Sustaining Organizations Through Harmonized Civic and Employee Identities: Implications for Employee Engagement and Voice Behavior

Abstract

In pursuit of sustainable management, organizations must not only balance economic, environmental, and social goals but also cultivate human-centered strategies that support long-term viability. Drawing on identity theory, this study introduces civic identity—an individual’s self-concept as a responsible member of society—into the workplace and examines how its relationship with employee identity (i.e., enhancement and conflict) influences sustainability mechanisms: work engagement, role expansion, and voice behavior. Two field studies were conducted using multi-source data from 339 employees at large conglomerates (study 1) and three-wave data from 121 employees at entrepreneurial firms (study 2). Multiple regression analyses revealed that identity enhancement positively predicted both work engagement and role expansion, which in turn strengthened voice behavior. In contrast, identity conflict showed relatively weaker negative effects, while supplementary analyses indicated that identity separation exerted more pronounced negative influences. Although identity conflict and separation did not significantly affect role expansion in study 2, the results across both studies were largely consistent. By incorporating a neglected nonwork identity that fosters sustainable employee behavior, this research expands the scope of organizational studies and sustainability science. Implications for integrating civic identity into sustainable management strategies are discussed.

1. Introduction

As sustainability has emerged as a guiding principle for organizations across sectors, there is growing recognition that human behavior and values play a central role in determining the success of sustainability efforts [1]. Traditionally, sustainable management has emphasized environmental protection, economic viability, and intergenerational equity [2]. However, achieving sustainability in complex organizations also requires building human systems that promote ethical, proactive, and adaptive behaviors in the workplace [3]. In this context, the identities employees hold—and the extent to which these identities are aligned or misaligned—are critical for understanding motivation and constructive behavior at work [4]. Yet, there remains a gap in understanding how deeply held identities shape behavioral outcomes that support sustainability goals.

In organizational research, employee identity—defined as identification with the organization—has long been a central topic [5,6]. More recently, however, scholars have highlighted the importance of multiple identities [7,8]. Most of these studies have focused on work-related identities, such as workgroup, organizational, or professional identities [9,10], while largely overlooking nonwork identities in the workplace. Even studies that address work–nonwork identity relationships [11,12,13] tend to emphasize personal or family identities (e.g., gender, morality, and parental roles). This approach has limitations, especially as the boundaries between work and nonwork domains become increasingly blurred. Employees are now influenced by diverse memberships and roles that transcend the traditional work–nonwork divide [14]. Consequently, both individuals and organizations must navigate the challenges of managing multiple identities simultaneously. It is therefore essential to examine how these dynamics unfold at work by expanding the scope to include a broader range of nonwork identities (see Figure 1).

To address this research gap, the present study drew on identity theory [15] and focused on employees’ civic identity—an important but underexplored nonwork identity in the organizational literature. Civic identity refers to how individuals define themselves as members of a community and take on civic roles, such as contributing to the welfare of society [16]. The importance of citizenship is increasingly emphasized globally and locally. International institutions like UNESCO have prioritized civic education [17], and individuals are engaging more actively with social issues through peer-to-peer digital platforms [18]. Furthermore, more organizations are aligning with societal goals and engaging in purpose-driven business activities [19,20]. These trends may encourage employees to incorporate civic values into their self-concepts. Despite its prominence in fields such as political science and sociology, civic identity remains largely overlooked in both sustainability and organizational research—although its inclusion could yield valuable insights into employee motivation and behavior.

In exploring the intersection of civic and employee identities, this study examined how employees experienced the relationship between their civic and employee identities, focusing on two key forms: identity enhancement and identity conflict. Identity enhancement refers to a perceived harmony or synergy between multiple identities, whereas identity conflict involves feeling torn between competing identity-based expectations [21,22]. Employees may experience enhancement when their work roles facilitate civic participation and conflict when their organizational roles hinder the expression of civic identity. Prior research suggests that identity enhancement and conflict can significantly influence employee attitudes and behaviors [23,24].

What, then, are the motivational and behavioral consequences of civic–employee identity relationships? Identity theory suggests that individuals with strong identities engage in two distinct role processes—“role playing” and “role making”—to validate and enact their identities [25,26]. In the workplace, employees not only perform their assigned roles (role playing) but also expand those roles by initiating related tasks (role making). Building on this framework, the present study proposes that civic–employee identity relationships affect two psychological processes: work engagement and role expansion. Work engagement—characterized by vigor, dedication, and absorption [27,28]—is viewed as role playing, reflecting the intensity of employees’ involvement in their assigned tasks. Role expansion refers to employees’ perceptions of broader work responsibilities, reflecting role making [29]. Furthermore, this study posited that both processes would promote employee voice behavior, a representative positive change-oriented behavior [30]. Voice behavior is particularly relevant to civic and employee identity configurations as it emphasizes contributing to and taking actions for a better organization.

In sum, this study examined how civic–employee identity enhancement and conflict influenced employees’ work motivation and voice behavior. To address this research question, two empirical studies were conducted in contrasting organizational contexts—large corporations (study 1) and entrepreneurial firms (study 2)—to capture variations in identity dynamics shaped by structural and resource differences. By incorporating a previously overlooked nonwork identity (i.e., civic identity), this research offers novel insights into organizational phenomena that traditional work identity frameworks cannot fully explain, thereby extending the literature on organizational behavior and sustainable management. In addition, the study contributes to the United Nations Sustainable Development Goals (SDGs), particularly Goals 8 (Decent Work and Economic Growth), 12 (Responsible Consumption and Production), and 16 (Peace, Justice and Strong Institutions) [31]. Through a cross-disciplinary approach that bridges organizational behavior and sustainability science, this research provides a broader understanding of how identity configurations relate to sustainable employee behavior.

