702078

research-article2017
JLOXXX10.1177/1548051817702078Journal of Leadership & Organizational StudiesKim and Beehr

Article
Journal of Leadership &

Self-Efficacy and Psychological Ownership

Organizational Studies
2017, Vol. 24(4) 466­–478
© The Authors 2017
Mediate the Effects of Empowering Reprints and permissions:
sagepub.com/journalsPermissions.nav

Leadership on Both Good and DOI: 10.1177/1548051817702078

https://doi.org/10.1177/1548051817702078
journals.sagepub.com/home/jlo

Bad Employee Behaviors

Minseo Kim1 and Terry A. Beehr1

Abstract
This study examined the potential effects of empowering leadership on followers’ in-role performance and deviant
behaviors via self-efficacy and psychological ownership over a 3-week period in a sample of 299 full-time employees
working in the United States. Results from structural equation modeling demonstrated that empowering leadership was
positively related to both self-efficacy and psychological ownership, which in turn were both negatively related to deviant
behaviors. Alternative model comparisons and bootstrapping both confirmed the mediation effects of self-efficacy and
psychological ownership. However, only one of the two mediators, self-efficacy, was positively related to followers’ in-
role performance. Together, these findings highlighted the important roles of self-efficacy and psychological ownership
explaining why empowering leadership may result in followers’ behaviors.

Keywords
empowering leadership, self-efficacy, psychological ownership, in-role performance, workplace deviance

Empowering leadership has received increasing attention thereby resulting in reduced performance (Cordery,
from researchers and practitioners alike, because of its Morrison, Wright, & Wall, 2010). Other studies find weak
potential to have positive effects on important work out- or no direct effects of empowering leadership on perfor-
comes, such as creativity, job satisfaction, citizenship mance (e.g., Ahearne, Mathieu, & Rapp, 2005; Srivastava,
behavior, and turnover intentions (Amundsen & Martinsen, Bartol, & Locke, 2006). If there are weak or no direct
2015; Dewettinck & van Ameijde, 2011; Fong & Snape, effects, however, there may be indirect effects.
2015; Raub & Robert, 2010; Zhang & Bartol, 2010). We suggest self-efficacy and psychological ownership as
According to the theory of psychological ownership (Van key mediating processes in the empowering leadership–
Dyne & Pierce, 2004), when employees have feelings of employee behavior relationships because they are motiva-
possession toward the organization, they become more tional in nature. Figure 1 illustrates a proposed model in
attached to, protective of, and responsible for it. This occurs which empowering leadership leads to both good (i.e., job
because empowering leaders provide employees with performance) and bad (i.e., deviance) employee behaviors
greater autonomy, power, responsibility, and development through two mediators. The model proposes effects for
support, leading subordinates to work independently, with empowering leadership because it leads to the psychological
increased feelings of competence, self-determination, states of self-efficacy and psychological ownership, which
meaningfulness, and impact (Amundsen & Martinsen, are the more proximal determinants of employees’ reactions.
2014b; Zhang & Bartol, 2010). As a result, employees are Self-efficacy, an individual’s belief in his or her capacity to
intrinsically motivated to work hard to achieve their work perform tasks successfully (Bandura, 1997), may be enhanced
goals, believing that their performance depends on their by empowering leader behaviors including motivation and
efforts, resulting in high productivity (Fong & Snape, 2015;
Vecchio, Justin, & Pearce, 2010). 1
Central Michigan University, Mount Pleasant, MI, USA
However, not all studies show these favorable results.
Instead, some studies suggest empowering leadership might Corresponding Author:
Minseo Kim, Department of Psychology, College of Humanities and
even have detrimental consequences, perhaps because Social and Behavioral Sciences, Central Michigan University, 233 Sloan
empowering leaders’ focus on high autonomy in decision Hall, Mount Pleasant, MI 48859, USA.
making and task delegation might increase task uncertainty, Email: [email protected]
Kim and Beehr 467

development support (e.g., coaching, modeling, and encour- performance, which refers to employees doing behaviors
aging initiative). Self-efficacy is inherently part of motiva- that are helpful to the organization’s effectiveness but that
tion from some theoretical viewpoints, such as expectancy are outside the formal requirements of their jobs. In-role
theory in which it is a form of the expectancy that one’s effort behavior in the workplace, a concept first introduced by
will lead to performance (e.g., Vroom, 1964). Leaders who Katz and Kahn (1966) is the performance of the core tasks
empower employees are giving them more control over their that constitute one’s job.
work, which should increase employees’ beliefs that they can Deviant behaviors of employees directed at both an indi-
perform the job. vidual and an organization (e.g., verbal harassment, spread-
The second mediator is another consequence of empow- ing rumors, and wasting resources) can be costly (Litzky,
ering leadership, psychological ownership. Psychological Eddleston, & Kidder, 2006; Robinson & Bennett, 1995).
ownership is employees’ feelings that the organization is Research on employee deviance suggests that some leader-
“theirs” and sense of shared responsibility toward its suc- ship styles, including charismatic, authentic, and ethical
cess (Mustafa, Martin, & Hughes, 2016; Pierce, Kostova, & leadership could be antecedents M. E. Brown & Treviño,
Dirks, 2001). Psychological ownership can be intrinsically 2006; Erkutlu & Chafra, 2013; van Gils, van Quaquebeke,
rewarding, because the success of the organization is a van Knippenberg, van Dijke, & De Cremer, 2015). The
reflection on the owners and employees who feel such own- present study, however, examines the role of empowering
ership can help it succeed by working hard. In expectancy leadership and focuses on mediators that explain its link to
theory terms, employees who feel ownership of the organi- both performance and deviance.
zation can expect to feel more proud if the organization is Although empowering leadership may conceptually
successful, and they can help make the organization suc- overlap with other positive forms of leadership such as
cessful with their good job performance. Empowering lead- transformational leadership and leader–member exchange,
ers may create an environment wherein followers can several studies established the distinctiveness of empower-
develop that sense of ownership by letting them make deci- ing leadership from other leader behaviors (e.g., Amundsen
sions or involving them in work processes, thereby leading & Martinsen, 2014b; Pearce et al., 2003; Tekleab, Sims,
to an increased responsibility. Yun, Tesluk, & Cox, 2008). Empowering leadership theory
The two mediators in the model, self-efficacy and psy- emphasizes the role of autonomy support, which is an
chological ownership, are related, but they correspond to important factor in generating intrinsic motivation, thereby
different parts of intrinsic motivation based on expectancy predicting job performance (Dysvik & Kuvaas, 2011;
theory (Vroom, 1964). Self-efficacy corresponds to Morgeson, Delaney-Klinger, & Hemingway, 2005). We
Expectancy I (sometimes just labeled expectancy), which is propose that empowering leadership may lead to more
the employees’ expectancy that they can perform well if desirable work behaviors and fewer undesirable behaviors
they try hard (effort). Psychological ownership represents by increasing the two mediation processes, self-efficacy
employees’ expectancy that they will experience a positive and psychological ownership.
intrinsic outcome (e.g., pride) if they perform well, which Weak relationships between empowering leadership and
corresponds to Expectancy II (sometimes labeled instru- employee behaviors may have been found in previous
mentality); as noted earlier, the sense of pride is expected to research because empowering leadership is a more distal
result from good performance in part because performance influence (Vecchio et al., 2010), having effects through the
is part of the success of the organization, which is “owned” more proximal mediators, as the model in Figure 1 sug-
by the employee. gests. The study contributes to the empowering leadership
A leader might be reluctant to empower subordinates literature by (a) examining bad as well as good behaviors
because empowerment constitutes control and freedom, and also (b) providing a theoretical explanation for why
which can enable employees to do bad, as well as good empowering leadership affects subordinates—it creates
behaviors. One recent study showed that empowering lead- psychological resources in the form of self-efficacy and
ership might have both positive and negative aspects in pre- psychological ownership, which are more proximal motiva-
dicting followers’ performance (Cheong, Spain, Yammarino, tors of employees’ behaviors.
& Yun, 2016). Contrary to that concern, however, we expect
that to the extent leaders’ empowering behaviors result in Relationships of Empowering Leadership With
self-efficacy and psychological ownership, employees
should be more motivated to do good and less motivated to
Self-Efficacy and Psychological Ownership
engage in bad behaviors. Therefore, we examined not only Generalized self-efficacy is the belief that one can achieve
employees’ in-role performance (good behavior), but also desired outcomes across a variety of tasks and situations
their deviant behaviors (bad behaviors). In-role perfor- (Bandura, 1997; Chen, Gully, & Eden, 2001). From the
mance refers to performance that is formally expected of perspective of social cognitive theory (Bandura, 1986,
the employee; it is often contrasted with extra-role 1997), self-efficacy can be enhanced by verbal persuasion
468 Journal of Leadership & Organizational Studies 24(4)

