Leading “us” towards greater commitment: the role of sharing decision-making power in collectivistic and individualistic cultures

Identity leadership describes how leaders influence others by making them feel part of a group (Haslam et al., 2025). Research has consistently found an association between identity leadership and team identification (e.g. Steffens et al., 2014; van Dick et al., 2018), with employees feeling more connected with their team when their managers display more identity leadership characteristics. There is also a link between effective leadership and shared decision-making power (e.g. Manz and Sims, 1987; Pearce and Conger, 2003). We argue that there is a similar association between identity leadership and participation in decision-making. Indeed, leadership training programmes based on the social identity approach to leadership describe that a core component of identity leadership is to involve employees in decision-making processes (Haslam et al., 2003, 2017).

Team identification is associated with several positive outcomes in the workplace. This includes lower levels of turnover intention (O’Reilly et al., 1989) and higher levels of social support (Steffens et al., 2017), team performance (Lembke and Wilson, 1998) and commitment (Tsui et al., 1992). Several leadership models suggest the importance of sharing the decision-making power (for distributed leadership, see Gronn, 2000, for shared leadership, see Pearce and Conger, 2003). Benefits from empirical studies include decreased levels of emotional distress (El Zein et al., 2019) and increased levels of affective commitment (i.e. emotional connection to the organisation; Knezović and Smajić, 2022).

Despite the potential importance of team identification and participation in decision-making on the role of commitment, existing studies on identity leadership have primarily focused on outcomes such as job satisfaction, burnout or organisational citizenship behaviour (e.g. Steffens et al., 2014; van Dick et al., 2021). Addressing this conceptual gap is important because by promoting team identification and shared decision-making, identity leadership may offer distinct pathways to commitment compared with other leadership models. The paper aims to investigate these associations by exploring the pathways from identity leadership to affective commitment via both team identification and participation in decision-making.

Furthermore, cross-cultural research on leadership has yielded mixed findings, with some studies identifying cross-cultural differences in leadership effectiveness (e.g. House et al., 2004; Gelfand et al., 2017), whilst others identified similarities in perceptions of leadership and similar associations between identity leadership and organisational outcomes (e.g. Bracht et al., 2023; Monzani et al., 2024). The present study aims to explore whether our conceptual model is generalisable across individualistic and collectivistic cultures. This study extends our knowledge of identity leadership and contributes to our understanding of how cultural values (i.e. collectivistic and individualistic) shape the pathways from identity leadership to affective commitment.

Inspired by the social identity theory (Tajfel and Turner, 1979) and the self-categorisation theory (Turner et al., 1987), identity leadership conceptualises leadership as a process derived from group membership rather than focusing only on the leader or on the dyadic relationships between the leader and their group members (Hogg, 2001). Haslam et al. (2025) developed the identity leadership model around four dimensions. Haslam and Reicher (2016) define identity leadership as “a group process that centers on a psychological bond between leaders and followers grounded in an internalized sense of their common group membership; that is, a sense of shared social identity or ‘we-ness’” (p. 23). The first dimension, identity prototypicality, represents the idea of “being one of us”. It means being an example of the essential and distinctive characteristics of the group, for instance, having shared characteristics and values with the group members (Haslam et al., 2025). The second dimension, identity advancement, represents the idea of “doing it for us”. More specifically, it corresponds to leaders who put the interests of the group before their own interests. An identity advancement leader tries to enhance the group by promoting its benefits and acting as a servant for the group (Steffens et al., 2014). The third dimension, identity entrepreneurship, represents the idea of “crafting a sense of us”. This implies that leaders act in order to create a shared idea, a shared meaning of what it means to be an “us” and what the group stands for (Steffens et al., 2014). The last dimension, identity impresarioship, represents the idea of “making us matter”. This means that leaders take actions and create activities which give meaning to the group. This will give group members the feeling that their group matters and it will also communicate to members of the outgroup that the group is important (Haslam et al., 2025).

Creating a shared sense of identity is beneficial for leaders and their organisations as well as for their group members. Firstly, it is expected that identification with the collective directs people to behave according to the collective goals (Mowday et al., 1979). Through a shared sense of identity, leaders can shape and direct the development of the team (i.e. being an identity entrepreneur), something which increases in-role performance and reduces turnover intentions (van Dick and Kerschreiter, 2016). Secondly, the shared sense of identity increases job satisfaction (van Dick and Kerschreiter, 2016) and also changes employees’ perception of situations. Consequently, this protects them from the potential negative consequences of demanding situations (Haslam and van Dick, 2011). Moreover, Mathieu and Zajac (1990) found that by initiating structures (i.e. being an identity impresario) such as establishing patterns of organisation and communication, a leader can increase organisational commitment among their group members (i.e. an individual’s feelings about how strongly they are identified and involved in their organisation; Mowday et al., 1979).

Organisational commitment has been conceptualised in different ways, including Meyer and Allen (1991) three-component model, which is the most widely used conceptualisation of organisational commitment. This model is comprised of three components: the emotional attachment to the organisation (i.e. affective commitment), the need for attachment to the organisation (i.e. continuance commitment) and the feeling of obligation towards the organisation (i.e. normative commitment). However, Meyer and Allen (1991) three-component model of commitment has been subject to criticism (e.g. Ko et al., 1997; Solinger et al., 2008), and there is debate in the literature about the conceptualisation of organisational commitment (e.g. Bentein et al., 2005; Cohen, 2007), including a particular emphasis on the importance of affective commitment (e.g. Bergman, 2006; Mercurio, 2015; Morrow, 2011). Accordingly, the present study focuses on affective commitment (i.e. “employee’s emotional attachment to, identification with, and involvement in the organization”; Meyer and Allen, 1991, p. 67) as the most relevant outcome of identity-based leadership processes. Although no study has investigated the association between identity leadership and affective commitment, other conceptualisations of leadership have been found to be positively associated with affective commitment (for instance, see Haque et al., 2021 for responsible leadership, see Lee and Ahmad, 2009 for leadership styles, see Ly, 2024 for inclusive leadership, see Ribeiro et al., 2018 for transformational leadership). Thus, we suggest that identity leadership is positively associated with affective commitment (Hypothesis 1).

