Leader-member exchange (LMX) 11 Items

1. Abstract

The Leader-Member Exchange (LMX) Multidimensional Scale, originally established by Robert C. Liden and John M. Maslyn (1998) as the LMX-MDM, represents a foundational psychometric instrument designed to capture the nuanced, social-exchange-based architecture of workplace dyadic interactions between managers and their subordinates. Departing from early unidimensional operationalizations of vertical dyad linkage—such as the widely used unidimensional Leader-Member Exchange 7-item scale (LMX-7)—the multidimensional framework conceptualizes high-quality interpersonal exchanges as multifaceted relationships constructed across discrete behavioral and psychological currencies. The scale delineates four distinct yet interrelated dimensions: Affect (mutual interpersonal attraction and positive emotional sentiment), Loyalty (public expression of mutual support, defense, and interpersonal fidelity), Contribution (perception of work-oriented activity and discretionary effort invested toward common objectives), and Professional Respect (admiration for professional competencies, operational knowledge, and vocational reputation). Administered primarily via a 7-point Likert response format across dyadic counterparts, the instrument provides researchers and organizational consultants with granular analytical capability. Psychometric investigations across diverse international samples demonstrate exceptional reliability, with subscale Cronbach’s alpha coefficients systematically meeting or exceeding α = .80, robust composite reliabilities, and strict configural, metric, and scalar measurement invariance. Extensive confirmatory factor analyses affirm that the hypothesized four-factor structure provides superior fit indices relative to competing single-factor or orthogonal representations. Demonstrating robust convergent, discriminant, and predictive validity, the scale effectively accounts for significant incremental variance in organizational citizenship behaviors (OCBs), employee turnover intentions, psychological empowerment, affective organizational commitment, and objective job performance ratings. This article provides an exhaustive academic appraisal of the instrument, dissecting its theoretical underpinnings, psychometric validation trajectories, structural configurations, and operational administration.

2. Keywords

Leader-Member Exchange, LMX-MDM, Dyadic Relationship Quality, Social Exchange Theory, Vertical Dyad Linkage, Affect, Loyalty, Contribution, Professional Respect, Psychometric Evaluation

3. Authors

The Multidimensional Leader-Member Exchange scale was formulated and validated by Robert C. Liden and John M. Maslyn. Their seminal validation study, entitled “Multidimensionality of Leader-Member Exchange: An Empirical Assessment through Scale Development,” was published in the Journal of Management in 1998 (Vol. 24, No. 1, pp. 43–72).

• Robert C. Liden, Ph.D.: Professor of Management at the University of Illinois at Chicago (UIC), College of Business Administration. Dr. Liden is an internationally celebrated authority in organizational behavior, leadership dynamics, servant leadership, and social exchange processes within organizational structures. He has served as Associate Editor for the Academy of Management Journal and has contributed foundational frameworks regarding dyadic supervisor-subordinate development.
• John M. Maslyn, Ph.D.: Professor of Management and Entrepreneurship at the Michael J. Coles College of Business, Kennesaw State University. Dr. Maslyn’s investigative agenda emphasizes dyadic workplace relationships, managerial coaching, organizational politics, interpersonal justice, and the psychological contracts binding organizational agents.

Institutional affiliations during the initial scale development were centered at the University of Illinois at Chicago, where extensive iterative pilot testing, psychometric sampling, and cross-validation procedures were conducted.

4. Purpose

The primary purpose of the multidimensional Leader-Member Exchange scale is to provide a theoretically grounded, empirically validated, and operationally rigorous measurement tool capable of isolating the distinct social exchange currencies that characterize supervisor-subordinate relationships. Historically, research operating under the Vertical Dyad Linkage (VDL) paradigm treated leader-member exchanges as an omnibus, singular continuum running from low-quality transactional supervision (out-group status) to high-quality transformational partnership (in-group status). While unidimensional scales like the LMX-7 (Graen & Uhl-Bien, 1995) achieved widespread utility, their aggregated structure obscured vital empirical distinctions. Specifically, treating LMX as a monolithic construct prevented researchers and practitioners from examining relationships characterized by uneven, asymmetric, or multidimensional developmental trajectories—for example, dyads characterized by elevated professional competence and operational output but completely devoid of interpersonal warmth, or conversely, dyads marked by deep mutual liking and interpersonal friendship unaccompanied by task-directed effort or technical competence.