Figure 1.  Hypothesized model.

(encouragement), direct mastery experience of accom- work environment, feelings of belongingness are likely to
plishing a task or skill, and observational learning. In the develop. Through participative goal setting and discus-
workplace, empowering leadership such as coaching (ver- sions, employees can strengthen their role within the orga-
bal persuasion) and modeling (observational learning) may nization, amplifying their sense of belongingness, a core
make subordinates feel more confident in their capability component of psychological ownership. An experience of
to perform their jobs, because they could learn how effec- control, which accompanies participation and influence,
tive performance can be attained through leaders’ guidance can make people feel responsible for the work and for the
and feedback, and by observing the work of their supervi- organization. Thus, a sense of belonging and feelings of
sor (e.g., vicarious learning). When empowering leaders organizational identification are related to the construct of
encourage employees to take part in their work-related psychological ownership.
decision making and increase their involvement, they have According to Pierce et al. (2001), people use a sense of
opportunities to expand their knowledge and learn from ownership for the purpose of defining themselves and
each other by exchanging information (Latham, Winters, & expressing their self-identity to others. Empowering leader
Locke, 1994). All of these processes contribute to fostering behaviors are likely to contribute to increasing the percep-
self-efficacy. More directly, empowering leaders should be tion of self-identity, because individuals have a drive to
familiar with their subordinates’ capabilities and encourage identity in work settings (Ashforth & Mael, 1989) and
the use of their competencies, which contributes to the develop feelings of psychological ownership through expe-
development of subordinates’ self-efficacy beliefs riences with the target organization (G. Brown, Crossley, &
(Amundsen & Martinsen, 2014a). Robinson, 2014). Empowering leaders are willing to pro-
vide a wealth of opportunities for their followers to invest
Hypothesis 1: Empowering leadership is positively themselves in job or work outcomes, and thereby to help
related to followers’ self-efficacy. gain a sense of connectedness within the organization.
Last, leaders’ behaviors, such as encouraging followers’
Empowering leadership not only helps increase follow- initiative and letting them make decisions, will lead to an
ers’ self-efficacy but it may also help followers develop increased sense of accountability or responsibility of fol-
feelings of psychological ownership, which is character- lowers. As employees engage in the decision-making pro-
ized by belongingness, self-identity related to the organiza- cess, they show high attention to their tasks and feel more
tion, and accountability (Avey, Avolio, Crossley, & responsible for their performance, which can affect psycho-
Luthans, 2009). An important part of an employee’s self- logical ownership. Thus, empowering leadership should
concept can be derived from membership or identity in an help influence three elements in the definition of psycho-
organization or group where organizational identity is logical ownership—belongingness, self-identity, and
defined as the perception of oneness with or belonging to accountability (Avey et al., 2009), all of which are predicted
an organization (Mael & Ashforth, 1992). Empowering to result in feelings of psychological ownership.
leaders pay attention to followers’ voice and encourage
them to engage more fully in their task activities. When Hypothesis 2: Empowering leadership is positively
employees feel that they are heard and have input into their related to followers’ psychological ownership.
Kim and Beehr 469

