The Challenge of Claiming Leadership for Younger Female Managers: Exploring Differences Between Employees’ Behaviors and Perceptions

Abstract

Leadership is critical to any enterprise, raising the question of when managerial leadership is accepted—specifically, whether managers communicate in ways that earn employee endorsement. The claiming and granting framework suggests managers can claim leadership, but employees may or may not grant it. Yet most research relies on retrospective evaluations and overlooks actual verbal behaviors. Moreover, responses to leadership claims likely depend on a manager's demographics, particularly age and gender. While studies often examine these cues in isolation, their interplay matters. To explore this, we fine-coded 37,277 verbal behaviors from 68 manager-employee dyads during workplace meetings. Male and female managers claimed leadership equally often. However, for female managers younger than their employees, claiming leadership was linked to lower post-meeting endorsement—but not to in-meeting granting. In contrast, older female managers received the highest post-meeting endorsement across all age-gender constellations when claiming leadership. For male managers claiming leadership, age was unrelated to endorsement. These findings highlight how subtle gender and age biases shape leadership acceptance: younger female managers, in particular, may face undermined authority without overt resistance. Raising awareness of these dynamics is key to fostering equitable leadership recognition.

When employees endorse their managers as leaders, it not only enhances managerial effectiveness but also fuels their motivation to seek more senior leadership roles (e.g., Mesdaghinia, 2014). In contrast, a lack of leader endorsement can have negative results for managers, such as elevated emotional exhaustion and increased turnover intentions (Dwivedi et al., 2021; Knudsen et al., 2009). These consequences pose a considerable economic risk to organizations given the high costs of recruitment and training (Bannister & Griffeth, 1986) and the inevitable loss of knowledge when managers (have to) leave (Eckardt et al., 2014). It is therefore critical for organizations to understand how daily interactions influence whether their managers receive endorsement from employees or not.

To this end, the claiming and granting framework of leadership (DeRue & Ashford, 2010) provides a useful theoretical lens. It suggests that managers may claim leadership (i.e., signal that they are willing to offer direction and to exhibit social influence) in their interactions with employees, and employees may or may not grant them leadership status. Leadership, in this view, is understood here not as a formal role but as a social process constructed between the manager and the employee (Fairhurst & Uhl-Bien, 2012). This relational understanding emphasizes the importance of studying concrete interactions—particularly in one-on-one meetings between managers and their direct reports (Flinchum et al., 2022; Mroz et al., 2018; Ruben & Gigliotti, 2016)—as critical sites where leadership is enacted. In such interactions, leadership endorsement—or the lack thereof—becomes visible through employees’ behavioral granting responses to managerial leadership claiming.

Thus far, however, extant research has predominantly relied on asking employees about aggregated hindsight perceptions of leadership behaviors and not investigated what is really happening in such meetings (Fischer et al., 2020). While this research has offered first valuable insights into employees’ post-meeting interpretations, it overlooks that, as managerial roles become more demographically diverse, these perceptions may be shaped by evaluative biases (Chin, 2010; Scott & Brown, 2006). Hence, we propose expanding the focus to a micro-level examination of the claiming and granting framework by asking whether all managers receive equal endorsement by employees for their leader claiming behaviors in such meetings. Shifting the analytical focus to distinguish employees’ immediate behavioral reactions from their later, cognitively filtered assessments of endorsement provide a more nuanced understanding of micro-inequities (cf., Rowe, 1990)—that is, subtle and often unconscious disparities in how leadership behaviors are granted and perceived in everyday interactions.

While the managerial role signals achieved status, micro-inequities often arise from demographic characteristics that serve as ascribed status cues, shaping how leadership behaviors are evaluated by employees. These ascribed status cues not only influence formal career outcomes but also subtly structure everyday interpersonal exchanges, shaping how requests, claims or behaviors generally are interpreted and responded to, particularly when multiple low-status demographic cues co-occur (Csillag et al., 2025).

In the context of leadership, gender has long been a focal ascribed status cue, with a persistent puzzle (Banks, 2023): although women tend to be perceived as showing more effective leadership behaviors (Buss et al., 2025; Paustian-Underdahl et al., 2024; Shen et al., 2025), their advancement into managerial roles remains limited (World Economic Forum, 2022). Some scholars have suggested that part of the explanation may lie in the tendency to focus solely on gender, which may oversimplify the so-called “acceptance challenge” (Stamper & McGowan, 2022). This perspective aligns with research on intersectional stereotypes (Petsko et al., 2022), which suggests that individuals are perceived and evaluated based on the combined influence of multiple demographic attributes rather than any single characteristic in isolation (Csillag et al., 2025).

In the context of leadership, age has been discussed as a particularly salient ascribed status cue that can shape employees’ willingness to endorse their manager (Stamper & McGowan, 2022). Age is both highly visible and strongly linked to leadership as traditionally, individuals who are relatively older than their employees have held higher positions in the organizational hierarchy (Daldrop et al., 2025; De Meulenaere et al., 2024; Fousiani et al., 2025). Taken together, these insights point to the need for a more nuanced perspective. We are therefore interested in how the interaction of a manager's relative age with their gender, influences employees’ behavioral and cognitive reactions to their manager's leader claiming behavior.

We argue that when a manager deviates more substantially from the stereotypical leader image by differing in not just one but two demographic characteristics related to ascribed status in the context of leadership, these characteristics will jointly extenuate the endorsement of their leadership. Specifically, incorporating our research in the claiming and granting framework of leadership (DeRue & Ashford, 2010), we propose that the interaction between being female and being younger amplifies the perceived lack of fit with leadership, therefore resulting in a more negative effect on employee's granting of leadership than each characteristic individually. Accordingly, a manager who claims leadership and is female and younger than their employee should experience the greatest difficulty in gaining leadership endorsement. We test our hypothesis by fine-coding relational control as expressed through the verbal leadership claiming and granting behaviors of managers and their employees within 68 work meetings (a total of 37,277 verbal behaviors) and by complementing these behavioral data with a post-meeting employee survey. In doing so, we make three contributions to the literature.

First, we seek to resolve ambiguous findings about reactions to women's behaviors (Paustian-Underdahl et al., 2024) by extrapolating on the idea of a “double jeopardy” effect, which posits that female managers who are younger than their employees, compared to the other age-gender constellations (i.e., female managers older than their employees and male managers irrespective of their relative age), will receive lower endorsement from their employees when claiming leadership in interactions. On the one hand, an intersectional perspective underscores how the interaction between gender and age shapes leadership perceptions in ways that cannot be fully understood by considering each characteristic separately. This lens adds important nuance to frameworks such as role congruity theory (Eagly & Karau, 2002), which has traditionally examined how women are penalized for violating gendered expectations, yet often treats women as a uniform group, overlooking within-group differences such as age. On the other hand, it holds practical relevance as organizations increasingly aim to promote upcoming female talent within the context of an aging workforce (Bax et al., 2023), which entails that understanding the unique leadership challenges faced by relatively younger women becomes imperative (McEldowney et al., 2009). Such goals are also in line with advancing gender equality and amplifying the voices of younger individuals as central pillars of the United Nations’ Sustainable Development Goals (United Nations, 2023).