In academic and applied research, the instrument serves several pivotal objectives:

• Disentangling Antecedents and Outcomes: It allows investigators to evaluate whether specific leader behaviors, subordinate traits, or contextual variables uniquely predict distinct facets of the relationship. For instance, demographic similarity or perceived deep-level value congruence may drive the Affect dimension, while cognitive ability and specialized technical expertise may specifically foster Professional Respect.
• Targeted Organizational Interventions: In diagnostic and executive development settings, the tool enables leadership consultants to identify precise relational deficits. Rather than simply diagnosing a dyad as exhibiting “low LMX,” the profile clarifies whether the dysfunction stems from deficient collaborative output (Contribution), perceived betrayal or lack of advocacy (Loyalty), absence of technical credibility (Professional Respect), or cold, hostile interpersonal dynamics (Affect).
• Testing Advanced Social Exchange Hypotheses: The scale facilitates nuanced structural equation modeling (SEM) to assess differential mediation and moderation pathways. Researchers can examine how specific exchange dimensions selectively mediate the relationship between leadership styles (e.g., transformational, servant, or abusive supervision) and employee workplace outcomes such as creative deviance, silent withholding of ideas, voice behavior, and operational burnout.

Ultimately, the instrument operationalizes the profound reality that organizational exchanges are mediated through distinct psychological and behavioral resources, establishing a comprehensive platform for micro-organizational inquiry.

5. Psychological Construct

The psychological construct assessed by this instrument is the quality of the Leader-Member Exchange (LMX), defined as the reciprocal social and psychological exchange relationship that develops within a hierarchical dyad through repeated, iterative cycles of interaction, role-making, and resource allocation. Liden and Maslyn (1998) expanded upon Dienesch and Liden’s (1986) initial theoretical formulation, which posited that social exchange relationships do not develop along a singular dimension, but rather through the exchange of distinct, qualitatively differentiated commodities. The scale identifies four primary subdimensions:

1. Affect

The Affect dimension represents the mutual interpersonal attraction and positive emotional sentiment members of the dyad experience toward one another, rooted primarily in interpersonal chemistry, personal liking, and generalized sociability rather than formal organizational duties. Individuals characterized by high affective exchange genuinely enjoy working together, appreciate each other’s personality characteristics, and view their dyadic partner as someone with whom they would gladly maintain a personal friendship. When affect is high, interactions are characterized by informal communication, warmth, empathy, and shared humor. Importantly, affect can develop independently of operational performance; a supervisor may experience deep personal affection for a subordinate who possesses only mediocre technical skills.

2. Loyalty

The Loyalty dimension assesses the degree to which dyadic partners publicly support, defend, and preserve each other’s character, decisions, and professional standing. Loyalty reflects organizational allegiance and interpersonal defense, signifying an unwritten covenant where the leader or member will defend the counterpart against external attacks, organizational politics, or superior reprimands, even in scenarios involving ambiguous information or honest mistakes. This dimension creates a psychological safety net. A high-loyalty relationship assures the individual that their dyadic counterpart will not exploit vulnerabilities, misrepresent their intentions to upper management, or abandon them during organizational crises.

3. Contribution

The Contribution dimension encompasses perception of the amount, direction, and quality of work-oriented activity that each dyadic partner puts forth toward the explicit or implicit goals of the dyad. Unlike the socio-emotional nature of Affect and Loyalty, Contribution is anchored in task-oriented behaviors, discretionary effort, and operational dependability. It reflects a willingness to perform duties that transcend formalized job descriptions, undertake difficult assignments, work overtime, and expend supplementary physical or cognitive effort to advance the partner’s agenda. In a reciprocal dynamic, a supervisor contributes developmental resources, mentorship, and political capital, while the subordinate reciprocates through elevated productivity, proactive problem-solving, and task dedication.

