Target similarity and gender difference: Effects of servant leadership on supervisory commitment and supervisory citizenship behavior

The concept of servant leadership has gained a significant stream of research for improving and promoting organizational performance (Canavesi & Minelli, 2022). While this unique leadership style leads others by serving and helping followers, it might develop followers’ positive attitudes and behaviors (Smallfield, Sun, van Dierendonck, & Liden, 2023). In the era of post-COVID-19, it might promote employee commitment, citizenship behavior, work-life balance and human capital for organizational sustainability (Batool, Mohammad, & Awang, 2022); and reduce employee depression, burnout and deviant behavior (Ruiz-Palomino, Yáñez-Araque, Jiménez-Estévez, & Gutiérrez-Broncano, 2022). In business practice, servant leadership is also adopted as the guiding philosophy for over 20% of Fortune magazine 100 companies, such as Disney, Southwest Airlines, Nordstrom, Wegmans and Starbucks (Blanchard & Conley, 2022).

While employee commitment has multiple targets such as organization or supervisors (Van Rossenberg, Cross, & Swart, 2022), supervisory commitment has a stronger impact on in-role and extra-role performance than organizational commitment (Becker, 2016). For this reason, followers might be more closely identified toward supervisors than the whole organization (Zhao, Liu, & Gao, 2016). Although Smallfield et al. (2023) theoretically propose servant leadership positively affects supervisor commitment, there is still a lack of empirical evidence. Similarly, the linkage between servant leadership and supervisory citizenship behavior has not been explored in the literature (Roberts, 2023b). Supervisory citizenship behavior is more specific and narrower than organizational and individual citizenship behaviors respectively (Rupp & Cropanzano, 2002). Specifically, in the literature review, individual citizenship behavior is more focused on co-workers than supervisors (Gao Héliot & Roberts, 2023; Roberts, 2023a).

Lavelle, Rupp, and Brockner (2007) developed the target similarity model by integrating multifocal perspectives of justice, social exchange and citizenship behavior for a better explanation about employee attitudes and behaviors in the workplace. This model proposes that subordinates might hold multiple, unique social exchange relationships with organization, supervisors, colleagues or customers. Compared to short-term economic exchange, social exchange is long-term and reflects mutual support, trust and commitment (Lavelle, Rupp, Manegold, & Thornton, 2015). Because of a supervisor’s fair treatment, subordinates promote high-quality social exchange. Then, they feel a high responsibility to the supervisor and are engaged in supervisor citizenship behavior. By applying the target similarity model (Lavelle et al., 2007), this study investigates the effects of servant leadership on supervisory commitment and supervisory citizenship behavior. While supervisory commitment might be considered to be a proxy of social exchange that leads to supervisory citizenship behavior in this model (Lavelle et al., 2007), its mediating role is also explored in this study.

While gender diversity in the board might promote business performance (Ghazi, Miramontes, & Brahme, 2023), the relationship of supervisory leadership behavior and gender is still one important topic in the era of COVID-19 (Eichenauer et al., 2022). Supervisors are facing leadership challenges such as work from home and workforce reductions. As female leaders might be better than their counterparts during times of crisis, leadership styles and stereotypes might explain gender differences in leadership (Bullough, Guelich, Manolova, & Schjoedt, 2022). Particularly, female supervisors are more likely to use servant leadership than male leaders (Xiu, van Dierendonck, & Lv, 2023). Specifically, their gender stereotypes with communion are more reasonable for servant leadership than male supervisors (Canavesi & Minelli, 2022). Therefore, leadership gender might moderate the linkage between servant leadership and job performance (Lemoine & Blum, 2021). For the research call of Tonoyan et al. (2022), this study compares supervisory gender groups to promote the relationship between servant leadership and supervisory commitment.

This study contributes to literature in several ways. First, by expanding the target similarity model (Lavelle et al., 2007), this study explores the impact of servant leadership on supervisory commitment and supervisory citizenship behavior. Second, this study analyzes the mediating role of supervisory commitment on the relationship between servant leadership and supervisory citizenship behavior. Third, with respect to the research call of Tonoyan et al. (2022), this study also compares gender groups in the linkage between servant leadership and supervisory commitment.