Second, we contribute to an emergent research stream that aims to bring widely cited but rarely operationalized theories to life (Hambrick, 2007; Köhler & Cortina, 2021) by drawing from the claiming and granting framework of leadership (DeRue & Ashford, 2010). To translate the theoretically proposed claiming and granting of leadership into observable communicative acts (DeRue & Ashford, 2010), we revisit a behavioral operationalization that has fallen into oblivion, namely relational control behaviors that assert leadership status (Courtright et al., 1989; Fairhurst et al., 1995). Noting that the advancement of theory and methods goes hand in hand (Fischer & Sitkin, 2023; Lê & Schmid, 2022), we are convinced that understanding the daily building blocks of leadership in concrete interactions can ultimately help advance explanations of higher-level phenomena such as the systematic underrepresentation of women in senior leadership positions (Banks, 2023; World Economic Forum, 2022).

Third, we add nuance to understanding leader endorsement by differentiating between granting leadership status in terms of (1) the employee's immediate verbal granting behaviors towards their manager during interactions and (2) the employee's cognitive granting, that is, the employee's self-reported endorsement of their manager in the leader role following the meeting. Differentiating between these two forms of granting (i.e., actual behavior versus self-reported perceptions) speaks to recent calls for advancing leadership research by clearly separating behavioral and perceptual constructs (Banks et al., 2023; Fischer, 2023; Paustian-Underdahl et al., 2024).

Theoretical Background and Hypothesis

Building on a social influence process perspective, DeRue and Ashford (2010) propose that leadership is not tied to a title or role; instead, it is co-constructed between two parties, often between managers and employees who can claim and grant leadership. Not everything a manager does is thus a claim of leadership. Similarly, while an employee is often expected to show compliance and follow, they can also resist or withhold deference, either openly or covertly (Falbe & Yukl, 1992; Van der Velde & Gerpott, 2023). The view of followership as defined by observable acts of leadership endorsement stands in contrast to a role-based approach, which equates followers with individuals in a position of relatively low formal power (e.g., Carsten et al., 2010; Kellerman, 2019). A process-focused perspective, by contrast, emphasizes the dynamic interaction process between managers and employees, highlighting how the exhibition of influence is constructed through ongoing claiming and granting within dyads (or larger groups).

Claiming leadership describes actions (e.g., verbal or non-verbal behavior) that people take to signal that they are willing to offer direction and to exhibit social influence (DeRue & Ashford, 2010). In line with the understanding of leadership as being constructed through communication (Fairhurst & Uhl-Bien, 2012), we focus specifically on verbal leader claiming behaviors. Although studies that empirically operationalize claiming leadership in a hierarchical set-up remain scarce, Marchiondo et al. (2015) offer a notable exception. The authors manipulated claiming leadership with two sentences (“I was planning to take the lead in this meeting, okay? I’ll present what I’m thinking and then we can discuss other alternatives”), which participants who should imagine being in an employee role saw or read as the expression of a male manager in an experimental vignette study. While this manipulation reflects the conceptual notion that claiming leadership means that a manager seeks to dominate and lead the conversation, natural interactions between managers and their employees at work are more complex. That is, both parties may use a wider variety and more subtle forms to claim and grant leadership. To address questions about employees’ reactions to leader claiming behaviors in daily interactions, a more comprehensive and fine-grained approach is thus needed.

We draw from the relational control coding literature (Courtright et al., 1989; Fairhurst et al., 1995) that aligns with the ideas of DeRue and Ashford (2010) in terms of the notion that managers and employees reciprocally define their relational influence in their interactions. Precisely, in the relational control coding literature, “up behaviors” signal that one person is asserting definitional rights about how influence is exhibited, which corresponds with the definition of claiming leadership. Notably, “up behaviors” can be identified in the context of the subsequent communicative act (which together form an “interact” - a sequence of two contiguous behaviors; Courtright et al., 1989; Fairhurst et al., 1995). To illustrate, claiming leadership manifests when an interaction partner asks a question that seeks to change the direction of the conversation or suggests how the conversation should proceed. However, this function of a question can be determined by considering what has been said after such behavior. For instance, if the leader claiming behavior is followed by an acceptance of such influence attempts, or rather a challenge (DeRue & Ashford, 2010). These behaviors are referred to as granting leadership or as “down” behaviors. Granting leadership can happen directly (e.g., through verbal behaviors such as supporting statements or active listening in the form of follow-up questions) or more indirectly (e.g., people internally accepting and supporting the other in their leader role). In formal organizations, such behaviors may also in part be subsumed under the label compliance, that is, a signal that an employee accepts the formal hierarchical structure and, thus, guidance by a manager. This idea of studying leadership as a process of verbally claiming and giving away control aligns with the core assumption of a co-constructed understanding of leadership (DeRue & Ashford, 2010), arguing that “a leader claim in isolation provides observers insufficient information to determine leadership influence” (Marchiondo et al., 2015, p. 895).

The function of leader claiming is that when a manager engages in such behavior, they signal leadership potential and should thus be more likely to be endorsed by their employees (Hemshorn de Sánchez et al., 2025; Marchiondo et al., 2015). Yet, a leader claiming behavior is more likely to succeed (i.e., elicit a granting reaction) if it matches the employees’ expectations of how a manager should behave and be in a leadership role (DeRue et al., 2009; Schyns et al., 2011). As employees’ expectations toward managers are fundamentally influenced by demographic cues associated with leadership and status, not all managers’ claiming behavior may equally result in employees behaviorally granting leadership and cognitively endorsing the manager in their leadership role.

Claiming and Being Granted Leadership in the Context of Demographic Stereotypes

A long history of research on gender stereotyping shows that people perceive men and women differently: Men tend to be evaluated as more agentic, that is, as possessing characteristics related to taking charge and being in control (e.g., such as being dominant or assertive), while women are perceived as more communal, that is, as possessing characteristics related to building and maintaining relationships (e.g., such as being supportive and understanding; Eagly et al., 2020; Haines et al., 2016; Hentschel et al., 2019; Paustian-Underdahl et al., 2024). Importantly, these stereotypes are not only descriptive but also prescriptive or normative: People tend to require men to be agentic but not too communal and women to be communal but not too agentic (Heilman et al., 2024; Prentice & Carranza, 2002; Sczesny et al., 2025).

Role congruity theory (Eagly & Karau, 2002) identified two main career challenges women face because of descriptive and prescriptive gender roles. First, leadership positions are stereotypically associated with agentic characteristics (Koenig et al., 2011), resulting in a high fit between perceptions of men and leadership but a weaker alignment with perceptions of women. Because women are seen as less consistent with the stereotypical image of leadership, they are often viewed less favorably in managerial roles and face biases and barriers as a result (Eagly & Carli, 2007; Heilman et al., 2024). Second, and most relevant for the rationale of this study, compared to male managers, prescriptive gender norms requiring women to be communal but not too agentic also apply to female managers in their daily job (Heilman et al., 1995; Heilman & Okimoto, 2007). In other words, in interactions with their employees, female managers are expected to be caring and understanding but not too dominant (Eagly et al., 1992; Ma et al., 2022). Conversely, dominant behaviors tend to be more valued when displayed by male managers. When women, including female managers, violate these gender norms by displaying highly agentic behavior—such as demonstrating the assertiveness typically expected of someone in a managerial role—they often experience social backlash in the form of less favorable evaluations (Eagly et al., 1992; Rudman et al., 2012; Schlamp et al., 2021; Williams & Tiedens, 2016). This is because others may perceive that they violate gender norms regarding how women should and should not behave.