2. Literature Review

2.1. Identity Theory and Civic–Employee Identity Relationship

As a representative theory on individual identity, identity theory [15,32] suggests that an individual or the self is a collection of identities resulting from multiple roles carried out in society (e.g., manager, chairman, parent). Individuals enact their role identities as a means of expressing themselves and gaining support and acceptance from members of their role set [32]. This process is called self-verification, and it suggests that people are motivated to verify or confirm their current self-views [33,34]. As a result, identities are valued by individuals and affect their thoughts, motivations, and behaviors in a given context. When an individual identifies with a role as an employee, it entails membership in the organization. In a similar vein, if a person identifies with a civic role, it brings the person an awareness of being a member in that society.

Since civic identity is a person’s self-concept as a citizen [16], if individuals have higher levels of civic identity, they will try to match their role expectations as citizens with their behaviors by contributing to social welfare and positive social change. Empirical studies have found that a strong sense of civic identity motivates people to assist their neighbors, volunteer, and pay taxes to provide for those who suffer from hardship [16,35]. Also, high levels of civic identity promote generalized reciprocity and interpersonal trust [36]. Although civic identity has not been adopted in organizational research, as Figure 1 shows, two identities—civic and employee—are more likely to be activated in the work domain simultaneously because the boundary between the two social domains—work and society—is blurred.

As mentioned earlier, recent research in multiple-identity research suggests that individuals psychologically experience their identities as either enhancing or conflicting [21,37,38,39]. Based on this perspective, civic–employee identity relationships can be conceptualized as two distinct types: identity enhancement and identity conflict. Furthermore, drawing on identity theory—which posits that once individuals identify with a certain social membership or role, they are motivated to enact their roles—this study examines two role-related psychological processes: work engagement as a form of role playing and role expansion as a form of role making. These two processes are proposed as key mechanisms through which identity relationships influence voice behavior. In sum, this study argues that identity relationships serve as sources of identity verification and that work engagement and role expansion mediate the link between identity experiences and proactive behavior in organizations.

2.2. Civic–Employee Identity Enhancement, Work Engagement, and Role Expansion

2.2.1. Identity Enhancement and Work Engagement

Prior research suggests that enacting multiple identities—rather than a single, isolated identity—provides individuals with greater meaning and energy, which can be invested in their work [40]. When employees perceive that the values associated with their civic and employee identities are complementary, this synergy generates positive affect and psychological resources [4,41]. Positive affect, in turn, broadens cognition, increases openness to diverse actions, and expands energy reserves [42]. Thus, employees experiencing identity enhancement between their civic and employee identities are likely to invest more cognitive and emotional resources in their roles, resulting in heightened work engagement.

In addition, identity enhancement promotes the fulfillment of authentic self-expression and self-actualization at work. Employees are intrinsically motivated to enact their “true selves” in their workplace roles [7,43]. When individuals can harmoniously express both their civic and employee identities, they perceive that the whole self is being acknowledged and expressed [34,44]. Furthermore, identity enhancement enables employees to draw upon skills, experiences, and resources from both identity domains, reinforcing a sense of wholeness and authenticity [45,46]. These processes collectively increase the perception that one’s full self is engaged in work, which fosters higher levels of work engagement [7,47].

2.2.2. Identity Enhancement and Role Expansion

When civic and employee identities are perceived as complementary, they are more easily co-activated at work [48]. Since civic identity is often expressed in the societal domain, enacting it within the organizational context implies crossing traditional role boundaries. When multiple identities are activated frequently and simultaneously, they form cognitive associations such that the activation of one primes the other [49]. This reinforcement process leads employees to view the organization and society as more integrated, allowing them to incorporate civic responsibilities into their workplace roles [14]. As a result, they redefine their work roles to include behaviors that benefit the organization and society, expanding the scope of what they perceive as their job responsibilities.

Cognitively, identity enhancement fosters a more holistic and flexible self-concept, enabling individuals to integrate diverse role expectations [12]. It promotes cognitive flexibility [46,50], facilitating the ability to take the perspectives of other roles fluidly. Additionally, the simultaneous validation of multiple identities creates psychological safety, which supports broader cognitive availability [51,52]. Although civic–employee identity relationships have rarely been empirically examined, recent research shows that identity synergy across team memberships enhances innovation [53]. Drawing on the preceding discussion, the following hypothesis is proposed:

Hypothesis 1.

Employees with high identity enhancement will show high (a) work engagement and (b) role expansion.

2.3. Civic–Employee Identity Conflict, Work Engagement, and Role Expansion

2.3.1. Identity Conflict and Work Engagement

When employees experience tension or incompatibility between their civic and employee role identities, they may face psychological insecurity and self-concept dissonance [52,54]. This internal struggle increases stress and negative affect [55], while draining psychological energy [40,54]. Individuals encountering identity conflict often engage in continuous self-regulation, oscillating between or suppressing competing identities [56,57]. Such regulation consumes significant cognitive and motivational resources [58,59], reducing the capacity for sustained focus and thereby diminishing work engagement.

Additionally, identity conflict limits opportunities for authentic self-expression and prevents the use of personal resources linked to multiple identities. When employees feel unable to enact one of their core identities—such as civic identity—they may experience a loss of authenticity [60], leading to feelings of alienation and emotional depletion [43,61]. This inauthenticity diminishes their ability to invest their whole selves at work. Furthermore, Carminati and Gao Héliot (2023) also proposed that in ethically charged situations, identity conflict, experienced by professionals as incompatible moral values, would negatively affect psychological well-being [11]. Ostermeier et al. (2023) also found that identity conflict is associated with both increased levels of emotional exhaustion and psychological distress and, consequently, increased turnover intentions [62].

2.3.2. Identity Conflict and Role Expansion

Employees who perceive conflict between civic and employee identities may experience anxiety and ambiguity over which role to prioritize [38]. Such conflict, characterized by contradictory cognitions [63], often leads to the suppression of one identity in favor of the other [37,64]. In most cases, employees may suppress their civic identity—a nonwork identity—in favor of their employee identity. This separation reinforces the perceived boundary between organizational and societal domains [48,65], discouraging employees from incorporating civic responsibilities into their work roles.

