Introduction to the special issue on theory testing and replications in leadership science: Part 2

In a testament to the need for replication and theory testing research in leadership science, and the scientific community's concordance of that need, the special issue devoted to leadership science received many submissions, and ultimately, accepted eight papers to be published across two issues of Journal of Management Scientific Reports (JOMSR): Volume 4, Issues 2 and 3. In our introduction to Part 1 in Volume 4, Issue 2 (Brouer et al., 2026), we provided an overview of the editorial team's motivation and goals for the special issue, as well as the four articles included in Part 1 (Volume 4; Issue 2). This introduction to Part 2 (Volume 4; Issue 3) provides an overview of the remaining four articles selected for the leadership science special issue.

Overview of Part 2

“Diverse Differentiation: Empirical Tests of an LMXD Configurations Framework” by Aiken and Locklear (2026), empirically tests portions of Buengeler et al.'s (2021) conceptual framework on leader–member exchange differentiation (LMXD), which refers to the extent to which leaders establish different quality relationships with team members. Arguing that researchers have treated the estimation of LMXD too simplistically, causing misalignment between theory and measurement and inconsistent findings, Buengeler et al. (2021) explicated three types of LMXD (i.e., separation, variety, and disparity). They maintain that each of these have unique measurement and operationalizations that are more or less appropriate for testing different theories, and thus, may be more or less appropriate for testing team outcomes.

To empirically test portions of this conceptual framework, Aiken and Locklear conducted two studies. In Study 1, they performed an exact reproducibility study, as well as, a constructive reproduction of one study by Cobb and Lau (2015). Specifically, using Buengeler's framework, they reanalyzed Cobb and Lau's data using disparity rather than separation estimations of LMXD to better align the operationalization of the LMXD with the justice-based theory utilized in the original research. In Study 2, Aiken and Locklear further test the Buengeler et al. framework using secondary data originally published by Kim et al. (2022) thereby also extending that study's original findings. In particular, Aiken and Locklear examined whether LMXD separation predicts team potency and performance better than using team LMX (operationalized as team median LMX). In both of Aiken and Locklear's studies, LMXD explained additional variance beyond central tendency scores, and the operationalization of LMXD significantly impacted team functioning. Surprisingly, empirical distinctions between differing operationalizations of LMXD were marginal, which raises concerns about redundancy and the discriminate validity between separation and disparity. It is laudable that Aiken and Locklear used existing data sets and conducted both reproducibility studies and the constructive extensions, testing original and new hypotheses. Moreover, advancing theory in LMXD, their findings provide a more nuanced view of the value in differentiating LMXD types and aligning theory with measurement, while also highlighting challenges in operationalizing these concepts that warrant future research.

The next paper by Broucher and Saef (2026) is titled “Boss Up, But How? Legitimating Versus Collaborative Claiming as Pathways to Gendered Differences in Leader Identity Strength.” This study provides a direct empirical test of DeRue and Ashford's (2010) leader identity development framework, which proposes that leadership is socially constructed through reciprocal claiming and granting processes, yet has received limited empirical scrutiny to date. Broucher and Saef test this theory by integrating social role and role congruity theories to examine how the leader identity development process unfolds differently for men and women. They distinguish between two forms of claiming: legitimating claiming, which involves asserting authority through formal roles and organizational policies, and collaborative claiming, which involves offering expertise and resources to encourage shared participation.

Using social role theory, Broucher and Saef (2026) argue that women using legitimate claims, which can be seen as more agentic, face greater social costs. This steers women leaders toward less effective collaborative claims. In a daily diary study of 77 leaders (575 daily surveys), Broucher and Saef found that women were significantly less likely than men to engage in legitimating claiming, while no gender difference emerged in collaborative claiming. Critically, legitimating claims were more effective at eliciting follower granting and strengthening leader identity, creating a cumulative identity advantage for men who can freely use the more effective strategy without backlash. This study exemplifies the kind of rigorous theory testing this special issue seeks to promote, subjecting a foundational but largely untested theoretical framework to direct empirical evaluation and revealing important boundary conditions related to gender in the leader identity construction process.