Relationships of Self-Efficacy and Psychological less negatively than did those with low self-efficacy when
Ownership With Behavioral Outcomes exposed to stressors such as work overload and long work
hours (Jex & Bliese, 1999). Additionally, Latham and
Empowering leadership can engender self-efficacy and psy- Frayne (1989) suggested a negative relationship between
chological ownership as depicted in Figure 1, but the model enhanced self-efficacy and still other undesirable behav-
also proposes that self-efficacy and psychological owner- iors such as absenteeism. Overall, high self-efficacy is
ship are likely to affect subordinates’ behaviors. In-role per- negatively related to bad behavior, and therefore we pro-
formance refers to basic acts that are required or expected pose that it is negatively related to employees’ bad or devi-
of members by the organization (Williams & Anderson, ant behaviors in the workplace.
1991) and is often referred to simply as job performance.
Self-efficacy is related to confidence in one’s ability. It is Hypothesis 4: Self-efficacy is negatively related to fol-
the employee’s belief or expectation that they can success- lowers’ deviant behaviors.
fully accomplish tasks if they exert effort, a feature of
expectancy theory of motivation (Vroom, 1964), and moti- Self-efficacy should result in good employee behaviors
vation should lead to better performance. Employees who because it helps employees feel confident that they can per-
see themselves as efficacious and feel better about their job form successfully, thereby gaining a sense of achievement,
through the positive experiences of work are likely to exert similar to an expectancy in expectancy theory, as noted ear-
considerable effort on their tasks, leading to successful in- lier. Similar to self-efficacy, psychological ownership also
role performance. As one would expect, thus, meta-analy- contributes to generating motivational drives to perform
ses have confirmed the positive relationship between well in response to increased responsibility over the job.
self-efficacy and task performance (Chen, Casper, & Doing the job well reflects favorably on one’s self-identity
Cortina, 2001; Stajkovic & Luthans, 1998). (G. Brown, Pierce, & Crossley, 2014). When employees
consider the job and organization as possessions (psycho-
Hypothesis 3: Self-efficacy is positively related to fol- logical ownership), their sense of responsibility and self-
lowers’ in-role performance. identity increase if the job and organization are successful.
Thus, they feel committed to help their job and organization
The model indicates that employees experiencing self- through high performance.
efficacy not only perform more good behaviors but they Employees with feelings of ownership tend to be more
also perform fewer bad behaviors. Self-efficacy is an ele- satisfied with their job and show more interest in the orga-
ment of employees’ core self-evaluations or favorable nization (Avey et al., 2009). If the organization is “theirs,”
assessments of one’s self-worth (e.g., Judge, Locke, & they should feel more responsible for its success, which
Durham, 1997), and behaving badly would contradict those they can help enhance with good job performance. Overall,
assessments. Therefore, reduction in deviant behaviors that when employees show greater attention to their job and
violate organizational norms and threaten the well-being of greater interest in the organization, their performance
the organization, its members, or both (Robinson & Bennett, should be improved.
1995) is another possible consequence of self-efficacy.
These behaviors include major offenses (e.g., stealing, bul- Hypothesis 5: Psychological ownership is positively
lying, and sabotaging), or less offensive behaviors includ- related to followers’ in-role performance.
ing publically criticizing the organization or gossiping
about coworkers, and employees’ self-efficacy may be neg- Parallel to the hypothesized effects of self-efficacy on
atively related to them because they tend to feel intrinsically deviant behaviors, psychological ownership may also serve
rewarded only for more positive accomplishment. to inhibit employees from engaging in deviant acts against
According to arguments by Fox and Spector (1999), the organization and other individuals in it. Employee devi-
employees may exhibit counterproductive behavioral ance occurring within the organization may result from a
responses when experiencing stressful events (e.g., failing failure of responsibility to others and a lack of belonging-
to achieve personal and organizational goals). Individuals ness to the work setting. Psychological ownership is likely
with high-efficacy beliefs may be less vulnerable to the to help overcome dysfunctional behaviors because it places
negative influence of stressors and react to stressful situa- greater responsibility on the employee for the welfare of the
tions with positive attitudes rather than responding with organization, raising an employee’s sense of self-identity
deviant behaviors. This is because they feel confident in and belongingness. Employees should not purposely harm
their abilities to effectively control and endure adverse the organization if they consider themselves to be its own-
situations; that is, to continue being successful by com- ers. Therefore, organization members who feel strong own-
pleting tasks well (Bandura, 1997). Supporting this ership for the organization will be less likely to engage in
assumption, army soldiers with high self-efficacy behaved dysfunctional behaviors toward it. For example, employees
470 Journal of Leadership & Organizational Studies 24(4)

who feel a sense of ownership for their organization criti- & Podsakoff, 2012). This amount of time between measure-
cize the organization less and disclose confidential com- ments has been used successfully in past leadership studies
pany information less (Van Dyne & Pierce, 2004). Similarly, and therefore should be an appropriate delay between data
when employees feel ownership toward a variety of targets collections (e.g., Demirtas, 2015; Neubert, Kacmar,
within the organization, they tend to believe that they have Carlson, Chonko, & Roberts, 2008; Zohar & Polachek,
the right to influence the targets in positive, but not negative 2015). To further reduce response bias effects, social desir-
ways, and feel more accountable to the targets (Pierce, ability was used as a control variable in case there was
Rubenfeld, & Morgan, 1991). Therefore, it is expected that socially desirable responding across variables. Initially, 606
feelings of ownership to lead organizational members to not participants completed the first survey. Of these, 375 (61%)
only adopt behaviors that are beneficial to the targets but also employees completed the second survey 3 weeks later. Due
to discourage them from exhibiting certain types of undesir- to missing data, unmatched codes Time 1 (T1) and Time 2
able behavior such as bullying and theft. Supporting these (T2), and low effortful responding (data from participants
assumptions, psychological ownership was found to be nega- who selected the same answers [80%] for scales and who
tively associated with counterproductive work behaviors in completed the survey more than twice as fast as the average
one previous study (Avey et al., 2009). Therefore, we expect time were removed), 76 participants’ data were eliminated,
it to be negatively related to bad or deviant behaviors in the resulting in a final sample of 299 participants answering
present study. both surveys. Of the total sample, 56.2% were female,
82.9% were White, and 66.1 % had at least a bachelor’s
Hypothesis 6: Psychological ownership is negatively degree. The sample consisted of relatively young employ-
related to followers’ deviant behaviors. ees: 62.0% were between 18 and 39 years, 17.8% between
40 and 49 years, 16.8% between 50 and 59 years, and 3.4%
for more than 60 years old. In terms of employment tenure,
Self-Efficacy and Psychological Ownership as
9.0% were employed for less than a year, 52.2% from 1 to 5
Mediators years, 18.4% from 6 to 10 years, and 20.4% for more than
The set of hypotheses is embedded within and form the 10 years. The participants were employees from a variety of
model in Figure 1, and testing the model constituted a industries (e.g., education, health care, and finance).
simultaneous test of the hypotheses. Especially key to the
study’s model, however, is the process of mediation. We Measures
propose that the reasons why empowering leadership is
related to employees’ in-role performance and deviant The first survey (T1) included measures of empowering
behaviors are employees’ psychological states of self-effi- leadership, general self-efficacy, psychological ownership,
cacy and psychological ownership, which mediate the rela- and demographics. The second survey (T2) measured in-
tionships of empowering leadership with both good and bad role performance, deviant behaviors, social desirability,
employee behaviors. Empowering leadership is linked to general self-efficacy, and psychological ownership. Thus,
the two outcomes to the extent that it may affect these the two mediators—self-efficacy and psychological owner-
employee states. ship—were measured at both times to use their average as
an estimate of their scores at the midpoint (designated as
Hypothesis 7: Employees’ self-efficacy and psycho- T1½) between T1 and T2. Temporally, mediators should
logical ownership mediate the relationships of empow- have their effects between the timing of the predictor and
ering leadership with in-role performance and deviant the criterion variables, which suggest a three-wave data col-
behaviors. lection process. Each data collection period in a research
project typically results in addition sample attrition, how-
ever, and to avoid excessive attrition we employed only two
Method measurement times 3 weeks apart but used the data to esti-
mate scores on the mediators between those two time peri-
Participants and Procedure ods. The mean of the mediator scores at the first and second
Participants were recruited using Amazon’s Mechanical time points estimates their scores between those time points
Turk, an online survey system connecting researchers and if their scores change over time, an assumption we support
survey respondents. Only full-time employees working in statistically below.
the United States were recruited for this survey. Data were Empowering leadership was measured using the 18-item
collected at two time points, 3 weeks apart, to measure the Empowering Leadership Scale (Amundsen & Martinsen,
outcomes separately from the predictors, which helps 2014a). Four items (α = .88) measured power sharing, six
reduce common method bias effects (Podsakoff, MacKenzie, items (α = .93) measured development support, and eight
Kim and Beehr 471