The research model is illustrated in Figure 1. Accordingly, servant leadership directly impacts supervisory citizenship behavior (H1). Then, supervisory commitment has a direct effect on supervisory citizenship behavior (H2). Servant leadership directly impacts supervisory commitment (H3). Moreover, supervisory commitment mediates the relationship between servant leadership and supervisory citizenship behavior (H4). Finally, a supervisory gender analysis was conducted to compare supervisor gender in the relationship between servant leadership and supervisory commitment (H5).

According to Greenleaf (1977), servant leaders first serve and then lead. Liden, Wayne, Zhao, and Henderson (2008) proposed seven essential characteristics of a servant leader: placing subordinates first, emotional healing, empowering, helping subordinates grow and succeed, having conceptual skills, behaving ethically and creating value for the community. Specifically, servant leaders with stewardship put their followers’ needs and desires before their own (van Dierendonck & Patterson, 2018). For long-term organizational goals, leaders take care of followers’ well-being, empower and train followers to become future leaders (Khan, Mubarik, Ahmed, Islam, & Khan, 2022). Additionally, servant leaders have wisdom and a vision for goal setting, effective support and leading followers (Al-Asfour, Charkasova, Rajasekar, & Kentiba, 2022). Behaving ethically, servant leaders might be open, fair and honest with followers (Sendjaya et al., 2020). Moreover, servant leaders promote value for the community outside their organization and encourage followers to do the same (Meuser & Smallfield, 2023).

Neubert, de Luque, Quade, and Hunter (2022) find that servant leadership is a leadership style globally that is linked to four global culture dimensions, namely power distance, humane orientation, assertiveness and institutional collectivism. Moreover, using different GLOBE culture clusters, Mittal and Dorfman (2012) point out that servant leadership dimensions of egalitarianism and empowerment are more strongly associated with European cultures than Asian cultures. However, servant leadership dimensions of empathy and humility are more strongly endorsed in Asian cultures than European cultures. In addition, McCune Stein, Bell, and Ai Min (2020) realize that the relationship between servant leadership and leader-member exchange is stronger for Anglo subordinates than Chinese. However, there are no significant differences in affective commitment, organizational citizenship behavior, job satisfaction, innovative work behavior and job performance.

Becker, Billings, Eveleth, and Gilbert (1996) defined supervisory commitment as identification with a supervisor’s personality and internalization of value congruence between subordinates and supervisors. This refers to the psychological attachment of subordinates to their supervisor. Followers without external pressures are free of their decision and responsibility. In other words, followers shape volitional dedication and establish positive supervisor-subordinate relationships in the workplace (Santana-Martins, Sanchez-Hernandez, & Nascimento, 2022).

Although supervisory commitment is a type of employment commitment, it is different and strongly related to organizational commitment (Becker, 2016). Specifically, organizational commitment of followers is the acceptance of organizational goals and values, extra effort on behalf of the organization and desire to remain with the organization (Mowday, Steers, & Porter, 1979). Supervisory commitment might be a stronger predictor of in-role, extra-role, job satisfaction, turnover and prosocial organizational behaviors than organizational commitment (Becker, 2016). Then, some studies have explored the mediating role of supervisory commitment for the linkages between organizational commitment and employee-related factors, such as psychological contract, empowerment and turnover (Kidron, 2018).

In the literature review, supervisory commitment might be promoted by several drivers, such as authentic leadership (Imam, Naqvi, Naqvi, & Chambel, 2020) and organizational communication (Holzwarth, Gunnesch-Luca, Soucek, & Moser, 2021). Similarly, supervisor’s justice, support and trust in the leader might lead to supervisory commitment (Akram, Kamran, Iqbal, Habibah, & Ishaq, 2018). Furthermore, supervisory commitment leads to several positive outcomes such as employee participation (Joo, Byun, Jang, & Lee, 2018), proactivity (Singh & Rangnekar, 2020), change readiness (Seggewiss, Straatmann, Hattrup, & Mueller, 2019), employee creativity (Imam et al., 2020), well-being and job satisfaction (Huyghebaert, Gillet, Becker, Kerhardy, & Fouquereau, 2017), job involvement and innovative behavior (Wang & Hou, 2023). Similarly, supervisory commitment decreases counterproductive work behavior and follower turnover (Smallfield et al., 2023).

