How Transformational Leadership Motivates Employee Involvement: The Roles of Psychological Safety and Traditionality

This study aims to explore how transformational leadership affects employee job involvement. Using the cross-sectional survey method, a questionnaire survey was conducted among 205 teachers in 15 universities located in Southern China. Hierarchical regression and bootstrapping technique were combined to test the hypothesis. The results reveal that psychological safety mediates the link between transformational leadership and job involvement. Moreover, the traditionality of university teachers strengthens the effect of psychological safety on job involvement. This study expands job involvement literature by examining the integrative effects of multiple resources, and all resources are aligned to extend the pool of resources owned by university teachers. These findings also bring some enlightenment on how to motivate university teachers to engage actively in their work.


Introduction
In the past few decades, job involvement has attracted a lot of attention in fields of management and applied psychology (e.g., Ju et al., 2021;Yeh, 2021). Job involvement was defined as the extent to which employees psychologically identify with their work (Brown, 1996). Highly involved employees are likely to devote more resources to their job, and pursue higher job performance (Saks, 2006). Given the potential significance of job involvement, many studies focused on identifying situational and personal factors that affect this positive work status (Sheikh et al., 2013;Yeh, 2021). Saks and Gruman (2014) proposed an integrated model, which suggests that leader behavior is a critical factor to promote employee involvement. Conservation of resource (COR) theory provides a critical perspective to understand employee involvement because it emphasizes the role of resources in reducing work-related stress and satisfying job demands (Hildenbrand et al., 2018). COR theory also highlights the importance of work context as offering various resources (Halbesleben et al., 2014), and leaders are always noted for providing followers with multiple valuable resources (Akdere & Egan, 2020;Piccolo & Colquitt, 2006). Accordingly, this study mainly focuses on how supervisors affect employee involvement. Transformational leadership has been convincingly shown to have a significant impact on employee outcomes (Judge & Piccolo, 2004), such as organizational commitment (Bashir et al., 2020), job satisfaction (Khan et al., 2020), and task performance (Lai et al., 2020). Surprisingly, there are relatively few studies on how transformational leadership affects employee involvement, especially in nonprofit organizations (Rana et al., 2016). In fact, employees in nonprofit organizations also need to engage in their work (Borst et al., 2020;Park et al., 2018). For example, universities are generally viewed as special nonprofit organizations, which plays an important role in cultivating talents (Thurlings et al., 2015;Zeng & Xu, 2020). Moreover, although transformational leadership has been suggested to affect job involvement through, for example, self-efficacy (Tims et al., 2011) and meaning in work (Ghadi et al., 2013), the influential mechanism of this relationship need to be further explored (Amor et al., 2020). Saks and Gruman (2014) proposed that the perceived psychological safety is an important explanation mechanism for leaders to influence employee engagement. Accordingly, we focus on how transformational leadership affects university teachers' job involvement in this study.
Specifically, we propose that transformational leadership motivates university teachers' job involvement due to its positive effect on psychological safety, and this relationship will be influenced by individual's personality (Do & Minbashian, 2020). Namely, transformational leadership does not fulfill all individuals' needs equally. This study focuses on the Chinese traditionality, and high traditionalists are more likely to fulfill responsibilities according to role norms than low traditionalists (Farh et al., 1997). Since subordinates' traditionality may strengthen the effect of leadership behaviors on job performance (Liu et al., 2013), we thus expect that transformational leadership has a more positive and indirect impact on job involvement via psychological safety for individuals with high-level traditionality.
The contributions of this research include the following aspects. First, we shed light on the transformational leadership-job involvement relationship by examining psychological safety as a potential mediating mechanism. This effort provides empirical support for the integrated model of employee engagement (Saks & Gruman, 2014). Second, this study incorporates individual's personality into the relationship between psychological safety and job involvement by examining traditionality as an important boundary condition, highlighting individual differences in the selection and implementation of contextual and personal resources. Third, most empirical studies on job involvement are conducted in business organizations, which limits the generalizability of research conclusions. Rana et al. (2016) found that Leadership styles are important predictors of job involvement among teachers. Taking university teachers as the research sample, this study will enrich the understanding of employee involvement in different organizational contexts.