The third article by Güntner and Nieberle (2026), “Determining the Who and When of Leading and Following: A Micro-Temporal Analysis of Reciprocated Claiming and Granting Patterns in Dyads,” also tests DeRue and Ashford's (2010) leader identity development framework. Rather than focusing on gender, Güntner and Nieberle note that this framework emphasizes leader and follower claiming and granting behaviors that can occur outside of formal roles. However, Güntner and Nieberle argue that people determine who leads and who follows through a process of reciprocal leader-follower behaviors, even within the context of formal roles. Using formal leader-follower dyads, they explore how formal leaders’ and followers’ claiming and granting behaviors are influenced by factors at the individual level such as formal roles and leadership structure schemas (LSS) and at the relational level (including convergence of LSS and interaction history). Güntner and Nieberle also evaluate whether reciprocated leadership claims increase leader identity salience, regardless of formal role.

Using a multi-method approach that combines survey-based measures and behavioral interaction coding and data from 86 individuals in 43 dyads (4,262 behavioral units), Güntner and Nieberle (2026) demonstrated that individuals in formal leader roles claimed leadership significantly more often than formal followers. However, LSS did not moderate these behaviors. Reciprocated claims and grants were not significantly associated with leader identity salience. Nonetheless, early interaction patterns were connected to similar behaviors later, highlighting the importance of relational history in shaping leadership dynamics. The use of multi-methods is to be commended, and these findings enhance understanding of leader identity formation in formal dyads, clarify the boundary conditions under which adaptive leadership theory applies, and provide practical guidance for fostering co-constructed leadership processes.

Finally, the article by Schleu et al. (2026) titled “From Star Player to Star Coach? The Longitudinal Validity of Performance-Based Promotion in the NBA” begins to address a fundamental question in leadership research about whether top-performing employees can become effective leaders who drive team success. Despite the prevalence of performance-based promotion in organizations, prior evidence on the link between employee performance to subsequent leader performance has been inconclusive. For instance, Schleu et al. (2024) found no systematic link between player and coach performance in professional soccer. Schleu and colleagues take this line of inquiry one step further by conducting a constructive replication in the National Basketball Association (NBA), a context that offers several methodological advantages over the use of professional soccer data by Schleu et al. (2024), including more reliable performance indicators due to the high volume of scoring, greater coach influence during games through timeouts and substitutions, and more balanced competition through salary cap regulations. Importantly, Schleu and colleagues also advance theory by pitting three competing perspectives against one another: the performance requirements perspective, the follower-centric perspective, and the Theory of Expert Leadership, the last of which predicts that expert knowledge acquired as high-performing players should increasingly benefit them as coaches, over time.

Using Bayesian structural equation modeling on the careers of all 329 NBA head coaches between 1947 and 2020, the results revealed only a weak relationship between player and subsequent coaching performance that did not strengthen with coaching experience. This indicates that former star players did not increasingly leverage their playing expertise to become better coaches over time. These findings support the performance requirements perspective and challenge the assumptions underlying performance-based promotion, suggesting that organizations should place greater emphasis on assessing leadership-relevant competencies rather than rewarding past performance with advancement into leadership roles. The authors should be commended for this constructive replication, as well as for testing competing perspectives. This work provides mounting evidence that an important real-world practice, performance-based promotion, might not produce the most effective leaders. In this way, constructive replication with competing tests can begin to more scientifically inform practice.

In sum, all four of the papers included in Part 2 of this special issue exemplify the importance of first time or repeated theory testing and refinements. Three of the four articles empirically examine previously untested frameworks or theoretical models of leadership, while the fourth assesses three competing perspectives. Despite the popularity of the leader identity development framework (DeRue & Ashford, 2010; cited 2,923 times on Google scholar as of 04/20/2026), empirical testing and support for its main propositions remain scarce. Using multi-methods, Güntner and Nieberle test the process of the leader identity development framework in the context of formal roles, as well as the impact of individual and relational level variables on the process and outcomes of claiming and granting. Exploring a different boundary condition, Broucher and Saef, examine the role of gender and varying types of leadership claiming (i.e., legitimate versus collaborative), finding that gender impacts the type of claiming used, as well as subsequent leadership identity. These two studies offer empirical evidence for nascent research on the leader identity development framework, while also using interesting and differing methodologies. Aiken and Locklear also test a conceptual framework, providing some of the first empirical evidence on how differing operationalizations of LMXD impact important team outcomes.

Lastly, engaging an underutilized method of theory testing, Schleu and colleagues assess competing hypotheses from three perspectives, finding that top performing NBA players cannot channel their success as players to leadership effectiveness at the coaching level. Thus, all four papers directly combat the issue of lack of theory testing in leadership science. This is a vital part of advancing science, which can ultimately be used to inform practice and policy (National Academies of Sciences, Engineering, and Medicine, 2019).