Team identification can be defined as “the process by which individual team members perceive themselves in terms of the values, goals, attitudes, and behaviors they share with other team members” (Janssen and Huang, 2008, pp. 70–71). It is expected that the behaviours of the members of a team will reflect the objectives and the values of their team (van Knippenberg, 2000). The three components (cognitive, evaluative, and emotional) of social identity help to describe how a team creates a “sense of oneness” (Ellemers et al., 1999). When members of a group have similar thoughts, feelings and behaviours, they are more successful in creating common attitudes (Haslam and van Dick, 2011). It is suggested that a leader’s behaviour can create a shared self-concept between them and their group members, which reinforces their team identification in their workgroups (Huettermann et al., 2014). Additionally, several recent studies have found a positive association between identity leadership and team identification (e.g. Bracht et al., 2023; van Dick et al., 2021). Therefore, we hypothesise that identity leadership is positively associated with team identification (Hypothesis 2).

Furthermore, research studies have found that the salience of a team identity is important for individuals: they become more willing to identify themselves with the team and conduct actions which are on behalf of the team (Brickson, 2000). Highly identified employees help and cooperate with other team members, engage in organisational citizenship behaviours (Janssen and Huang, 2008) and increase social support (Steffens et al., 2017). Having highly identified employees is also beneficial for organisations. It decreases turnover intentions and absenteeism (O’Reilly et al., 1989) and increases altruism (Chattopadhyay, 1999) and team performance (Lembke and Wilson, 1998). Moreover, team identification also increases commitment to the organisation (Tsui et al., 1992). Several studies have shown that when individuals are strongly identified with their team members, they tend to act in line with their group membership status and this leads to more commitment to their team (Doosje et al., 1995; Ellemers et al., 1997). Research has found that team identity has a positive association with affective commitment (e.g. Bruggeman, 2006; Gautam et al., 2004, for a meta-analysis, see Riketta, 2005). Thus, we hypothesise that team identification is positively associated with employees’ levels of affective commitment toward their organisation (Hypothesis 3). Based on Hypotheses 2 and 3, we suggest that team identification acts as a mediator between identity leadership and affective commitment.

When creating a shared social identity, a primary objective of identity leaders is to integrate the individual social identities, including their goals and values. This may be difficult for leaders and managers on their own, as they often fail to correctly perceive both an individual’s self-defined group memberships and actual relationships between group members (e.g. Brewer, 1995; Peteraf and Shanley, 1997). To develop a shared identity, it is therefore important to encourage group members to actively participate in group processes and decision-making, in order to identify goals and values that are strongly related to an individual’s social identity (e.g. Lewin, 1956; van Knippenberg and van Schie, 2000). Leadership development programmes based on identity leadership, such as the ASPIRe model or the 5R leadership development programme, place employee participation in the centre of the identity development process (Haslam et al., 2003, 2017, 2023; Peters et al., 2013). Furthermore, a leader who is “one of us” is more likely to reduce the psychological distance between the leader and the group members (Haslam et al., 2025) and thus create an environment in which employees are inclined to participate in decision-making processes (i.e. “sharing reaching a conclusion with others-subordinates, peers, or supervisors”, Knoop, 1995, p. 379). In fact, sharing an identity with the leader has been shown to increase willingness to participate in activities that lead the group forward, and this seems to be a mediator between shared identity and performance (Slater et al., 2019). In line with research on empowering leadership (e.g. Amundsen and Martinsen, 2014; Manz and Sims, 1987), shared leadership (e.g. Nicolaides et al., 2014; Pearce and Conger, 2003), and participative leadership (e.g. Somech, 2006; Wang et al., 2022), we hypothesise that identity leadership is positively associated with employees’ levels of participation in decision-making processes (Hypothesis 4).

The development of a shared social identity without actively including group members in collective decisions may lead to disengagement instead of commitment (e.g. Jetten et al., 2002; Wanous et al., 2000; Terry et al., 2001), whilst encouraging participation leads to a sense of ownership and procedural justice and thereby increases commitment (e.g. Haslam et al., 2002; Wegge and Haslam, 2003). For example, previous studies have found participation in decision-making to be positively associated with employees’ levels of affective commitment (e.g. Knezović and Smajić, 2022). This positive association was found in various settings including health care settings (Liu et al., 2015), government agencies (Scott-Ladd et al., 2006), the public sector (Kumar and Saha, 2017) and in a situation of strategic change (Lines, 2004). Therefore, we hypothesise that participation in decision-making is positively associated with affective commitment (Hypothesis 5). Thus, based on Hypotheses 4 and 5, we suggest that participation in decision-making is a mediator between identity leadership and affective commitment.