Theoretical Background and Hypothesis Development
To better understand how transformational leadership affects university teachers' job involvement, we draw on COR theory (Halbesleben et al., 2014;Hobfoll et al., 2018), which has been proven useful for understanding employee involvement (e.g., Chen & Fellenz, 2020;Malik & Garg, 2020). The basic principle of COR theory is to motivate individuals to obtain new resources and protect existing resources (Halbesleben et al., 2014), involving objects, personal characteristics, or energies (Hobfoll, 1989). Ten Brummelhuis and Bakker (2012) further divided resources into personal and contextual resources. Personal resources are close to one's self (i.e., energy or attention), whereas contextual resources depend on social context in which an individual is located (i.e., working conditions). Also, they pointed out that contextual resources bring about different work outcomes due to their influence on personal resources, and these relationships will be affected by personal key resources.
According to COR theory, we expect that transformational leadership, as a contextual resource, is associated with increased job involvement due to the linking mechanism of psychological safety. Although we conceptualize psychological safety as a personal resource because it refers to the degree to which an individual perceive he is trusted and can trust others around him/her (Kahn, 1990), we also acknowledge that individuals use resources differently. Accordingly, we propose that the impact of transformational leadership on job involvement via psychological safety is strengthened by the traditionality of university teachers, which constitutes a key resource that affects the use of external resources.

Transformational Leadership and Job Involvement
Transformational leadership focuses on increasing employees' participation in organizational goals (Bass, 1985), which generally involves inspirational motivation, idealized influence, individual consideration, and intellectual stimulation (Bass & Avolio, 1990b). Employees who work under transformational leadership are motivated by their leaders, and their performance exceeds their expectations (Yukl, 1989). The literature shows that transformational leadership plays a vital role in motivating employees to engage in their work (e.g., Lai et al., 2020;Tims et al., 2011). Job involvement was defined as the extent to which employees psychologically identify with their work (Brown, 1996), which can help us understand how people think of their job (Ju et al., 2021). Saks and Gruman (2014) proposed that leadership styles (e.g., LMX, empowering leadership, and transformational leadership) are critical sources for employee involvement. These positive leadership styles transform into corresponding leadership behaviors, which affect employees' job involvement through various shared pathways (Decuypere & Schaufeli, 2020). Previous studies also showed that employees will engage in their work when their individualized needs are considered by their leaders (e.g., Harter et al., 2002;Sheikh et al., 2019).
According to COR theory, job resources such as supervisory support trigger a motivational process that leads to job involvement. Employees who have sufficient resources will be actively engaged in work (Halbesleben et al., 2009). Support from supervisors is generally seen as an important work resource (Bakker & Demerouti, 2007), which is a critical predictor for employees' job involvement (Bakker & Demerouti, 2008). Given that transformational leadership is always highlighted as a contextual characteristic (Braun et al., 2013), we conceptualize it as a contextual resource that should expand the pool of resources owned by employees (Halbesleben, 2006). We can expect that university teachers under transformational leadership are actively involved in their work to gain sufficient contextual resources that leaders provide. Specifically, transformational leaders will care about employees' individual needs and motivate them with a higher purpose and mission (Hildenbrand et al., 2018), which constitutes a resource that promotes employees to perform beyond their expectations. Employees who receive support and guidance from transformational leaders are likely to perceive their work as challenging and then will actively participate in tasks (Tims et al., 2011). Accordingly, we propose: Hypothesis 1 (H1) Transformational leadership is positively related to university teachers' job involvement.