In addition to these core theoretical tenets, role congruity theory and related conceptual lenses advocate for contextual or personal factors that exacerbate or mitigate the incongruity between women and leadership positions (Badura et al., 2018; Eagly & Karau, 1991, 2002; Heilman et al., 2024; Paustian-Underdahl et al., 2014). In line with an intersectionality perspective (Petsko et al., 2022), it is thus an oversimplification to assume that female managers are uniformly confronted with gender-based resistance. Rather, stereotypes toward gender and leadership can change over time (Eagly et al., 2020; Powell et al., 2021), pointing to the need for more nuanced theorizing that considers additional visible demographic cues associated with ascribed status (Csillag et al., 2025) to determine how female managers are evaluated and treated. In this regard, intersectionality refers to the interplay between demographic characteristics in predicting relevant outcomes (Dennissen et al., 2020), which is often used to explain why only the combination of several demographic cues indicating low ascribed status results in detrimental outcomes (Csillag et al., 2025).

In the context of leadership, one of the most immediately visible demographic cues associated with leadership is a person's age (Howard, 2000)—particularly a manager's age in relation to the employee's own age (De Boom & De Meulenaere, 2024; Shore et al., 2003; Tsui et al., 2002). A manager's age relative to an employee's age is a signal to employees about whether leadership status is deserved (Bilinska & Wegge, 2016; De Meulenaere et al., 2024). This relational understanding of relative age differences, rather than chronological age, corresponds well with our understanding of leadership as a social process constructed between the manager and the employee (Fairhurst & Uhl-Bien, 2012). More specifically, relational demography research (Tsui et al., 1995, 2002) emphasizes the relevance of relative differences in human categories for individual outcomes (Tsui & Gutek, 1999) and emphasizes that it is not merely the manager's age alone that signals leadership status over another individual, but rather the disparity in age between them and their employee.

To illustrate this situation, employees compare their career progress with others in the organization to judge whether they are “on track” (cf., Lawrence, 1984). A manager younger than the employee signals that the employee's or the manager's career does not conform to these implicit age norms (Weidner et al., 2024). As people are motivated to maintain positive views of themselves (Dunning et al., 2004), they will likely attribute the negative evaluations on their manager and be motivated to believe that a manager who is younger than themselves may not be justified or qualified for leadership (cf., Buengeler et al., 2016). Providing empirical support for this conceptual reasoning, scholars showed that older employees experience more negative emotions, such as being angry or disgusted, when they work for managers who are younger than themselves and argued that these negative emotions occur due to comparisons and perceived status incongruence between the lower age of the manager and their higher hierarchical position (Kunze & Menges, 2017). Further, comparatively older employees expect less from younger managers and have lower perceptions of their leadership performance (Collins et al., 2009).

We bring together these arguments to propose that manager gender and manager-employee age differences need to be considered in combination as their interplay results in outcomes that are not apparent when each demographic is considered in isolation. Building on the idea of congruity-mitigating and -enhancing factors proposed by role congruity theory and related research (Badura et al., 2018; Eagly & Karau, 1991, 2002; Paustian-Underdahl et al., 2014), we propose that the interaction between manager's gender and the manager-employee age difference will shape how employees react to leader claiming behaviors. Since both female gender and relatively younger age are less stereotypically associated with leadership, we propose that younger female managers experience a double jeopardy of these two characteristics (Berdahl & Moore, 2006; Sanchez-Hucles & Davis, 2010). These visible departures from stereotypical leader characteristics may lead to heightened penalties for claiming leadership, resulting in the lowest levels of leadership endorsement. Relatively older female managers fit the implicit age norm assumption that managers are more experienced than their employees (Tsui et al., 1995) and should thus be protected from unfavorable employee reactions when they claim leadership. Male managers who are younger than their employees still have the advantage of the male gender as a high-status characteristic that matches the ideal of what leaders look like (Powell & Graves, 2003; Roberson et al., 2007). Only when managers are both female and relatively younger will gender and relative age negatively predict the degree to which employees grant leadership ¹ .

Notably, some research on the intersection of gender and race suggests that younger female managers may benefit from evaluative advantages compared to their male or older counterparts (see outgroup/subordinate male target hypothesis, Navarrete et al., 2010; Veenstra, 2013; and the invisibility advantage, Livingston et al., 2012). In the domain of gender and age, a study by Scheuer and Loughlin (2020) found that older employees rated younger female managers more favorably than their younger male counterparts. This study asked participants to read vignettes about managers who decided to cut an employee's pay after the employee was late several times. In this setup, participants may be inclined to respond in ways they believe are expected of them, particularly when their responses carry no real-world consequences (Wulff et al., 2023; see also Khademi et al., 2021; Lonati et al., 2018; Sturm & Antonakis, 2015). Supporting this idea, prior research has demonstrated that people's decisions in actual situations can differ markedly from those made in hypothetical ones (FeldmanHall et al., 2012). Employees focus on what manager behaviors mean for them or how these behaviors make them feel. Their evaluations are more subjective and linked to their personal experience.

In sum, while we acknowledge that the findings on intersectionality are mixed such that the combination of several status-related demographic cues can amplify or mitigate stereotypes and reactions (Hall et al., 2019), we propose that in the case of leader claiming behavior shown by younger and older male and female managers in real workplace meetings, younger female managers are more likely to experience the drawbacks compared to a vignette study. Indeed, unlike participants in an experimental vignette setting (e.g., Marchiondo et al., 2015; Scheuer & Loughlin, 2020), who might be worried about being perceived as biased and who experience no consequences for socially desirable responses, direct employees are deeply embedded in the relational dynamics of the workplace and actually experience the consequences of their behavior. Here, we expect that a relatively younger female manager who claims leadership in an interaction with their employee will be granted less leadership during the interaction itself (i.e., immediate verbal granting behavior) and will be less endorsed as a leader following the interaction (i.e., subsequent cognitive granting of leadership). In other words, the disadvantages related to the low ascribed status of being a woman and being relatively younger manifest in a double jeopardy effect (Berdahl & Moore, 2006) in the form of low immediate leadership granting and low subsequent cognitive leader endorsement. ² In contrast, we do not expect a negative relationship when the manager possesses only one non-stereotypical demographic characteristic. Our conceptual model is summarized in Figure 1.

Hypothesis 1. For female (but not male) managers, manager-employee age difference moderates the positive relationship between managers’ leader claiming behaviors and employees’ immediate verbal granting behaviors during a meeting interaction, such that the relationship is weaker for female managers who are relatively younger than their employees.

Hypothesis 2. For female (but not male) managers, manager-employee age difference moderates the positive relationship between managers’ leader claiming behaviors and employees’ cognitive granting following a meeting interaction (i.e., leader endorsement), such that the relationship is weaker for female managers who are relatively younger than their employees.

Figure 1.

Conceptual Model.

Method

Study Design

We recruited managers and employees to record a routine one-on-one meeting and to complete a questionnaire before and after the meeting. We collected data from 73 ³ manager-employee dyads from organizations in different sectors: information technology (22.1%), service (16.2%), the public sector (11.8%), manufacturing (7.4%), and ‘other’ sectors (42.5%; e.g., energy industry, freelance, non-profit). However, four dyads were excluded because they did not follow the instructions (i.e., they included three people in the meetings), and one dyad was excluded because the manager selected ‘other’ when reporting their gender. Therefore, our final dataset consisted of 68 manager-employee dyads.