Beyond identity suppression, conflict also hampers cognitive flexibility and the ability to adopt expanded role definitions. Heightened anxiety from identity conflict restricts attentional allocation and adaptability across tasks [66], while increasing sensitivity to risk and pessimistic evaluations. This leads employees to narrow their behavioral repertoire and resist expanding their roles. Identity conflict can also provoke identity threat, which further narrows cognition and creativity [57,67]. Although empirical research on civic identity remains limited, Yang et al. (2021) found that identity conflict between work and social media identities was negatively related to work performance [68]. Similarly, civic–employee identity conflict may cause employees to perceive expectations for organizational participation as misaligned with—or even irrelevant to—their formal roles.

Hypothesis 2.

Employees with high identity conflict will show low (a) work engagement and (b) role expansion.

2.4. Work Engagement, Role Expansion, and Voice Behavior

A defining feature of voice behavior is its constructive and prosocial nature [69]. Voice is primarily motivated by a desire to improve the organization or help others within it [70]. Employees who are enthusiastic and dedicated to their work are more likely to engage in proactive efforts to maintain or enhance work conditions [71]. Given that voice behavior is both change-oriented and discretionary, it demands a high level of psychological motivation and personal resources. Work engagement—characterized by energy, dedication, and absorption—provides such resources, enabling employees to invest extra effort [72]. Although empirical research on this link remains limited, existing studies support the association. For example, Cheng et al. (2014) and Duan et al. (2022) found that work engagement positively predicts voice behavior [73,74].

Previous research suggests that while behaviors like organizational citizenship behavior (OCB) may extend beyond formal job roles, employees differ in whether they perceive such behaviors as in-role or extra-role [75,76]. When employees view these behaviors as in-role—due to broader or expanded role definitions—they are more likely to engage in them [75]. Employees with high levels of role expansion are more inclined to take ownership of problems and contribute beyond formal responsibilities, often because they internalize these behaviors as integral to their jobs [77]. This may also be influenced by normative pressures to meet implicit expectations or gain social rewards [78]. Research supports that in-role perceptions significantly increase the frequency of discretionary behaviors [75,79].

While much of the literature on role expansion focuses on OCB, some studies link it to proactive and change-oriented behaviors. For instance, Chiaburu (2008) posited that employees who view voice as part of their role are more likely to process and act on improvement-related issues [80]. Parker et al. (2006) found that a broader role definition predicts greater proactivity [77], and McAllister et al. (2007) showed that role expansion leads to increased taking-charge behaviors [81]. Based on the theoretical arguments and empirical evidence, this study posits that identity enhancement will be positively associated with voice behavior, mediated by work engagement and role expansion. When employees perceive alignment between their civic and employee identities, they are more engaged and more likely to adopt expansive role definitions, thus facilitating voice. Conversely, identity conflict is expected to reduce voice behavior by depleting work engagement and narrowing role perceptions.

Hypothesis 3.

Employees with high (a) work engagement and (b) role expansion will show high voice behavior.

Hypothesis 4.

Employees with high identity enhancement will show high voice behavior via (a) work engagement and (b) role expansion.

Hypothesis 5.

Employees with high identity conflict will show low voice behavior via (a) work engagement and (b) role expansion.

In sum, Figure 2 represents the theoretical model of the study.

To rigorously examine the proposed theoretical model, this study adopted a multi-study research design involving two independent field studies with samples from distinct organizational contexts—large corporations and entrepreneurial firms. The decision to conduct two studies was motivated by both theoretical and methodological considerations. From a theoretical standpoint, testing the model across different organizational contexts—namely, large corporations and entrepreneurial firms—allows for constructive replication and offers insight into how identity-based mechanisms operate under varying structural and cultural conditions. Organizational scale and structure can significantly shape identity dynamics, such as the salience of civic values, employee role engagement, and the perceived compatibility between personal and professional identities. From a methodological perspective, employing two distinct study designs enhances both the internal and external validity of the findings. The second study not only replicates the core model but also addresses potential limitations of the first study by introducing a longitudinal survey design to reduce common method bias. Together, the two studies provide a robust and contextually sensitive test of how civic–employee identity relationships influence sustainable work behavior.

3. Study 1: Method

3.1. Participants and Procedure

Study 1 was conducted among full-time employees and their immediate supervisors in large Korean conglomerates across diverse industries. Employees working at large conglomerates were chosen because they were more likely to identify with their organizations. Moreover, while making a profit is important for successful companies, showing concern for society is also required due to the high level of media attention they receive, which can lead to employees experiencing identity enhancement and conflict. Inclusion criteria required that participants had been employed in their current roles for at least six months to ensure sufficient organizational experience.

Executive directors and managers at large for-profit companies who were alumni from a university in Korea were contacted and asked for their participation. Eleven companies agreed to participate in the survey, and their directors and HR managers helped to distribute the questionnaires. To reduce common method variance, data were collected from two sources—supervisors and subordinates. Each supervisor was given a survey package that contained questionnaires for the leader and group members. Group members were asked to answer questions about identity relationships, work engagement, and role expansion. After one week, the leaders assessed to what extent each member engaged in voice behavior. All participants were offered a gift for their participation (approximately USD 4) and were promised that their responses would be kept confidential.

To ensure adequate statistical power and methodological rigor, the sample size for the empirical analyses was determined through a Monte Carlo simulation approach. This method allowed for precise estimation of the minimum required sample size based on the proposed model and anticipated effect sizes. Surveys were distributed to 470 employees (70 supervisors and 400 group members), which exceeded the minimum number of participants of 200 required for appropriate estimation of this research model [82,83]. Responses came from 70 group leaders and 364 group members, yielding a response rate of 92.3%. Outliers were checked [84], and omitting outliers resulted in 339 subordinate–supervisor matched data for all analyses. The dropped responses were not significantly different from the valid responses with regard to their demographic information. To account for the missing data, full information maximum likelihood (FIML) estimation was used. Participants worked in different industries, such as financial services, IT, and consumer product manufacturing. The average age of the participants was 36 (SD = 4.75), with an average tenure of 8.5 years (SD = 26), and the percentage of male participants was 55%.