4. Professional Respect

The Professional Respect dimension captures the perception of the degree to which each dyadic partner has built a reputation, within or outside the organization, of excelling at their line of work. It represents a cognitive appraisal of the other’s technical competence, historical track record, professional proficiency, and specialized domain knowledge. Unlike Affect, which is emotional, Professional Respect is purely evaluative and competence-based. A subordinate can harbor immense professional respect for an abrasive, emotionally distant supervisor due to their peerless strategic intellect and industry acumen; conversely, a leader may hold deep professional respect for a subordinate’s specialized technological prowess while maintaining an entirely impersonal, transactional emotional relationship.

6. Theoretical Framework

The theoretical architecture of the Multidimensional Leader-Member Exchange scale is grounded at the convergence of three foundational paradigms within sociology and organizational psychology: Social Exchange Theory, Role Theory, and Vertical Dyad Linkage Theory.

Social Exchange Theory and the Norm of Reciprocity

The primary bedrock of LMX theory resides in Social Exchange Theory as articulated by George Homans (1958) and Peter Blau (1964). Blau demarcated a fundamental distinction between economic exchange and social exchange. Economic exchanges are predicated on formal, explicit contractual obligations, immediate quid-pro-quo compensation, tangible resources (e.g., monetary wages for predefined labor), and clear legal recourse in the event of default. In contrast, social exchanges involve the trade of unspecified, non-monetized socio-emotional resources, long-term and diffuse future obligations, and absent legal mechanisms for enforcement.

Underpinning social exchange is Alvin Gouldner’s (1960) universal norm of reciprocity, which dictates that individuals are psychologically and socially obligated to return benefits received and refrain from injuring those who have conferred benefits upon them. In hierarchical organizational contexts, leaders possess differential control over resources (e.g., task autonomy, inside information, mentoring, emotional support, and budgetary latitude). When a leader allocates these discretionary resources to a subordinate beyond the baseline employment contract, the subordinate experiences cognitive indebtedness, motivating the return of valued behaviors—such as heightened loyalty, extended effort, and organizational citizenship. Liden and Maslyn incorporated Foa and Foa’s (1974, 1980) Resource Theory of Social Exchange, which classifies exchange resources along dimensions of particularism (the extent to which the value of a resource is tied to the specific person providing it) and concreteness. By doing so, they established that Affect, Loyalty, Contribution, and Professional Respect function as qualitatively distinct exchange currencies governed by varying degrees of particularism and formality.

Vertical Dyad Linkage (VDL) and Role Theory

The structural foundation of LMX emerged from early Vertical Dyad Linkage research pioneered by Fred Dansereau, George Graen, and William Haga (1975). Prior to VDL, leadership researchers operated under the unexamined assumption of the Average Leadership Style (ALS)—the premise that leaders display a uniform, homogeneous behavioral pattern toward all subordinates within a collective work group. VDL challenged this premise empirically, demonstrating that leaders systematically develop highly differentiated, idiosyncratic relationships with individual subordinates.

This dynamic was subsequently synthesized with Role Theory (Kahn et al., 1964; Graen, 1976). The developmental lifecycle of an exchange relationship progresses through three progressive phases:

1. Role Taking: The initial sampling phase, wherein the leader initiates formal task requests and evaluates the subordinate’s basic technical competencies and behavioral reliability within contractual boundaries.
2. Role Making: The negotiation phase, wherein the dyad explores unstructured interactions. The leader offers informal opportunities and psychological support; the subordinate decides whether to accept and reciprocate with discretionary effort and interpersonal loyalty. This is the stage where multidimensional currencies are first exchanged and negotiated.
3. Role Routinization: The stabilization phase, wherein predictable patterns of mutual expectations, social support, and professional respect become deeply embedded. The dyad achieves a stable equilibrium characterized either as an in-group (high LMX across multidimensional currencies) or out-group (contractually restricted, low LMX) partnership.