The study was reviewed and approved by the Scientific and Ethical Review Board (VCWE) by Vrije Universiteit Amsterdam (#2018-126R1). Considering that the data collection approach is not often used in a field setting as companies are concerned about the effort and data security (Hemshorn de Sánchez et al., 2022), we applied different recruitment strategies to be able to obtain this sample. Taking particular care to highlight the privacy of the data, participants were contacted through the network of the authors and their research teams, as well as informed about the study through social media networks (e.g., LinkedIn, and Viva Engage) and networking events. Most participants who agreed to participate worked in Dutch and German organizations. In order to be included in our sample, participants had to video or audio record a routine one-on-one meeting between themselves and their manager (a hierarchical power difference in a workplace setting, e.g., daily supervisor or project manager) or employee. Furthermore, the meeting had to consist only of those two individuals, and it could be face-to-face or in a virtual setting.

Once the dyads agreed to participate, we sent them an email with concrete instructions and a Frequently Asked Questions document. Specifically, they (1) received information on how to video or audio record their routine meeting, (2) received a code to depersonalize their data, (3) were asked to fill in the consent form and complete a short questionnaire. The FAQ document included details such as the minimum required length of the recording, password-protecting the recording, and securely sending the video or audio record file to the research team. Additional information was provided to ensure participants felt comfortable recording their actual one-on-one meetings. This included details about the research team conducting the study, confirmation of the study's ethical approval, and assurances regarding data confidentiality. We emphasized that data would be aggregated across participants, guaranteeing that individual information would remain completely confidential and never be disclosed. After the one-on-one meeting, employees were asked to complete one more questionnaire that included our study constructs of interest.

We subsequently used the participants’ code to link their questionnaire data with the data obtained from the video or audio record. With regard to the latter, we coded the communication of the manager-employee dyads using INTERACT software (Mangold, 2010). This software allows coders to (1) cut the tape into sequences of messages (which results in timestamped data), (2) note who was speaking, and (3) assign a corresponding behavioral code based on the relational control coding scheme described in the next section. An expert coder comprehensively trained four research assistants on using the INTERACT software and the coding scheme until they reached a satisfying Cohen's kappa value (Cohen, 1960) of at least Ƙ = .70.

Sample Demographics

The manager-employee composition of the 68 dyads was as follows: 20 dyads with male managers and male employees, 15 dyads with male managers and female employees, 6 dyads with female managers and male employees, and 27 dyads with female managers and female employees. Of the 68 managers, 35 were male, and 33 were female. The average age of the manager was 39.27 years (SD = 13.06). Managers had an average of 8.65 years of experience in a leadership role (SD = 9.10) and an organizational tenure of 7.74 years (SD = 7.70). Of the 68 employees, 26 were male, and 42 were female. The average age of the employees was 31.14 years (SD = 11.81). Employees had an average organizational tenure of 3.72 years (SD = 6.22). In terms of the topics of the one-on-one meetings, they were typically about providing feedback, assigning tasks, or providing project updates. On average, meetings lasted 37.56 min (SD = 14.89).

Relational Control Coding Scheme

To recap the conceptual foundation of our behavioral coding approach, whether a statement is claiming leadership, granting leadership or just neutral information sharing can only be determined when considering the embeddedness in the conversation context. This conceptual assumption underlying the understanding of claiming and granting leadership renders most existing leader behavior coding schemes (e.g., Gerpott et al., 2019; Meinecke & Lehmann-Willenbrock, 2015; Schlamp et al., 2021) inapplicable because these schemes only consider verbal behavior in isolation. In contrast, the relations control coding scheme (Fairhurst, 1990) used for this study considers the conversational context. It consists of 50 behavioral categories that we applied to code the verbal behaviors occurring in the manager-employee meetings. The total number of coded behaviors from the 68 meetings was 37,277. The utilized relations control coding scheme by Fairhurst (1990), adapted from Rogers and Frace (1975), has been employed in past research to better understand control within dyadic and group situations in marital couples (Rogers, 2001), student project teams (Abraham et al., 2019) and organizations (Courtright et al., 1989; Fairhurst et al., 1987, 1995).

When applying the relational coding scheme for behavior coding, each utterance that is made during the meeting by the manager or the employee receives three digits: 1) one for the speaker (e.g., 1 = manager), 2) one for the grammatical form (e.g., 1 = assertion), and 3) one for the response form (e.g., 1 = support). Table 1 displays all possible digits, that together result in one specific code for each utterance. The combination of the grammatical form and the response form determines the relational control move of the speaker (see Table 2 as well as Fairhurst, 1990; Fairhurst et al., 1995; Rogers & Frace, 1975). The relational control move can either be one up (↑) if the interaction partner asserts their status (i.e., claiming), one down (↓) if the interaction partner endorses the other's status (i.e., granting), and one across (→) if it is a neutral, leveling speech.

Table 1.

Relational Control Coding Scheme.

Digit 1	Digit 2	Digit 3
1 = Manager	1 = Assertion	1 = Support
2 = Employee	2 = Question	2 = Non-support
	3 = Successful talk-over	3 = Extension
	4 = Unsuccessful talk-over	4 = Answer
	5 = Other statements	5 = Instruction
		6 = Order
		7 = Disconfirmation
		8 = Topic Change
		9 = Backchannel/Active listening
		10 = Other

Note. Each utterance receives three digits that reflect (1) the speaker (i.e., manager or employee), (2) the grammatical form, and (3) the response form. The combination of (2) + (3) results in 50 different types of relational control behaviors that can be coded in terms of their control direction.

Table 2.

Relational Control Behaviors Categorized by Control Direction (Claiming vs. Granting).

	Support	Nonsupport	Extension	Answer	Instruction	Order	Disconfirmation	Topic change	Backchannel/ Active listening	Other
Assertion	↓	↑	→	↑	↑	↑	↑	↑	→	→
Question	↓	↑	↓	↑	↑	↑	↑	↑	↓	↓
Successful talk-over	↓	↑	↑	↑	↑	↑	↑	↑	↑	↑
Unsuccessful talk-over	↓	↑	→	↑	↑	↑	↑	↑	→	→
Other statements	↓	↑	→	↑	↑	↑	↑	↑	→	→

Note. The combination of the grammatical and response form determines the relational control move. The relational control statements can either be one up (↑) if the interaction partner asserts their status (i.e., claiming), one down (↓) if the interaction partner endorses the other's status (i.e., granting), and one across (→) if it is neutral, leveling speech.

To provide an example illustrating how Tables 1 and 2 are connected, the phrase “Don’t you see that this is a mediocre deliverable?” from the manager would be coded as 122 (1 = manager, 2 = question, 2 = non-support) (see Table 1). Based on Table 2, one can see that this combination of 2 = question and 2 = non-support represents claiming. The employee's phrase “That is an excellent idea” would be coded as 211 (2 = employee, 1 = assertion, 1 = support). This combination represents a granting behavior (see Table 2). Appendix 1 provides an example transcript that outlines the coding approach in more detail for further illustrative purposes.

Measures

Leader Claiming Behavior During the Meeting Interaction

In the utilized relational control coding scheme (Abraham et al., 2019; Fairhurst et al., 1995; Rogers & Frace, 1975), 33 types of statements (i.e., the combination of the grammatical form and the response form) are classified as verbal leader claiming behavior (see Table 2). Examples of these statements are non-supportive questions (e.g., “Didn’t I ask you to inform me before talking to the suppliers?”) or assertion instructions (e.g., “Please send me the report so I can review it before the meeting”). We included the sum of managers’ leader claiming behaviors in the analysis.

