ORIGINAL RESEARCH

published: 24 September 2020

doi: 10.3389/fpsyg.2020.537756

The Impact of Paradoxical

Leadership on Employee Voice
Behavior: A Moderated Mediation
Model
Xiyuan Li, Ying Xue* , Hao Liang and Dong Yan
Economics and Management School of Wuhan University, Wuhan University, Wuhan, China

Paradoxical leadership is associated with positive behavioral outcomes. However, the

link between paradoxical leadership and voice behavior is not comprehensively studied
in extant literature. This paper builds a theoretical model to reveal how paradoxical
leadership facilitates promotive and prohibitive voice behavior of employees, drawing
upon social cognitive theory and regulatory focus theory. We proposed a moderated
mediation model that employees’ voice behavior is related to paradoxical leadership
through self-efficacy and psychological safety. With data from 268 leader – employee
pairs of questionnaires, this study conducted a structural equation model to test the
conceptual framework. The results show that (a) leader’s paradoxical leadership related
to employee’s promotive and prohibitive voice behaviors positively; (b) employee’s self-
Edited by:
Jon Gruda, efficacy and psychological safety mediate the extent of effect the superior’s paradoxical
Maynooth University, Ireland leadership has on subordinate’s voice behavior; (c) the more obvious the subordinate’s
Reviewed by: promotion focus orientation, the stronger the mediating effect of self-efficacy; and
Ruiping Zhang,
Zhengzhou University, China
(d) the more obvious the subordinate’s prevention focus orientation, the weaker the
Hakan Erkutlu, mediating effect of psychological safety. These conclusions reveal the influencing
Nevşehir Hacı Bektaş Veli University,
mechanism of a superior’s paradoxical leadership on a subordinate’s voice behavior.
Turkey
It expands paradoxical leadership-related studies, enriches studies related to the field of
*Correspondence:
Ying Xue “leader – employee voice behavior,” and highlights the relationship between the duality of
xxyer77@gmail.com paradoxical leadership behavior on employees with different regulatory focus orientation
with a new perspective.
Specialty section:
This article was submitted to Keywords: paradoxical leadership, voice behavior, self-efficacy, psychological safety, regulatory focus
Organizational Psychology,
a section of the journal
Frontiers in Psychology
INTRODUCTION
Received: 25 February 2020
Accepted: 24 August 2020
Voice behavior (Van Dyne and LePine, 1998) has attracted plenty of attention in the field
Published: 24 September 2020
of organizational behavior. With the advent of artificial intelligence and the rapid growth of
Citation: technology, the business environment is continually shifting. In order to ensure that organizations
Li X, Xue Y, Liang H and Yan D
endure, it is necessary for superiors in an organization to take a broad and long-term view and
(2020) The Impact of Paradoxical
Leadership on Employee Voice
observe the overall situation. Primarily, every employee is supposed to utilize their own wisdom,
Behavior: A Moderated Mediation intelligence, and perform appropriate actions to provide clear advice because empirical studies have
Model. Front. Psychol. 11:537756. found that voice behavior is critical for sustainable development and organizational innovation
doi: 10.3389/fpsyg.2020.537756 (Morrison, 2014). However, voice behavior may be understood as challenging the status quo and

Frontiers in Psychology | www.frontiersin.org 1 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

as something that may have negative consequences (Son, 2019). In order to test the proposed model, a pairing study
Thus, many employees may be cautious about making a voice was conducted, and it was found that paradoxical leadership
behavior, be reticent, or are likely to reflect before offering positively relates to employees’ promotive and prohibitive voice
suggestions. As a result, many scholars focus on researching behavior, and employee’s self-efficacy and psychological safety
how to make employees actively engage in voice behaviors has a mediating role in such relationship. It was found that
(Weiss and Morrison, 2018). the more obvious the employee’s promotion focus, the stronger
Leadership has been recognized as a key factor in the analysis the mediating effect of self-efficacy, and the more obvious the
of voice behavior-related risks and decision-making (Giessner employee’s prevention focus, the weaker the mediating effect of
and Schubert, 2007; Liu et al., 2013). Numerous scholars have psychological safety. This study makes important contributions
studied the relationship between leadership and employee voice to the literature on employee voice, enriching the study related
behaviors (LePine and Van Dyne, 2001; Detert and Burris, to the field of “leader – employee voice behavior.” Additionally,
2007). Leadership factors can be divided into positive and this study constructed the influencing mechanism of paradoxical
negative. Positive factors include ethical, authentic paternalistic, leadership on employee voice behavior and expanded paradoxical
transformational, inclusive, servant leadership (Detert and leadership-related studies. Lastly, the research provided the study
Burris, 2007; Hsiung, 2012; Zhang et al., 2015; Chen and Hou, of the influence of the duality of paradoxical leadership behavior
2016; Yan and Xiao, 2016; Weiss et al., 2017) that can stimulate on employees with different regulatory focus orientation with a
employee voice behavior. The negative factors, such as abusive new perspective, which is of great significance for both theoretical
leadership (Xu et al., 2015), can be detrimental. However, these development and practical management.
studies focused on specific characteristics of leaders but did
not explain how they considered managing an evolving and
complicated environment. There are no relevant studies in this LITERATURE REVIEW AND
field that have examined the relationship between paradoxical HYPOTHESES
leadership and employee voice behavior (see Table 1). Many
authors had examined the influence of psychological safety on
Paradoxical Leadership and Employee
the relationship between leaders and employee voice behaviors
(Liang et al., 2012, see Table 1). However, the courage required Voice Behavior
to engage in voice behavior was determined by employees’ trust Zhang et al. (2014) conceptualized paradoxical leadership
in the organizational environment and their confidence in voice behavior based on the unity of opposites in the philosophy of
effectiveness, specifically the employee’s self-efficacy (Bandura, Yin-Yang, which has attracted the attention of many scholars
1985). Lastly, there is little agreement among academics about and practitioners. According to them, paradoxical leadership
how employees, with different regulatory focus tendencies, suggests that the behavior of management, while appearing
react to paradoxical leadership behaviors. Therefore, a gap paradoxical, is internally connected and could manage the needs
in the literature exists concerning the duality of paradoxical of organizational development and the individual needs of
leadership behavior with employees with different regulatory employees. Paradoxical leadership primarily has five dimensions:
focus orientations. (1) Combining self-centeredness with other-centeredness,
This research aims at assessing the relationship between (2) Maintaining both distance and closeness,
paradoxical leadership and employee voice behavior, examining (3) Treating subordinates uniformly while allowing
the mediated effects of psychological safety and self-efficacy and individualization,
the moderated effects of promotion and prevention focuses. (4) Enforcing work requirements while allowing flexibility,
It was propounded that the integrated social cognitive theory and
(Bandura, 1993) and regulatory fit theory (Higgins, 2000) explain (5) Maintaining decision control while allowing autonomy
how paradoxical leadership affects employee voice behavior. (Zhang et al., 2014, p. 541).
According to social cognitive theory, people observe and study
an environment, which affects their behavior by influencing their Voice behavior is defined as “promotive behavior that
objective, self-efficacy, and other factors. Paradoxical leadership emphasizes the expression of constructive challenge intended to
can consider the needs of all sides in an organization and improve rather than merely criticize,” and as “making innovative
motivate employees to solve challenging difficulties. By studying suggestions for change and recommending modifications to
paradoxical leadership behaviors, employees could make the standard procedures even when others disagree” (Van Dyne and
correct and timely decision, increasing their self-efficacy and, LePine, 1998, p. 109). It is an active behavior that reflects the
hence, provide advice. Therefore, in addition to psychological employees’ orientation to participate in discussing organizational
safety, this study selected self-efficacy as a mediating variable reform and to propose constructive opinions. However, it
of the model to explain why paradoxical leadership is related has occasionally been necessary for the employees to identify
to employee voice behavior. Individuals with a dissimilar difficulties and challenge others to improve organizational,
regulatory focus had their own regulating system, which included operational efficiency, as employee voice behavior entails certain
promotion and prevention focuses (Higgins, 1997). It is argued social risks (Milliken et al., 2003; Detert and Burris, 2007;
that the employee regulatory focus moderates the relationship Detert and Trevino, 2010; Grant, 2013). According to Liang et al.
between paradoxical leadership and employee voice behavior. (2012), voice behavior can be categorized as promotive and

Frontiers in Psychology | www.frontiersin.org 2 September 2020 | Volume 11 | Article 537756

 Li et al. Paradoxical Leadership and Voice Behavior

TABLE 1 | Empirical research on leadership and voice behavior.