Previous research has also established the existence of significant cultural differences between countries. Culture can be defined as “the collective programming of the mind that distinguishes the members of one group or category of people from another” (Hofstede, 2001, p. 9). Cultural values shape how individuals define the self, relate to groups, interpret authority and evaluate leadership behaviours, thereby influencing both leadership processes and their consequences (House et al., 2004; Gelfand et al., 2017). Hofstede (2001) noted that one aspect in which cultures differ is their individualistic or collectivistic orientation. In fact, several scholars have suggested that individualism and collectivism are the fundamental characteristics of culture (Hofstede et al., 2010; Triandis, 1995; Lam et al., 2002). Individualistic orientation refers to the degree to which an individual values independence and standing out from the group rather than belonging to a community. People from individualistic cultures tend to see themselves as more independent (Markus and Kitayama, 1991). Some of the most individualistic cultures include the United States, the United Kingdom and the Netherlands (Hofstede et al., 2010). Collectivistic orientation corresponds to the opposite: people prioritise group loyalty, commitment, conformity and fitting in with the group over standing out as an individual (Hofstede, 2001). People from collectivistic cultures tend to see themselves as more interdependent (Markus and Kitayama, 1991). Some of the most collectivistic cultures include Guatemala, China and Portugal (Hofstede et al., 2010). In individualist cultures, leadership behaviours that successfully cultivate a shared sense of “we” may have a stronger impact, as they actively transform loosely connected individuals into a psychologically meaningful collective. Accordingly, team identification may represent a more central mechanism through which identity leadership translates into positive organisational outcomes in individualistic cultures. In collectivistic cultures, affective commitment may already be high without the need for a leader to cultivate it, as employees may already feel a sense of shared identity and social embeddedness within the group. Leadership behaviours that reinforce collective identity may be beneficial for both individualistic and collectivist cultures, although their added value on organisational outcomes such as commitment might be lower where identification levels are already high due to cultural norms (Earley, 1993).

Cultural differences are also likely to shape the meaning and impact of participation in decision-making. Individualistic cultures place greater emphasis on personal agency, voice and self-determination (Snibbe and Markus, 2005). Employees in such contexts may expect to be involved in decisions affecting their work and may interpret participation as a signal of respect, empowerment and inclusion. As a result, participation in decision-making is likely to be more strongly associated with positive work attitudes, including affective commitment. By contrast, collectivistic cultures are often characterised by higher acceptance of hierarchical authority and greater emphasis on role clarity and social harmony (Hofstede et al., 2010). In such contexts, employees may view decision-making as the leader’s responsibility, or they may prioritise group harmony over individual “voice” (Sagie and Aycan, 2003). Consequently, participation in decision-making may play a less prominent role in shaping affective commitment, even when leaders encourage involvement.

Findings from cross-cultural studies in organisational contexts have found mixed results for the role of culture (for a review, see Tsui et al., 2007). On the one hand, differences in collectivistic orientation affect ingroup behaviours and expectations (Cox et al., 1991) and there is evidence of cross-cultural variation in leadership effectiveness (e.g. House et al., 2004; for reviews, see Gelfand et al., 2017; Hanges et al., 2016). For instance, in a meta-analysis, Agag et al. (2025) found that collectivism moderates the association between transformational leadership and organisational outcomes. These findings suggest that certain behaviours might be more desirable and effective in one culture compared to another. On the other hand, there are numerous cross-cultural studies conducted in organisational settings that found minimal or no cross-cultural differences (e.g. Agarwal et al., 1999; Sosik and Jung, 2002; Spector et al., 2001). For instance, Walumbwa et al. (2005) found that the association between transformational leadership and organisational commitment was moderate in both individualistic and collectivistic cultures (e.g. r = 0.41 in the United States and r = 0.42 in China). Furthermore, cross-cultural studies found similar associations between identity leadership and organisational outcomes across different cultures (e.g. Bracht et al., 2023; Monzani et al., 2024; van Dick et al., 2021). Comparing the associations of different conceptualisations of leadership, Bracht et al. (2023) found that collectivism was a significant moderator of the association between social identification and innovative behaviours in models including transformational leadership and authentic leadership. However, collectivism was not a significant moderator of this association when their model used identity leadership or leader-member exchange quality instead. These findings suggest that different conceptualisations of leadership may lead to different pathways to organisational outcomes. Also, it highlights that the role of culture and, more specifically, the levels of individualism/collectivism, is unclear and needs to be investigated further. As findings from previous cross-cultural studies are mixed, and the model presented in this paper has not been investigated before, it is unclear whether we could expect cross-cultural differences. Thus, we have no a priori hypothesis regarding the cross-cultural aspect of our data, but instead explore it as a research question to be resolved empirically (exploratory research question).

Fundamentally grounded in social identity theory (Tajfel and Turner, 1979) and the self-categorisation theory (Turner et al., 1987), identity leadership (Haslam et al., 2025) provides a strong theoretical framework for the conceptual model of this study. There is still a lack of research regarding the pathways through which a leader adopting the identity leadership approach achieves positive organisational outcomes among employees. This paper aims to help fill this gap through investigating the association between identity leadership and affective commitment, and two potential pathways that may connect these concepts.

The first hypothesised mechanism, via team identification, reflects employees’ emotional and cognitive attachment towards their team. Managers with high levels of identity leadership facilitate the creation of a shared self-concept that leads to employees feeling more committed to their organisations. The second hypothesised mechanism, via participation in decision-making, reflects employees’ feelings of autonomy and agency. Managers with high levels of identity leadership encourage group members to actively define the group's goals and processes. In doing so, identity leaders foster a sense of ownership and reinforce employees’ feelings of involvement and commitment. These two pathways, based on shared emotion and shared agency, represent the conceptual model of this study. Additionally, this research project aims to test whether the hypothesised model is applicable to both collectivistic and individualistic cultures.

