Voice behaviour refers to employees’ concerns, thoughts, and recommendations regarding work-related issues, intending to change or improve the current work situation (Liang et al., 2012). It is a discretionary, prosocial, and constructive extra-role behaviour. The predictors of voice behaviour include individual characteristics (Starzyk & Sonnentag, 2019), attitudes, and emotions (Wang et al., 2022), as well as leaders’ style, behaviour, and personality (Weiss et al., 2018).

Voice behaviour is optimized when employees perceive a strong fit between themselves and their job or organisation (Erdogan et al., 2020), and when their work calling aligns with that of their leaders, although research on the latter remains insufficient.

Calling congruence involves a psychological process within the framework of person-environment fit theory (Cable & Edwards, 2004), by which the alignment in calling between leaders and employees can bolster employee perceived insider status. Perceived insider status refers to an individual’s perception of belonging as either an insider or outsider within the organisation, reflecting their relationship with the organisation and the degree of personal space and acceptance they feel they have acquired within it (Masterson & Stamper, 2003). With insider status, employees feel like they are part of the work team and act according to leadership expectations to help their organisations thrive (Zhang et al., 2022). The psychological process from perceived insider status to voice behaviour involves the conservation of resources theory, indicating that employees with a stronger sense of insider status are more likely to engage in voice behaviour. The conservation of resources theory posits that individuals must acquire and preserve personal strengths and social connections, with those possessing more resources being more likely to invest in resource-enhancing activities (Hobfoll et al., 2018). On one hand, individuals use their existing resources to protect themselves, while on the other hand, they actively construct and maintain resources to cope with potential resource losses (Hobfoll et al., 2018). Employees with a strong perception of insider status possess more resources, which they may leverage to build social networks that help them integrate more effectively into their work (Li et al., 2014; Liu et al., 2024). To preserve and enhance these resources, they may engage in voice behaviour to align with leaders’ expectations, in order to achieve higher performance evaluations and social standing, ultimately leading to resource enhancement.

What has been lacking from the literature is an examination of how dyadic calling congruence between employee and leader impacts follower voice behaviour through perceived insider status.

According to the level of leader calling and employee calling, we propose four different matching scenarios as shown in Table 1: leader high-employee high, leader high-employee low, leader low-employee low, leader low-employee high. Leader high-employee high and leader low-employee low fall under the category of congruence, leader high-employee low and leader low-employee high fall under that of incongruence.

This study will address the following: first, between the two congruence scenarios, whether perceived insider status is higher in a high–high one than in a low–low one; and second, between the two incongruence scenarios, whether perceived insider status is higher when employees are at a higher level of calling than leaders in comparison to the opposite.

Person-environment fit theory refers to the compatibility that arises when individual characteristics align with those of the work environment (Cable & Edwards, 2004). With a work calling fit, supplementary fit occurs between leaders and employees. Under different matching conditions, there can be either gains or losses of resources, which can affect employee behaviour. In case of a high degree of calling congruence between the employee and the leader, an employee may exhibit greater cognitive situational adaptability (Cable & Edwards, 2004). This alignment of calling promotes the formation of a “value consensus” and a shared communication system between the leader and subordinates, enhancing predictability in each other’s behaviour, reducing ambiguity, and minimizing resource consumption. This facilitates supplementary fit for resource gains, thereby encouraging employees to acquire insider status.

In case of a low degree of calling congruence between the employee and the leader, although alignment exists, it fails to meet the supplementary fit compatibility condition. Specifically, the leader has a low sense of responsibility and an absence of purpose, meaning, and work internal motivation, which will be transmitted to subordinates during interactions (Xie et al., 2019). Moreover, subordinates also exhibit a low sense of purpose, lacking internal motivation and meaning in their work, making it difficult to handle daily varied and complex tasks (Tosti-Kharas et al., 2024), inevitably diminishing subordinate perceived insider status.

In cases of incongruence where leaders have a low sense of calling but subordinates have a higher one, motivated subordinates may invest more energy into work, complete unattractive tasks, and derive enjoyment from them (Tosti-Kharas et al., 2024), potentially elevating their perceived insider status. In contrast, in situations of incongruence where leaders have a high calling but subordinates have a low calling, leaders may exhibit strong prosocial behaviour and motivation to complete organisational tasks, while employees are less likely to actively engage in their work (Kim et al., 2023; Lian et al., 2021; Sawhney et al., 2020), potentially reducing their perceived insider status.