Study Sample Antecedent Mediates/Moderates Control Variables

Variables

• Walumbwa and • 291 employees on 58 teams of a Ethical Leadership •/ • Gender, tenure and education
Schaubroeck (2009); government R&D institution in • Psychological Safety • Leader’s span, idealized influence
Walumbwa et al. Taiwan • Group Conscientiousness leadership
(2012), Chen and Hou • 894 employees and their 222 • Group size, idealized influence
(2016) immediate supervisors in a large leadership
financial institution in the
southwestern United States
• 316 nurses working at a large
medical center in the United States
• Wong et al. (2010); • 70 work groups with 404 Authentic • Leader–Member Exchange, • Gender, age, education, marital
Hsiung (2012), Liang salespersons of chain stores in Leadership Positive Mood status, organizational tenure, group
(2017) Taiwan • Organization-Based Self-Esteem size
• 457 subordinates and 90 • Trust in Manager, Work • Supervisor sex, education level,
supervisors from 90 functional work Engagement age, work group tenure, and
groups in a large public sector subordinate–supervisor dyad
organization in Taiwan tenure, group size, psychological
• 280 registered nurses working in safety
acute care teaching and community • Gender, employment status, type of
hospitals in Ontario hospital, education, specialty area
• Ning et al. (2012); Chan • 402 leader–subordinate paired data Paternalistic • Leader–Member Exchange, Status • Subordinates’ age, gender,
(2014), Zhang et al. from four companies located in Leadership Judgment educational level
(2015) Beijing in Mainland China • Information Sharing • Sex, age, education levels,
• 202 leader–follower dyads of a • Collectivism, Team Collectivism organizational tenure, dyadic
manufacturing factory in the leader–follower tenure
People’s Republic of China • Age, gender, education, and
• 31 supervisors and 245 beauticians marriage
of a large beauty chain in Shenzhen
and Zhuhai City in China
• Detert and Burris • 3153 crew members in 105 Transformational • Perceived Psychological Safety • Tenure, ethnicity, job type, gender,
(2007); Liu et al. (2010), restaurants Leadership • Social Identification, Personal hours worked per week, job shift,
Liu and Liao (2013); • 324 part-time MBA students from Identification proactive personality, employee
Duan et al. (2016) two universities in China • Leader Voice Expectation, Voice attitudes, having ideas
• 394 matched pairs of subordinates Role Perception •/
and their direct leader in • Power Distance, Structure Distance • Gender, age, and working tenure
Southeastern China with leader, psychological safety,
• 923 leader and follower dyads in felt responsibility to change
China and Australia • Time with leader, ethnicity
• Carmeli et al. (2010); • 126 participants participated in the Inclusive Leader Language Team size, total number of leaders’
Weiss et al. (2017), Qi hospital’s 1 day simulation training Leadership • Psychological Safety utterances
and Liu (2017) sessions • Caring Ethical Climate • Tenure, age
• 150 employees who were • Gender, age, education, tenure
employed in the R&D units of 8
knowledge-intensive organizations
that develop advanced
technological products
• 329 employees from 105 teams of
enterprises from six major cities in
China
• Walumbwa et al. • 473 matched supervisor Servant Leadership • Psychological Safety • Gender, age, politics status, tenure,
(2010); Hunter et al. subordinate dyads questionnaires • Procedural Justice Climate, highest educational
(2013), Yan and Xiao • 815 full-time employees from seven Employee Self-efficacy, Employee •/
(2016) multinational companies operating Commitment to the supervisor, • Number of employees per store
in Kenya Service Climate
• 337 executives from 385 stores in • Service Climate
the US
• Burris et al. (2008); Xu •152 employees working in the Abusive Leadership • Emotional Exhaustion • Subordinates’ gender,
et al. (2015) service industry in Macau, People’s • Psychological Attachment organizational tenure, psychological
Republic of China safety, affective commitment
• 499 restaurant managers in • Organization tenure, position
United States tenure, hours per week, distributive
justice, alternative employment,
have ideas, psychological safety,
futility

Frontiers in Psychology | www.frontiersin.org 3 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

prohibitive. Promotive voice behavior refers to the ability of Mediating Effects of Self-Efficacy
employees to freely express their innovative ideas and suggestions Self-efficacy is defined as “people’s beliefs of their capabilities to
to increase operational efficiency (Liang et al., 2012), while organize and execute the courses of action required to manage
prohibitive voice behavior occurs when employees present their prospective situations” (Bandura, 1995, p. 2). It is believed that
concerns and misgivings about problems associated with the the stronger the employee’s self-efficacy, the more likely they are
status quo of their organization and propose related suggestions to take action and be confident in their decisions (Okpozo et al.,
(Liang et al., 2012). 2017; Hans and Gupta, 2018). If an employee’s self-efficacy was
Leadership behavior is considered a key factor for employees high, they would make greater efforts and insist action is taken in
in their analysis of the risks caused by voice behavior (Detert and the face of possible negative results and social risks (Parker et al.,
Burris, 2007). Employees adopt different voice behavior strategies 2006). Therefore, a correlation exists between higher self-efficacy
for different leadership styles (Mowbray et al., 2015). This and the likelihood to provide advice.
study asserts that paradoxical leadership is related to employee According to Wang et al. (2015), leadership is one of the
voice behavior positively. Based on social cognitive theory, the most crucial external factors that influence and promotes
organizational environment influences employees’ cognition. The employee self-efficacy. Specifically, paradoxical leadership
management style, personality traits, and behavioral pattern balances the relationship between self-centeredness and other-
of superiors affect employees’ cognition and influence their centeredness by making employees feel respected, which
behaviors further (Detert and Burris, 2007). Additionally, strengthened their self-belief in their own abilities. This
paradoxical leadership makes the employees feel that their contributes to the formation of employee self-efficacy (Owens
work is respected and enthuses them to provide suggestions and Hekman, 2012). Besides, paradoxical leaders that maintain
(Liang et al., 2012). Paradoxical leaders maintain a sound an appropriate distance from employees establish a good
relationship with employees motivating them (Mowbray et al., relationship with them, facilitating greater leader – employee
2015), allowing them to be individuals, helping arouse their interaction. A higher number of interactions imply that more
passion for work and job engagement, and promoting them information was shared, hence further increasing employee
to offer advice and suggestions for organizational development self-efficacy (Gardner et al., 2005). Furthermore, paradoxical
(Wang and Jiang, 2015; Lapointe and Vandenberghe, 2018). leadership provides employees space to be more individual
Paradoxical leadership allows flexibility at work, which can with increased support, allowing for flexible working, while
facilitate employee voice behavior (Zhang et al., 2014), stimulate providing a fairer environment for both leaders and employees.
employees’ sense of responsibility, and increase their intrinsic These factors increase employee self-efficacy. Additionally,
motivation to provide advice (Zhang and Bartol, 2010). Hence, paradoxical leadership ensures that employers maintain
the five dimensions of paradoxical leadership influence employee control over decision-making while moderately empowering
voice behavior positively. employees. When employees perceive greater autonomy, their
According to social cognitive theory, the perceived attitudes intrinsic motivation for work strengthened, enhancing their
of employees influence their behaviors. If employees believe self-efficacy (Ou et al., 2014). Lastly, leaders’ own behavior
that managers are concerned about their benefits and interests, sets an example for employees (Walumbwa et al., 2011) and
then they would conduct extra-role performance behaviors, balances uncertain management environments, contradictions,
such as voice behavior (Walumbwa et al., 2010). Paradoxical and tensions, addressing numerous challenges and increasing
leadership balances the conflicts between organizational and employee self-efficacy.
employee development and allows employees to feel supported, Leadership behaviors influence employee self-efficacy and play
autonomous, and responsible. All these factors contributed to a vital role in the relationship between the leader and voice
employee voice behavior. behavior (Chen and Hou, 2016). Previous studies have proven
Lastly, paradoxical leadership does not balance the that self-efficacy mediates the relations between ethical and
contradictions in organizational management without flexibility, paternalistic leadership types and voice behavior (Wang et al.,
openness, and a willingness to learn (Zhang et al., 2015). In 2015). Given these discussions, the following hypotheses are
the process of having a good relationship with employees, acknowledged:
paradoxical leaders also tend to be role models for employees.
Employees can learn how to balance the contradictory Hypothesis 2a: Employee’s self-efficacy mediates the
relationship between benefits and risks of voice behavior and relationship between paradoxical leadership and promotive
promote them to express their opinions at an appropriate time. voice behavior.
Based on the above discussions, this study proposes the
following hypotheses: Hypothesis 2b: Employee’s self-efficacy mediates the
relationship between paradoxical leadership and prohibitive
voice behavior.
Hypothesis 1a: Superior’s paradoxical leadership positively
related to employee promotive voice behavior.
Mediating Effects of Psychological Safety
According to previous literature, employees might weigh
Hypothesis 1b: Superior’s paradoxical leadership positively or consider the advantages and disadvantages of voice
related to employee prohibitive voice behavior. behavior because of its potential negative effects. According