The Mediating Role of Psychological Safety
Psychological safety was defined by Kahn (1990) as "feeling able to show and employ one's self without fear of negative consequences to self-image, status or career" (p. 708). This definition mainly focuses on individuals' perception, which highlights the importance of perceiving minimized interpersonal risk (Frazier et al., 2017). As a positive personal resource, psychological safety is a major explanation mechanism for how contextual resources affect various workrelated outcomes (ten Brummelhuis & Bakker, 2012). Following the integrated model of employee involvement (Saks & Gruman, 2014), we expect that psychological safety is a potential mediating mechanism for why transformational leadership motivates employee involvement.
The positive leadership styles signal to employees that they are valued, which is a critical source of psychological safety (Yin et al., 2019). It is reasonable that supervisory support is an important predictor of psychological safety (Frazier et al., 2017). In keeping with COR theory, transformational leadership can extend the pool of resources owned by employees (Halbesleben, 2006), which can facilitate the establishment of psychological safety as a personal resource (Detert & Burris, 2007). Specifically, transformational leaders tend to stimulate intellectually and encourage followers to challenge assumptions (Zhang et al., 2011), which provides a safe working environment where employees are encouraged to express themselves (Carmeli et al., 2014). Moreover, individualized consideration represents the notion that the specific, individualized needs of each need to be considered (Avolio & Bass, 1995). This means that two-way communication is encouraged by transformational leaders (Bass & Riggio, 2006). That is, individuals will perceive high levels of psychological safety when their leaders engage in individualized consideration and intellectually stimulating behaviors (Detert & Burris, 2007). As Edmondson (1999) stated, employees under leaders who are supportive, coaching-oriented, and open to questions and challenges should enhance their psychological safety.
Additionally, some research also suggested work engagement is an important outcome of psychological safety (Frazier et al., 2017). Kahn (1990) identified psychological safety as a necessary condition for employee involvement, and subsequent works have provided support for Kahn's argument (e.g., Edmondson & Lei, 2014;May et al., 2004). Therefore, we can expect that psychological safety is a personal psychological resource that enables employees to actively engage in their job tasks. The perception of psychological safety reflects a favorable emotional experience, which helps employees devote more effort to their work (Christian et al., 2011). Thus, psychological safety represents a resource that enables employees to meet job demands well, thereby promoting job involvement (Demerouti et al., 2001). Moreover, psychological safety may also bring about new resources (Singh et al., 2018). Employees who work in a safe environment are willing to share knowledge and establish connections with others, thereby harvesting various valuable resources, such as support from leaders or colleagues.
In summary, we propose that transformational leadership as a contextual resource enhances job involvement via its positive influence on psychological safety as a personal psychological resource. The behaviors of individualized consideration and intellectual stimulation related to transformational leadership can make employees perceive a safe working environment. This kind of good emotional experience related to psychological safety should, in turn, encourage employees to engage actively in their work. As shown above, we make the following hypothesis: Hypothesis 2 (H2) Psychological safety mediates the transformational leadership-job involvement relationship.