Supervisory citizenship behavior is citizenship behavior toward the supervisor of followers which helps supervisors in the workplace socially and psychologically (Rupp & Cropanzano, 2002). Compared to in-role performance, this extra performance is discretionary and not directly rewarded (Williams & Anderson, 1991). It directly benefits the supervisor and indirectly contributes to the organization. Consequently, supervisory citizenship behavior might increase work performance and organizational effectiveness and decrease turnover and absenteeism (Organ, 2018). Several studies have focused on positive drivers of supervisory citizenship behavior, such as ethical leadership (Yam, Fehr, Burch, Zhang, & Gray, 2019), talent status and management identification (Wikhamn, Asplund, & Dries, 2021), workaholism and family-support supervisor behavior (Pan, 2018). Conversely, authoritarian leadership might decrease supervisory citizenship behavior (Ahmad Bodla, Tang, Van Dick, & Mir, 2019). Furthermore, supervisory citizenship behavior might be positively related to leader-member exchange (Wulani, Handoko, & Purwanto, 2022).

According to the target similarity model (Lavelle et al., 2007), servant leadership might promote supervisory citizenship behavior through the mediation of leader-member exchange, trust and justice. Servant leadership with ethical behaviors and community contributions might promote social exchange and justice in the workplace (Yasir & Jan, 2023). Therefore, the stewardship of servant leaders might promote an ethical climate and well-being for followers (Wang et al., 2022). Then, serving behaviors beyond the call of duty might lead to supervisory citizenship behavior for reciprocity (Gnankob, Ansong, & Issau, 2022). Thus, the following hypothesis is proposed:

By applying the target similarity model (Lavelle et al., 2007), while supervisory commitment might be considered to be a proxy for social exchange relationship, supervisory commitment might directly lead to supervisory citizenship behavior. By the supervisor’s fair treatment, subordinates foster social exchange and a high responsibility to the supervisor; in turn, they are engaged in supervisor citizenship behavior (Lavelle et al., 2015). In other words, through person-supervisor fit, leader-member exchange might promote helping behavior toward supervisors (Zhang, Lam, & Deng, 2017). The thoughts lead to the following hypothesis:

With the positive characteristics of stewardship, follower development and ethical behaviors, servant leadership might encourage social exchange, fairness and trust; and discourage social loafing in work groups (Kauppila et al., 2022). Therefore, based on the target similarity model (Lavelle et al., 2007), servant leadership might promote supervisory commitment. Specifically, while servant leaders share a common identity and stimulate a sense of shared supervisory identity (Zhao et al., 2016), they might promote high-quality supervisor–subordinate relationships through personal identification (Lv et al., 2022). Consequently, supervisor-subordinate promotion fit leads to commitment to the supervisor (Linando & Halim, 2023). In addition, servant leadership encourages followers’ empowerment, work engagement, intrinsic motivation and autonomy (David, Johnson, Meng, & Lopez, 2021). This leads to trust and commitment to the supervisor (Smallfield et al., 2023). Likewise, servant leadership leads to supervisory commitment through supervisory procedural, interpersonal and informational justice (Akram et al., 2018).

The mediating role of supervisory commitment in the relationship of servant leadership and supervisory citizenship behavior is premised on the preceding discussions. Specifically, according to the target similarity model (Lavelle et al., 2007), servant leadership with justice and social exchange promotes supervisory commitment; in turn, supervisory commitment might directly lead to supervisory citizenship behavior. Then, the hypotheses are proposed:

There are gender stereotypes in leadership (Smith, Eriksson, & Smith, 2021). Male leaders are determined, in self-control, willing to take risks, competitive and self-confident. They have lower scores on individualized consideration than female leaders (Abarca & Majluf, 2021). Female leaders are socially competent, dialog-oriented and helpful. Additionally, while male leaders emphasize mistakes, failures and problem waiting; female leaders are favored in followers’ needs, success and development (Blake-Beard, Shapiro, & Ingols, 2020). The servant leadership style is adopted by female leaders more frequently as compared to their male counterparts. Their servant leadership is authentic and compassionate (Ghazi et al., 2023).