In this cross-cultural cross-sectional study, we hypothesise that identity leadership is positively associated with affective commitment (H1) and that this association is mediated by both team identification and participation in decision-making. More specifically, we hypothesise that higher levels of identity leadership are associated with higher levels of team identification (H2) which, in turn, are associated with higher levels of affective commitment (H3). Furthermore, we hypothesise that higher levels of identity leadership are associated with higher levels of participation in decision-making (H4) which, in turn, are associated with higher levels of affective commitment (H5; the full model is presented in Figure 1). Finally, we explore whether the model is generalisable to both collectivistic and individualistic cultures (exploratory research question).

We specified three inclusion criteria: (1) participants had to work as an employee, (2) they had to work in a team of at least three people and (3) their team needed to have a supervisor. Our sample was composed of 741 participants. Participants who had more than 10% of missing values or who did not seem to be fully engaged with the study based on their completion time were excluded (n = 43) (Meade and Craig, 2012). They either answered too quickly (in 3 min or less) or not quickly enough (in 24 h or more). The final dataset was composed of 698 participants. Participants were aged from 18 to 65 years old (M = 31.6, SD = 9.6) with a slightly greater percentage of women (n = 363; 52.0%). Regarding their nationality, participants were Portuguese (22.8%), German (12.7%), British (12.6%), Turkish (9.8%), Bulgarian (8.3%), French (7.7%), Dutch (6.9%), Spanish (6.7%), Mexican (5.0%) and the rest from additional countries (7.5%). The most represented educational level was an undergraduate degree (39.5%), followed by a postgraduate degree (29.2%), then a high school diploma or equivalent (22.5%). Most of the sample was in full-time employment (77.1%) and with a permanent contract (65.3%). The average team was composed of 12.7 people (SD = 16.5) and the mean amount of time participants had spent within their team was 4.7 years (SD = 5.7).

Country clusters were created based on Hofstede’s (2001) classification, where countries were assigned an individualism (IDV) score on a continuum from 1 (collectivism) to 100 (individualism). Countries with an IDV score equal to 60 or above were classified as individualistic cultures (e.g. the United Kingdom [IDV = 89]), whilst those with a score equal to or below 40 were classified as collectivistic cultures (e.g. Portugal [IDV = 27]). Countries closer to the mid-point (i.e. with an IDV score between 41 and 59, such as Spain [IDV = 51]) were classified as neither individualistic nor collectivistic and were kept for the analyses based on the full sample but excluded from the country cluster analyses. Based on the 10 observations per parameter recommendation (Bentler and Chou, 1987; Bollen, 1989) and our hypothesised model composed of 31 parameters to estimate, we aimed for a sample of 310 participants per country cluster to reach sufficient power. This was achieved for both the collectivistic (n = 326) and individualistic clusters (n = 315).

Participants were asked to evaluate the level of identity leadership of their team’s current supervisor using the Identity Leadership Inventory (IL; Steffens et al., 2014). The scale consists of 15 items, reflecting the four dimensions of identity leadership: identity prototypicality (e.g. “This leader embodies what the group stands for”), identity advancement (e.g. “This leader stands up for the group”), identity entrepreneurship (e.g. “This leader devises activities that bring the group together”), and identity impresarioship (e.g. “This leader makes people feel as if they are part of the same group”). Participants answered on a 7-point Likert scale (1 = not at all; 7 = completely). The scale showed excellent internal consistency reliability (McDonald’s omegas for each country range between 0.93 and 0.98).

We assessed affective commitment (AC) using the affective commitment items of the organisational commitment scale by Meyer et al. (1993). The affective commitment subscale, which reflects the emotional attachment to the organisation, is composed of 6 items (e.g. “I would be very happy to spend the rest of my career with this organisation”). Participants answered on a 7-point Likert scale (1 = strongly disagree; 7 = strongly agree). The scale showed a good internal consistency reliability (McDonald’s omegas for each country range between 0.84 and 0.92).

We used the team identification scale to assess how much participants identified with their team (TI; Doosje et al., 1995). This scale has 4 items that were rated on a 7-point Likert scale (1 = not at all; 7 = extremely). Due to the nature of this study, we adapted the items to be relative to the organisational team (e.g. “I identify with other members of my team”). The scale showed a very good internal consistency reliability (McDonald’s omegas range for each country between 0.86 and 0.95).

We measured how much the participants felt that their supervisor allows the members of their team to take part in decision-making processes for different aspects of their work by using the participation in decision-making scale (PDM; Witt et al., 2000). The six items refer to six different types of decisions in an organisational context (e.g. “Your team performance appraisal review”). Participants answered on a 5-point Likert scale (1 = we don’t discuss things very much as his/her decisions are usually adopted; 2 = we don’t discuss things very much and the group make most of the decisions; 3 = we discuss things in a great deal and his/her decision is usually adopted; 4 = we discuss things in a great deal and the group decision is usually adopted; 5 = we discuss things in a great detail and come to a decision based on consensus regarding the issue). The scale showed a good internal consistency reliability (McDonald’s omegas for each country range between 0.78 and 0.90).

Participants were asked to provide their gender, age, nationality, level of education, work field, type of employment (part-time or full-time), type of contract (temporary or permanent), country in which they were working, team size and number of years they spent working in this team.

The scales that were not available in the target languages were translated following the back-translation technique (Brislin, 1970) ^[1]. The ILI had already been validated in Dutch, English, French, German, Spanish and Turkish (Steffens et al., 2014; van Dick et al., 2018) and the affective commitment had been validated in Dutch, English, French, German, Portuguese and Turkish (Süß, 2007; Nascimento et al., 2008; de Gilder et al., 1997; Vandenberghe et al., 2001; Wasti, 2002).