items (α = .91) measured motivational support. Example Example items include “Said something hurtful to someone
items are “My leader gives me authority over issues within at work” (interpersonal deviance) and “Taken an additional
my department” (power sharing), “My leader shows me how or longer break than is acceptable at your workplace” (orga-
I can improve my way of working” (development support), nizational deviance). Respondents were asked to rate the
and “My leader listens to me” (motivational support), rated extent they had participated in each of the behaviors over
on a 7-point Likert-type scale from 1 (strongly disagree) to 7 the past year using a 7-point Likert-type scale from 1 (never)
(strongly agree). Reliability for the full scale was .96. to 7 (daily). Reliability for the full scale was .93.
General self-efficacy was measured with the eight-item Control variables were social desirability and gender,
(T1 α = .92; T2 α = .94) New Generalized Self-Efficacy because responses of in-role performance and deviant
Scale developed by Chen, Gully, and Eden (2001). An behaviors could especially be susceptible to social desir-
example item is “When facing difficult tasks, I am certain ability biases, and gender may also influence employee
that I will accomplish them” rated on a 5-point Likert-type deviant behaviors. Two meta-analyses suggested that males
scale from 1 (strongly disagree) to 5 (strongly agree). The are more aggressive than females, displaying more deviant
mean absolute difference in self-efficacy from T1 to T2 was behaviors (Berry, Ones, & Sackett, 2007). Social desirabil-
0.28 with a standard deviation of 0.33. A one-sample t test ity was measured at T2 using the five-item (α = .74) Socially
showed that the average absolute change was significant, Desirable Response Set (Hays, Hayashi, & Stewart, 1989).
t(298) = 14.53, p < .01, indicating that employees’ scores An example item is “No matter who I’m talking to, I am
had changed during this 3-week interval. Note that because always a good listener.” Respondents were asked to indicate
absolute differences were tested, the result does not mean the degree to which each statement is true or false on a
that the scores increased; some may have increased and 5-point Likert-type scale from 1 (definitely true) to 5 (defi-
some decreased. Because there were changes from T1 to nitely false). Because social desirability was measured from
T2, the mean of T1 and T2 was used to estimate the score the same source as the other variables (the employee), using
between T1 and T2. it as a control variable helped control not only for the
Psychological ownership was assessed using six items socially desirability construct but also for common method
(T1 α = .86; T2 α = .87) from Avey et al. (2009), which variance, similar to the marker variable approach (Podsakoff
reflected the three dimensions of psychological ownership. et al., 2012).
The original psychological ownership scale had eight items,
but two of them referred directly to self-efficacy, which
overlapped with the wording of items in the self-efficacy
Results
scale described above. Therefore, those two items were not Means, standard deviations, reliabilities, and correlations of
used in the psychological ownership measure. Example all the variables are presented in Table 1. Because empow-
items are “I would challenge anyone in my organization if I ering leadership and psychological ownership were highly
thought something was done wrong” (accountability; T1 α correlated (r = .70, p < .01), a confirmatory factor analysis
= .78; T2 α = .75), “I feel I belong in this organization” using LISREL 8.8 was run to test the fit of the two-factor
(belongingness; T1 α = .83; T2 α = .88), and “I feel being a model (using the three dimensions for empowering leader-
member in this organization helps define who I am” (self- ship and three dimensions for psychological ownership as
identity; T1 α = .77; T2 α = .80). Items were rated on a indicators). The strong correlation could have meant that
6-point Likert-type scale ranging from 1 (strongly disagree) the measures were overlapping in some artefactual way
to 6 (strongly agree). The mean absolute difference in psy- because participants were not distinguishing between the
chological ownership from T1 to T2 was 0.52 with a stan- constructs. The two-factor model produced a reasonable fit,
dard deviation of 0.49. A one-sample t test showed that the however, χ2(8, N = 299) = 27.72, p < .01; root mean square
average absolute change was significant, t(298) = 18.62, p of approximation [RMSEA] = .09; comparative fit index
< .01, indicating that many employees’ scores had changed [CFI] = .99; nonnormed fit index [NNFI] =.98; incremental
during this 3-week interval. Therefore, the mean of T1 and fit index [IFI] = .99, providing evidence of discriminant
T2 was used to estimate the score between T1 and T2. validity of empowering leadership and psychological own-
In-role performance was assessed using Williams and ership, in spite of the high correlation. A confirmatory fac-
Anderson’s (1991) seven-item job performance scale (α = tor analysis of the two mediators of self-efficacy and
.87). An example item is “I adequately complete assigned psychological ownership was also run to statistically con-
duties” rated on a 7-point Likert-type scale from 1 (strongly firm the differences between these constructs. The result
disagree) to 7 (strongly agree). revealed the two mediator constructs in the present study to
Workplace deviance was measured with the 19 items be distinct, χ2(8, N = 299) = 38.76, p < .01; RMSEA = .11;
developed by Bennett and Robinson (2000). Interpersonal CFI = .98; NNFI = .96; IFI = .98.
deviance was measured with 7 items (α = .87), and organi- Empowering leadership, the predictor in the model, was
zational deviance was measured with 12 items (α = .89). positively correlated to the two mediators, self-efficacy
472 Journal of Leadership & Organizational Studies 24(4)

Table 1.  Means, Standard Deviations, Reliabilities, and Correlations.

Variables M SD 1 2 3 4 5 6 7
1.  Empowering leadership (T1) 4.90 1.15 .96  
2.  Self-efficacy (T1 + T2 average) 4.09 0.55 .49** .92/.94  
3. Psychological ownership 4.35 0.89 .70** .51** .86/.87  
(T1 + T2 average)
4.  In-role performance (T2) 6.09 0.95 .14* .45** .18** .87  
5.  Deviant behavior (T2) 1.73 0.86 −.16** −.28** −.25** −.35** .93  
6.  Social desirability (T2) 3.60 0.75 .10 .28** .14* .30** −.39** .74  
7.  Gender (T1) 1.56 0.50 −.04 .06 −.10 .03 −.14* .09 —

Note. N = 299. Reliabilities are in italics on the diagonal. Gender: Male = 1; Female = 2.
*p < .05. **p < .01.