Furthermore, according to the role congruity theory (Eagly & Heilman, 2016), female leaders possess more communion and less agency than equivalent male leaders. Ghazi et al. (2023) found that female servant leaders have more communion than male servant leaders. On the other hand, male servant leaders have more agency problems than female servant leaders (Xiu et al., 2023). Eichenauer et al. (2022) found male supervisors have more agentic behaviors than their peers in managing their subordinates in the crisis context of the Covid-19 pandemic.

Moreover, leader gender might moderate the linkage between servant leadership and team effectiveness (Duff, 2013). Moreover, female leaders are scored higher than male leaders in the relationships of servant leadership, prosocial motivation and job performance (Lemoine & Blum, 2021). Therefore, this study hypothesizes the following:

The empirical setting of this study is the retail industry in Vietnam. This country has a transitional economy and Confucian culture. Leadership self-development is considered a useful business practice for new opportunities and uncertainties in a transitional economy. However, the lack of servant leadership might negatively impact job satisfaction, work engagement, leadership growth and well-being of followers (Khatri, Dutta, & Kaushik, 2022). Because of cultural differences between global East and West, servant leadership and leader gender have a different effect on followers' attitudes and behaviors in countries representing those regions (Wang et al., 2022). While most studies have been conducted in Western countries, an empirical study from a country in Asia (Vietnam) offers a global perspective to that topic.

The data were collected from 478 salespeople. The gender of salespeople was 65.7% female and 34.3% male. By age group of salespeople, 64.9% of the sample were 18–30 years old, 32.6% were 31–40 years old and 2.5% were above 40 years old. Based on the educational background of salespeople, 43.5% had a bachelor’s degree or higher and 56.5% had a school diploma or below. According to job tenure of salespeople, 18.8% were below 1 year, 48.5% were 1–4 years and 32.6% were above 4 years. Furthermore, the gender of supervisors was 61.9 % female and 38.1% male. By age group of supervisors, 16.7% of the sample were 18–30 years old, 71.1% were 31–40 years old and 12.2% were above 40 years old. The statistics showed that the sample covered a variety of salespeople and supervisors in terms of gender, age, education and job tenure. Thus, this sample was appropriate for further analysis.

The questionnaire was first prepared in English and translated into Vietnamese via a translation and back-translation process (Douglas & Craig, 2007). The collaborative approach was undertaken by two university academics. After comparing the two English versions, mismatches were discussed and adjusted using the Vietnamese version. The pretest was conducted through interviews with five salespeople to refine the wording of Vietnamese questionnaire items. To encourage the participation of these informants, the respondents are anonymous and small gifts (i.e. coupons for coffee or breakfast) were used for them.

The interviews were held at home appliance and electronics specialty stores in eight provinces in Vietnam. A total of 478 questionnaires were collected from the salespeople. The survey was based on convenience sampling. The response rate was 79.67% for a total of 600 distributed questionnaires. For each case, salespeople’ practical behaviors were asked for servant leadership, supervisory commitment and supervisory citizenship behavior. Furthermore, demographic information of salespeople and supervisors was collected for this investigation.

The seven-item scale of Liden et al. (2015) was adapted to measure servant leadership. Supervisory commitment was measured using the five items from Cheng, Jiang, and Riley (2003). The five-item scale of Rupp and Cropanzano (2002) was used to measure supervisory citizenship behavior. All scales were rated on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

First, an exploratory factor analysis (EFA) was applied to all scales together for a preliminary assessment of dimensionality, convergent and discriminant validity. The results indicated the factor structure fully matched the design. The items were also loaded mainly on the designated factors. However, two items must be eliminated due to low loadings (below 0.5). The EFA factor loadings of the 15 remaining items ranged from 0.673 to 0.821.

Then, confirmatory factor analysis (CFA) was performed to assess convergent and discriminant validities. The model included three constructs and their respective items. The significant factor loadings of the 15 remaining items ranged from 0.638 to 0.853 in Table 1 (Hair, Black, Babin, & Anderson, 2019). The test for normality showed that the skewness values ranged from −1.036 to −0.134. The kurtosis values ranged from −0.305 to 1.714. These absolute values were within the criteria of skewness and kurtosis, below 3 and 8 respectively.