Employee Granting Behavior During the Meeting Interaction

In the utilized relational control coding scheme (Abraham et al., 2019; Fairhurst et al., 1995; Rogers & Frace, 1975), eight types of statements are classified as granting behaviors (see Table 2). Examples of these statements are supportive assertions (e.g., “This is a good plan”) or extension questions (e.g., questions that continue the flow of the conversation, such as “And when is the deadline for this project?”). In the analysis, we included the sum of employees’ granting behaviors that followed managers’ leader claiming behaviors (i.e., transitional frequencies; Lehmann-Willenbrock et al., 2011) to measure immediate verbal granting behaviors during an interaction.

Employee Cognitive Granting (i.e., Leader Endorsement)

We assessed employees’ cognitive granting of leadership via questionnaire before the meeting (used as a control variable) and following the meeting (used as the outcome in terms of employees’ subsequent leader endorsement). Specifically, employees responded to three items (Giacalone et al., 1988; Michener & Lawler, 1975): “I am willing to have this leader continue in his/her role as a team leader,” “I support this leader in his/her role as a team leader,” and “I think this leader is legitimately in his/her role as a team leader.” Employees indicated their answers on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). Cronbach's alpha values for leader endorsement before and after the meeting were α = .80 and α = .89, respectively.

Gender

Managers and employees reported their gender in a questionnaire before the meeting.

Age Difference

Managers and employees reported their ages in a questionnaire before the meeting. We calculated the age difference of each manager-employee dyad by subtracting the employee's age from the manager's age. Positive values indicate that the manager is older than the employee; negative values indicate that the manager is younger than the employee. For example, if the manager is 54 years old and the employee is 31, the age difference would be 23 years; if the manager is 31 and the employee is 54 years old, the age difference would be 23 years.

Results

Relational Control Coding Validation

We first sought to provide additional evidence for the validity of the coding scheme by running an independent validation study. We investigated whether naive individuals can reliably differentiate between the proposed leader claiming and granting behaviors. We, therefore, recruited 100 participants via Amazon CloudResearch (Litman et al., 2017). After a short explanation and training round with six example statements, participants received a list of 18 written communicative behaviors (i.e., behaviors that occur in manager-employee interactions in similar forms). Participants were asked to classify these model behaviors into either claiming or granting behaviors according to the relational coding scheme. We coded participants’ incorrect answers as 0 and correct answers as 1. The mean correctness score for coding claiming behaviors was .82 (SD = 0.38); for coding granting behaviors, it was also .82 (SD = 0.37). These results substantiate trust in the coding approach: when trained only with a few example statements, raters from the general population classified most of the behaviors in the same way as our trained research assistants. Note that the four research assistants who were conducting the coding in the actual study underwent much more rigorous and comprehensive coder training, and we can reasonably assume that their coding was even more accurate. A detailed description of the relational control coding validation is included in Appendix 2 of the manuscript.

Preliminary Analyses

Using our data set of 68 manager-employee meetings, we coded 19,060 manager behaviors (i.e., unstandardized behaviors). Of these, 3,104 were claiming (i.e., upward) behaviors, 4,235 were granting (i.e., downward), and 11,721 were neutral (i.e., across) statements by the managers. For employees, we coded 18,217 behaviors. Of these, 2,739 were claiming behaviors, 3,871 were granting behaviors, and 11,607 were neutral.

Table 3 displays means ⁴ , standard deviations, and correlations between our focal variables (manager gender, manager-employee age difference ⁵ , the managers’ leader claiming behaviors, the employees’ granting behaviors during the meeting, and employees’ leader endorsement of their manager as reported after the meeting) and the control variables (employees’ leader endorsement before the meeting ⁶ , employee gender ⁷ ). There was no correlation between manager gender and leader claiming behaviors, and an exploratory t-test indicated that male (M = 73.49, SD = 43.50) and female managers (M = 67.77, SD = 37.11) showed similar levels of leader claiming behaviors, t(66) = .58, p = .56.

Table 3.

Descriptive Statistics and Intercorrelations of Study Variables.

Variable	M	SD	(1)	(2)	(3)	(4)	(5)	(6)
1. Manager gender	1.48	0.50
2. Manager-employee age difference	9.47	11.94	−.001
3. Leader claiming behaviors	70.72	40.33	−.07	−.14
4. Granting behaviors	12.81	18.91	−.15	−.07	.38**
5. Leader endorsement (i.e., cognitive granting)	4.81	0.42	−.17	.17	−.04	.06
6. Pre-leader endorsement (control)	4.74	0.45	−.13	.09	−.16	.04	.67**
7. Employee gender (control)	1.62	0.49	.40**	.02	−.05	−.18	−.14	−.11

Note. *p < .05, **p < .01 (two-tailed). All variables were measured at the individual level (N = 68 dyads). Manager and employee gender was coded as 1 = female, 2 = male. The mean values for verbal behaviors refer to the standardized behaviors per person, per 60-min period.

Hypothesis Testing

To test Hypothesis 1, which proposed that the interaction between manager gender and manager-employee age difference moderates the positive relationship between female (but not male) managers’ leader claiming behaviors and employees’ immediate verbal granting behaviors, we used model 3 of SPSS PROCESS (Hayes, 2015). PROCESS results are displayed in Table 4. Results showed that the proposed three-way interaction between manager gender, manager-employee age difference, and the manager leader claiming behavior was not significant (b = −.02, p = .16; Cohen f² = 0.03; see Table 4). This means that we did not find support for employees’ immediate verbal granting behaviors toward their managers as being influenced by the combined effects of manager gender and manager-employee age difference. Hence, the data did not provide support for Hypothesis 1. Notably, a regression analysis showed a direct positive relationship between managers’ leader claiming behaviors and employee granting behaviors (when the moderators manager gender and manager-employee age difference are not included in the analysis), b = .18, t(66) = 3.31, p = .001. This indicates that the managers’ leader claiming behaviors were generally positively associated with employees’ immediate verbal granting behaviors.

Table 4.

Results for the Three-Way Interactions of Manager's Leader Claiming Behavior, Manager Gender and Manager-Employee Age Difference on the Dependent Variables.

					95% CI
Model	b	SE	t	p	Lower	Upper
Employee Granting Behaviors During the Meeting Interaction
Managers’ leader claiming behaviors (LCB)	−.01	.19	−.04	.97	−.40	.38
Manager gender (Gen)	−11.85	11.58	−1.02	.31	−35.04	11.33
Manager-employee age difference (AgeDiff)	−1.56	1.31	−1.19	.23	−4.18	1.05
LCB × Gen	.14	.14	.98	.33	−.14	.42
LCB × AgeDiff	.02	.015	1.49	.14	−.01	.05
Gen × AgeDiff	1.01	.82	1.24	.22	−.62	2.65
LCB × Gen × AgeDiff	−.02	.011	−1.43	.16	−.04	.01
Employee gender (control)	−6.10	4.97	−1.23	.22	−16.04	3.83
Employee Cognitive Granting (i.e., Leader Endorsement) Following the Meeting Interaction
Managers’ leader claiming behaviors (LCB)	.006	.003	2.08	.04 *	.0002	.013
Manager gender (Gen)	.23	.19	1.24	.22	−.14	.61
Manager-employee age difference (AgeDiff)	.02	.02	.76	.45	−.03	.059
LCB × Gen	−.005	.002	−2.42	.02*	−.01	−.001
LCB × AgeDiff	−.0004	.0002	−1.69	.09	−.0009	.0001
Gen × AgeDiff	−.020	.013	−1.53	.13	−.047	.006
LCB × Gen × AgeDiff	.0005	.0002	2.66	.01*	.0001	0009
Pre-leader endorsement (control)	.64	.08	7.70	.00**	.48	.81
Employee gender (control)	.03	.08	−.37	.71	−.19	.13

Note. *p < .05, **p < .01. All variables were measured at the individual level (N = 68 dyads).