The Moderating Role of Traditionality
Contingency theory points out that the effectiveness of leadership is affected by employee characteristics or personality (Fiedler, 1964). Some literature discusses the role of followers' personalities in the reaction to transformational leadership (Buil et al., 2019;Frieder et al., 2019). To further explore whether all individuals can benefit equally from various resources in terms of increased job involvement, the traditionality of university teachers is considered as a boundary condition. This also provides empirical support for the argument that the impact of psychological safety on employee involvement is influenced by individual differences (Saks & Gruman, 2014).
Although Chinese societies have experienced great changes as a result of economic reforms, traditional values still have a critical influence on managerial practices (Fu et al., 2010). Traditional value was defined by Schwartz (1992) as the acceptance of traditional ideas and customs. Yang et al. (1989) developed the construct of Chinese traditionality and identified five dimensions, namely obedience to authority, filial piety, conservation and patience, fatalism and defense, and male dominance. This construct was introduced into the field of organization and management, mainly focusing on the dimension of obedience to authority (Farh et al., 1997). Farh et al. (2007) believed that high traditionalists are likely to fulfill the expectations and responsibilities given to them by social roles, while low traditionalists mainly follow the principle of inducing-contribution balance. In keeping with COR theory, we argue that traditionality constitutes a key resource, which has a great influence on the use of other resources. Halbesleben et al. (2014) proposed that the value of resources depends on the degree to which individuals match their respective resources. This argument is consistent with the concept of a key resource, which determines how the individual chooses and utilizes resources and affects the efficiency of resource usage (ten Brummelhuis & Bakker, 2012). From this perspective, we argue that traditionality affects the extent to which individuals benefit from psychological safety, leading to increased job involvement for individuals with high levels of traditionality. Specifically, high traditionalists are more sensitive to leaders' authority (Hui et al., 2007) and likely to accomplish assigned tasks effectively (Liu et al., 2013), which reflects a stable individual trait that is independent of situational factors when they engage in proactive behaviors. That is, employees with stable individual traits will perceive the working environment as being psychologically safe (Chan, 2006). Moreover, high traditionalists are likely to meet the expectations of social roles (Farh et al., 1997), which is similar to Kahn's (1990) argument that employees' engagement is the harnessing of organization membership to perform their work roles. This resulting fit between the perception of psychological safety and high traditionality should lead to increased job involvement. On the contrary, low traditionalists are less sensitive to hierarchical relationships between supervisors and subordinates (Hui et al., 2007), and are mainly based on the principle of inducing-contribution balance (Farh et al., 2007). That is, low traditionalists place more emphasis on cognitive than emotional influences, and they will benefit less from the perception of psychological safety. We further propose a moderated mediation model (see Figure 1). Therefore, we propose: Hypothesis 3 (H3) Traditionality moderates the indirect effect of transformational leadership on job involvement via psychological safety in such a way that the indirect effect will be greater for individuals with high-than lowlevel traditionality.

Procedure and Participants
We tested our hypotheses using data collected from from 15 universities in Southern China. Before beginning the formal survey, we obtained ethical approval from our institutional review boards. We recruited survey participants using snowball sampling through their social network of acquaintances. Among the 300 questionnaires distributed through the professional online questionnaire platform, we received 205 valid answers. At the end of the investigation, each participant was paid 20 RMB through WeChat (about $ 3.098 USD).

Measures
All measurement items are selected from mature scales, and we made some small modifications to the measurement items to make them more suitable for the research situation. Except for the control variables, all variables are measured on a seven-point Likert scale.

Transformational Leadership
We measured transformational leadership using four items for inspirational motivation and four items for individualized consideration (Bass & Avolio, 1990a). As has prior research (Detert & Burris, 2007), we combined these two subcomponents into an overall measure of transformational leadership. Sample items are "Gets me to look at problems from many

Psychological Safety
The three items validated by Detert and Burris (2007) based on the earlier work of Edmondson (1999) were used. A sample item is "It is safe for me to speak up around here." Cronbach's alpha was .85 in this study.

Job Involvement
Job involvement was measured using Kanungo's (1982) 10-item scale. A sample item is "The most important things that happen to me involve my present job." Cronbach's alpha was .93 in this study. Traditionality. Traditionality was measured with five items developed by Farh et al. (1997). A sample item is "The best way to avoid mistakes is to follow the instructions of a senior." Cronbach's alpha was .78 in this study.
Control variables. Previous research has shown that job involvement is affected by sociodemographic variables (e.g., Ju et al., 2021;Yeh, 2021). Therefore, the gender, age, education level, and job title of the participants are controlled in this study.

Common Method Bias Tests
If all the variables are collected from the same respondents, common method bias will be a potential problem (Hew et al., 2018;Loh et al., 2022;Wong et al., 2020). We sent an e-mail to the participants to explain the research purpose and emphasize the anonymity of the survey. We adopted two statistical analyses to test common method bias (Lew et al., 2020). First, Harman's single-factor analysis was used to test the common method bias, and result showed that variance explained by the single factor is 36.55% (below 50%). Second, the alternative method proposed by Liang et al. (2007) also confirms that common method bias is not a serious concern, as the ratio of the average substantive variance to the average method variance is relatively small. Accordingly, we can infer that common method bias is not very serious in this study.