We conducted an online cross-sectional study. Participants were recruited over the internet via social media (e.g. LinkedIn, Facebook) using the snowball technique, and via Prolific – an online platform where participants get paid to participate in online studies. After providing their informed consent, eligibility for the study was checked and if they fulfilled the inclusion criteria, they were asked to answer the questionnaire. At the end of the questionnaire, we provided a debrief which informed participants about the purpose of the study. This study was approved by the ethics committee of the University of Reims Champagne-Ardenne.

To control for common method variance, we used different scale formats and different scale endpoints, as well as ensuring that the data were collected anonymously (Podsakoff et al., 2003). To reduce acquiescence bias, some of the items were reverse-coded (Podsakoff et al., 2003).

All analyses were done with the R software (R Core Team, 2025). We used structural equation modelling (Bollen, 1989; Kline, 2023) and the lavaan package (Rosseel, 2012) to analyse our data. Following the recommendation of Little et al. (2002), we used three parcels instead of items to estimate the latent variables. Therefore, before the actual modelling, we built parcels using the “single-factor” method (Landis et al., 2000; Matsunaga, 2008). However, for TI we used the items as parameters instead of parcels as the scale is composed of only four items. The estimations were done with the bias-corrected and accelerated bootstrap (BCa; Fritz and MacKinnon, 2007; MacKinnon et al., 2004). All reported coefficients are standardised.

In addition, following the recommendations of Kline (2023) and Schermelleh-Engel et al. (2003), we used several fit indices to interpret the model fit in general. First, we used Chi-square (χ²) and its associated p-value. We also included different classes of goodness-of-fit criteria, which were the Comparative Fit Index (CFI; Bentler, 1990), the Tucker–Lewis Index (TLI; Tucker and Lewis, 1973), the Root Mean Square Error of Approximation (RMSEA; Steiger, 1990) with a 90% confidence interval and the Standardised Root Mean Square Residual (SRMR; Jöreskog and Sörbom, 1989).

The mean values and standard deviations, as well as the correlations between the scales, for the whole sample and for both country clusters, are depicted in Table 1. Since the absolute values of skewness and kurtosis were slightly higher than 1, indicating univariate non-normality for some variables (Kline, 2023) and the Mardia’s multivariate test (Mardia, 1970) indicated multivariate non-normality, we used the Satorra–Bentler correction in all estimations and model test statistics (Satorra and Bentler, 1988). The McDonald’s omega showed good levels of internal consistency reliability (≥0.70). Factor loadings showed good levels of construct validity (≥0.70). Inter-scale correlation showed good levels of concurrent validity (most correlations were medium [≥0.30] to large [≥0.50]; Cohen, 1992).

We used a stepwise approach to test our hypotheses. First, we started with a null model which we then expanded to the full model depicted in Figure 1. The null model, Model 1, contained only identity leadership and affective commitment. In Model 2, we added team identification as a mediator of the association between identity leadership and affective commitment. In Model 3, we added participation in decision-making as a second mediator of the association between identity leadership and affective commitment, since we were interested in the effect participation in decision-making could have in addition to team identification. The results of the analyses, including the model fit statistics, as well as the standardised path coefficients, the standard errors and the p-values, are displayed in Table 2. The standardised path coefficients are also reported in Figure 1.

Model 1 showed good model fit statistics and identity leadership was significantly associated with affective commitment (β = 0.35, p < 0.001). Similar to Model 1, Model 2 showed good model fit. In Model 2, there was a significant direct effect of identity leadership on affective commitment (β = 0.17, p = 0.001). The results also showed a significant effect of identity leadership on team identification (β = 0.58, p < 0.001) and a significant effect of team identification on affective commitment (β = 0.32, p < 0.001). Furthermore, the indirect effect of identity leadership on affective commitment through team identification was significant, β = 0.17, BCa CI [0.10, 0.23], p < 0.001. The total effect on affective commitment was 0.33 and therefore the ratio of indirect effect to direct effect was 0.53 (Wen and Fan, 2015). Therefore, team identification was a partial mediator of the association between identity leadership and affective commitment.

The fit indices of Model 3 were also good. In Model 3, there were significant direct effects of identity leadership on affective commitment (β = 0.13, p = 0.025), team identification (β = 0.58, p < 0.001) and participation in decision-making (β = 0.55, p < 0.001). The effect of team identification on affective commitment remained significant (β = 0.31, p < 0.001), suggesting that team identification partially mediated the association between identity leadership and affective commitment, β = 0.18, BCa CI [0.11, 0.24], p < 0.001. The ratio of indirect effect to the total effect was 0.52. However, the effect of participation in decision-making on affective commitment was not significant (β = 0.07, p = 0.153). In addition, the indirect effect of identity leadership on affective commitment through participation in decision-making was non-significant, β = 0.04, BCa CI [−0.01, 0.09], p = 0.155. The ratio of indirect effect to total effect was 0.11.

To investigate if there were any differences between country clusters, we first looked at the correlations between the variables in each cluster. The results are shown in Table 1. In both country clusters, there were medium correlations between participation in decision-making and team identification and large correlations between identity leadership and participation in decision-making and between identity leadership and team identification. However, the correlations between affective commitment and the three other variables were different in the two clusters of countries. In the collectivistic cluster, the correlations between identity leadership, participation in decision-making and team identification, on the one hand, and, affective commitment, on the other hand, were all small. However, in the individualistic cluster, there was a medium correlation between participation in decision-making and affective commitment and large correlations between affective commitment and both identity leadership and team identification. This pattern became even more salient when we fitted our final model, Model 3, to each country cluster separately. The path estimates, as well as indirect effects, are displayed in Table 3. The standardised path coefficients of the collectivistic and individualistic clusters are also reported in Figure 1.