Frontiers in Psychology | www.frontiersin.org 4 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

to Kahn (1990), individuals who had high psychological safety and mediating variable. Hence, this study illustrates the following
were assured of their status and hence provided advice about hypotheses:
possible misunderstandings, negative impressions, and the
positive effects caused by the external environment, without Hypothesis 3a: Employees’ psychological safety mediates the
concern. With psychological safety, employees’ misgiving about relationship between paradoxical leadership and promotive
the possible negative effects of voice behavior was lessened, voice behavior.
and they felt assured about expressing their own opinion.
Furthermore, Liang et al. (2012) also indicated that the higher Hypothesis 3b: Employees’ psychological safety mediates the
the psychological safety based on organizational cognition was, relationship between paradoxical leadership and prohibitive
the more likely employees were to work on voice behaviors. voice behavior.
Based on the social cognitive theory, employees’ motives,
attitudes, and behaviors are influenced by the environment. Moderating Effects of Regulatory Focus
When making the decision to provide advice, employees carefully According to a study performed by Higgins (1997), there are two
evaluate whether the surrounding environment is favorable for self-regulating systems – promotion focus and prevention focus.
voice behavior, as employees were more likely to provide advice Individuals with unique self-concept and different regulatory
in a supportive and superior interpersonal environment (Carmeli focus orientation have a different emotional experience, cognitive
et al., 2009). When leaders were uninterested in employee voice style, and action. Individuals with greater promotion focus pay
behavior, employees sensed inconsistencies with their leader’s more attention to vision, expectation, and gains; are more
expectation and hence generated psychological insecurity. But sensitive to rewards; and reveal the pursuit of their “ideal self.”
when leaders presented more instructive, supportive, and open People who are with a prevention focus concentrated on duty,
characteristics, had a sound relationship with employees, and responsibility, and losses; are more sensitive to punishment; and
took their individual needs into account, they gained the trust reveal their pursuit of the rules and the realization of their “moral
of their employees and increased their psychological safety self.”
(Liang et al., 2012). According to regulatory fit theory, if external influences
Detert and Burris (2007) realized that the psychological matched the characteristics of individual regulatory focus
safety of employees mediated and explained the relations orientation, the external influence was strengthened. Specifically,
between the openness of the leader and the voice behavior if employees reasoned that the leadership style was consistent
of the employee. When employees felt their opinions were with their regulatory focus orientation, it may stimulate the
supported by their superiors, their psychological safety increased, generation of a fitted regulator, influence the employees’
encouraging them to express their views (Nembhard and cognition, and strengthen their behavioral motive. If employees
Edmondson, 2006). When leaders meet both the needs of believe that leadership style is inconsistent with their regulatory
the organization and employees, they respect the employees’ focus orientation, it would be challenging to generate regulatory
individuality, maintain their self-esteem and confidence, and fit and the employees’ cognition, and managers would fail to
capitalize on their advantages and abilities. These characteristics influence their behavioral motive (Henker et al., 2015).
directly facilitate harmonious interpersonal relationships with In relation to the effect of paradoxical leadership on employee
employees, build a supportive and open environment for behavior, superior’s paradoxical leadership centered on balancing
leaders and employees, improve employees’ psychological safety, the contradictions and tensions between organizational
and encourage them to provide advice to paradoxical leaders and employee needs, making employees feel the different
(Mowbray et al., 2015). characteristics of the leaders. Specifically, paradoxical leadership
Several studies have examined the link between leadership and is employee-centered and provides employees with moderate
psychological safety. For instance, Walumbwa and Schaubroeck flexibility, empowerment, and individuation. Such behaviors
(2009) stated that psychological safety involves perceiving and helped employees pursue their “ideal self.” This self-concept
experiencing a high degree of interpersonal trust. Further, prompts employees to implement behavioral strategy, which
psychological safety takes a central stage in the context of work corresponds with employee promotion focus orientation,
environment wherein employees develop mutual trust as well enhances their self-efficacy, and increases their confidence to
as comfort level in working with each other, both of which perform voice behavior. Conversely, paradoxical leadership
indirectly affect their voice behavior. Yang et al. (2019) also ensures employees meet their job requirements, controls
mentioned that psychological safety is a key differentiating factor decision-making, and ensures that all employees are treated
that encourages creativity at workplace and provides additional fairly. These behavioral characteristics are closely aligned with
insights about the paradoxical leadership behavior. Authors employees’ “moral selves.” This self-concept prompts employees
have also suggested that by undermining hierarchical barriers, to perform “avoidant” behavioral strategies, which accords with
leaders encourage a climate of psychological safety and trust the prevention focus orientation of employees. For employees
wherein employees feel more confident and safer not only while with an intensive prevention focus, their behavioral strategy
sharing suggestions but also while raising any work-related remained more cautious and conservative and was less affected
issues (Edmondson, 2003; Nembhard and Edmondson, 2006; by the external environment (Lee et al., 2000). Paradoxical
Hsiung, 2012). In the abovementioned studies, the authors have leadership contradicted the clear rules, responsibilities, and
considered psychological safety as a unidimensional construct duties that they observed and was more likely to be seen

Frontiers in Psychology | www.frontiersin.org 5 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

as a violation of the rules, resulting in lower self-efficacy, They represented five age groups, namely, 18–25 years (4.4%),
psychological safety, and a decrease in their voice behaviors. 26–30 years (30.0%), 31–40 years (46.7%), 41–50 years (12.2%),
Hence, the following hypotheses are proposed based on and over 50 years (6.7%). Among the employees, 51.1% were
previous literature: males and 48.9 were females. They represented five age groups,
namely, 18–25 years (31.0%), 26–30 years (42.5%), 31–40 years
Hypothesis 4a: Employee’s promotion focus positively
(20.1%), 41–50 years (6.0%), and over 50 years (0.4%).
moderates the mediating effect of employee’s self-efficacy in
the relationship between paradoxical leadership and voice
behavior; the higher the promotion focus, the stronger the Ethics Statement
mediating effect of self-efficacy. Prior to commencing the data collection, ethical approval was
applied and was approved by the Research Ethics Committee of
Hypothesis 4b: Employee’s prevention focus negatively the University. According to our research design, the study did
moderates the mediating effect of employee’s psychological not violate any legal regulations or common ethical guidelines.
safety in the relationship between paradoxical leadership In order to ensure that this study has followed ethical principles,
voice behaviors; the higher the prevention focus, the weaker the research purpose of the study is introduced, and consent
the mediating effect of psychological safety. was obtained before completing the hard copy questionnaires.
Figure 1 illustrates the research model and all hypotheses of Additionally, we emphasized that all the participants could reject
this research. any questions or withdraw from the study at any time. Lastly,
their anonymity and confidentiality were assured.