Table 2.  Summary of Model Fit Indices.

Model test χ2 df RMSEA CFI NNFI IFI Δχ2 Δdf ∆p

Measurement model 546.31 174 .08 .95 .94 .95  
Hypothesized model 583.79 181 .09 .95 .94 .95  
Alternative model 1: Direct path from 580.68 180 .09 .95 .94 .95 3.11 1 .08
empowering leadership to in-role performance
Alternative model 2: Direct path from 581.28 180 .09 .95 .94 .95 2.51 1 .11
empowering leadership to deviant behaviors

Note. df = degrees of freedom; RMSEA = root mean square of approximation; CFI = comparative fit index; NNFI = nonnormed fit index; IFI =
incremental fit index. χ2 Values for the structural models are significant at p < .01.

(r = .49, p < .01; Table 1) and psychological ownership (r = (two consisting of three items each and one of two items).
.70, p < .01); the predictor was also related to the criteria Similarly, the seven responses from the in-role performance
(empowering leadership to in-role performance, r = .14, p < were parceled into three indicators (two consisting of two
.05, and to deviant behaviors, r = −.16, p < .01). The two items each and one of three items). Finally, for the five-item
mediators were related to the criteria: Self-efficacy was social desirability scale, the items were used as indicators.
related to in-role performance (r = .45, p < .01) and deviant All the model fit indices are shown in Table 2. The
behaviors (r = −.28, p < .01), and psychological ownership hypothesized model fit the data moderately well, χ2(181, N
was related to in-role performance (r = .18, p < .01) and = 299) = 583.79, p < .01; RMSEA = .09; CFI = .95; NNFI
deviant behaviors (r = −.25, p < .01). = .94; IFI = .95. Standardized coefficients are in Figure 2.
Empowering leadership was positively related to self-effi-
cacy (β = .54, p < .01) and psychological ownership (β =
Hypotheses and Model Testing .74, p < .01), supporting Hypotheses 1 and 2. The results
The correlations in Table 1, as noted above, were consistent also supported Hypothesis 3, that self-efficacy would be
with all hypotheses. Structural equation modeling was used positively related to in-role performance (β = .44, p < .01),
to further test the hypotheses. For full structural equation as well as Hypothesis 4, that self-efficacy would be nega-
modeling with latent variables, subscales were used as man- tively related to deviant behaviors (β = −.15, p < .01).
ifest indicators of the two measures that had three subscales However, Hypothesis 5 was not supported, because the path
(empowering leadership and psychological ownership). For coefficient for psychological ownership to in-role perfor-
deviant behaviors, which had only two subscales, four indi- mance (β = −.01, ns) was not significant. Hypothesis 6, that
cator variables were used: two parcels of interpersonal psychological ownership would be negatively related to
deviance (one consisting of three items and one of four deviant behaviors was supported (β = −.19, p < .01).
items) and two parcels of organizational deviance (two con- Regarding the two control variables, gender and social
sisting of six items each). For the two other variables with desirability, the association between gender and deviant
more than five items (self-efficacy, in-role performance, behaviors was very weak (r = −.14, p < .05). We therefore
and social desirability), the item-parceling method was included only social desirability as a control variable in the
used. The eight items measuring self-efficacy were ran- analyses presented in Figure 1: It was moderately correlated
domly parceled to form three indicators of a latent variable with in-role performance (r = .30, p < .01) and deviant
Kim and Beehr 473

Figure 2.  Structural equation model with standardized coefficients.

Note. All paths in structural model analysis are significant at p < .01 except the path from psychological ownership to in-role performance (β = −.01, ns).

behaviors (r = −.39, p < .01). We note, however, that more Mediation and Bootstrapping
complete analyses showed that neither control variable sig-
nificantly affected the conclusions in the present study. In Furthermore, for more rigorous tests of the mediation pro-
summary, the results suggested that empowering leadership posed in Hypothesis 7, bootstrapping analyses were calcu-
had indirect effects on deviant behaviors via both self-effi- lated using PROCESS macro, a computational tool for
cacy and psychological ownership, but empowering leader- mediation analysis (Hayes, 2013). Table 3 presents the
ship had effects on in-role performance only through one direct effects and bootstrapped estimates for the indirect
mediator, self-efficacy. effects with 95% confidence intervals. Following the rec-
The fit of the overall model is consistent with the media- ommendations of Preacher and Kelley (2011), k2 (Kappa-
tion hypothesis (Hypothesis 7). For further evidence about squared) is also reported as a mediation effect size. The k2 is
mediation, two alternative overall models were tested: one not sensitive to sample size, because the k2 is the ratio of the
model adding a direct path from empowering leadership to indirect effect to the maximum possible size of the indirect
in-role performance (Alternative Model 1), and a second effect given the constraints of the data (Hayes, 2013). To
model adding a direct path from empowering leadership to determine the criteria for describing the magnitude of effect
deviant behaviors (Alternative Model 2). If mediation is a sizes, Cohen’s guidelines defining small (.01), medium
key feature in the model, the alternative models with direct (.09), and large (.25) effect sizes were considered (Preacher
paths would not significantly improve the fit, and that is & Kelley, 2011). Both self-efficacy and psychological own-
what happened. The additional paths did not significantly ership significantly mediated the relationships between
improve or change any fit indices, Δχ2(1, N = 299) = 3.11, p empowering leadership and the two outcomes, because
= .08, for Alternative Model 1; Δχ2(1, N = 299) = 2.51, p = their confidence intervals did not include a zero, supporting
.11, for Alternative Model 2 (Table 2). Therefore, given no mediation effects.
noticeable improvement in fit, the originally hypothesized As an example, the model included empowering lead-
mediation model was retained as the ideal model because it ership as a predictor and in-role performance as an out-
was the most parsimonious. come, and this relationship was mediated by psychological
474 Journal of Leadership & Organizational Studies 24(4)

Table 3.  Results of Bootstrapping Tests for Estimating Indirect Effects with 95% Confidence Intervals (CI).