From the CFA of the refined measurements, the model yielded satisfactory fit indices. Specifically, χ²/df = 322.872/87 = 3.711 was below 4; comparative fit index (CFI) = 0.929 and Tucker–Lewis index (TLI) = 0.914 exceeded 0.9 and root mean square error approximation (RMSEA) = 0.075 was below 0.08 (Hu & Bentler, 1999). As presented in Table 1, composite reliabilities of three constructs were from 0.820 to 0.877 and above 0.7 (Hair et al., 2019). The average variance extracted (AVE) of all constructs ranged from 0.505 to 0.551 and above the threshold value of 0.50 (Fornell & Larcker, 1981). Therefore, the results indicated significant reliability and convergent validity.

Correlations between pairs of constructs had values ranging from 0.576 to 0.685 (Table 2). The correlations between factors should not be higher than 0.85 for discriminant validity (Hair et al., 2019). Furthermore, the square roots of AVEs were larger than the factor correlations (of its rows and columns) and also supported discriminant validity (Fornell & Larcker, 1981). In summary, the measurement scales of the concepts were satisfactory in terms of reliability, convergent validity and discriminant validity.

According to the anonymity of the respondents, the response biases were controlled in self-reports of salespeople. Furthermore, the Harman single-factor CFA, which is the most widely known approach, was undertaken (Fuller, Simmering, Atinc, Atinc, & Babin, 2016). In particular, the CFA results showed very low fit indices (χ²/df = 940.705/90 = 10.452; CFI = 0.744; TLI = 0.702 and RMSEA = 0.141). Therefore, common method variance (CMV) was not a major source of variation in the observed items.

Structural equation modeling (SEM) was used to test this hypothesized model. Then, the mediating effects were assessed using the bias-corrected bootstrapping technique (Preacher & Hayes, 2008). Finally, two groups of male and female supervisors are compared for the relationship of servant leadership and supervisory commitment (Byrne, 2016).

Firstly, the results suggest the hypothesized mediation model fits the data well. Specifically, χ²/df = 322.872/87 = 3.711; CFI = 0.929; TLI = 0.914 and RMSEA = 0.075 (Hu & Bentler, 1999). Furthermore, the standardized coefficients and significance of p-values were obtained. Generally, all of the hypotheses were supported at p ≤ 0.05 (Table 3). For H1, the direct path from servant leadership to supervisory citizenship behavior was significantly positive (γ1 = 0.277; p < 0.001, respectively). For H2, supervisory commitment had a direct impact on supervisory citizenship behavior (β2 = 0.437; p < 0.001). For H3, the direct path from servant leadership to supervisory commitment was significantly positive (γ3 = 0.686; p < 0.001).

Moreover, the bootstrapped 95% confidence interval (CI) around the effects did not contain zero. For H4, the mediation of supervisory commitment was also significant for servant leadership and supervisory citizenship behavior (β4 = 0.300; p < 0.001). Therefore, H4 was supported as presented in Table 4.

Secondly, to test the moderating effect of supervisor gender on the relationship between servant leadership and supervisory commitment, a multiple group analysis using SEM was employed. Two stages of analysis were conducted. First, the path between servant leadership and supervisory commitment was not constrained for the two samples (female and male). Next, the constraint was set equal for both groups. In particular, no constraints were set for the measurement models.

As presented in Table 5, the effects of servant leadership on supervisory commitment were significant in the female group (γ_female = 1.127; p < 0.001) and in the male group (γ_male = 0.363; p < 0.001). Then, the difference between the two groups was significant for this linkage with Δχ²₍₁₎ = 27.907 (p < 0.001). Thus, H5 was supported.

The study provides a better understanding of the processes or mechanisms through which servant leadership relates to supervisory commitment and supervisory citizenship behavior. First, in line with Greenleaf’s explanation (Crippen, 2022), servant leadership is optimal for followers, supervisors and organizations. In particular, supervisors and followers might take care and help each other in order to develop their organizations. By expanding the target similarity model (Lavelle et al., 2007), this study identifies the significant direct impact of servant leadership on supervisory commitment and supervisory citizenship behavior (0.686 and 0.277 respectively). While previous studies have focused on servant leadership and its organizational targets, this study makes a theoretical contribution by specifying supervisory targets as the consequences of servant leadership. The target similarity model proposes that followers hold distinctive exchange relationships with multiple referents. However, the targets toward supervisor might provide a closer vision for servant leadership in the workplace. Moreover, this study points out that servant leadership indirectly affects supervisory citizenship behavior through supervisory commitment.