To test Hypothesis 2, which proposed that manager-employee age difference moderates the positive relationship between female (but not male) managers’ leader claiming behaviors and the employees’ cognitive granting (i.e., leader endorsement), we again used PROCESS model 3 (see Table 4; Hayes, 2015). In addition to controlling for employee gender, we also controlled for employees’ leader endorsement before the meeting (Bernerth & Aguinis, 2016). The three-way interaction between manager leader claiming behaviors, manager gender, and manager-employee age difference on leader endorsement was positive and significant as the bootstrapping interval did not include zero (b = .0005, p = .008, Cohen f² = 0.12) ⁸ . In line with Hypothesis 2, the extent to which employees cognitively granted leadership to their managers (as manifested in the post-meeting leader endorsement scores) in response to their manager's leader claiming behaviors was influenced by the interplay of their manager's gender and their age difference between the employee and their manager.

To follow up on the significant three-way interaction, we calculated the conditional effects. Specifically, we calculated the impact of male and female managers’ claiming behaviors on leader endorsement at low (1SD below the mean, corresponding to managers being two years younger than their employee) and high (1SD above the mean, corresponding to managers being 21 years older than their employee) levels of the manager-employee age difference ⁹ . For male leaders, the conditional effect of leader claiming behavior on leader endorsement was not significant, irrespective of any age difference value; that is, neither for managers who were younger than their employees by 1SD (b = .0008; p = .58) nor for managers who were older than their employees by 1SD (b = .0026; p = .19). This means that there was no relationship between the employees’ leader endorsement of male managers, leader claiming behavior and manager-employee age difference.

For female managers, the conditional effect of managers’ leader claiming behavior on leader endorsement was negative and significant when managers were younger than their employees by 1SD (b = −.006; p = .008). In other words, leader claiming behaviors of female managers younger than their employees decreased employees’ endorsement of them as leaders. In contrast, the conditional effect was positive and significant when female managers were older than their employees by 1SD (b = .007; p = .005). These findings indicate that a higher (versus lower) number of leader claiming behaviors of older female managers increased employees’ endorsement of them as leaders.

We followed the steps proposed by Dawson and Richter (2006) to analyze the slope differences. The slope for female managers who were younger than their employees by 1SD differed significantly from the other slopes: that of female managers who were older than their employees by 1SD, t(66) = 4.01, p < .001, that of male managers who were older than their employees by 1SD, t(66) = −2.78, p = .008, and that of male managers who were younger than their employees by 1SD, t(66) = −2.62, p = .01. In summary, these results provide support for Hypothesis 2. The data show that leader claiming behaviors resulted in lower endorsement for female managers who are younger than their employees compared to older female managers and compared to male managers irrespective of their age.

As an exploratory analysis, we also considered the slope for female managers who were older than their employees by 1SD and found that when it differed significantly not only from female managers who were younger than their employees by 1SD but also from male managers who were younger than their employees by 1SD, t(66) = 2.77, p = .007. These findings indicate that older female managers were endorsed more than younger managers, irrespective of gender. The slope difference for female managers who are older than their employees by 1SD compared to that of male managers who are older than their employees by 1SD was insignificant, t(66) = 1.72, p = .091. None of the other slopes differed significantly from each other.

Figure 2 illustrates the interaction effect between leader claiming behavior, managers’ gender, and manager-employee age difference on employees’ leader endorsement.

Figure 2.

Interaction Effect between Leader Claiming Behavior, Managers’ Gender, and Manager-employee Age Difference on Employees’ Leader Endorsement.

To gain a nuanced understanding of the specific manager-employee age difference that influences the relationship between manager gender and leader claiming behavior on employee endorsement, we conducted the Johnson-Neyman analysis. The Johnson-Neyman analysis identifies the point at which age difference values in the interaction of manager gender and leader claiming behaviors on leader endorsement change from statistically nonsignificant to significant (Bauer & Curran, 2005; Hayes & Matthes, 2009). Results (see Figure 3) show that the conditional effect of claiming behavior and manager gender on leader endorsement was significant when managers were (1) less than 3.1 years older than their employees or (2) more than 30 years older than their employees. Combining these results with the slopes analysis presented above, we conclude that leader claiming by a female manager was negatively associated with leader endorsement when the manager was younger and similar in age to the employee. In addition to these predicted findings, we also found that leader claiming by a female manager was positively associated with leader endorsement when the manager was substantially older than the employee.

Figure 3.

Results of Johnson-Neyman Analysis: Point at which Age Difference Values in the Interaction of Manager Gender and Leader Claiming Behaviors on Leader Endorsement Change from Statistically Nonsignificant to Significant.

Discussion

We proposed that the extent to which a manager is granted leadership status by their employee for engaging in leader claiming behaviors can be better understood by considering the interaction of both the manager's gender (i.e., being female or male) and the age difference between the manager and their employee (i.e., being relatively younger). Using relational control coding to capture verbal behavior in typical one-on-one meetings between managers and employees with over 37,000 behaviors, we found that male and female managers exhibited the same amount of leader claiming behaviors in daily workplace meetings with their employees. During these meetings, managers’ leader claiming behaviors were overall positively associated with employees’ immediate verbal granting behaviors during the meeting, and this relationship was not affected by the managers’ gender or their age difference with employees. These results were not in line with Hypothesis 1, as we expected that female managers relatively younger than their employees would receive fewer immediate verbal granting behaviors.

In contrast, we found that cognitive granting – i.e., the employees’ leader endorsement after the meeting – varied depending on the demographic characteristics of the manager. For male managers, there was no relationship between leader claiming behavior and leader endorsement, irrespective of whether managers were older or younger than their employees. For female managers, age differences with employees mattered: In line with Hypothesis 2, for female managers who were younger than their employees, leader claiming behaviors were negatively linked to their employees’ endorsement of them as leaders. Interestingly, we further found that for female managers who were relatively older than their employees, leader claiming behaviors were more strongly associated with cognitive granting, compared to relatively younger male and female managers.