MATERIALS AND METHODS Measures

Paradoxical Leadership
Participants and Procedure This study utilized a paradoxical leadership scale designed by
A total of 360 leader – employee pairs of questionnaires were Zhang et al. (2015). This scale consisted of 22 items and
distributed in several cities in China, including Beijing, five dimensions, and its Cronbach’s α was 0.897. As a general
Hangzhou, Wuhan, Anhui, Zhengzhou, Shenzhen, and research practice, a value of 0.70 or higher for Cronbach’s
Guangzhou, to white-collar workers from a variety of industries alpha qualifies a research instrument for self-consistence and
such as finance, communication, manufacturing, education, acceptance. Employee participants were asked to report their
and real estate. leader’s paradoxical leadership on a five-point Likert scale. An
After removing the respondents with regular responses, example item was: “Uses a fair approach to treat all employees
missing items, and ineligible details, we finally obtained 268 uniformly, but also to treat them as individuals.” Response
valid response pairs of questionnaires that were used for analysis. options ranged from (1) strongly disagree to (5) strongly agree.
These leaders are presently in work relationship with these
employees or have had worked in past. Due to this, the data Self-Efficacy
collected from these leaders and employees helped in reaching The general self-efficacy scale designed by Schwarzer et al. (1997)
more insightful results based on their behavioral experiences. In was adopted in this study. This scale contained 10 items, and its
the sample of leaders, 63.3% were males, and 36.7% were females. Cronbach’s α was 0.946. Employee participants self-reported their

FIGURE 1 | Illustrates the research model and all hypotheses of this research.

Frontiers in Psychology | www.frontiersin.org 6 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

self-efficacy on a five-point Likert scale. An example item was: adopted: Firstly, Amos 20.0 software was used to perform
“I will be able to achieve most of the goals that I have set for the confirmatory factor analysis to investigate the discriminant
myself.” Response options ranged from (1) strongly disagree to validity of the variables. Then, this study utilized Herman’s single-
(5) strongly agree. factor test to conduct common method biases analysis. Following
that, SPSS 22.0 was adopted to conduct descriptive statistical
Psychological Safety analysis. Lastly, the study utilized Mplus 7.0 to verify hypotheses.
A five-item scale designed by Liang et al. (2012) was used to
measure psychological safety. Employee participants were asked Confirmatory Factor Analysis
to reported their psychological safety on a five-point Likert scale. In order to investigate the discriminant validity of the variables in
Its Cronbach’s α was 0.889. An example item was: “In my work the theoretical model, Amos 20.0 software was utilized to perform
unit, I can express my true feelings regarding my job.” Response the confirmatory factor analysis. A factor packing method was
options ranged from (1) strongly disagree to (5) strongly agree. used to build models with different factors, and they were
Promotive Voice Behavior compared in terms of goodness of fit. The results in Table 2
This study adopted the five-item scale designed by Liang et al. indicate that the seven-factor model was the most effective.
(2012) to measure promotive voice behavior, and its Cronbach’s
α was 0.925. Leader participants rated their employee’s promotive
Common Method Biases Analysis
voice behavior on a five-point Likert scale. An example item was: Two measures were utilized to remit the effect of common
“Proactively develop and make suggestions for Nos that may method bias in this study. On the one hand, experienced
influence the unit.” Response options ranged from (1) strongly academicians were invited to check all items on our
disagree to (5) strongly agree. questionnaire and revise explanations. Respondents were
allowed to, anonymously, comment when completing the
Prohibitive Voice Behavior questionnaires. On the other hand, Herman’s single-factor test
The five-item scale designed by Liang et al. (2012) was was conducted on the survey data, and a factor analysis was
utilized in this study, and its Cronbach’s α was 0.859. Leader performed on all questionnaire items with SPSS v22.0. The initial
participants rated their employee’s prohibitive voice behavior on characteristic value of 11 factors was higher than 1, the explained
a five-point Likert scale. An example item was: “Advise other total variation was 70.69%, and the first factor was 21.19%,
colleagues against undesirable behaviors that would hamper which was lower than totality by 50%, so common method biases
job performance.” Response options ranged from (1) strongly were acceptable.
disagree to (5) strongly agree.
Descriptive Statistical Analysis
Regulatory Focus SPSS. v22.0 software was utilized for the descriptive analysis
This study uses the regulatory focus scale designed by Zhou et al. of the variables in the study, and the results are presented in
(2012). This scale consists of seven items and two dimensions. Table 3. The results revealed that there were significant positive
Promotion focus has four items. An example item is “In general, correlations between paradoxical leadership and (a) employee
I am focused on achieving positive outcomes in my life.” The promotive voice behavior (r = 0.158, P < 0.01), (b) employee
Cronbach’s α was 0.875. Prevention focus included three items, prohibitive voice behavior (r = 0.158, P < 0.01), (c) employee
for example, “I often worry that I will fail to accomplish my work self-efficacy (r = 0.181, P < 0.01), and (d) employee psychological
goals.” The Cronbach’s α was 0.863. Employee participants were safety (r = 0.169, P < 0.01). These were consistent with the
asked to rate their regulatory focus on a five-point Likert scale hypotheses proposed in this study.
(“1” = “strongly disagree” and “5” = “strongly agree”).
Hypothesis Testing
Control Variables
Mplus 7.0 was utilized to construct a structural equation model,
This study collects demographics of leaders and employees,
and the SEM fitness indexes of χ2 /df, RMSAEA, SRMR, CFI,
including gender, working years, and age, to use as control
and TLI (χ2 /df = 1.949, RMSAEA = 0.060, SRMR = 0.033,
variables and recoded every variable. In terms of gender variable,
CFI = 0.969, and TLI = 0.919) all met the recommended values,
a male was coded as “1” and female was coded as “2.” For work
which indicated that the model constructed in this study was
experience, less than 3 years’ experience was coded as “1,” 4–
reasonable and reliable, which could be analyzed in the next
5 years was coded as “2,” 6–10 years was coded as “3,” 11–20 years
step. The structural model with path coefficients is presented in
was coded as “4,” and over 20 years was coded as “5.” In relation to
Figure 2.
age, those aged 18–25 were coded as “1,” aged 26–30 were coded
It can be seen in Figure 2 that the path coefficient of
as “2,” aged 31–40 were coded as “3,” aged 41–50 were coded as
paradoxical leadership on promotive voice behavior was positive
“4,” and those over 50 were coded as “5.”
and significant (β = 0.334, p < 0.01), supporting hypothesis
1a. The path coefficient of paradoxical leadership on prohibitive
DATA ANALYSIS AND RESULTS voice behavior was positive and significant (β = 0.228, p < 0.01),
which supported hypothesis 1b.
To establish the cause and relationship among the variables Table 4 illustrated the results of mediating effects, which is
conceptualized in this study, the following procedures were based on the bootstrapping method. As shown in Table 4, the

Frontiers in Psychology | www.frontiersin.org 7 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

TABLE 2 | Results of confirmatory factor analysis.