3.2. Measures

Unless otherwise indicated, Likert-type scales were used ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). All English items were translated into Korean using back-translation procedures. All measured are provided in the online Appendix A.

Identity enhancement and conflict. Identity relationship variables were measured with two four-item scales adapted from existing measures [21,85]. The items were worded to reflect civic and employee identity. A sample item for identity enhancement is “I feel that the responsibilities associated with my employee role make it easy to fulfill my responsibilities as a citizen”. A sample item for identity conflict is “I often experience conflict between my identity as a citizen of society and an employee of the organization”.

Work engagement. The 9-Item Utrecht Work Engagement Scale (UWES-9) [86] was used. The scale measured three subdimensions of work engagement: vigor, dedication, and absorption. Sample items are “At work, I feel bursting with energy”, “I feel happy when I am working intensely”, and “I am immersed in my work”.

Role expansion. Following previous studies [78], employees rated the extent to which they viewed the same items that supervisors completed regarding voice behaviors as part of their in-role behaviors. For voice behavior, four items were used, and a sample item is “It is definitely part of my role at work to proactively voice out constructive suggestions that help the unit reach its goals”.

Voice behavior. A five-item voice behavior scale developed by Liang, Farh, and Farh (2012) was used [87]. Supervisors rated their employees’ voice behavior. Sample items include “This particular group member proactively develops ideas and makes suggestions about issues that may influence the unit”.

Control variables. To consider the possibility that employees’ characteristics might influence voice behavior, education level and tenure were included as control variables. For example, employees with higher levels of education may in general have more ideas to voice [88]. In addition, employees who have stayed longer in their organizations or have more experience might feel more comfortable speaking up at work [89].

4. Study 1: Results

4.1. Preliminary Analysis

Data were analyzed using SPSS 27.0 and Mplus 8.1 (the dataset is available at https://osf.io/rm6vg/) (accessed on 15 May 2025). Prior to testing the hypotheses, confirmatory factor analyses (CFAs) were performed to check the validity of the study variables. The hypothesized five-variable model (i.e., identity enhancement and conflict, work engagement, role expansion, and voice) demonstrated acceptable fit to the data (X² = 519.10, df = 286, CFI = 0.955, TLI = 0.949, SRMR = 0.049, RMSEA = 0.050) and a better fit than alternative models. Also, all items were well loaded to their latent variables (p < 0.001), and standardized factor loadings were above 0.61.

Although the data were collected from two sources, some variables were collected in a cross-sectional design and through self-report. Since this increased a concern about common-method variance, both procedural and statistical efforts were made to reduce the possibilities [90]. Questionnaires were kept concise and short, increasing the clarity of expression and minimizing redundancies. After data collection, a common latent factor technique was used, and only 0.06 of average squared loading of the items on the common method factor was found. Also, the significance level of all factor loadings did not change significantly from the original model.

Table 1. Descriptive statistics and correlations (study 1).

Variable	M	SD	1	2	3	4	5	6	7	8
1. Education (years)	16.04	1.71
2. Tenure (years)	8.53	6.69	−0.12 *
3. Identity enhancement	3.44	0.71	0.07	0.25 **	(0.89)
4. Identity conflict	2.06	0.83	−0.06	−0.03	−0.27 **	(0.93)
5. Identity separation	3.50	0.60	0.22 **	0.12 *	0.44 **	−0.21 **	(0.81)
6. Work engagement	3.56	0.63	0.07	0.11	0.31 **	−0.13 **	0.38 **	(0.77)
7. Role expansion	3.64	0.77	0.17 **	0.15 **	0.28 **	−0.18 **	0.37 **	0.24 **	(0.93)
8. Voice behavior	3.23	0.74	0.06	0.09	−0.32 **	−0.12	−0.26 **	0.53 **	0.66 **	(0.88)

N = 339 * p < 0.05, ** p < 0.01.

shows the descriptive statistics and correlations among the study variables. Also, Cronbach’s alpha coefficients are reported in parentheses. The mean value of identity conflict (M = 2.06) was much lower than that of identity enhancement (M = 3.44), and there was a small negative correlation between identity enhancement and conflict (r = −0.27, p < 0.01). All the correlations of hypothesized relationships showed expected directions.

4.2. Hypothesis Tests

To estimate the hypothesized model, multiple regression analyses were performed. Hypotheses 1a and 1b predicted that identity enhancement would be positively related to work engagement and role expansion, respectively. As Model 1 and Model 3 of

Table 2. Multiple regression results for study 1.

	Work Engagement		Role Expansion		Voice Behavior
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Education level	0.07 ***	0.06 ***	0.02	0.02	0.07 ***	0.07 *	0.07 ***	0.05 *
Tenure	0.00	0.00	0.01	−0.00	0.00 *	0.00 *	0.01 *	0.01 *
Identity enhancement	0.36 ***	0.34 ***	0.21 ***	0.18 ***	0.21 ***	0.08	0.21 ***	0.08
Identity conflict	−0.07 *		−0.12 **		−0.10 *	−0.06
Identity separation		−0.07 *		−0.12 **			−0.07 +	−0.03
Work engagement						0.28 ***		0.29 ***
Role expansion						0.18 **		0.18 **
F	28.96 ***	29.53 ***	9.41 ***	9.69 ***	11.29 ***	13.15 ***	10.94 ***	12.97 ***
R2	0.27	0.27	0.11	0.11	0.13	0.20	0.12	0.20

Note: N = 339. Unstandardized coefficients are reported. + p < 0.10. * p < 0.05. ** p < 0.01. *** p < 0.001 (one-tailed).

show, the coefficients between identity enhancement and work engagement (b = 0.36, p = 0.000) and role expansion (b = 0.21, p = 0.000) were positive and significant, supporting Hypothesis 1. Hypotheses 2a and 2b predicted the negative effects of identity conflict on work engagement and role expansion, respectively. The results showed that identity conflict had a negative relationship with work engagement (b = −0.14, p = 0.033) and role expansion (b = −0.12, p = 0.003), supporting Hypothesis 2.