In Model 3, we found a strong effect of identity leadership on team identification and on participation in decision-making, for both country clusters. However, the direct effect of team identification on affective commitment (collectivistic cluster: β = 0.22, p = 0.002; individualistic cluster: β = 0.36, p < 0.001) and the indirect effect of identity leadership on affective commitment through team identification (collectivistic cluster: β = 0.13, p = 0.003; individualistic cluster: β = 0.20, p < 0.001) were descriptively stronger in the individualistic cluster. Similar to the results of Model 3 when using the full sample, the direct effects of identity leadership and participation in decision-making on affective commitment in the collectivistic cluster were not significant (β_{IL → AC =} 0.03, p = 0.738, β_{PDM → AC =} 0.03, p = 0.691). However, these two effects were significant in the individualistic cluster (β_{IL → AC =} 0.28, p = 0.001, β_PDM → AC = 0.18, p = 0.013) and there was a small significant indirect effect of identity leadership on affective commitment through participation in decision-making (β = 0.10, p = 0.015). Therefore, on the one hand, we found a full mediation of identity leadership through team identification and no effect of participation in decision-making on affective commitment in the collectivistic cluster, and on the other hand, we found a direct effect of identity leadership on affective commitment and partial mediations of both team identification and participation in decision-making on the association between identity leadership and affective commitment in the individualistic cluster.

The first aim of this study was to investigate the association between identity leadership and affective commitment among employees whilst considering the mediating role of team identification and participation in decision-making. Overall, the findings provide empirical support for the social identity approach to leadership (Haslam et al., 2025). We first hypothesised that identity leadership would be positively associated with affective commitment (H1). The results support this hypothesis, as there was a medium positive association between identity leadership and affective commitment. Drawing on the social identity approach to leadership (Haslam et al., 2025), this finding confirms that leaders who represent and advance a shared “we” are able to foster stronger affective commitment among their team members. This is in line with previous studies using other conceptualisations of leadership, which found positive associations with affective commitment (Lee and Ahmad, 2009; Ly, 2024; Haque et al., 2021; Ribeiro et al., 2018). It also contributes to the growing body of literature which suggests that identity leaders are beneficial for both individuals and organisations as they improve levels of job satisfaction, perceived team support, task cohesion and organisational citizenship behaviours (e.g. Steffens et al., 2014; van Dick et al., 2018). Increasing employees’ affective commitment could be a way for organisations to increase employees’ performance (e.g. Ribeiro et al., 2018; for meta-analyses, see Meyer et al., 2002; Riketta, 2002), as such managers following identity leadership principles would be valuable for organisations.

Turning to Hypotheses 2 and 3, we postulated that team identification would mediate the association between identity leadership and affective commitment. More precisely, we hypothesised that higher levels of identity leadership would be associated with higher levels of team identification (H2) which, in turn, would be associated with higher levels of affective commitment (H3). These hypotheses were supported by the results. Our findings suggest that team identification mediates the association between identity leadership and affective commitment. Thus, the more a leader displays the characteristics of an identity leader, the more their team members feel identified with the group and committed to their organisation. These findings are consistent with previous studies which found that identity leadership increases team identification (e.g. Bracht et al., 2023; van Dick et al., 2021) and that team identity is positively associated with affective commitment (e.g. Riketta, 2005).

Regarding Hypotheses 4 and 5, we postulated that participation in decision-making would mediate the association between identity leadership and affective commitment. More precisely, we hypothesised that higher levels of identity leadership would be associated with higher levels of participation in decision-making (H4) which, in turn, would be associated with higher levels of affective commitment (H5). As expected, we found identity leadership to be associated with participation in decision-making. However, participation in decision-making was not associated with affective commitment. Our results did not support our hypothesis that participation in decision-making would mediate the association between identity leadership and affective commitment. As such, the more a leader displays characteristics of an identity leader, the more their team members perceive themselves as participating in decision-making processes. However, this did not translate into greater commitment towards their organisation when the overall sample was included in the analyses. The positive association between identity leadership and participation in decision-making is consistent with programmes designed to establish identity leadership in organisations (Haslam et al., 2003, 2017). For instance, the ASPIRe model, “places an emphasis on employees' involvement in organizational decision-making” (Haslam et al., 2003) and the 5R leadership development programme “is designed to include and mobilize followers” (Haslam et al., 2017). The non-significant path between participation in decision-making and affective commitment was unexpected as the literature suggests that affective commitment increases in parallel with the degree of participation of employees in job-related decisions (e.g. Knezović and Smajić, 2022; Lines, 2004). This association might be affected by other factors, such as culture, which was investigated in our exploratory research question.

As described in the introduction, the second aim of this paper was exploratory and investigated whether this model would be applicable to both collectivistic and individualistic cultures (Hofstede et al., 2010). We recruited participants from 18 different countries. The countries were split in two country clusters based on their level of individualism/collectivism: the collectivistic cluster (Bulgaria, Mexico, Portugal, Turkey and some additional participants from several other collectivistic countries) and the individualistic cluster (France, Germany, the Netherlands, the United Kingdom and a few participants from other individualistic countries).

On the one hand, in the collectivistic cluster, the association between identity leadership and affective commitment was fully mediated by team identification. On the other hand, our results showed that, in the individualistic cluster, the association between identity leadership and affective commitment was partially mediated by both team identification and participation in decision-making.