Model CMIN/df RMSEA GFI CFI TLI

Seven-factor PL,PE,PS,V1,V2,F1,F2 1.489 0.043 0.911 0.976 0.971

Six-factor PL,PE + PS,V1,V2,F1,F2 3.153 0.09 0.824 0.893 0.872
Five-factor PL,PE + PS,V1 + V2,F1,F2 3.402 0.095 0.806 0.878 0.857
Four-factor PL,PE + PS,V1 + V2,F1 + F2 5.402 0.128 0.731 0.772 0.739
Three-factor PL + PE + PS,V1 + V2,F1 + F2 6.047 0.137 0.709 0.735 0.7
Two-factor PL + PE + PS,V1 + V2 + F1 + F2 9.025 0.173 0.597 0.574 0.523
Single-factor PL + PE + PS + V1 + V2 + F1 + F2 12.295 0.206 0.509 0.398 0.329

PL, Paradoxical Leadership; PE, Self-efficacy; PS, Psychological Safety; V1, Promotive Voice Behavior; V2, Prohibitive Voice Behavior; F1, Promotion Focus; F2, Prevention
Focus. “ + ” refers to “combine two factors into a single factor.”

TABLE 3 | Results of descriptive statistical analysis.

Variable M SD PL PE PS V1 V2 F1 F2

Paradoxical leadership 3.888 0.475 1

Self-efficacy 3.628 0.735 0.181** 1
Psychological safety 3.590 0.828 0.169** 0.258** 1
Promotive voice behavior 3.796 0.776 0.206** 0.320** 0.250** 1
Prohibitive voice behavior 3.713 0.699 0.158** 0.346** 0.287** 0.752** 1
Promotion focus 3.794 0.811 0.139* 0.456** 0.115 0.120* 0.192** 1
Prevention focus 3.039 0.949 0.176** 0.117 0.075 0.145* 0.116 0.013 1

PL, Paradoxical Leadership; PE, Self-efficacy; PS, Psychological Safety; V1, Promotive Voice Behavior; V2, Prohibitive Voice Behavior; F1, Promotion Focus; F2, Prevention
Focus. *p < 0.05. **p < 0.01.

95% confidence interval for bias correction of the direct effect On the basis of the mediating effects, promotion focus and
of self-efficacy on paradoxical leadership and promotive voice prevention focus were taken as moderated variables, and the
behavior was [0.053, 0.242] (which did not include 0). This result SEM was reconstructed. The fitness indexes of χ2 /df, RMSAEA,
indicated that the indirect effect of paradoxical leadership on SRMR, CFI, and TLI (χ2 /df = 1.944, RMSAEA = 0.059,
promotive voice behavior was significant, which means that self- SRMR = 0.029, CFI = 0.965, and TLI = 0.896) all met
efficacy mediated the relationship between paradoxical leadership the recommended values, which indicated that the model
and promotive voice behavior, supporting hypothesis 2a. The constructed in this study was reasonable and reliable, which could
95% confidence interval for bias correction of the direct effect be analyzed in the following step.
of self-efficacy on paradoxical leadership and prohibitive voice In Table 5, when high promotion focus orientation was
behavior was [0.081, 0.239] (which did not include 0). This result obvious, the mediating effect of employee self-efficacy in
indicated that the indirect effect of paradoxical leadership on the relationship between paradoxical leadership and employee
prohibitive voice behavior was significant, which means that self- promotive voice behavior (Effect = 0.148, Boot 95% CI = [0.077,
efficacy mediated the relationship between paradoxical leadership 0.372]) and employee prohibitive voice behavior was significant
and prohibitive voice behavior, supporting hypothesis 2b. (Effect = 0.179, Boot 95% CI = [0.084, 0.287]). When low
The 95% confidence interval for bias correction of the promotion focus orientation was obvious, the mediating effect
direct effect of psychological safety on paradoxical leadership of employee self-efficacy in the relationship between paradoxical
and promotive voice behavior was [0.134, 0.347] (which did leadership and employee promotive voice behavior was non-
not include 0). This result indicated that the indirect effect significant (Effect = 0.084, Boot 95% CI = [−0.075, 0.260]).
of paradoxical leadership on prohibitive voice behavior Additionally, its mediating effect in the relationship between
was significant, which means that psychological safety paradoxical leadership and employee prohibitive voice behavior
mediated the relationship between paradoxical leadership was non-significant (Effect = 0.083, Boot 95% CI = [−0.038,
and promotive voice behavior, supporting hypothesis 3a. The 0.240]). This indicated that the focus of the promotion positively
95% confidence interval for bias correction of the direct effect of regulated the mediating effect of self-efficacy in the relationship
psychological safety on paradoxical leadership and prohibitive between paradoxical leadership and employee voice behavior.
voice behavior was [0.045 0.162] (which did not include 0). When high prevention focus orientation was significant,
This result indicated that the indirect effect of paradoxical the mediating effect of employee psychological safety in the
leadership on prohibitive voice behavior was significant, which relationship between paradoxical leadership and employee
means that psychological safety mediated the relationship promotive voice behavior (Effect = −0.107, Boot 95%
between paradoxical leadership and promotive voice behavior, CI = [−0.296, −0.001]) and prohibitive voice behavior
supporting hypothesis 3b. was significant (Effect = −0.156, Boot 95% CI = [−0.375,

Frontiers in Psychology | www.frontiersin.org 8 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

FIGURE 2 | Path Coefficients of the Hypothesis Model.

−0.011]). When low prevention focus orientation was Theoretical Implications

significant, the mediating effect of employee psychological Initially, the study conducts an examination of the relationship
safety in the relationship between paradoxical leadership and between paradoxical leadership and employee voice behavior,
employee promotive voice behavior (Effect = −0.024, Boot 95% expanding the previous studies pertaining to paradoxical
CI = [−0.118, 0.039]) and employee prohibitive voice behavior leadership. Although prior research has examined the
was non-significant (Effect = −0.032, Boot 95% CI = [−0.141, relationship between leadership and employee voice behaviors
0.062]). This suggests that prevention focus reversely moderates (LePine and Van Dyne, 2001; Detert and Burris, 2007), there is
the mediating effect of psychological safety in the relationship still a scarcity of studies in this area. The empirical results of this
between paradoxical leadership and employee voice behavior. study reveal that paradoxical leadership can promote employee
Therefore, hypotheses 4a and 4b were supported.

DISCUSSION TABLE 4 | Results of mediating effect.

Based on social cognitive theory and regulatory fit theory, Effect Estimate SE Est./SE BC 95% CI
this study examines the influencing mechanism of paradoxical
Lower Upper
leadership on employee voice behavior, proposed by Zhang et al.
(2015). It discusses how employee self-efficacy and psychological PL-PV1
safety mediated and explained it by analyzing the regulating Total effect 0.210 0.089 2.363 0.036 0.383
effect of employee regulatory focus orientation and sample Direct effect −0.012 0.095 −0.128 −0.203 0.175
data acquired from the survey. Finally, it draws conclusions Indirect effect
corresponding to the model. Moreover, the proposed model TOTAL 0.222 0.052 4.300 0.109 0.323
explained and provided good fit to the data. First, it was PL-PS-PV1 0.084 0.032 2.659 0.134 0.347
found that both promotive voice behavior and prohibitive voice PL-PE-PV1 0.138 0.048 2.905 0.053 0.242
behavior can be encouraged by practicing paradoxical leadership, PL-PV2
demonstrating positive relationship. Second, we found that Total effect 0.149 0.067 2.236 0.025 0.294
there is an indirect effect of paradoxical leadership on voice Direct effect −0.098 0.070 −1.395 −0.226 0.043
behavior through self-efficacy and psychological safety. Third, Indirect effect
when the moderating effect of regulatory focus of employees TOTAL 0.247 0.046 5.333 0.171 0.348
was considered into the model, it was revealed that promotion PL-PS-PV2 0.098 0.029 3.322 0.045 0.162
focus has a positive moderating effect on mediating effect of self- PL-PE-PV2 0.149 0.041 3.657 0.081 0.239
efficacy and prevention focus has an inverse moderating effect on PL, Paradoxical Leadership; PE, Self-efficacy; PS, Psychological Safety; V1,
mediating effect of psychological safety. Promotive Voice Behavior; V2, Prohibitive Voice Behavior.