Hypotheses 3a and 3b predicted that high work engagement and role expansion would show positive relationships with voice behavior, respectively. As Model 6 of Table 2 shows, work engagement (b = 0.28, p = 0.000) and role expansion (b = 0.18, p = 0.002) were positively related to voice behavior even after controlling for identity relationship variables. Therefore, Hypothesis 3 was supported.

Hypothesis 4 predicted that identity enhancement would be positively related to voice behavior via work engagement and role expansion. Path analyses were conducted to confirm the significance of indirect relationships by using the PROCESS macro version 4.2. Indirect effects for each of 5000 bootstrapped samples were computed, and the bias-corrected 95% confidence intervals of the estimated effects were constructed. As

Table 3. Summary of indirect effects (studies 1 and 2).

	Path		Estimate	95% Confidence Interval
				Lower Level	Upper Level
	Identity enhancement → Work engagement	→ Voice	0.102	0.048	0.207
	→ Role expansion		0.037	0.010	0.165
Study 1	Identity conflict    → Work engagement	→ Voice	−0.019	−0.045	0.002
	→ Role expansion		−0.022	−0.048	−0.004
	Identity separation   → Work engagement	→ Voice	−0.021	−0.048	−0.001
	→ Role expansion		−0.021	−0.048	−0.004
	Identity enhancement → Work engagement	→ Voice	0.057	0.008	0.134
	→ Role expansion		0.172	0.072	0.289
Study 2	Identity conflict    → Work engagement	→ Voice	−0.024	−0.073	0.013
	→ Role expansion		−0.005	−0.102	0.089
	Identity separation   → Work engagement	→ Voice	−0.047	−0.010	−0.009
	→ Role expansion		−0.027	−0.105	0.054

Bootstrap = 5000, unstandardized coefficients are reported.

shows, the indirect effect of identity enhancement on voice through work engagement was positive (estimate = 0.102, 95% CI [0.048, 0.207]), supporting Hypothesis 4a. For the relationship between identity enhancement and voice behavior, the indirect effect through role expansion was positive (estimate = 0.037, 95% CI [0.010, 0.165]), supporting Hypothesis 4.

Hypothesis 5 predicted that work engagement and role expansion would mediate the relationship between identity conflict and voice behavior. The indirect effect of identity conflict on voice via work engagement was negative but non-significant (estimate = −0.019, 95% CI [−0.045, 0.002]), rejecting Hypothesis 5a. Finally, the indirect effect of identity conflict via role expansion was negative (estimate = −0.022, 95% CI [−0.048, −0.004]), supporting Hypothesis 5b.

4.3. Supplementary Analysis

The main analyses demonstrated that identity enhancement and identity conflict influence voice behavior through two distinct mechanisms—work engagement and role expansion. However, the data revealed that participants reported substantially lower levels of identity conflict (M = 2.06) compared with identity enhancement (M = 3.44), and both the direct and indirect effects of identity conflict were relatively weak. This raised the possibility that identity conflict alone may not fully capture the dynamics at play when civic and employee identities were perceived as incompatible. To further explore this issue, the analysis drew on the work–nonwork identity literature, which proposes identity separation as an alternative coping strategy individuals use when managing incompatible identities.

Unlike identity conflict, which involves active tension between competing identities, identity separation entails the creation of spatial, temporal, or psychological boundaries that compartmentalize identities across domains [65]. This strategy allows individuals to avoid conflict by distancing one identity from the other, rather than resolving the incompatibility directly. Given the low prevalence and weak effects of reported identity conflict in the current sample, it is plausible that individuals who would otherwise report high identity conflict have either exited the organization or are employing identity separation as a way to mitigate tension. Accordingly, identity separation was included in a supplementary analysis as a theoretically grounded alternative to identity conflict, in order to explore its distinct effects on work-related outcomes.

Identity separation was measured using a three-item scale [46,91], including the item “I keep everything about being a citizen and being an employee separate” (Cronbach’s α = 0.86). The average score for identity separation (M = 2.79) was notably higher than that of identity conflict. As shown in Model 2 and Model 4 of Table 2, identity separation significantly and negatively predicted work engagement (b = −0.07, p = 0.013) and role expansion (b = −0.12, p = 0.002). Furthermore, the indirect effects of identity separation on voice behavior through both mediators were also negative and statistically significant (work engagement, estimate = −0.021, 95% CI [−0.048, −0.001]; role expansion, estimate = −0.021, 95% CI [−0.048, −0.004]).

Additionally, to assess the robustness of the results and evaluate the effect sizes of the identity relationship constructs on the mediators, Cohen’s f² values were calculated. The largest effects were observed for identity enhancement, with values ranging from 0.03 to 0.22, indicating small to moderate magnitudes. The smallest effects were associated with identity conflict, with f² values ranging from 0.01 to 0.02. Identity separation showed modest effects, with values ranging from 0.02 to 0.03. Furthermore, scatter plots for central relationships in the model (e.g., identity relationships → work engagement; role expansion) are included in Appendix B (Figure A1), which illustrates the relative strength and direction of each identity-based relationship. These findings suggest that identity separation may be a more prevalent and functionally distinct mechanism from identity conflict, and thus, separating one’s civic and employee identities may suppress voice behaviors at work. Figure 3 displays the results of the path analysis for the extended model. Finally, robustness checks were conducted by comparing the hypothesis test results with and without the inclusion of control variables. The analyses without control variables yielded results consistent with those of the full model. These findings suggested that the inclusion of control variables did not materially alter the main conclusions of the study.