Confirmatory Factor Analysis (CFA)
To assess the appropriateness of our measurement model, we conducted a CFA through Mplus 7.4 software. The results indicated that the hypothesized four-factor model yielded fit indexes within an acceptable range (χ 2 = 672.63, df = 291, χ 2 /df = 2.31, TLI = 0.89, CFI = 0.88, RMSEA = 0.08, SRMR = 0.08), fit the data better than any of other models (see Table 1). Although the TLI and CFI (less than 0.90) of the four-factor model in this study does not meet the standard in traditional studies, relevant study (Taasoobshirazi & Wang, 2016) believes that whether the acceptance of the model cannot completely rely on these indicators, and it is also acceptable if the fit indexes of the hypothetical model is significantly better than other models (Liang et al., 2018). These results suggest that these variables are distinguishable empirically.

Descriptive Statistical Analysis
The means, standard deviations, and correlations of transformational leadership, psychological safety, traditionality, and job involvement are presented in Table 2. Transformational leadership is positively correlated with psychological safety and job involvement (p < .05), and psychological safety is positively correlated with job involvement (p < .05). Furthermore, we make an assessment of the normality of these variables by nonparametric tests. The results show that the variables in this study conform to normal distribution. These results provided initial support for our hypotheses.

Hypothesis Testing
The correlation coefficients of all variables are below .5, suggesting there is no serious collinearity in this study. As shown in Table 3, we only put control variables in M1, and added transformational leadership to M2. The results showed that transformational leadership had a significant positive impact on job involvement (β = .48, p < .01). Therefore, Hypothesis 1 was supported. To test the mediating role of psychological safety, we combined hierarchical regression analysis and bootstrapping analysis. Following Baron and Kenny's (1986) method, transformational leadership had a significant positive effect on psychological safety (β = .41, p < .01), as shown in M9 of Table 3. Meanwhile, psychological safety was positively associated with job involvement (β = .455, p < .01) in M3. The results for M4 further showed that psychological safety had a significantly positive impact on job involvement (β = .357, p < .01) when both transformational leadership and psychological safety were regressed on job involvement. Based on these results, psychological safety partially mediated the relationship between transformational leadership and job involvement. Therefore, Hypotheses 2 was supported. In this study, we proposed that traditionality moderates the relationship between psychological safety and job involvement. We applied Muller et al.'s (2005) approach for testing moderated mediation and found that transformational leadership was significantly associated with work engagement (β = 0.50, p < .01) in M5. Meanwhile, transformational leadership had a significantly positive impact on psychological safety (β = 0.34, p < .01) in M8. Further, psychological safety was significantly associated with job involvement (β = .23, p < .01) in M6. Lastly, the interaction between psychological safety and traditionality was significantly associated with job involvement (β = .13, p < .05) in M7. Based on these results, Hypotheses 3 was supported. To describe the moderating effect of traditionality, we plotted the statistically significant interaction. The relationship between psychological safety and job involvement was stronger when traditionality was above average (Mean + 1 SD) than when it was below average (Mean -1 SD) (see in Figure 2).
According to Edwards and Lambert (2007), we further adopted the moderated path analysis to test the moderated mediation effect (H3). The indirect effect of transformational leadership on job involvement through psychological safety varied significantly according to the different levels of traditionality (Δβ = .18, p < .01). Specifically, the indirect effect of transformational leadership on job involvement through psychological safety was stronger when traditionality was high (β = .20, p < .01) than when it was low (β = .02, n.s.). Thus, H3 is supported.

Conclusions and discussion
Drawing on COR theory, we proposed a resource model that considers the joint effect of transformational leadership, psychological safety, and traditionality on job involvement. Our findings support this proposed model and the premise that  transformational leadership is significantly associated with university teachers' job involvement, and this relationship is mediated by psychological safety. Moreover, the traditionality of university teachers strengthens these relationships, indicating that the effect of transformational leadership on employee involvement depends, to a certain extent, on subordinate's personality.