7. Validity

Extensive psychometric investigations have established the robust construct, convergent, discriminant, and criterion-related validity of the multidimensional LMX scale across diverse organizational, occupational, and cross-cultural contexts.

Construct and Structural Validity

During the original scale development by Liden and Maslyn (1998), initial item pools were derived through literature review and inductive critical incident techniques. Subsequent exploratory factor analyses (EFA) and rigorous confirmatory factor analyses (CFA) across multiple independent organizational samples (consisting of both working undergraduate students and full-time organizational employees across financial, educational, and manufacturing sectors) systematically demonstrated that the theoretical four-factor model provided an exceptionally superior fit to empirical data compared to alternative nested models. While a single-factor unidimensional representation exhibited poor fit indices (e.g., Tucker-Lewis Index [TLI] < .70, Root Mean Square Error of Approximation [RMSEA] > .15), the four-factor model yielded excellent structural parameters:

• Comparative Fit Index (CFI) values consistently ranged between .95 and .99.
• Non-Normed Fit Index (NNFI / TLI) consistently exceeded .94.
• Root Mean Square Error of Approximation (RMSEA) values remained strictly below .06, with 90% confidence intervals confirming tight model fit.
• Standardized Root Mean Square Residual (SRMR) values were below .05.

Convergent and Discriminant Validity

Convergent validity was confirmed via strong, statistically significant correlations between the four subscales and established unidimensional measures of LMX, most notably the Graen and Uhl-Bien (1995) LMX-7 scale. While the subscales correlated positively with LMX-7 (correlations generally ranging from r = .52 to r = .78, p < .001), they remained distinct, demonstrating that while they tap into the broader social exchange domain, they do not collapse into redundant indicators.

Discriminant validity was established through nested model CFA comparisons and average variance extracted (AVE) versus shared variance assessments (Fornell-Larcker criterion). The four dimensions demonstrated distinct empirical separability from related leadership and organizational constructs, including:

• Perceived Organizational Support (POS): Showing that employees distinctly differentiate their exchange with their immediate supervisor from their broader exchange with the overarching organization.
• Transformational Leadership: Confirming that dyadic relationship quality is conceptually and empirically distinct from the leader’s generalized, charismatic behavioral style.
• Supervisor-Directed Organizational Citizenship Behavior (OCB-S): Establishing that behavioral reciprocation by subordinates is distinct from the perceptual quality of the dyadic relationship itself.

Criterion-Related and Predictive Validity

The multidimensional scale exhibits formidable predictive validity, accounting for significant incremental variance in organizational outcomes over and above traditional unidimensional scales. Longitudinal and multisource studies demonstrate that:

• The Contribution dimension uniquely predicts objective job performance metrics, supervisor ratings of competence, and task performance (β values ranging from .22 to .38, p < .01).
• The Loyalty dimension serves as a potent negative predictor of employee turnover intentions, psychological withdrawal, and unauthorized absenteeism, while positively predicting affective organizational commitment and whistle-blowing support.
• The Affect dimension is strongly predictive of organizational citizenship behaviors directed at individuals (OCB-I), informal mentoring, and perceived interpersonal climate quality.
• The Professional Respect dimension accounts for significant variance in subordinate psychological empowerment, role clarity, and willingness to accept critical developmental feedback.

8. Reliability

The internal consistency and temporal stability of the multidimensional LMX scale have been verified across hundreds of independent empirical investigations spanning various cultural and linguistic demographics.

Internal Consistency (Cronbach’s Alpha)

In the foundational validation by Liden and Maslyn (1998), internal consistency coefficients for the subscales demonstrated strong reliability across both development and validation samples:

• Affect: Cronbach’s α consistently ranged from .86 to .90 across samples.
• Loyalty: Cronbach’s α consistently ranged from .74 to .85 across samples.
• Contribution: Cronbach’s α consistently ranged from .74 to .82 across samples.
• Professional Respect: Cronbach’s α consistently ranged from .89 to .92 across samples.