Theoretical Implications

Our study has several implications for theory about gender, age, and leadership. To begin with, it is noteworthy that in a field setting, we find effects of managers’ combined status-related demographic cues (i.e., gender and age) on cognitive granting (i.e., overall leader endorsement) as reported by employees after the meeting but not on employees’ immediate verbal granting behaviors during the meeting. Lower levels of granting leadership directly and potentially quite saliently during the meeting would indicate to the manager that their leadership is challenged. Contrarily, lower levels of leader endorsement, as reported by employees right after the meeting, means that the manager is less accepted as a leader in a more indirect and concealed manner. This difference suggests that although neither female nor male managers younger than their employees tend to be directly confronted with non-support by their employees after claiming leadership, younger female manager's claiming of leadership may often be questioned behind their back, which is harder to address. We did not expect to find no influence of the manager's combined status cues on employees’ cognitive granting in response to the manager's leader claiming behaviors. Instead, we anticipated that female managers relatively younger than their employees would receive fewer immediate verbal granting behaviors based on research demonstrating prejudice against female leaders (Eagly & Karau, 2002; Scott & Brown, 2006). As such, the findings underscore the critical distinction between perceived and actual behaviors of female managers, which aligns with findings from older research on non-verbal behavior (Butler & Geis, 1990) but ever since then has received only very limited attention (Paustian-Underdahl et al., 2024). Indeed, leadership theories allowing to make explicit predictions about differences in perceptions and behaviors in the context of demographic cues remain remarkably absent despite repeated calls to disentangle actual and perceived leader behavior from its outcomes when developing more nuanced, less tautological, and ultimately more useful theory (Antonakis, 2017; Banks et al., 2023; Fischer & Sitkin, 2023). In that regard, our study provides a novel contribution to further understanding leader claiming and granting processes, using the behavioral building blocks proposed by DeRue and Ashford's claiming/granting framework (2010) combined with demographic information and perceptual measures. A potential interpretation is that this divergence of behavioral versus cognitive reaction could be related to employees wanting to act in line with gender equality ideals (i.e., wanting to react in a fair and accepting manner to managers independent of their demographic characteristics). In contrast, employees’ cognitive interpretations of managers’ leader claiming behaviors could be guided more by prescriptive gender and age stereotypes.

Interestingly, we found that female managers who are substantially older than their employees are more strongly endorsed in their leadership role when displaying a high (versus low) number of leader claiming behaviors. This finding could, on the one hand, be interpreted in the context of work on selection effects and the double standard of competence (Foschi, 1996; Rosette & Tost, 2010). Specifically, older female managers who are still in managerial roles may profit from their behavior being interpreted as especially suited for leadership, having had to repeatedly exceed higher performance thresholds and persist in roles where they may have been consistently confronted with gendered stereotypes. Alternatively, this cognitive endorsement advantage may stem from reduced perceptions of role incongruity, such that relatively older women are increasingly visible in leadership roles in daily life and especially in the media (Sloane, 2023; Williams, 2021). More prevalent narratives of older female managers actively claiming leadership may foster stronger perceptions of role congruity between older women and leadership positions compared to their younger counterparts.

The identified negative reaction towards younger female managers and the positive reaction towards older female managers claiming leadership also contributes to the growing field of research discussing when agency displayed by female managers or women in general leads to an advantage rather than a disadvantage for them (Eareckson & Heilman, 2024; Hentschel et al., 2018; Ma et al., 2022). Ma et al. (2022) discuss that agency displayed by women can lead to both advantages and disadvantages. They argue that the evaluation of a behavior that can be perceived as agentic, such as claiming leadership, depends on the specific type of agency attributed by the observer. Specifically, this evaluation depends on whether or not the type of agency is perceived as prescribed (i.e., desirable or appropriate) or proscribed (undesirable or inappropriate). Women who show agency in the prescribed terms of competence and independence are particularly positively evaluated; they “overachieve” agency requirements that tend to be lower for women than for men. However, women who show agency in proscribed terms of dominance are particularly negatively evaluated for their violation. Our research extends this idea by suggesting that the perception of agency as prescribed or proscribed may depend not only on the actor's gender but also on their age. While claiming leadership might be seen as dominance (a proscribed behavior) for relatively younger female managers, claiming leadership might be seen as leadership competence (a prescribed behavior) for older women. As a result, employees may react differently depending on the age of a female manager.

In terms of a broader theoretical perspective, our approach to considering gender and age in combination can be connected to recent advancements in a lens-based approach to intersectional stereotypes (Petsko et al., 2022), which suggests that observers (e.g., employees) evaluate others (e.g., their managers) through one currently activated social lens (Petsko et al., 2022). This lens, defined as an identity-specific schema for categorizing others, can be complex because, depending on the context, the observer may merge several intersectional identities (and stereotypes) of the evaluated person. Recent conceptual work tried to specify this complex approach to intersectional stereotypes by suggesting that an age-by-gender lens is particularly likely to be activated in the leadership context (Thrasher, 2022). As such, our work can be seen as an empirical addition to this newly emerging research stream.

Lastly, our verbal behavior analysis offers a novel methodological contribution to the science of actual behaviors beyond the ritualized use of questionnaires (Fischer et al., 2020). Previous research has mainly focused on survey data to measure leader behaviors and leadership styles in the context of gender and age (Buengeler et al., 2016; Lanaj & Hollenbeck, 2015). Considering the abundance of academic and practitioner articles debating the leadership styles of men and women, it is surprising that so few studies have actually measured male and female leaders’ verbal behavior, relying instead on proxies such as self- or other-reported data (Paustian-Underdahl et al., 2024). Literature that did consider actual participation suggests that talking a lot may be beneficial for both male and female managers (Badura et al., 2018; Loignon et al., 2025). However, it has been suggested that not only communication quality but also the different types of participation (i.e., communication content, e.g., Gerpott et al., 2019) should be taken into account to fully understand how leaders are accepted by their employees. Despite these calls, extant research on communication content, to our knowledge, has not yet studied the combined role of gender and age in the context of the claiming/granting framework of leadership. The relational control coding scheme that we applied in this research expands previously utilized interaction coding approaches in leadership research that only considered the content of a single focal behavior (Gerpott et al., 2019; Hemshorn de Sánchez et al., 2025), which overlooks that managers and employees communicate reciprocally (Decuypere & Pircher Verdorfer, 2022). Instead, we rejuvenate the idea of control moves in interactions, allowing us to capture behaviors by considering both the content and their embeddedness in the interaction (i.e., their response form). Specifically, granting and claiming leadership does not happen in a vacuum but contains the idea of an interaction counterpart that reacts to such control moves. Conceptually, such an approach reflects the importance of considering the communication context to understand power dynamics in one-on-one meetings between managers and their employees. This aspect is often neglected (Flinchum et al., 2022). Empirically, we offer a new approach to validate coding studies (see Appendix 1). While it is common to solely report the training of coders according to a coding scheme and then report their agreement, we additionally validated our coding by surveying whether naïve participants with only a brief introduction to the coding would come to similar conclusions when presented with the material. The high scores obtained here should lend trust to the developed coding and suggest it would also hold when employed in other research groups. As such, this novel approach offers a way for future research to make qualitative coding research more replicable.

Practical Implications

First, the results provide preliminary evidence indicating that employees may still provide a lower endorsement rating after an interaction with relatively younger (and same-age) female managers who verbally claim leadership during a meeting. Hence, the lack of leader endorsement might occur “behind the back” of relatively younger female managers. This divergence between immediate behavioral granting and cognitive granting at a later time could potentially disrupt the formation of a leader identity when younger female managers learn about their lower endorsement in subsequent explicit evaluations or via workplace gossip (Varty et al., 2021). Such a disruption could be particularly damaging, as people can be reluctant to lead when they fear their image might be harmed by seeming bossy, unqualified, or different from others (Lee Cunningham et al., 2023). As such, it seems warranted to make younger female managers aware that though they may not recognize immediate detrimental behavioral reactions, they may still be confronted with subtle stereotypical evaluations that negatively impact their assessment ratings. Early and continued 360-degree feedback can help younger female managers, particularly in their leadership development, to have open conversations about non-endorsement (Ely et al., 2011; Knipfer et al., 2017). In that regard, the Human Resource Management departments should ensure that 360-degree feedback results are used solely for developmental purposes rather than for promotion (or similar) decisions, as the feedback can be subject to bias and subjectivity (Atwater et al., 2002).