Theoretical Implications
This study provides several theoretical contributions to the related literature. Drawing on COR theory, this study contributes to the consideration of the integrative effect of various resources on job involvement. All resources (i.e., transformational leadership, psychological safety, and traditionality) are aligned in extending the pool of resources owned by university teachers, which highlights their relevance for job involvement and provides a coherent theoretical model. Moreover, we revealed that psychological safety is a mediating variable of the transformational leadership-job involvement relationship, which expands the theory by showing that contextual resources (e.g., transformational leadership) contribute to various work outcomes (e.g., job involvement) due to its positive influence on personal resources (e.g., psychological safety). We also provided an empirical test for Saks and Gruman's (2014) proposition that psychological safety is one of the important psychological conditions for employee involvement.
Finally, we identified individual's personality as a critical boundary condition and showed that job involvement's increasing effect on psychological safety depends on the traditionality of university teachers. According to COR theory, we revealed that transformational leadership constitutes a stronger resource effect on work engagement via psychological safety for individuals with a high-level of traditionality. Thus, we contributed to the theory by showing that individual difference is an important boundary condition in the relationship between psychological condition and employee involvement.

Practical Implications
Employee involvement is an important factor in the sustainable development of an organization (Rich et al., 2010). This study contributed to managerial practice by revealing transformational leadership as a pathway to increasing resources of job involvement. That is, appropriate ways to activate the motivations of job involvement are to train leaders in terms of transformational leadership (Dvir et al., 2002) and to consider the use of relevant instruments of transformational leadership as tools in leader selection. Moreover, this study showed that transformational leadership can enhance subordinates' psychological safety. Therefore, organizations and leaders should strive to build a safe work environment and encourage employees to speak up or express themselves freely. Lastly, we also revealed that individual's personality (e.g., traditionality) may affect the effectiveness of leadership. Managers need to be aware of cultural boundaries when using western theories, particularly in a Confucian society such as China, where individuals tend to have high-level traditionality (Wang & Zhang, 2020). Organizations should recruit employees with high level of traditional values, so as to translate valuable resources into high performance.

Limitations and Future Research
Several limitations also need to be acknowledged. First, we adopted a cross-sectional research design in this study. One of the limitations of this method is that it is difficult to identify causal relationships among variables. Longitudinal field studies can be used to gather data repeatedly over a period, clarifying whether causal relationships exist. Second, the data of this study was self-reported and single-sourced, which may cause a CMV issue (Podsakoff et al., 2003). Future research can collect information about job involvement by supervisor evaluation. Third, this study mainly focuses on university teachers in China. The understanding of traditional values may differ across industries or countries, indicating some caution concerning the generalizability of the findings. Future studies can validate our findings with cross-cultural or multi-industry samples.

Transformational Leadership
My supervisor gets me to look at problems from many different angles. My supervisor challenges me to think about old problems in new ways. My supervisor challenges me to be innovative in my approach to work assignments. My supervisor encourages me to be an independent thinker. My supervisor commends me when I achieve my goals. My supervisor gives me positive feedback when I perform well. My supervisor gives me special recognition when my work is very good. My supervisor ackownledges improvement in my quality of work.

Psychological Safety
If you make a mistake on this team, it is often held against you (reverse scored). It is safe to take a risk on this team. No one on this team would deliberately act in a way that would undermine my efforts.

Job Involvement
The most important things that happen to me involve my present job. To me, my job is only a small part of who I am. I am very much involved personally in my job.

Traditionality
When people are in dispute, they should ask the most senior person to decide who is right. Those who are respected by parents should be respected by their children. The best way to avoid mistakes is to follow the in-structions of senior persons. Before marriage, a woman should subordinate herself to her father; after marriage, to her husband. The chief government official is like the head of a household, the citizen should obey his decisions on all state matters.

Authors' Note
Guangyi Xu is now affiliated to School of Business, Ningbo University, Ningbo, China.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors thank for the funding from the National Science Foundation of China (71872066, 72172047). Guangdong Philosophy and Social Sciences Planning Project (GD21CJY02). Guangdong Educational Science Planning and Research Project (2020GXJK405). The Innovation and Strengthening University Project of Guangdong Pharmaceutical University (2018WTSCX053).