Subsequent meta-analytic reviews and large-scale organizational field studies (e.g., Greguras & Ford, 2006; Dulebohn et al., 2012) confirm that the composite reliability (CR) and Cronbach’s alpha values for all four dimensions uniformly exceed the conventional psychometric threshold of .70, typically hovering in the .80 to .92 range. Inter-item correlations within each respective subscale remain within the recommended .40 to .75 interval, indicating high item homogeneity without problematic redundancy.

Temporal Stability (Test-Retest Reliability)

Investigations examining the longitudinal invariance of the instrument indicate stable test-retest reliability over temporal intervals ranging from two weeks to six months. In organizational settings where dyadic composition remained unchanged, test-retest correlations for stable relationships typically fall between r = .68 and r = .82 (p < .001). Lower longitudinal stability is appropriately observed during structural reorganization or active managerial transition, confirming that the scale accurately captures real-time dynamic shifts in dyadic relationship evolution.

9. Factor Analysis

The factor structure of the scale was rigorously evaluated during its inception via a multi-stage analytic protocol combining Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

In Phase 1 of scale creation, an initial pool of potential items was administered to an exploratory sample. Principal Axis Factoring (PAF) and Principal Component Analysis (PCA) utilizing both orthogonal (Varimax) and oblique (Promax) rotations were conducted. Oblique rotation was prioritized because the underlying dimensions of a social exchange relationship were theoretically anticipated to correlate moderately with one another.

The scree plot examination, Kaiser’s criterion (eigenvalues > 1.0), and parallel analysis clearly substantiated a robust four-factor solution. The four extracted factors accounted for over 65% of the total cumulative variance. Items loaded cleanly onto their respective target constructs—Affect, Loyalty, Contribution, and Professional Respect—with standardized factor loadings uniformly exceeding .60, and cross-loadings on non-target factors remaining well below the .30 threshold. Items demonstrating weak factor loadings or substantial cross-loadings were excised, resulting in a refined set of items representing the underlying dimensions.

Confirmatory Factor Analysis (CFA)

In Phase 2, CFA was implemented using structural equation modeling software (e.g., LISREL, AMOS, Mplus) on independent validation samples to contrast the hypothesized four-factor oblique structure against plausible alternative configurations:

1. One-Factor Model: All items loaded onto a single, undifferentiated global LMX factor (χ²/df > 6.5, CFI < .72, RMSEA > .14). This model was decisively rejected.
2. Two-Factor Model: Combining socio-emotional dimensions (Affect and Loyalty) into one factor and task-related dimensions (Contribution and Professional Respect) into a second factor (χ²/df > 4.2, CFI < .83, RMSEA > .11). This configuration failed to achieve acceptable fit.
3. Four-Factor Orthogonal Model: The four dimensions modeled as completely uncorrelated factors (χ²/df > 5.1, CFI < .78, RMSEA > .13). This was rejected, as the subscales naturally share common variance within an overall dyadic relationship.
4. Hypothesized Four-Factor Oblique Model: Four distinct, correlated factors (χ²/df < 2.1, CFI = .97, TLI = .96, RMSEA = .048, SRMR = .039). Chi-square difference tests (Δχ²) confirmed that the four-factor oblique model fit the empirical data significantly better than all alternative nested specifications (p < .001).

Standardized factor loadings (λ) for the items onto their designated latent factors ranged from .68 to .92, confirming strong indicator reliability.

10. Instrument / Measurement Tool

The multidimensional Leader-Member Exchange instrument is structured as an objective, self-report psychometric inventory designed for administration to organizational employees (subordinate-perspective reporting on the leader) or in parallel dyadic formats (leader-perspective reporting on the subordinate).