Frontiers in Psychology | www.frontiersin.org 9 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

TABLE 5 | Results of moderated mediation analysis.

Mediators Dependent variables Moderators Level Effect Boot SE Boot 95% CI

Lower Upper

PE V1 F1 M−1 SD 0.084 0.064 −0.075 0.260

M 0.112 0.052 0.036 0.294
M + 1 SD 0.148 0.058 0.077 0.372
V2 F1 M−1 SD 0.083 0.069 −0.038 0.240
M 0.109 0.054 0.017 0.261
M + 1 SD 0.179 0.062 0.084 0.287
PS V1 F2 M−1 SD −0.024 0.038 −0.118 0.039
M −0.083 0.053 −0.242 0.016
M + 1 SD −0.107 0.051 −0.296 −0.001
V2 F2 M−1 SD −0.032 0.050 −0.141 0.062
M −0.094 0.046 −0.248 −0.033
M + 1 SD −0.156 0.041 −0.375 −0.011

PL, Paradoxical Leadership; PE, Self-efficacy; PS, Psychological Safety; V1, Promotive Voice Behavior; V2, Prohibitive Voice Behavior; F1, Promotion Focus; F2,
Prevention Focus.

voice behavior. When leaders tend to engage into paradoxical leadership on employee voice behavior with the help of
leadership behavior, employees tend to engage in both promotive conclusive evidences. The findings obtained verify that
and prohibitive voice behaviors. This finding contributes to individuals with a dissimilar regulatory focus behave differently,
the development of a more comprehensive account of the which dovetail with Higgins’s (1997) study. Besides, there is
relationship between paradoxical leadership and voice behavior. little agreement among academicians about how employees
The study also contributes to the existing literature about with different regulatory focus tendencies react to paradoxical
how a paradoxical style of leadership relates to employee leadership behaviors as argued by Lee et al. (2000) and Henker
voice behavior by highlighting the mediating effects of self- et al. (2015). Through empirical testing in this study, it was
efficacy and psychological safety and moderating effects of revealed that promotion focus has a positive moderating effect
regulatory focus in a collective manner rather than studying on mediating effect of self-efficacy and prevention focus has an
different factors individually and by integrating the social inverse moderating effect on mediating effect of psychological
cognitive theory and regulatory focus theories. The empirical safety. This is in contrast to the arguments presented in academic
results suggest that paradoxical leadership balances the conflicts literature. For instance, as per the regulatory fit theory, the voice
between organizational and employee development and allows behavior would be strengthened in cases where leadership style
employees to feel supported, autonomous, and responsible. This was consistent with their regulatory focus orientation and vice
can be explained further with the help of social cognitive versa (Henker et al., 2015).
theory that the perceived attitudes of employees are connected In addition, it was revealed that employees who practice high
to their behaviors. If employees believe that managers are regulatory focus orientation, both promotion and preventive,
concerned about their benefits and interests, then they would when their regulatory focus matches with leadership style,
conduct extra-role performance behaviors, such as voice behavior may stimulate the generation of a fitted regulator, employees’
(Walumbwa et al., 2010). cognition, and strengthen their behavioral motive. In contrast,
Despite verifying the mediating effects of psychological safety, at low levels of regulatory focus orientation, it would be
the results of this study are strongly supportive of the mediating challenging to generate regulatory fit with employees’ cognition,
effects of self-efficacy as well. The majority of previous studies and managers would fail to influence their behavioral motive
examining the relationship between leadership and employee (Henker et al., 2015). Moreover, the literature also suggested that
voice behavior focused on the mediating effect of psychological paradoxical leadership is employee-centric and helps employees
safety. However, this study explains both employee self- to pursue their “ideal self.” These findings thus contribute to
efficacy and psychological safety mediated between paradoxical the development of a better understanding of the relationship
leadership and employee voice behavior. This conclusion reveals between paradoxical leadership and voice behavior.
how paradoxical leadership is related to employee voice behavior,
enriching and perfecting voice behavior-related theories. This Practical Implications
finding supports the results conducted by Mowbray et al. (2015), This study discusses the influencing mechanism of paradoxical
which highlights that self-efficacy can mediate the relationship leadership on employee voice behavior and constitutes an
between voice behavior and another leadership despite ethical expansion of the studies on the topic. The conclusions drawn
and paternalistic leadership (Wang et al., 2015). offer some guidance for business management practice.
Lastly, this study explored that regulatory focus moderates Firstly, managers should improve the characteristics of
the mediating effect in influencing the mechanism of paradoxical paradoxical leadership behaviors. Currently, the business

Frontiers in Psychology | www.frontiersin.org 10 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

environment is constantly shifting and managements practice an intensive level of regulatory focus, their behavioral
face uncertainty and complicated challenges. Managers in strategy remains more cautious and conservative and was less
organizations are unavoidably confronted with numerous affected by the external environment (Lee et al., 2000). For
dilemmas and paradoxes. Their ability to deal with these are, example, employees with intensive prohibitive behavior may see
thus, increasingly important to the effectiveness of leadership. paradoxical leaders’ competing behaviors as a clear contradiction
While managers see these challenges from a paradoxical of rules, which may undermine their self-efficacy, psychological
perspective that originate in the paradox theory, the way of safety, and consequently their voice behavior.
thinking will expand their cognition, integrate complexities,
survey the status quo from the perspective of long-term strategic Limitations and Recommendation
development, and balance the contradictions and tensions with Limitations of this study should be addressed in the future.
appropriate management behaviors. For example, in today’s First, this study was conducted from the perspective of employee
competitive world, stress is common among the employees that psychological cognition, self-efficacy, and psychological safety,
originate from rising competing demands. Such demands include which may be considered insufficient. In the “black box” of the
operating in the short term while planning for the long term, relationship between paradoxical leadership and employee voice
dealing with local scenarios while acting globally, and competing behavior, there are other interpreting mechanisms, such as the
as well as collaborating with other firms to remain profitable yet perspective of employees’ self-determination and leader–member
socially and environmentally friendly. The results obtained thus exchange. These factors were not included in this study, and
suggest that leaders embracing paradoxes help employees and related studies may wish to consider these perspectives in the
subordinates in making sense of such demands and voicing out future. Secondly, due to a number of restrictions, this study used
their opinion to improve processes and structures. cross-section data, making it challenging to examine the validity
Secondly, organizations are suggested to stress the building of of the theoretical model in a long-term period. Although data
the organizational atmosphere in order to improve employees’ were collected from various sources and included both managers
psychological safety and self-efficacy. Managers should adjust and employees to reduce common method biases, this study
their management style, constantly optimize their management could not obtain longitudinal, multi-temporal data and did not
work, attend to employees’ job performance, guide their work verify the long-term effect of paradoxical leadership on employee
based on employees’ characteristics, provide employees moderate voice behavior. Future studies may consider these aspects to
flexibility and autonomy, place emphasis on the building of team extend this research.
culture, provide positive feedback about employee behaviors, and
execute suitable incentives. These approaches will strengthen
employees’ confidence, improve the relationships between leaders DATA AVAILABILITY STATEMENT
and employees, improve employees’ psychological safety, and
improve employees’ psychological cognition. For instance, The original contributions presented in the study are included
paradoxical leadership provides employees space to be more in the article/supplementary material, further inquiries can be
individual with increased support, allowing for flexible working, directed to the corresponding author.
while providing a fairer environment for both leaders and
employees. These factors increase self-efficacy of employees,
which in turn boosts their voice behavior. ETHICS STATEMENT
Thirdly, the way of managing employees should lay
emphasis on adjusting measures to different employees Prior to commencing the data collection, ethical approval was
and implement proper management strategy according to applied and was approved by the Research Ethics Committee of
personal characteristics of employees. Managers lead employees the University. According to our research design, the study did
differently because of their different individual characteristics; not violate any legal regulations or common ethical guidelines. In
hence, managers should pay attention to their subordinates order to ensure that this study has followed the ethical principles,
as much as possible, implement proper management strategy, the research purpose of the study is introduced, and consent
and stimulate personal potential to the maximum extent, for was obtained before completing the hard copy questionnaires.
example, offering employees promotion to focus more on Additionally, we emphasized that all the participants could reject
autonomy, encouraging employees to be more active in voice any questions or withdraw from the study at any time. Lastly,
behavior, providing employees with prevention measures that their anonymity and confidentiality were assured.
focus more on care and personalized instructions, making job
requirements clearer and more specific, relieving their doubts
about risks, and promoting their psychological safety. AUTHOR CONTRIBUTIONS
Lastly, paradoxical leadership is employee-centered and
provides employees with moderate flexibility, empowerment, and XL conceived the theoretical framework, organized data
individuation. Such behaviors helped employees pursue their collection, and supervised this study. YX contributed to the
“ideal self.” This self-concept prompts employees to implement research idea and data analysis, and composed the manuscript.
behavioral strategy, which corresponds with employee promotion HL and DY collected the data, analyzed the statistics, and
focus orientation, enhances their self-efficacy, and increases their edited the manuscript. All authors contributed to the article and
confidence to perform voice behavior. However, if employees approved the submitted version.