Direct
effects Indirect effects

Predictor Mediator Outcome β (p) ab SE 95% CI abcs k2

Empowering Self-efficacy In-role performance −.09 (.07) .20 .04 [.13, .29] .25 .23
leadership Deviance −.03 (.58) −.09 .02 [−.15, −.05] −.13 .11
Psychological In-role performance .02 (.77) .10 .04 [.02, .18] .12 .08
ownership Deviance .01 (.25) −.14 .05 [−.24, −.05] −.18 .13

Note. N = 299. β = c′ (direct effect). ab = unstandardized indirect effect. SE = bootstrap standard error. ab, SE, and 95% CI were obtained from 10,000
bootstrap samples. abcs = completely standardized indirect effect. k2 = indirect effect/maximum possible mediation.

ownership. As seen in the third row of Table 3, the direct is, employees’ performance can be enhanced if leaders can
effect from empowering leadership to in-role performance increase followers’ feelings of self-efficacy by offering
was not significant (c′ = .02, p = .77). However, the indi- coaching and professional challenges with high standards,
rect effect of empowering leadership on in-role perfor- which can be components of empowering leadership.
mance via psychological ownership was significant (ab = No significant direct effect was found between empow-
.10, confidence interval [95%]: lower limit = .02, upper ering leadership and either in-role performance or deviant
limit = .18, k2 = .08), and the effect size of k2 can be inter- behaviors, consistent with the mediation model. Some other
preted as small with reference to Cohen’s standard. research found weak or nonsignificant relationships of
However, we advise caution when interpreting this par- empowering leadership with criteria (e.g., Srivastava et al.,
ticular finding, because the standardized path coefficient 2006), and this is because empowering leadership is a more
from psychological ownership to in-role performance in distal predictor; its effects occur only to the extent that it
the path diagram (Figure 2) was not significant. Note that leads to more proximal influences on employee behavior
the bivariate correlation of these two variables in Table 1 such as the two mediators identified in the present study.
was significant (r = .18, p < .01). The nonsignificant rela- The present study, therefore, contributes to the empowering
tionship between psychological ownership and in-role leadership literature by explaining how and why it may
performance in Figure 2 may be a statistical artifact caused affect followers’ behaviors.
by a suppression effect that occurred when two or more Regarding the specific outcome variables in the present
variables (self-efficacy and psychological ownership in study, past research examining the impact of empowering
the present study) jointly predicted one outcome (in-role leadership on contextual performance including deviant
performance). behaviors is scarce, although some studies have suggested
positive relationships between empowering leadership and
positive organizational citizenship behaviors (Fong &
Discussion Snape, 2015; Raub & Robert, 2010). Unlike past research
The primary purpose of the present study was to explain focusing only on positive employee behaviors, the present
why empowering leadership could affect employees’ in- study examined the relationship of empowering leadership
role performance and deviant behaviors by examining the with workplace deviance. By stimulating self-efficacy and
potential mediating role of self-efficacy and psychological psychological ownership, empowering leaders may inhibit
ownership, thereby shedding light on the mechanisms by unfavorable behaviors.
which empowering leadership influences subordinates’ Additionally, by leaders’ role-modeling and letting sub-
behaviors. The results generally supported the hypothesized ordinates actually make decisions, the overall meaning of
relationships: Empowering leadership may affect these the study’s mediators can be conceived as subordinates
employee behaviors to the extent that it leads to these two experiencing a sense of responsibility and as a result, mak-
psychological states. Subordinates with empowering lead- ing positive behavioral contributions to the organization.
ers encouraging their initiative and responsibility toward Given that employees’ deviant behaviors can impede orga-
the job experienced enhanced feelings of self-efficacy and nizational effectiveness, development of employees’ self-
psychological ownership (including facets of self-identity, efficacy and psychological ownership can be potential
belongingness, and accountability), and thereby engaged in human resource management strategies to counteract work-
fewer deviant behaviors. In addition to reducing the bad place deviance and promote individual performance in
employee behaviors (deviance), self-efficacy also contrib- organizations. If employees feel lack of belongingness to
utes to good employee behaviors in the form of in-role per- their organization, low self-efficacy, and blurred self-iden-
formance, although psychological ownership may not. That tity and accountability, the organization may fail to thrive.
Kim and Beehr 475

Both deviant behaviors and in-role performance, therefore, (Podsakoff et al., 2012) and (b) the results were the same
are important workplace behaviors that may be affected by even controlling for social desirability (and gender). As
empowering leadership. noted earlier, because social desirability was measured
Contrary to theoretical assumptions about one of the with the same method as the other variables in the study, its
study’s mediators, psychological ownership, the present use as a control variable helps control for both the effects
study did not support a positive unique relationship for it of social desirability and for common method effects.
with in-role performance. Psychological ownership thus Nevertheless, it would be recommended that ratings of
may transfer empowering leaders’ effects to dampen bad behavioral outcomes such as in-role performance and devi-
employee behaviors (deviance) but not necessarily to ant behaviors should be collected from supervisors,
increase good ones (performance). One possible explana- coworkers, or objective sources to further reduce common
tion is that a sense of ownership may more strongly lead to method effects.
defending than to striving. That is, employees experiencing Relatedly, the leadership variable is a measure of the per-
ownership will not hurt the organization (will not perform ceptions of the employees about their supervisors’ empow-
deviant behaviors), but they also do not strive very much to ering behaviors rather than a clearly objective measure of
help it through improved performance. This owner defense the behaviors. One of the most objective operationaliza-
mechanism needs to be tested by further research, however, tions of empowering leadership would be to manipulate the
before concluding that it exists, because the result may be leaders’ behaviors in an experiment. Future research using
inconsistent. One previous study similarly examined these experimental manipulations of the leaders’ behaviors could
relationships, without finding positive links between psy- provide stronger evidence about the effects of empowering
chological ownership and general employee performance leader behaviors.
(Mayhew, Ashkanasy, Bramble, & Gardner, 2007). By con- Although the timing of the measures’ data collection was
trast, in a study by Brown, Pierce, and Crossley (2014), consistent with a causal interpretation of empowering lead-
feelings of ownership significantly affected sales perfor- ership causing self-efficacy and psychological ownership,
mance. If the relationship is inconsistent, that implies pos- which then causes in-role behavior and deviant behavior,
sible moderator effects. Therefore, there is a further need to only a true experiment with manipulated variables and ran-
examine these relationships in order to replicate the positive dom assignment could provide strong evidence for causa-
effect of psychological ownership on employee perfor- tion. The interpretation of causation in the model must
mance or to explain possible moderator effects; such mod- therefore rest on the theoretical rationales for the hypothe-
erators might include individual differences (e.g., ses and the gathering of measures in the hypothesized
ego-defensive personality) and the nature of the job; for sequence. Thus, future research would benefit from adopt-
example, to what extent is in-role performance possible in ing an experimental or quasi-experiment method, especially
some jobs where maximum performance is limited by con- manipulating the empowering leadership variable (e.g., in
straints (e.g., amount of job resources available or assem- the form of leader training) in order to more firmly establish
bly-line work wherein the employee cannot work faster causation.
than the line moves)? The present study included both positive and negative
Overall, empowering leadership can lead to employees’ behavioral outcomes, but there are many other potential cri-
experiences of psychological ownership and feelings of teria that need to be explored to provide better understanding
self-efficacy in today’s organizations. In organizations of the contribution of empowering leader behaviors in the
exercising empowering leadership style and practicing workplace. Some examples of variables for future research
employees’ development support, employees feel responsi- include other consequences of empowering leadership, such
ble for their actions, improving their performance and as employees’ life satisfaction, work happiness, and physical
decreasing workplace-deviant behaviors. Therefore power and psychological health. Autonomy provided by empower-
sharing, fostering autonomy, motivational support, recogni- ing leadership is closely related to job stressor appraisals,
tion, guidance, and modeling (i.e., empowering leadership) which in turn is likely to have an impact on employees’ well-
are important for shaping a sense of self-efficacy and psy- being. Future research also could explore boundary condi-
chological ownership in employees. tions that accentuate or mitigate the strength of relations
hypothesized in the present study. For example, employees
having strong need for autonomy or internal locus of control
Limitations and Future Research
may respond more favorably to empowering leadership.
One limitation of this study is the use of single-source data, Additionally, certain types of autonomy such as the discre-
risking problems such as inflated relationships due to com- tion to schedule work and to choose the work methods used
mon method variance. That effect is unlikely in the present to accomplish jobs may result in different reactions and
study, however, because (a) the predictor, mediator, and behaviors of employees. Future research, therefore, is
criterion variables were not gathered at the same time encouraged to examine the effect of level or type of
476 Journal of Leadership & Organizational Studies 24(4)