One possible explanation for these findings is that people from collectivistic countries feel a stronger level of connection to others (e.g. Markus and Kitayama, 2010). As such, they may feel a higher level of belonging to their organisation and feel more committed to it. Moreover, findings from a meta-analysis found that in-group collectivism was associated with higher levels of affective commitment (Fischer and Mansell, 2009). Thus, it is possible that, for people from collectivistic countries, having an identity leader has less impact on their level of commitment because there is less room for improvement in commitment (i.e. ceiling effect). This explanation may be supported by the fact that people in the collectivistic cluster (M = 4.4; SD = 1.6) had higher levels of affective commitment than people in the individualistic cluster (M = 4.0; SD = 1.5). In addition, the countries included in our collectivistic cluster had lower levels of gross national income (GNI) per capita than the countries included in the individualistic cluster. Fischer and Mansell (2009) found that GNI was negatively associated with affective commitment – meaning that lower-income countries had higher levels of affective commitment than higher-income countries. Thus, economic factors may have contributed to this ceiling effect.

Moreover, people from individualistic cultures tend to prioritise autonomy, independence and standing out from the group, whilst people from collectivistic cultures emphasise group harmony, interdependence and being part of the group (Hofstede, 2001). Therefore, it may be more important for people from individualistic cultures to have a “voice” in decisions affecting their work and to take part in the decision-making processes. In individualistic cultures, people may prefer active, assertive and more confrontational conflict resolution strategies (Ohbuchi et al., 1999). They may value having a role in making decisions (Kitayama et al., 2004; Snibbe and Markus, 2005) and also have more confidence in their own decisions (Mann et al., 1998). In contrast, collectivistic cultures emphasise social harmony and “saving face”. The act of participating in decisions, which may require challenging assumptions, voicing dissent or taking responsibility for a failed decision, can threaten this harmony. Employees may participate nominally but hold back their true thoughts and input to avoid conflict or making the leader lose face (Triandis, 1995). As such, employees may be less inclined to seek influence over decisions or may perceive extensive participation as unnecessary or misaligned with normative leader–follower roles. Additionally, participation in decision-making may be more common in individualistic cultures (Sagie and Aycan, 2003). Thus, in those cultures, people may have greater expectations regarding their level of autonomy and their role in making decisions that affect their jobs. As such, when these expectations are not met, they may feel less committed to their organisation.

Another possible explanation is that the differences found between collectivistic and individualistic cultures may come from differences in terms of power distance among these cultures. Power distance is one of the Hofstede dimensions measuring the “extent to which the less powerful members of institutions and organisations within a country expect and accept that power is distributed unequally” (Hofstede et al., 2010, p. 61). Most of the countries which comprised the collectivistic cluster also had higher levels of power distance (e.g. Bulgaria: 70; Mexico: 81; Portugal: 63) than the countries which comprised the individualistic cluster (e.g. Germany: 35; the Netherlands: 38; the United Kingdom: 35). In countries with low levels of power distance, employees may expect to be consulted about decisions that affect their work, feel more confident in helping to make these decisions, and feel more empowered to contradict their manager (Hofstede et al., 2010). In countries with high levels of power distance, the employees seem to accept their position as subordinates and are less willing to participate and to take part in decision-making processes (Khatri, 2009). In these countries, employees may expect to be told what to do, consider the manager-employee relationship as unequal, have less involvement in the decision-making process, and be less likely to disagree with their manager (Hofstede et al., 2010). When a leader delegates decision-making authority, it can be interpreted as a sign of the leader’s incompetence. The leader could be perceived as avoiding responsibility or lacking the knowledge to make the decision (Sagie and Aycan, 2003). Employees may also be reluctant to challenge the leader’s authority or direction, making their participation non-genuine or superficial. Hence, strategies which aim to increase employees’ participation might not have an impact on their commitment towards their organisation as it is not something that they are looking for. Whilst in countries with a low power distance, the employees are more willing to strive to equalise the distribution of power. Thus, they might be more willing to take on more responsibilities and more power in decision-making processes. Therefore, actively encouraging participation might increase their level of commitment towards their organisation as it is something which is in accordance with their aspirations.

Importantly, our findings suggest that participation in decision-making is not inherently more effective in collectivistic cultures simply because they are “collective”. Instead, its effectiveness depends on whether participation aligns with culturally shared expectations about agency, authority, and the appropriate distribution of decision-making power. Thus, participation functions as a commitment-enhancing mechanism primarily in contexts where individual voice and autonomy are culturally valued. Our results highlight that identity leadership operates through culturally contingent pathways. In collectivistic cultures, affective commitment is primarily grounded in shared identity and belonging, whereas in individualistic cultures, commitment is shaped by a combination of shared identity and shared agency.

We acknowledge that this study has several limitations. First, this is a cross-sectional study, which prevents us from making causal inferences based on this data (Antonakis et al., 2010). Future studies should use longitudinal or experimental designs to investigate the directionality of the effects. However, while we cannot confirm the directionality of the associations in our study, we can cautiously assume that the hypothesised direction of associations between most of our variables is supported based on previous research. For example, behaviours and characteristics of leaders can lead to higher levels of commitment (e.g. Mathieu and Zajac, 1990), leaders creating a concept of a shared self for their group members can have a positive impact on team identification (Huettermann et al., 2014), identifying oneself with group members also leads to more commitment towards the organisation (e.g. Doosje et al., 1995), and high levels of participation in decision-making are known to have a positive impact on commitment (e.g. Kahnweiler and Thompson, 2000).

Even though online self-report surveys have been found to be a reliable methodology (e.g. Chang and Vowles, 2013), there might have been some undetected influences associated with this methodology. Thus, it would be valuable to investigate if these findings are supported by research using different methodologies. For instance, experimental vignettes could provide the high internal consistency of an experimental setting as well as the high external consistency of a survey design (Atzmüller and Steiner, 2010; Evans et al., 2015). Furthermore, interviews or field observations could also shed light on the association between identity leadership, participation in decision-making, team identification and commitment in specific settings.