Second, the results indicated that male and female managers engage in leader claiming behaviors at similar levels. This aligns with previous research demonstrating minimal gender-based differences in concrete behaviors (Mehl et al., 2007; Schlamp et al., 2021) and challenges the widely held assumption that men and women communicate and lead differently (cf., Obenauer et al., 2024). Such beliefs are reinforced by popular narratives that encourage women to “lean in,” like in structural organizational interventions that focus on teaching women how to adjust their behaviors (Wong et al., 2025), or unconscious bias trainings that promote the notion of “female leadership” as distinct from male leadership (Noon, 2018). However, despite the anticipated positive outcomes of such diversity initiatives, supporting evidence remains limited (Gierke et al., 2025) and lacks clear, unambiguous backing (Caleo & Heilman, 2019; Dover et al., 2020). Given these findings, organizations may need to reconsider how they frame discussions and training around women's leadership styles. For example, rather than emphasizing assumed gender differences, leadership development programs could be more effective by focusing on the behaviors that have empirically shown to increase follower effectiveness, regardless of gender.

Limitations and Future Directions

First, while our sample is beyond that found in previous dyadic interaction analysis research (e.g., Gerpott et al., 2019; Meinecke & Lehmann-Willenbrock, 2015; Schlamp et al., 2021), our sample is on the low end of the optimal size spectrum, particularly when considering the complexity of our analyses. That is, although we coded a large number of behaviors, the focus of our analysis is the dyad, and thus the sample size is relatively small. Hence, we caution against overinterpreting non-significant results as definitive evidence of the absence of an effect in the entire population. Although it would be advisable for future research to replicate our findings with larger samples, we know that getting access to the recordings of “real” workplace meetings between managers and their employees is challenging because many companies are concerned about data protection issues (cf. Güntner et al., 2023). Accordingly, future research may be well advised to creatively think about alternative ways to get access to communication data. For example, extensive, anonymized online communication text data from work-related chat protocols or data collected via apps that allow for live-coding speaker voices in interactions between managers and employees could be used to gain access to more extensive data sets of actual workplace behavior (cf. Buengeler et al., 2017; Wenzel & Van Quaquebeke, 2018). By the same token, such new data gatherings may also serve to check the robustness of the obtained results. For instance, building on the observed discrepancies between objective data and perceptions, scholars could additionally gather data on employees’ perceived age differences in relation to their leader (Avery et al., 2007; Gerpott et al., 2021).

Second, we acknowledge the possibility that our study design and setting could impact employees’ behaviors during the meeting. That is, recording the meetings might have influenced participants’ behavior, particularly regarding potential hesitancy to challenge a manager. However, when coding the recordings, we observed instances of non-socially desirable behaviors, such as employees making negative remarks, talking badly about others, and challenging their leaders. This suggests that participants did not exhibit excessive self-censorship due to the presence of recording devices (see also Güntner et al., 2023). Moreover, our study predominantly included participants from Dutch and German organizations, where research suggests that power distance and, consequently, automatic deference to authority is moderate to low (Chhokar et al., 2007). Studies in other cultural contexts could serve to triangulate the current results as employees might have a more compliance-oriented approach to followership in such settings.

Third, in our sample, a disproportionately higher number of female managers were paired with female employees compared to male employees, resulting in an imbalance in the gender composition of dyads. Therefore, we conducted additional exploratory analyses to check whether the interaction of manager-employee gender similarity, manager-employee age difference, and manager claiming behavior predicted employees’ verbal or cognitive granting. None of the results involving employee gender were significant (see footnote 7). Our reasoning for only controlling for employee gender and not developing a hypothesis for it was grounded in both conceptual and empirical work. From a conceptual perspective, role congruity theory (Eagly & Karau, 2002) suggests that gender biases in leader endorsement stem from societally shared expectations about what a leader should look like. Because these stereotypes are widely shared across society, both men and women are similarly exposed to them and socialized into learning them. As a result, we anticipated that employee gender would not play a meaningful role in granting leadership during the interaction or in cognitively endorsing leadership afterward. Empirically, previous research also often found that perceiver gender does not significantly influence backlash against women in positions of power (e.g., Brescoll, 2011; Davison & Burke, 2000; Van Gils et al., 2018). However, some studies indicate that employee gender may matter in certain contexts, such as evidence showing that leader-member gender similarity was negatively associated with absence (Pelled & Xin, 1997) or positively linked to the quality of working relationships (Vecchio & Brazil, 2007). Due the gender imbalance in our study, future research could thus further explore more balanced gender distributions, potentially not just in dyads but also in the broader team context—a point we turn to next.

Specifically, we focused on the intersection of manager gender and relative age in interactions in dyads, a unique level of analysis that has also been called the “forgotten level” in organizational behavior research in general (Kenny et al., 2006) and leadership research in particular (Flinchum et al., 2022). Nevertheless, our research could be expanded to the team level to explore whether similar or different dynamics emerge when several employees interact with a male or female manager. In the team context, leader group prototypicality—i.e., the extent to which the manager represents the group—may play an important role, which entails that a younger female manager leading several female employees may be more prototypical for the group than a male manager. In such a female-dominated context, with female managers being more common, leadership may be perceived as more in line with the female gender role and, thus, the congruity between women and leadership is increased (Eagly & Karau, 2002). Indeed, managers perceived to be more prototypical of the group receive stronger leader endorsement (Hogg, 2001) and are perceived as more effective (Van Knippenberg, 2011). In a team setting, it is thus important to explore gender- and age-based team prototypes in combination with team composition and manager characteristics to fully unpack the complex interplay of diversity cues (Weidner et al., 2024).

Lastly, we analyzed the manager's verbal behavior as a predictor of leader endorsement based on utilizing the relational control coding scheme. Evidently, future research with different behavioral measures could help us further understand potential differential effects of leader behavior on employee's behavioral versus cognitive reactions. Moreover, leadership can also be claimed and granted via non-verbal behaviors. Future research could collect comprehensive data on verbal and non-verbal cues (i.e., multimodal behavioral pattern analysis; Santangelo et al., 2020) to explain claiming and granting leader status. Linking these multimodal interaction patterns to DeRue and Ashford's claiming/granting theory (2010) makes their model more complex but potentially more concrete, ultimately contributing to making this impactful theory more testable (Lehmann-Willenbrock & Hung, 2024). While such complex analytical approaches were previously unfeasible due to the high data analytical complexity, recent advancements in Artificial Intelligence have introduced machine learning algorithms capable of managing these data sets. These developments offer promising avenues to further advance our understanding of the building blocks of leadership (Van Quaquebeke & Gerpott, 2023).

Conclusion

Our study contributes to the claiming and granting framework of leadership and provides insights into the intersectional influence of age and gender on leader endorsement in manager-employee dyads. We hope our research on how leadership is claimed and (behaviorally and cognitively) granted in interactions in daily workplace meetings will inspire researchers and practitioners to further explore biases resulting from subtle micro-level interactions and how they may accumulate to prevent equality in organizations.