autonomy on employees’ behaviors, which may offer further Amundsen, S., & Martinsen, Ø. L. (2014b). Self–other agreement
understanding of how working autonomously may drive in empowering leadership: Relationships with leader effec-
desired work outcomes as well as psychological states. tiveness and subordinates’ job satisfaction and turnover inten-
In addition to destructive deviant behaviors or construc- tion. Leadership Quarterly, 25, 784-800.
Amundsen, S., & Martinsen, Ø. L. (2015). Linking empower-
tive citizenship behaviors, empowering leadership may
ing leadership to job satisfaction, work effort, and creativity:
elicit constructive deviant behaviors—defined as “behav-
The role of self-leadership and psychological empowerment.
iors that violate organizational norms with the intent of Journal of Leadership & Organizational Studies, 22, 304-323.
helping the organization” (Galperin, 2012, p. 2989). Vadera, Ashforth, B. E., & Mael, F. (1989). Social identity theory and the
Pratt, and Mishra (2013) proposed psychologically empow- organization. Academy of Management Review, 14, 20-39.
ered employees were likely to exhibit constructive deviant Avey, J. B., Avolio, B. J., Crossley, C. D., & Luthans, F.
behaviors including searching for innovative methods to (2009). Psychological ownership: Theoretical extensions,
achieve work goals, even if the methods might be consid- measurement and relation to work outcomes. Journal of
ered “incorrect” by the organization’s current practices and Organizational Behavior, 30, 173-191.
procedures. Bandura, A. (1986). Social foundations of thought and action: A
social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New
Conclusion York, NY: Freeman.
Bennett, R. J., & Robinson, S. L. (2000). Development of a mea-
We examined how empowering leadership style may affect sure of workplace deviance. Journal of Applied Psychology,
subordinates’ behaviors by testing potential explanations 85, 349-360.
for effects of empowerment in the form of two mediating Berry, C. M., Ones, D. S., & Sackett, P. R. (2007). Interpersonal
variables: self-efficacy and psychological ownership. deviance, organizational deviance, and their common cor-
Empowering leaders who give subordinates autonomy and relates: A review and meta-analysis. Journal of Applied
support for pursuing unstructured tasks make them feel Psychology, 92, 410-424.
more personally accountable and more engaged in work Brown, G., Crossley, C., & Robinson, S. L. (2014). Psychological
processes; this can result in employees’ behaviors that are ownership, territorial behavior, and being perceived as a team
favorable for the organization due to their feelings of self- contributor: The critical role of trust in the work environment.
Personnel Psychology, 67, 463-485.
efficacy and psychological ownership. Leaders developing
Brown, G., Pierce, J. L., & Crossley, C. (2014). Toward an under-
subordinates’ self-efficacy and psychological ownership
standing of the development of ownership feelings. Journal of
may improve task performance and decrease deviant behav- Organizational Behavior, 35, 318-338.
iors. In summary, findings suggested that self-efficacy and Brown, M. E., & Treviño, L. K. (2006). Socialized charismatic
psychological ownership are critical mediating processes leadership, values congruence, and deviance in work groups.
because empowering leadership may not directly affect Journal of Applied Psychology, 91, 954-962.
subordinates’ behavioral responses. Employees’ psycho- Chen, G., Casper, W. J., & Cortina, J. M. (2001). The roles of
logical states explain why empowering leadership affects self-efficacy and task complexity in the relationships among
their behaviors. cognitive ability, conscientiousness, and work-related perfor-
mance: A meta-analytic examination. Human Performance,
14, 209-230.
Declaration of Conflicting Interests
Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a
The author(s) declared no potential conflicts of interest with new general self-efficacy scale. Organizational Research
respect to the research, authorship, and/or publication of this Methods, 4, 62-83.
article. Cheong, M., Spain, S. M., Yammarino, F. J., & Yun, S. (2016).
Two faces of empowering leadership: Enabling and burden-
Funding ing. Leadership Quarterly, 27, 602-616.
Cordery, J. L., Morrison, D., Wright, B. M., & Wall, T. D. (2010).
The author(s) received no financial support for the research,
The impact of autonomy and task uncertainty on team perfor-
authorship, and/or publication of this article.
mance: A longitudinal field study. Journal of Organizational
Behavior, 31, 240-258.
References Demirtas, O. (2015). Ethical leadership influence at organizations:
Ahearne, M., Mathieu, J., & Rapp, A. (2005). To empower or not Evidence from the field. Journal of Business Ethics, 126, 273-
to empower your sales force? An empirical examination of the 284.
influence of leadership empowerment behavior on customer Dewettinck, K., & van Ameijde, M. (2011). Linking leadership
satisfaction and performance. Journal of Applied Psychology, empowerment behaviour to employee attitudes and behav-
90, 945-955. ioural intentions: Testing the mediating role of psychological
Amundsen, S., & Martinsen, Ø. L. (2014a). Empowering leader- empowerment. Personnel Review, 40, 284-305.
ship: Construct clarification, conceptualization, and valida- Dysvik, A., & Kuvaas, B. (2011). Intrinsic motivation as a moderator
tion of a new scale. Leadership Quarterly, 25, 487-511. on the relationship between perceived job autonomy and work
Kim and Beehr 477