• Test Type: Psychometric rating scale; social exchange and leadership dyad diagnostic tool.
• Administration Format: Paper-and-pencil, computer-assisted web interview (CAWI), or embedded within broad organizational climate surveys.
• Administration Time: Approximately 3 to 5 minutes.
• Target Population: Working adults within organizational hierarchies who have an identifiable, direct supervisory or subordinate reporting relationship.
• Item Count: 12 standardized items (often adapted to 11 or 12 items depending on dyadic operationalization; the complete standard multidimensional battery utilizes 12 items across 4 dimensions).
• Response Scale: 7-point Likert response format scored as follows:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Slightly Disagree
  • 4 = Neutral (Neither Agree nor Disagree)
  • 5 = Slightly Agree
  • 6 = Agree
  • 7 = Strongly Agree
• Dimensional Allocation of Items:
  • Professional Respect: Items 1, 3, 6
  • Contribution: Items 2, 4, 7
  • Affect: Items 5, 8, 9
  • Loyalty: Items 10, 11, 12
• Scoring and Computational Rules:
  • All items are framed in a positive, direct direction; there are no reverse-coded items in the standard version.
  • Subscale Scores: Calculated by computing the mathematical mean (or sum) of the items corresponding to each respective dimension. Mean scoring (ranging from 1.00 to 7.00) is preferred to preserve interpretability relative to the original Likert anchors.
  • Global LMX Score: If an overall relationship quality metric is desired, researchers may compute the grand mean across all items, provided higher-order factor modeling justifies combining the subscales.

11. Permissions & Fee and Test Year

The scale was formally published in 1998 by Robert C. Liden and John M. Maslyn in the Journal of Management. The copyright for the original article and its contents is held by the Southern Management Association and published by SAGE Publications.

• Academic and Non-Commercial Research Usage: Consistent with standard academic conventions, the instrument items may be utilized, reproduced, and administered by university-affiliated researchers, graduate students, and non-commercial investigators for empirical research without payment of licensing fees, provided that full and accurate academic citation is accorded to Liden and Maslyn (1998).
• Commercial and Consultative Applications: Organizational consultants, corporate entities, commercial test publishers, and for-profit diagnostic platforms intending to package the tool into commercial diagnostic batteries, commercial software platforms, or fee-for-service consulting programs should secure formal copyright clearance or licensing verification through SAGE Publications / RightsLink.

12. References

• Blau, P. M. (1964). Exchange and power in social life. John Wiley & Sons.
• 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. Organizational Behavior and Human Performance, 13(1), 46–78. https://doi.org/10.1016/0030-5073(75)90005-7
• Dienesch, R. M., & Liden, R. C. (1986). Leader-member exchange model of leadership: A critique and further development. Academy of Management Review, 11(3), 618–634. https://doi.org/10.5465/amr.1986.4306242
• Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. (2012). A meta-analysis of antecedents and consequences of leader-member exchange: Integrating the past with an eye toward the future. Journal of Management, 38(6), 1715–1759. https://doi.org/10.1177/0149206311415280
• Foa, U. G., & Foa, E. B. (1974). Societal structures of the mind. Charles C. Thomas.
• Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25(2), 161–178. https://doi.org/10.2307/2092623
• 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. The Leadership Quarterly, 6(2), 219–247. https://doi.org/10.1016/1048-9843(95)90036-5
• Greguras, G. J., & Ford, J. M. (2006). An examination of the multidimensionality of supervisor and subordinate perceptions of leader-member exchange. Journal of Occupational and Organizational Psychology, 79(3), 433–465. https://doi.org/10.1348/096317905X53859
• Homans, G. C. (1958). Social behavior as exchange. American Journal of Sociology, 63(6), 597–606. https://doi.org/10.1086/222355
• Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D., & Rosenthal, R. A. (1964). Organizational stress: Studies in role conflict and ambiguity. John Wiley & Sons.
• Liden, R. C., & Maslyn, J. M. (1998). Multidimensionality of leader-member exchange: An empirical assessment through scale development. Journal of Management, 24(1), 43–72. https://doi.org/10.1177/014920639802400105

13. Items of the Scale