The participants were 43 leaders matched with 173 subordinates in China’s services industry. Among the leaders, approximately 62.79% were male, the average age was 38.37 (SD = 7.90) years. Among the employees, approximately 68.21% were male, and the average age was 25.23 (SD = 9.36) years.

Our study used internationally established scales with translation/back-translation procedure (Brislin, 1980) for English-based measures into Chinese. All scales were scores on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree).

The research ethics committee of Henan University approved the study. Participants consented to the data collection at two different time points. We ensured the confidentiality of participants’ data. They completed the surveys online.

This study conducted a discriminant validity test among the three factors of employee evaluation, and the analysis results are shown in Table 2. Among the various factor models, the three-factor model showed the best fit (χ² = 449, df = 321, RMSEA = 0.05, CFI = 0.97, IFI = 0.96, SRMR = 0.04) compared to the other two models, indicating good discriminant validity among the three variables in this study.

To test the hypotheses, we used the polynomial regression and response surface methodology (Edwards & Parry, 1993). This method examines the exact nature and extent to which congruence between predictor variables relates to the outcome variables (Edwards & Cable, 2009). By analyzing the three-dimensional surfaces, we gained a clearer interpretation of the data. The specific analysis is the mediator variable-perceived insider status (PIS) was regressed on the control variables and the five polynomial terms, that is, leader calling (LC), employee calling (EC), leader calling squared (LC²), leaders calling times employee calling (LC × CH), and employee calling squared (EC²). The values of LC and EC were mean-centered. The regression equation is as follows:

PIS = b₀ + b₁LC + b₂EC + b₃LC² + b₄LC × EC + b₅EC² + e

To test the hypotheses, it is necessary to calculate the regression coefficients of the regression equation, as well as the slopes and curvatures of the response surface along the congruence line (X = Y) and the incongruence line (X = −Y). Specifically, the slope of the congruence line (b1 + b2) is used to test Hypothesis 1, when the value of b₁ + b₂ is significant and positive, Hypothesis 1 is supported. While the slope of the incongruence line (b1 − b2) is used to test Hypothesis 2, when the value of b1 − b2 is significant and negative, Hypothesis 2 is supported. For Hypothesis 3 testing, following the recommendation of Edwards and Cable (Edwards & Cable, 2009), the raw data are multiplied by regression coefficients in polynomial regression, and then summed to create a block variable, regression analysis and bootstrapping are used for testing.

All study variables’ means, standard deviations, and correlations are presented in Table 3. Leader calling and employee calling were significantly related to employees’ perceived insider status and voice behaviour.

Hypothesis 1 predicted that employee perceived insider status is higher when employees and leaders are aligned at a high level of calling as opposed to when they are aligned at a low level. As shown in Table 4, the slope along the congruence line is significant and positive (0.598, p < 0.01), indicating that the high-high congruence condition has higher perceived insider status than the low-low congruence condition. The response surface in Figure 2 also indicates that PIS is higher at the rear corner (high/high congruence) than at the front corner (low/low congruence), thus supporting Hypothesis 1.

For testing the asymmetric incongruence effect (Hypothesis 2), the slope of the incongruence line (b₁ − b₂) can be used for detection. As shown in Table 4, the slope along the incongruence line is significant and negative (−0.702, p < 0.01), indicating that when an employee’s calling is higher than his or her leader’s, perceived insider status increases more sharply than it does when the employee calling is lower than the leader. This asymmetrical effect is also shown in Figure 2, in which perceived insider status is higher at the left corner than at the right corner, thus supporting Hypothesis 2.

To test the mediating effect of perceived insider status on the relationship between calling congruence and voice behaviour (Hypothesis 3), we used the block variable approach recommended by Edwards and Cable (Edwards & Cable, 2009). As shown in Table 5, the indirect effect was positive (0.51), and the 95% bias-corrected bootstrap confidence interval excluded zero (0.41, 0.61), providing support for Hypothesis 3.