performance. European Journal of Work & Organizational employee behavior. Journal of Applied Psychology, 93, 1220-
Psychology, 20, 367-387. 1233.
Erkutlu, H., & Chafra, J. (2013). Effects of trust and psychological Pearce, C. L., Sims, H. P., Jr., Cox, J. F., Ball, G., Schnell, E.,
contract violation on authentic leadership and organizational Smith, K. A., & Trevino, L. (2003). Transactors, transform-
deviance. Management Research Review, 36, 828-848. ers and beyond: A multi-method development of a theoretical
Fong, K. H., & Snape, E. (2015). Empowering leadership, psy- typology of leadership. Journal of Management Development,
chological empowerment and employee outcomes: Testing a 22, 273-307.
multi-level mediating model. British Journal of Management, Pierce, J. L., Kostova, T., & Dirks, K. (2001). Toward a theory
26, 126-138. of psychological ownership in organizations. Academy of
Fox, S., & Spector, P. E. (1999). A model of work frustration– Management Review, 26, 298-310.
aggression. Journal of Organizational Behavior, 20, 915-931. Pierce, J. L., Rubenfeld, S. A., & Morgan, S. (1991). Employee
Galperin, B. L. (2012). Exploring the nomological network of ownership: A conceptual model of process and effects.
workplace deviance: Developing and validating a measure of Academy of Management Review, 16, 121-144.
constructive deviance. Journal of Applied Social Psychology, Podsakoff, P. M., MacKenzie, S. P., & Podsakoff, N. P. (2012).
42, 2988-3025. Sources of method bias in social science research and rec-
Hayes, A. F. (2013). Introduction to mediation, moderation, and ommendations on how to control it. Annual Review of
conditional process analysis: A regression-based approach. Psychology, 63, 539-569.
New York, NY: Guilford Press. Preacher, K. J., & Kelley, K. (2011). Effect size measures for
Hays, R. D., Hayashi, T., & Stewart, A. L. (1989). A five-item mediation models: Quantitative strategies for communicating
measure of Socially Desirable Response Set. Educational and indirect effects. Psychological Methods, 16, 93-115.
Psychological Measurement, 49, 629-636. Raub, S., & Robert, C. (2010). Differential effects of empower-
Jex, S. M., & Bliese, P. D. (1999). Efficacy beliefs as a moderator ing leadership on in-role and extra-role employee behaviors:
of the impact of work-related stressors: A multilevel study. Exploring the role of psychological empowerment and power
Journal of Applied Psychology, 84, 349-361. values. Human Relations, 63, 1743-1770.
Judge, T. A., Locke, E. A., & Durham, C. C. (1997). The disposi- Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant
tional causes of job satisfaction: A core evaluations approach. workplace behaviors: A multi-dimensional scaling study.
Research in Organizational Behavior, 19, 151-188. Academy of Management Journal, 38, 555-572.
Katz, D., & Kahn, R. L. (1966). The social psychology of organi- Srivastava, A., Bartol, K. M., & Locke, E. A. (2006). Empowering
zations. New York, NY: John Wiley. leadership in management teams: Effects on knowledge shar-
Latham, G. P., & Frayne, C. A. (1989). Self-management training ing, efficacy and performance. Academy of Management
for increasing job attendance: A follow-up and a replication. Journal, 49, 1239-1251.
Journal of Applied Psychology, 74, 411-416. Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-
Latham, G. P., Winters, D. C., & Locke, E. A. (1994). Cognitive related performance: A meta-analysis. Psychological Bulletin,
and motivational effects of participation: A mediator study. 124, 240-261.
Journal of Organizational Behavior, 15, 49-63. Tekleab, A. G., Sims, H. P., Jr., Yun, S., Tesluk, P. E., & Cox, J.
Litzky, B. E., Eddleston, K. A., & Kidder, D. L. (2006). The (2008). Are we on the same page? Effects of self-awareness
good, the bad, and the misguided: How managers inadver- of empowering and transformational leadership. Journal of
tently encourage deviant behaviors. Academy of Management Leadership & Organizational Studies, 14, 185-202.
Perspectives, 20, 91-103. Vadera, A. K., Pratt, M. G., & Mishra, P. (2013). Constructive
Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater: deviance in organizations integrating and moving forward.
A partial test of the reformulated model of organizational Journal of Management, 39, 1221-1276.
identification. Journal of Organizational Behavior, 13, 103- Van Dyne, L., & Pierce, J. L. (2004). Psychological ownership
123. and feelings of possession: Three field studies predicting
Mayhew, M. G., Ashkanasy, N. M., Bramble, T., & Gardner, J. employee attitudes and organizational citizenship behavior.
(2007). A study of the antecedents and consequences of psy- Journal of Organizational Behavior, 25, 439-459.
chological ownership in organizational settings. Journal of van Gils, S., van Quaquebeke, N., van Knippenberg, D., van Dijke,
Social Psychology, 147, 477-500. M., & De Cremer, D. (2015). Ethical leadership and follower
Morgeson, F. P., Delaney-Klinger, K., & Hemingway, M. A. organizational deviance: The moderating role of follower
(2005). The importance of job autonomy, cognitive ability, moral attentiveness. Leadership Quarterly, 26, 190-203.
and job-related skill for predicting role breadth and job per- Vecchio, R. P., Justin, J. E., & Pearce, C. L. (2010). Empowering
formance. Journal of Applied Psychology, 90, 399-406. leadership: An examination of mediating mechanisms within
Mustafa, M., Martin, L., & Hughes, M. (2016). Psychological a hierarchical structure. Leadership Quarterly, 21, 530-542.
ownership, job satisfaction, and middle manager entrepre- Vroom, V. H. (1964). Work and motivation. New York, NY:
neurial behavior. Journal of Leadership & Organizational Wiley.
Studies, 23, 273-287. Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and
Neubert, M. J., Kacmar, K. M., Carlson, D. S., Chonko, L. B., organizational commitment as predictors of organizational
& Roberts, J. A. (2008). Regulatory focus as a mediator of citizenship and in-role behaviors. Journal of Management,
the influence of initiating structure and servant leadership on 17, 601-617.
478 Journal of Leadership & Organizational Studies 24(4)

Zhang, X., & Bartol, K. M. (2010). Linking empowering lead- Author Biographies
ership and employee creativity: The influence of psycho-
Minseo Kim is an I/O Psychology PhD student at Central
logical empowerment, intrinsic motivation, and creative
Michigan University. Her research interests include occupational
process engagement. Academy of Management Journal, 53,
stress, leadership, motivation, and contextual performance.
107-128.
Zohar, D., & Polachek, T. (2015). Using event-level data to test Terry A. Beehr is a professor of psychology and member of the
the effect of verbal leader behavior on follower leadership I/O Psychology faculty at Central Michigan University. His
perceptions and job performance: A randomized field experi- research interests include occupational stress, retirement, leader-
ment. Group & Organization Management, 22, 1-31. ship, motivation, and careers.