As respecting the call of Tonoyan et al. (2022), this study points out the gender difference in the relationship between servant leadership and supervisory commitment. Specifically, female supervisors are scored higher than male supervisors on the positive relationship between servant leadership and supervisory commitment. Therefore, this study contributes the feminist perspective of servant leadership to the role congruity (Eagly & Heilman, 2016). Compared to male supervisors, female supervisors might be more likely related to the style of servant leadership. Specifically, the stereotype of female leaders with communion is more closely related to servant leadership than male leaders (Ghazi et al., 2023). Therefore, servant leadership promotes equality and breaks the prejudice toward female leaders.

With the growth of studies on servant leadership in global scope and Asia Pacific countries, the empirical setting in Vietnam provides evidence for a theoretical contribution. In particular, servant leadership is necessary to promote positive outcomes in the service industry (Khatri et al., 2022). Therefore, this topic may be developed for global knowledge.

Servant leadership is an effective leadership style for application in business practice. It might develop employees’ well-being in the workplace and life satisfaction. Servant leaders may reduce followers’ deviance and turnover intention. In the era of post-Covid19, managers with servant leadership are servant first and survival later. Some adaptive leadership tactics are used for sustainable development.

Beyond organizational targets, managers with servant leadership might promote the positivity of supervisory commitment and supervisory extra-role behaviors. Followers’ flourishing, in turn, might increase store-level profits. Furthermore, managers might perform servant leadership in several direct and indirect ways to enhance supervisory positive outcomes. Therefore, their will to serve followers first might promote supervisory citizenship behavior through the mediating mechanism of supervisory commitment.

Furthermore, organizations should consider different methods for developing servant leadership. For example, the human resource department might have training programs and funding for managers about servant leadership. The lack of training might negatively impact the functionality of leaders, such as caring, empowerment or follower development. Some useful skills should be considered for training servant leadership such as harmony-thinking and soft-listening. Dual-training with recognition and implementation may be useful for servant leader development.

In the era of post-Covid 19, organizations should promote gender diversification and women participation in the diverse team of leaders. Women with servant leadership might move forward to become top business leaders and maximize the potential of frontline positive outcomes. Beyond the styles of transformational and democratic leadership, female managers might find servant leadership acceptable and effective for practical performance. Their entrepreneurs with ethical foundations are the servant leaders of the market and beyond. While female managers show higher scores for the effect of servant leadership on supervisory commitment, they might act as a lubricant for this positive linkage. The competence of female managers is developed with the stereotype of communion and avoids the agency problems of male leaders; in turn, that promotes organizational development with compassionate love, authenticity, forgiveness and well-being of servant leadership.

In this study, we explore the effects of servant leadership on followers’ desirable outcomes, specifically supervisory commitment and supervisory citizenship behavior. We also investigate the mediation of supervisory commitment in the relationship between servant leadership and supervisor citizenship behavior. Furthermore, this study compares supervisory gender groups for the linkage between servant leadership and supervisory commitment. Then, the findings support all direct and indirect effects of servant leadership on supervisory commitment and supervisory citizenship behavior. We also find that the positive relationship between servant leadership and supervisory commitment is stronger for female supervisors than male supervisors. While researchers and practitioners continue to look for ways to enhance servant leadership and followers’ positive outcomes, we expect our findings may promote future research in this field.

The study also has some limitations. The first involves the use of self-report measures and a single source of information. The followers are asked to report themselves and to rate the servant leadership of their supervisors. This situation might lead to common method variance bias. Although common method bias is not evident in the current study, future research should use multiple respondents (e.g. supervisors or co-workers). Additionally, the relationships between servant leadership and follower outcomes might be different in various service industries. The findings are from one nation and one service industry (i.e. retail industry in Vietnam). To promote generalizability, cross-national studies should be explored in various service industries, such as banking, airlines and healthcare.

Moreover, cross-sectional design in the field is known to limit the ability to infer causality. Cross-sectional mediated effects are more biased than longitudinal mediated effects. Therefore, longitudinal research should be conducted to enhance the causality of the relationships in the research model. Additionally, other styles of leadership (e.g. transformational leadership or authentic leadership) were not included in the present study. Therefore, future studies should consider these factors for comparison and variable control. Then, their validity and generalizability might be further promoted.