Frontiers in Psychology | www.frontiersin.org 11 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

REFERENCES Lee, A. Y., Aaker, J. L., and Gardner, W. L. (2000). The pleasures and pains of
distinct self-construals: the role of interdependence in regulatory focus. J. Pers.
Bandura, A. (1985). Social Foundations of Thought and Action: A Social Cognitive Soc. Psychol. 78, 1122–1134. doi: 10.1037/0022-3514.78.6.1122
Theory. Englewood Cliffs, NJ: Prentice Hall. LePine, J. A., and Van Dyne, L. (2001). Voice and cooperative behavior
Bandura, A. (1993). Perceived self-efficacy in cognitive development and as contrasting forms of contextual performance: evidence of differential
functioning. Educ. Psychol. 28, 117–148. doi: 10.1207/s15326985ep2802_3 relationships with Big Five personality characteristics and cognitive ability.
Bandura, A. (1995). Self-Efficacy in Changing Societies. New York, NY: Cambridge J. Appl. Psychol. 86, 326–336. doi: 10.1037/0021-9010.86.2.326
University Press. Liang, J., Farh, C. I. C., and Farh, J. L. (2012). Psychological antecedents of
Burris, E. R., Detert, J. R., and Chiaburu, D. S. (2008). Quitting before leaving: the promotive and prohibitive voice: a two-wave examination. Acad. Manag. J. 55,
mediating effects of psychological attachment and detachment on voice. J. Appl. 71–92. doi: 10.5465/amj.2010.0176
Psychol. 93, 912–922. Liang, S. G. (2017). Linking leader authentic personality to employee voice
Carmeli, A., Brueller, D., and Dutton, J. E. (2009). Learning behaviors in behaviour: a multilevel mediation model of authentic leadership development.
the workplace: the role of high-quality interpersonal relationships and Eur. J. Work Organ. Psychol. 26, 1–10. doi: 10.1080/1359432X.2017.1293656
psychological safety. Syst. Res. Behav. Sci. 26, 81–98. doi: 10.1002/sres.932 Liu, S. M., and Liao, J. Q. (2013). Transformational leadership and speaking up:
Carmeli, A., Reiter-Palmon, R., and Ziv, E. (2010). Inclusive leadership and power distance and structural distance as moderators. Soc. Behav. Pers. 41,
employee involvement in creative tasks in the workplace: the mediating role 1747–1756. doi: 10.2224/sbp.2013.41.10.1747
of psychological safety. Creat. Res. J. 22, 250–260. doi: 10.1080/10400419.2010. Liu, W., Tangirala, S., and Ramanujam, R. (2013). The relational antecedents of
504654 voice targeted at different leaders. J. Appl. Psychol. 98, 841–851. doi: 10.1037/
Chan, S. C. (2014). Paternalistic leadership and employee voice: does information a0032913
sharing matter? Hum. Relat. 67, 667–693. doi: 10.1177/0018726713503022 Liu, W., Zhu, R. H., and Yang, Y. K. (2010). I warn you because I like you: voice
Chen, S. Y., and Hou, Y. H. (2016). The effects of ethical leadership, voice behavior behavior, employee identifications, and transformational leadership. Leadersh.
and climates for innovation on creativity: a moderated mediation examination. Q. 21, 189–202. doi: 10.1016/j.leaqua.2009.10.014
Leadersh. Q. 27, 1–13. doi: 10.1016/j.leaqua.2015.10.007 Milliken, F., Morrison, E. W., and Hewlin, P. (2003). An exploratory study of
Detert, J. R., and Burris, E. R. (2007). Leadership behavior and employee voice: employee silence: issues that employees don’t communicate upward and why.
is the door really open? Acad. Manag. J. 50, 869–884. doi: 10.5465/amj.2007. J. Manag. Stud. 40, 1453–1476. doi: 10.1111/1467-6486.00387
26279183 Morrison, E. W. (2014). Employee voice and silence. Annu. Rev. Organ. Psychol.
Detert, J. R., and Trevino, L. K. (2010). Speaking up to higher-ups: how supervisors Organ. Behav. 1, 73–197. doi: 10.1146/annurev-orgpsych-031413-091328
and skip-level leaders influence employee voice. Organ. Sci. 21, 249–270. doi: Mowbray, P. K., Wilkinson, A., and Tse, H. H. M. (2015). An integrative review of
10.1287/orsc.1080.0405 employee voice: identifying a common conceptualization and research agenda.
Duan, J. Y., Li, C. W., Xu, Y., and Wu, C. H. (2016). Transformational leadership Int. J. Manag. Rev. 17, 382–400. doi: 10.1111/ijmr.12045
and employee voice behavior: a Pygmalion mechanism. J. Organ. Behav. 38, Nembhard, I. M., and Edmondson, A. C. (2006). Making it safe: the effects of leader
650–670. doi: 10.1002/job.2157 inclusiveness and professional status on psychological safety and improvement
Edmondson, A. C. (2003). Speaking up in the operating room: how team leaders efforts in health care teams. J. Organ. Behav. 27, 941–966. doi: 10.1002/jo
promote learning in interdisciplinary action teams. J. Manag. Stud. 40, 1419– b.413
1452. doi: 10.1111/1467-6486.00386 Ning, H. Y., Zhou, M. J., Lu, Q., and Wen, L. Q. (2012). Exploring relationship
Gardner, W. L., Avolio, B. J., Luthans, F., May, D. R., and Walumbwa, F. between authority leadership and organizational citizenship behavior in China:
(2005). Can you see the real me? A self-based model of authentic leader and the role of collectivism. Organ. Citizensh. Behav. 6, 231–244. doi: 10.1108/
follower development. Leadersh. Q. 16, 343–372. doi: 10.1016/j.leaqua.2005. 17506141211236677
03.003 Okpozo, A. Z., Gong, T., Ennis, M. C., and Adenuga, B. (2017). Investigating the
Giessner, S. R., and Schubert, T. W. (2007). High in the hierarchy: how vertical impact of ethical leadership on aspects of burnout. Leadersh. Organ. Dev. J. 38,
location and judgments of leaders’ power are interrelated. Organ. Behav. Hum. 1128–1143. doi: 10.1108/lodj-09-2016-0224
Decis. Process. 104, 30–44. doi: 10.1016/j.obhdp.2006.10.001 Ou, Y., Tsui, S., Kinicki, A. J., Waldman, D. A., Xiao, Z. X., and Song, J. W.
Grant, A. M. (2013). Rocking the boat but keeping it steady: the role of emotion (2014). Humble chief executive officers’ connections to top management team
regulation in employee voice. Acad. Manag. J. 56, 1703–1723. doi: 10.5465/ integration and middle managers’ responses. Admin. Sci. Q. 59, 34–72. doi:
amj.2011.0035 10.1177/0001839213520131
Hans, S., and Gupta, R. (2018). Job characteristics affect shared leadership: the Owens, B. P., and Hekman, D. R. (2012). Modeling how to grow: an inductive
moderating effect of psychological safety and perceived self-efficacy. Leadersh. examination of humble leader behaviors, contingencies, and outcomes. Acad.
Organ. Dev. J. 39, 730–744. doi: 10.1108/lodj-03-2018-0101 Manag. J. 55, 787–818. doi: 10.5465/amj.2010.0441
Henker, N., Sonnentag, S., and Unger, D. (2015). Transformational leadership and Parker, S. K., Williams, H. M., and Turner, N. (2006). Modeling the antecedents of
employee creativity: the mediating role of promotion focus and creative process proactive behavior at work. J. Appl. Psychol. 91, 636–652. doi: 10.1037/0021-
engagement. J. Bus. Psychol. 30, 235–247. doi: 10.1007/s10869-014-9348-7 9010.91.3.636
Higgins, E. T. (1997). Beyond pleasure and pain. Am. Psychol. 52, 1280–1300. Qi, L., and Liu, N. (2017). Effects of inclusive leadership on employee voice
doi: 10.1037/0003-066x.52.12.1280 behavior and team performance: the mediating role of caring ethical climate.
Higgins, E. T. (2000). Making a good decision: value from fit. Am. Psychol. 55, Front. Commun. 2:8. doi: 10.3389/fcomm.2017.00008
1217–1230. doi: 10.1037/0003-066x.55.11.1217 Schwarzer, R., Born, A., Iwawaki, S., and Lee, Y. M. (1997). The assessment of
Hsiung, H. H. (2012). Authentic leadership and employee voice behavior: a multi- optimistic self-beliefs: comparison of the Chinese, Indonesian, Japanese, and
level psychological process. J. Bus. Ethics 107, 349–361. doi: 10.1007/s10551- Korean versions of the general self-efficacy scale. Psychologia 40, 1–13.
011-1043-2 Son, S. J. (2019). The role of supervisors on employee voice behavior. Leadersh.
Hunter, E. M., Neubert, M. J., Perry, S. J., Witt, L. A., Penney, L. M., and Organ. Dev. J. 40, 85–96. doi: 10.1108/lodj-06-2018-0230
Weinberger, E. (2013). Servant leaders inspire servant followers: antecedents Van Dyne, L., and LePine, J. A. (1998). Helping and voice extra-role behaviors:
and outcomes for employees and the organization. Leadersh. Q. 24, 316–331. evidence of construct and predictive validity. Acad. Manag. J. 41, 108–119.
doi: 10.1016/j.leaqua.2012.12.001 doi: 10.5465/256902
Kahn, W. A. (1990). Psychological conditions of personal engagement and Walumbwa, F. O., Hartnell, C. A., and Oke, A. (2010). Servant leadership,
disengagement at work. Acad. Manag. J. 33, 692–724. doi: 10.5465/256287 procedural justice climate, service climate, employee attitudes, and
Lapointe, É, and Vandenberghe, C. (2018). Examination of the relationships organizational citizenship behavior: a cross-level investigation. J. Appl.
between servant leadership, organizational commitment, and voice and Psychol. 95, 517–529. doi: 10.1037/a0018867
antisocial behaviors. J. Bus. Ethics 148, 99–115. doi: 10.1007/s10551-015-3 Walumbwa, F. O., Mayer, D. M., Wang, P., Wang, H., Workman, K., and
002-9 Christensen, A. L. (2011). Linking ethical leadership to employee performance:

Frontiers in Psychology | www.frontiersin.org 12 September 2020 | Volume 11 | Article 537756

Li et al. Paradoxical Leadership and Voice Behavior

the roles of leader–member exchange, self-efficacy, and organizational Yan, A., and Xiao, Y. G. (2016). Servant leadership and employee voice behavior:
identification. Organ. Behav. Hum. Decis. Process. 115, 204–213. doi: 10.1016/j. a cross-level investigation in China. SpringerPlus 5, 1595–1605. doi: 10.1186/
obhdp.2010.11.002 s40064-016-3264-4
Walumbwa, F. O., Morrison, E. W., and Christensen, A. L. (2012). Ethical Yang, Y., Li, Z., Liang, L., and Zhang, X. (2019). Why and when paradoxical leader
leadership and group in-role performance: the mediating roles of group behavior impact employee creativity: thriving at work and psychological safety.
conscientiousness and group voice. Leadersh. Q. 23, 953–964. doi: 10.1016/j. Curr. Psychol. 1, 1–12. doi: 10.2224/sbp.9211
leaqua.2012.06.004 Zhang, X., and Bartol, K. M. (2010). Linking empowering leadership and employee
Walumbwa, F. O., and Schaubroeck, J. (2009). Leader personality traits and creativity: the influence of psychological empowerment, intrinsic motivation,
employee voice behavior: mediating roles of ethical leadership and work and creative process engagement. Acad. Manag. J. 53, 107–128. doi: 10.5465/
group psychological safety. J. Appl. Psychol. 94, 1275–1286. doi: 10.1037/a001 amj.2010.48037118
5848 Zhang, Y., Huai, M., and Xie, Y. (2015). Paternalistic leadership and employee voice
Wang, D., Gan, C., Wu, C., and Wang, D. (2015). Ethical leadership and employee in China: a dual process model. Leadersh. Q. 26, 25–36. doi: 10.1016/j.leaqua.
voice: employee self-efficacy and self-impact as mediators. Psychol. Rep. 116, 2014.01.002
751–767. doi: 10.2466/01.07.pr0.116k29w9 Zhang, Y., Waldman, D. A., Han, Y. L., and Li, X. B. (2014). Paradoxical leader
Wang, R., and Jiang, J. (2015). How abusive supervisors influence employees’ behaviors in people management: antecedents and consequences. Acad. Manag.
voice and silence: the effects of interactional justice and organizational J. 58, 538–566. doi: 10.5465/amj.2012.0995
attribution. J. Soc. Psychol. 155, 204–220. doi: 10.1080/00224545.2014.99 Zhou, Q., Hlrst, G., and Shhipton, H. (2012). Context matters: combined influence
0410 of participation and intellectual stimulation on the promotion focus–employee
Weiss, M., Kolbe, M., Grote, G., Spahn, D. R., and Grande, B. (2017). We can do it! creativity relationship. J. Organ. Behav. 33, 894–909. doi: 10.1002/job.779
Inclusive leader language promotes voice behavior in multi-professional teams.
Leadersh. Q. 29, 389–402. doi: 10.1016/j.leaqua.2017.09.002 Conflict of Interest: The authors declare that the research was conducted in the
Weiss, M., and Morrison, E. W. (2018). Speaking up and moving up: how voice absence of any commercial or financial relationships that could be construed as a
can enhance employees’ social status. J. Organ. Behav. 40, 1–15. doi: 10.1002/ potential conflict of interest.
job.2262
Wong, C. A., Spence Laschinger, H. K., and Cummings, G. G. (2010). Authentic Copyright © 2020 Li, Xue, Liang and Yan. This is an open-access article distributed
leadership and nurses’ voice behaviour and perceptions of care quality. J. Nurs. under the terms of the Creative Commons Attribution License (CC BY). The use,
Manag. 18, 889–900. doi: 10.1111/j.1365-2834.2010.01113.x distribution or reproduction in other forums is permitted, provided the original
Xu, A. J., Loi, R., and Lam, L. W. (2015). The bad boss takes it all: how abusive author(s) and the copyright owner(s) are credited and that the original publication
supervision and leader–member exchange interact to influence employee in this journal is cited, in accordance with accepted academic practice. No use,
silence. Leadersh. Q. 26, 763–774. doi: 10.1016/j.leaqua.2015.03.002 distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Psychology | www.frontiersin.org 13 September 2020 | Volume 11 | Article 537756