From servant leadership to organizational citizenship behavior: A theoretically grounded moderated mediation framework for Chinese private enterprises

China’s private enterprises are key contributors to the national economy but face heightened uncertainty and competitive pressure, making extra-role behaviors such as organizational citizenship behavior (OCB) increasingly vital. OCB refers to discretionary employee actions that are not formally rewarded yet benefit organizational functioning. Prior work links leadership—particularly servant leadership (SL)—to higher OCB, but the underlying mechanisms and boundary conditions remain underexplored in Chinese private firms where paternalistic and authoritarian styles may discourage autonomous OCB. This study investigates the intrinsic processes through which SL promotes OCB, focusing on relational identification and perceived organizational support (POS) as mediators, and examines workplace loneliness as a negative moderator that may weaken these effects. The research addresses how SL affects OCB via relational identification and POS (including sequential mediation), and how workplace loneliness moderates these mediated pathways, offering actionable insights for Chinese private enterprises.

The study is grounded in three complementary theories: Social Exchange Theory (SET) posits reciprocal exchanges, explaining how leaders’ supportive, serving behaviors elicit employees’ reciprocation through OCB and highlighting POS as a key mediator. Social Identity Theory (SIT) explains how servant leaders’ care, empathy, and developmental focus strengthen employees’ relational identification with the leader, aligning followers’ goals with leaders’ aims and prompting OCB. Psychological Contract Theory (PCT) addresses unwritten expectations and shows how negative emotions like workplace loneliness can disrupt trust and perceived reciprocity, thereby weakening relational bonds and OCB. Building on these theories, the literature review formulates hypotheses: H1, SL positively relates to OCB; H2, relational identification mediates the SL–OCB link; H3, POS mediates the SL–OCB link; H4, SL influences OCB via sequential mediation through relational identification then POS; and H5, workplace loneliness negatively moderates the mediated relationship, weakening SL’s indirect effect on OCB. The review also notes limited prior work integrating relational identification with POS in a sequential framework and highlights evidence suggesting that stronger relational identification can enhance perceptions of leader and organizational support, thereby encouraging OCB.

Design and sample: A cross-sectional survey was conducted in Zhejiang Province, China, an area dense with private enterprises. Convenience sampling was used. After expert pretest (3 experts) and pilot test (30 respondents), email invitations went to 20 companies; seven agreed to participate. Data collection via web link/QR code over six weeks yielded 335 valid responses (187 male, 148 female). Measures: All scales used established instruments. Servant leadership (independent variable) used Liden et al. (2015) 7-item short form (α=0.893). OCB (dependent variable) used Ahmad and Zafar (2018) 5-item scale (α=0.818). Relational identification used Sluss et al. (2012) 4-item scale (α=0.809). POS used a 6-item scale adapted from Eisenberger et al. (1990, 1997) via Albalawi et al. (2019) (α=0.897). Workplace loneliness used Lam and Lau (2012) 4-item instrument adapted from the UCLA Loneliness Scale (α=0.939). Analysis: SPSS was used for descriptives and normality; SmartPLS for PLS-SEM. Measurement model assessment included indicator reliability, composite reliability, convergent validity (AVE>0.50), and discriminant validity (HTMT<0.85). Common method bias was assessed via full collinearity VIFs (<3.3). Model fit indices reported were SRMR and NFI (SRMR=0.066, NFI=0.836). Structural model tested hypothesized paths using bootstrapping with 10,000 resamples. Collinearity diagnostics were acceptable (inner VIFs ≤2.436).

- Descriptives: Means of constructs exceeded 3.0; skewness and kurtosis within acceptable ranges. Correlations among SL, relational identification (RI), POS, and OCB were positive and significant; workplace loneliness (WL) showed small negative correlations with RI and nonsignificant with others. - Measurement model: All outer loadings ≥0.708; composite reliability acceptable across constructs; AVE>0.50 established convergent validity. HTMT values met the <0.85 criterion, confirming discriminant validity. - Model fit and CMB: SRMR=0.066 (<0.08) and NFI=0.836 (>0.80) indicated satisfactory fit. Full collinearity VIFs <3.3 suggested CMB not a concern. - Hypothesis tests (PLS-SEM, 10,000 bootstraps): • H1 (SL→OCB): β=0.296, t=5.725, p<0.01, supported. • H2 (SL→RI→OCB): Indirect β=0.146, t=4.445, p<0.01, supported (partial mediation). • H3 (SL→POS→OCB): Indirect β=0.105, t=3.258, p<0.01, supported (partial mediation). • H4 (SL→RI→POS→OCB): Sequential indirect β=0.045, t=2.812, p<0.01, supported. • H5 (WL moderates SL→RI affecting the sequential mediation): Interaction β=−0.025, t=2.206, p<0.05, supported; the positive SL→RI effect is stronger at low WL and weaker at high WL. Overall, all five hypotheses were supported, evidencing both parallel and sequential mediations and a negative moderation by workplace loneliness.

Findings show servant leadership directly enhances OCB and indirectly does so via relational identification and POS. Servant leaders’ caring, supportive behaviors strengthen followers’ relational identification, aligning follower goals with leadership aims and encouraging discretionary contributions. Employees also infer organizational care from leaders’ support, boosting POS and reciprocation via OCB in line with SET. Notably, a sequential mechanism operates: SL increases relational identification, which elevates POS, culminating in higher OCB. However, workplace loneliness diminishes the impact of SL on relational identification, weakening the mediated pathways to OCB. Practically, organizations should develop servant leadership capabilities, cultivate high-quality leader–member relationships, and bolster systemic signals of support. Addressing employees’ emotional states—especially reducing workplace loneliness via team-building, open communication, and social support—can help ensure the benefits of servant leadership are realized.

The study offers a theoretically grounded moderated mediation model linking servant leadership to OCB. Servant leadership promotes OCB through relational identification and POS, including a verified sequential pathway. Workplace loneliness weakens these effects, clarifying when SL may not uniformly translate into higher OCB. These insights extend theory by integrating SET, SIT, and PCT and provide practical guidance for enhancing citizenship through leadership, support, and attention to employees’ emotional states. Future research should further probe additional reasons negative emotions may attenuate SL’s effects and how leaders can adapt to mitigate such impacts.

- Focus on relational identification at the interpersonal level; other identifications (e.g., organizational, leader identification) were not examined. - Sample limited to Zhejiang Province and a convenience sample; generalizability is constrained. Future work should expand regions and sample size. - Cross-sectional design precludes strong causal inferences; longitudinal designs are recommended. - Contextual factors (organizational culture, industry, geography) were not incorporated; future work should integrate these to enrich understanding. - Although well-established scales were used, future research could refine measures tailored to the Chinese context and employ qualitative approaches for deeper insights.