To enhance the generalizability and address the limitations of study 1, a constructive replication was conducted in a distinct organizational context—entrepreneurial firms. These firms differed significantly from large corporations in terms of structure, employee experience, and identity dynamics. Their smaller size and formative nature tended to foster stronger identification and a heightened awareness of social value creation as employees were more directly involved in shaping the firm’s direction [92]. Examining the model in this setting offered a valuable contrast and deepened our understanding of how identity mechanisms functioned across varying work environments. Furthermore, while study 1 used multisource data, it relied on single-timepoint measures for key constructs, raising concerns about common method variance. To mitigate this, study 2 employed a three-wave survey design with temporal separation between predictor, mediator, and outcome variables, thereby strengthening causal inference and internal validity.

5. Study 2: Method

Participants, Procedure, and Measures

Study 2 was conducted among full-time employees working in small to medium-sized entrepreneurial firms in Korea, recruited through industry networks and professional contacts. Similar inclusion criteria were applied to ensure comparability. Participants were from two technology-based entrepreneurial firms. They were all certified as Innobiz (Innovation + Business) companies by the Korean government, which referred to small- to medium-sized enterprises (SMEs) equipped with a competitive edge and innovation capabilities. In collaboration with administrative staffs, all employees from the companies’ various departments were invited to participate in the survey. To reduce potential biases and improve temporal separation, data were collected using a three-wave survey design over several weeks, with key variables measured at different time points. Also, questionnaires were kept concise and short, increasing the clarity of expression and minimizing redundancies.

At time 1, web-based surveys were distributed to 140 employees, measuring identity relationships. This was similar to the sample size of 150 for estimating research hypotheses [82,83]. Three weeks later after time 1, a second survey was distributed to those who participated in the time 1 survey. The questionnaire included measures of work engagement and role expansion. Finally, three weeks later (time 3), the same participants were invited to participate in the survey and were asked to assess their voice behavior. Employees were paid USD 4 for their participation in the survey. Based on statistics checking for outliers, seven cases were removed. Using FIML estimation again, the final sample consisted of 121 employees. Of the participants, about 45% were women, the average age of the respondents was 33.5 years (SD = 6.3), and the average length of tenure at the company was 3.34 years (SD = 2.69).

Identity relationships, work engagement, role expansion, and voice variables were assessed using the same scales as in study 1, and in the case of voice behavior, item 5 was removed from the questionnaire in order to keep the questionnaires concise and lower the burden for the participants. The same control variables were included.

6. Study 2: Results

6.1. Preliminary Analysis

For the hypothesis tests, the same analytic approach was used as in study 1. To assess the validity of the study variables, CFAs were conducted. The hypothesized five-variable model demonstrated acceptable fit to the data (X² = 249.05, df = 138, CFI = 918, TLI = 0.899, RMSEA = 0.082, SRMR = 0.080). Also, all items were well loaded to their latent variables (p < 0.001), and standardized factor loadings were above 0.55. To check the possibility of common method variance, a common latent factor technique was used, and only 0.12 of average squared loading of the items on the common method factor was found.

Table 4. Descriptive statistics and correlations (study 2).

Variable	M	SD	1	2	3	4	5	6	7	8
1. Education (years)	14.93	1.75
2. Tenure (years)	3.34	2.69	−0.31 **
3. Identity enhancement	3.04	0.83	0.02	0.02	(0.87)
4. Identity conflict	2.48	0.93	−0.07	0.10	−0.15	(0.92)
5. Identity separation	3.22	0.97	0.00	−0.14	−0.21 *	0.34 **	(0.89)
6. Work engagement	3.38	0.57	0.13	0.18	0.28 **	−0.15	−0.36 **	(0.86)
7. Role expansion	3.40	0.65	0.00	0.12	0.26 **	−0.01	−0.21 *	0.46 **	(0.88)
8. Voice behavior	3.23	0.74	0.06	0.09	−0.32 **	−0.12	−0.26 **	0.53 **	0.66 **	(0.88)

N = 121 * p < 0.05, ** p < 0.01.

shows the descriptive statistics and correlations for all study 2 variables. The mean value of identity conflict (M = 2.48) was lower than those of identity enhancement (M = 3.04) and identity separation (M = 3.22). All the correlations of hypothesized relationships showed expected directions except for the relationships with identity conflict.

6.2. Hypothesis Tests

Hypotheses 1a and 1b predicted that identity enhancement would be positively related to work engagement and role expansion, respectively. As Model 1 and Model 3 of

Table 5. Multiple regression results for study 2.

	Work Engagement		Role Expansion		Voice Behavior
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Education level	0.08 *	0.07 *	−0.02	−0.02	0.05	0.04	0.05	0.03
Tenure	0.05 **	0.04 *	0.04 +	−0.04 +	0.03	−0.01	0.02	−0.01
Identity enhancement	0.18 **	0.14 *	0.31 ***	0.29 ***	0.27 ***	0.04	0.24 **	0.03
Identity conflict	−0.08 +		−0.01		−0.07	−0.04
Identity separation		−0.16 **		−0.05			−0.13 *	−0.06
Work engagement						0.31 **		0.29 **
Role expansion						0.56 ***		0.56 ***
F	4.93 **	6.96 ***	5.10 ***	5.23 ***	3.44 *	16.12 ***	4.13 **	16.31 ***
R2	0.17	0.22	0.17	0.18	0.12	0.50	0.14	0.50

N = 121. Unstandardized coefficients are reported. + p < 0.10. * p < 0.05. ** p < 0.01. *** p < 0.001 (one-tailed).

show, the coefficients between identity enhancement and work engagement (b = 0.18, p = 0.003) and role expansion (b = 0.31, p = 0.000) were positive and significant, supporting Hypothesis 1. Hypotheses 2a and 2b predicted the negative effects of identity conflict on work engagement and role expansion, respectively. The results showed that identity conflict had a negative but marginally significant relationship with work engagement (b = −0.08, p = 0.096), providing marginal support for Hypothesis 2a. However, its relationship with role expansion was non-significant (b = −0.01, p = 0.446), thus not supporting Hypothesis 2b.

Hypotheses 3a and 3b predicted that high work engagement and role expansion would show positive relationships with voice behavior, respectively. As Model 6 of Table 5 shows, work engagement (b = 0.31, p = 0.004) and role expansion (b = 0.56, p = 0.000) were positively related to voice behavior even after controlling for identity relationship variables. Therefore, Hypothesis 3 was supported.

