (LMX) and Trust Dimensions as Distinct Constructs

[DELTA] Competing Model [chi square] df [chi square] One-factor model 1723.61 135 Two-factor model (LMX and trust) 1024.75 134 698.86 *** Three-factor model (LMX and 744.59 132 280.16 ***

two-dimensional trust) Competing Model [DELTA]df RMSEA SRMR CFI One-factor model .27 .15 .79 Two-factor model (LMX and trust) 1 .16 .09 .86 Three-factor model (LMX and 2 .10 .05 .93

two-dimensional trust) Note: RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual; CFI = comparative fit index. *** p <.001. Table 2 Means, Standard Deviations, and Intercorrelations of Study Variables Variable M (SD) 1 2 3 4 1. LMX 2.53 -- 2. Calculus-based trust 3.75 .14 * -- 3. Identification-based trust 3.26 .29 ** .51 ** -- 4. Relationship duration 18.61 .06 .02 .01 --

(in months) Note: LMX = leader-member exchange. * p <.05. ** p <.01. Table 3 Hierarchical Regression Analysis of Linear and Nonlinear Calculus-Based Trust Terms Influencing Leader-Member Exchange Quality Variable Model 1 Model 2 Model 3 Step 1: Control

Identification-based trust .29 *** .33 *** .33 *** Step 2: Main effect

Calculus-based trust -.06 -.08 Step 3: Quadratic effect

Calculus-based trust squared -.03 Step 4: Cubic effect

Calculus-based trust cubed [DELTA][R.sup.2] .00 .00 Adjusted [R.sup.2] .09 *** .08 .07 F 17.88 *** 9.10 *** 6.08 *** Variable Model 4 Step 1: Control

Identification-based trust .32 *** Step 2: Main effect

Calculus-based trust .10 Step 3: Quadratic effect

Calculus-based trust squared -.28 * Step 4: Cubic effect

Calculus-based trust cubed -.40 * [DELTA][R.sup.2] .02 * Adjusted [R.sup.2] .09 * F 5.63 *** * p <.05. *** p <.001. Table 4 Hierarchical Regression Analysis of Linear and Nonlinear Identification-Based Trust Terms Influencing Leader-Member Exchange Quality Variable Model 1 Model 2 Model 3 Step 1: Control

Calculus-based trust .17 ** -.06 -.06 Step 2: Main effect

Identification-based trust .33 *** .33 *** Step 3: Quadratic effect

Identification-based trust squared .01 Step 4: Cubic effect

Identification-based trust cubed [DELTA][R.sup.2] .06 *** .00 Adjusted [R.sup.2] .03 ** .08 *** .07 F 5.91 ** 9.10 *** 6.05 *** Variable Model 4 Step 1: Control

Calculus-based trust -.06 Step 2: Main effect

Identification-based trust .19 Step 3: Quadratic effect

Identification-based trust squared .05 Step 4: Cubic effect

Identification-based trust cubed .17 [DELTA][R.sup.2] .01 Adjusted [R.sup.2] .07 F 4.94 *** ** p <.01 *** p<.001.