Although the sample was composed of participants from 18 countries, we did not have enough countries with a large enough sample size to conduct multi-level analyses by nesting the participants within countries (Maas and Hox, 2005). Additionally, our participants were independent of each other, as they were coming from different teams and organisations. Therefore, it was not possible to investigate team-level differences in the observed associations. Future research could use field experiments (Eden, 2021) to investigate whether these associations differ depending on the industry, characteristics of specific organisations, or characteristics of specific teams.

This paper focuses on one outcome only – affective commitment to their organisation. Future studies should test whether outcomes such as job satisfaction or burnout would lead to similar findings, building on the non-significant effect of participation in decision-making on affective commitment in the collectivistic cluster. One could argue that even though participation in decision-making does not increase affective commitment in collectivistic countries, it could increase job satisfaction and decrease burnout (e.g. Pacheco and Webber, 2016; Pijpker et al., 2020).

Furthermore, although we recruited participants from 18 different countries, most of them were European. Consequently, this limits the generalisability to other cultures (e.g. African, Asian; Henrich et al., 2010).

This paper has several theoretical and practical implications. First, it offers new light on the association between identity leadership and employees’ levels of affective commitment, as well as the role of team identification and participation in decision-making. It expands our knowledge of identity leadership and suggests that it plays an important role in promoting affective commitment in organisations. Second, the cluster comparison results suggest differences between collectivistic and individualistic cultures. These results highlight that participation in decision-making is not a universal mediator of the association between identity leadership and affective commitment. This emphasises that mechanisms associated with employees’ positive outcomes in organisations may differ based on cultural characteristics and supports the importance of considering the influence of culture when investigating leadership (e.g. van Dick et al., 2018). Therefore, future cross-cultural research on leadership could investigate employees’ expectations about their supervisor, across different cultures. As mentioned previously, it is possible that levels of power distance influence the willingness of employees to take part in decision-making processes. Fourth, although this paper does not aim to validate any scales in new languages, the translation and adaptation of several measures (e.g. team identification) in several languages (e.g. French, German) can be useful for future research.

The paper also has several practical implications for managers. The findings of this paper are in line with the wider literature on identity leadership, including large-scale cross-cultural studies (e.g. Bracht et al., 2023; van Dick et al., 2021) and longitudinal studies (e.g. Frenzel et al., 2023; Steffens et al., 2018), which emphasises the positive effect of identity leadership in organisations. Affective commitment – which we found to be associated with organisations that have higher levels of identity leadership – is beneficial for organisations as it is associated with positive outcomes such as lower turnover intention, greater job performance and greater job satisfaction (for meta-analyses, see Meyer et al., 2002; Riketta, 2002). Our findings indicate that engaging in identity leadership behaviours may be a useful technique for managers and organisations (Haslam et al., 2025) to promote affective commitment. Identity leadership can be actively cultivated in organisations through targeted training programmes, leadership awareness initiatives, and everyday managerial practices. These leadership development initiatives can explicitly focus on the four core dimensions of identity leadership (Haslam et al., 2025), providing managers with concrete tools to build and sustain a shared sense of “us” within their teams.

First, identity prototypicality (“being one of us”) can be developed by encouraging leaders to demonstrate alignment with the values, norms, and experiences of their team members. Training programmes can help leaders reflect on how their behaviour, language and decision-making signal group membership – for example, by using inclusive language (e.g. “we” rather than “I”; Steffens and Haslam, 2013) or limiting hierarchical distinctions such as office size or appearance (i.e. clothing).

Second, identity advancement (“doing it for us”) can be strengthened by training leaders to prioritise collective interests over personal gain. Organisations can encourage leaders to visibly advocate for their team’s needs, protect team members in times of uncertainty, and communicate how decisions serve the collective rather than individual agendas. For instance, they could make recognition and rewards contingent on member contributions to team performance (Yukl and Gardner, 2020).

Third, identity entrepreneurship (“crafting a sense of us”) can be developed by equipping leaders with skills to actively construct and maintain a shared understanding of what the group stands for. Practical interventions may include facilitating team workshops to define shared goals and values or organising team meetings in distinctive ways, such as a standing team meeting (van Dick and West, 2013).

Finally, identity impresarioship (“making us matter”) can be fostered by training leaders to create structures, rituals and activities that give tangible expression to the group’s identity. This may involve organising regular team meetings, celebrating collective achievements, designing meaningful onboarding practices or creating symbols that reinforce team distinctiveness. This could also be achieved by articulating an attractive vision for the team (Yukl and Gardner, 2020).

Organisations may implement these principles through established identity-based leadership development programmes, such as the ASPIRe model or the 5R leadership development programme (Haslam et al., 2003, 2017), which combine reflection, collective sense-making and action planning. Importantly, such programmes emphasise awareness of group processes and encourage leaders to see leadership as a relational and collective phenomenon rather than an individual attribute.

Furthermore, in individualistic cultures, managers could actively share the decision-making power by providing autonomy, delegating responsibilities and seeking input from employees. In collectivistic cultures, they should focus on reinforcing group identity and shared purpose. In these cultures, managers seeking participation in decision-making should be particularly mindful of cultural norms. For instance, they could use collective participation that seeks to promote group decisions and aims to find consensus. This could allow managers to gather valuable inputs while keeping the harmony of the group.

Identity leadership programmes emphasise the importance of creating a shared identity and of sharing the decision-making power with employees (Haslam et al., 2003, 2017). The results of this paper support the idea that having a greater sense of shared identity is associated with greater levels of affective commitment across both collectivistic and individualistic cultures. However, sharing the decision-making power could have a greater impact on affective commitment in individualistic cultures than in collectivistic cultures.