Hypothesis 4 predicted that identity enhancement would be positively related to voice behavior via work engagement and role expansion. As Table 3 shows, the indirect effect of identity enhancement on voice through work engagement was positive (estimate = 0.057, 95% CI [0.008, 0.134]), supporting Hypothesis 4a. For the relationship between identity enhancement and voice behavior, the indirect effect through role expansion was positive (estimate = 0.172, 95% CI [−0.242, 0.006]), supporting Hypothesis 4b. Hypothesis 5 predicted that work engagement and role expansion would mediate the relationship between identity conflict and voice behavior. The indirect effects of identity conflict on voice via work engagement (estimate = −0.019, 95% CI [−0.045, 0.002]) and via role expansion (estimate = −0.022, 95% CI [−0.048, −0.004]) were negative but non-significant, rejecting Hypothesis 5.

6.3. Supplementary Analysis

Since the effects of identity conflict were weak as in study 1, a supplementary analysis was conducted with an identity separation variable. As shown in Model 2 and Model 4 of Table 5, identity separation had a stronger negative effect on work engagement than identity conflict (b = −0.16, p = 0.002) and had the same insignificant effect on role expansion (b = −0.05, p = 0.247). Finally, work engagement negatively mediated the relationship between identity separation and voice (estimate = −0.047, 95% CI [−0.105, −0.009]), while role expansion did not show a significant indirect effect (estimate = −0.027, 95% CI [−0.105, 0.054]). Figure 4 displays the results of the path analysis for the model.

To assess the robustness of the results and evaluate the effect sizes of the on the mediators, Cohen’s f² was calculated. The largest effect sizes were observed for identity enhancement, ranging from 0.05 to 0.16, indicating small to moderate effects. The smallest effects were associated with identity conflict, with f² values ranging from 0.00 to 0.02. Identity separation showed medium-level effects, with f² values ranging from 0.01 to 0.09. Moreover, scatter plots for central relationships in the model (e.g., identity relationships → work engagement; role expansion) are included in Appendix B (Figure A2). These findings suggested that identity separation may be a more prevalent and functionally distinct mechanism from identity conflict. Finally, robustness checks were conducted by comparing the hypothesis test results with and without the inclusion of control variables. The analyses without control variables yielded results consistent with those of the full model.

7. Discussion

In pursuit of sustainable management, organizations must not only balance economic, environmental, and social goals but also cultivate human-centered strategies that support long-term viability. Drawing on identity theory, this study introduced civic identity—an individual’s self-concept as a responsible member of society—into the workplace and examined how its relationship with employee identity (i.e., enhancement and conflict) influences sustainability mechanisms: work engagement, role expansion, and voice behavior. Two field studies were conducted using multi-source data from 339 employees at large conglomerates (study 1) and three-wave data from 121 employees at entrepreneurial firms (study 2). Across two empirical studies, the findings provided empirical support for the importance of identity alignment in shaping motivational and change-oriented behaviors in diverse organizational contexts.

Specifically, identity enhancement significantly and positively predicted both work engagement and role expansion, which in turn led to increased voice behavior in both large corporations and entrepreneurial firms. Scholars have called for adopting multiple identity perspectives in studying voice behavior [93] because empirical evidence regarding the relationship between employee identity and voice has shown mixed results [94,95]. The two empirical studies showed that employees with high identity enhancement spoke up for the organization more frequently as they were more engaged in their work and took broader responsibilities at work. In contrast, employees with high identity conflict and separation between civic and employee identities decreased employee voices by reducing their work engagement and role breadth.

Notably, in both studies, the effects of identity conflict were relatively weak, compared with the effects of identity enhancement. Several reasons may explain this finding. First, employees may respond to civic–employee identity conflict in a variety of ways. For instance, prior studies suggest that in some cases, employees prioritize one primary identity [38], and if employees prefer civic identity at the workplace in conflict situations, they may try to express the identity, which would likely not result in negative influences on work engagement or role expansion. Second, employees who experience identity conflict might not be able to stay in their organization. The respective mean values of identity conflict were low (2.07 in study 1; 2.48 in study 2), compared with those for identity enhancement (3.44 in study 1; 3.04 in study 2). This result implied that employees who experienced identity conflict had already quit their jobs, and fewer employees with a high level of identity conflict remained.

There was also the possibility that employees had a different type of civic–-–employee identity relationship or changed their identity relationships, such as occurs with identity separation. As our supplementary analyses in both studies showed, identity separation had stronger effects on work engagement and role expansion than did identity conflict. This result could indicate that employees in a state of identity conflict might have at least tried to verify both of their identities—civic and employee—despite the difficulties that existed. Once they separated their identities, however, they might have given up or not tried to verify one or all of their identities, which increased stronger, negative effects on work outcomes. Further research should be conducted to investigate this in depth.

Finally, with respect to role expansion, this study found mixed results because study 2 (in entrepreneurial firms) showed non-significant effects of identity conflict and separation on role expansion, which led to a rejection of the mediation hypotheses of role expansion. There may be several reasons to explain these results, including the small sample size of study 2. Operating within complex ecosystems, entrepreneur firms strive to develop and deliver new products under conditions of extreme uncertainty, and this fosters ambiguity in employees’ tasks and roles [96]. In addition, startup organizations have limited human resources and meager financial resources. Therefore, employees are expected to perform multiple roles simultaneously [97,98]. Since role expansion may occur frequently or naturally in such an environment, it is possible that the effects of reducing role expansion were not found even when employees were in a state of identity conflict or separation. An HR director who used to work for large corporations and who currently works for a venture company commented on this finding:

“…we have not put in place appropriate systems, resources, and clear role boundaries between employees. As you know, small firms are busy places, and employees need to deal with different and broad tasks…”.

These reflections underscore the importance of further investigating the relationship between work engagement and voice behaviors in entrepreneurial firms. Based on the above comment, for example, future research might examine the role of context in influencing the dynamics of civic–employee identity relationships and their verification in the workplace.

7.1. Implications for Theory and Research

This study offers new insights into sustainable management by revealing how the interplay between civic and employee identities contributes to psychological engagement, role expansion, and change-oriented behavior in the workplace. The results have theoretical implications for sustainability science and the organizational behavior field. First, the research has meaningful implications in that the role of citizens has been increasingly highlighted in contemporary society [99] as individuals and organizations express societal concerns and try to make a positive impact on society [19,20]. By introducing civic identity—a nonwork identity—into the sustainability and organizational study conversation, this research enriches our understanding of how individual identity configurations shape organizational sustainability outcomes. Existing studies have largely focused on work-related identities (e.g., professional or organizational identities), often overlooking the broader social self that employees bring to their roles [14].

The findings indicated that when employees experienced a synergy between their civic and employee roles, they demonstrated greater work engagement and a stronger sense of ownership that extended beyond formal job descriptions—both of which are critical for fostering adaptability, innovation, and social responsibility within organizations. Furthermore, the harmonization of civic and employee identities facilitates voice behavior, which functions not merely as a communicative act but as a core mechanism for sustaining transparency, ethical responsiveness, and collaborative problem-solving. Given that previous research on the relationship between employee identity and voice has yielded mixed findings [94,95], this study responds to scholarly calls to incorporate multiple identity frameworks in examining voice behavior [93].

Moreover, research findings suggest that employee behavior related to sustainability not only is a function of institutional policies or leadership but is deeply rooted in the employee’s own internal identity structure. This expands the conceptualization of sustainable human resource management [100] by integrating psychological and sociological perspectives of identity [37]. The research also was particularly meaningful in that the empirical evidence was confirmed in the context of both large corporations and entrepreneurial firms.

7.2. Implications for Practice

Practically, organizations seeking to promote sustainable employee behavior must go beyond encouraging mere compliance or performance by creating environments where civic identity is welcomed and integrated into daily work life. Managers should attend to how employees experience the relationship between their civic and employee identities, aiming to foster harmony while minimizing conflict and separation. Prior research has demonstrated that identity enhancement and conflict are malleable and can be influenced by organizational interventions [24], suggesting that managers have the capacity to shape employees’ experiences of their multiple identities. Accordingly, this study proposes three practical strategies—centered on organizational leadership, job design, and interpersonal dynamics within teams—to help align civic and employee identities.

First, at the organizational or executive level, it is important to avoid pressuring employees into engaging in ethically questionable or socially undesirable practices for the sake of financial performance. Leaders should also cultivate a culture that does not encourage silence or passive compliance in the face of such practices. Instead, organizations can foster identity enhancement by reinforcing socially responsible goals and values in their strategic direction and daily operations. When organizational leaders explicitly prioritize mutual growth and ethical practices, employees will be more likely to experience motivation and engagement.

Second, from a job design perspective, organizations should help employees recognize the social value inherent in their work—particularly how their tasks contribute to the well-being of others, including clients, colleagues, or the broader community. Enhancing employees’ awareness of their positive impact on others can promote a greater sense of purpose and alignment with their civic identity. When employees feel their work positively influences others, they will be more likely to experience motivation and engagement.

Third, in terms of leadership and team dynamics, a psychologically safe environment—where inappropriate behavior such as coercive supervision, manipulation, or unethical demands from leaders or colleagues is absent—is critical for reducing identity conflict and encouraging proactive engagement. When team leaders and coworkers refrain from imposing undue pressure or engaging in misconduct, it may become easier for employees to internalize and enact prosocial values in the workplace.

7.3. Limitations and Future Directions

This study was not without limitations. First, this study used a cross-sectional design and single-source data in study 2. This study tried to improve the validity of the findings by reducing common-method bias through procedural and statistical methods. Nevertheless, there still remains a possibility of biased results. Therefore, while the theoretical model suggests that identity relationships influence work engagement, role expansion, and voice behavior, the findings should be interpreted with caution, and further research is needed to address the causal relationship.

Second, future studies should attempt to capture specific sources of and reasons for civic–employee identity enhancement and conflict. This study was more interested in how people responded to identity enhancement and conflict than what made them experience identity enhancement and conflict specifically. Future research could focus on specific triggers of civic–employee identity relationships in the workplace. Relatedly, it would be interesting to investigate how identity relationships and identities can change over time. This study treated identities and their relationships as being relatively static in this study, but they can have dynamic natures. For example, employees can integrate different identities to create a new identity [39]. For this type of inquiry, qualitative methods can be suggested in future research.

Finally, this study only examined identity relationships between two identities—civic and employee. However, there are other identities to be considered in the workplace, such as occupational, relational, or team identities. Recent multiple identity studies have also started to investigate the relationships among more than two identities [12]. Thus, extending the civic–employee identity relationship to more identities will be beneficial for understanding more complex phenomena regarding multiple identities in the workplace. Moreover, this study assumed that work and society domains were independent. However, there is another view that society and the organization can have a nested relationship such that organizations belong to a broader society [49]. Therefore, future research may benefit from investigating the effects of different types of relationships between work and society domains.

8. Conclusions

This study highlights the importance of harmonizing civic and employee identity relationships in promoting sustainable employee behavior within organizations. The findings demonstrate that civic identity can positively influence voice behavior when integrated with employees’ work roles, primarily through the mediating roles of work engagement and role expansion. Conversely, identity conflict and separation may hinder voice behaviors, underscoring the complex interplay between personal and professional identities in the workplace. These results contribute to a more nuanced understanding of identity dynamics at work and offer a foundation for future research on sustainable human resource practices and employee behaviors that align personal and organizational values. From a practical perspective, organizations should consider ways to acknowledge and support employees’ civic identity as part of their workplace experience. By recognizing employees not only as workers but also as citizens, managers and policymakers can cultivate a workplace environment that promotes both individual flourishing and collective sustainability.