Curr Psychol (2016) 35:77–82

DOI 10.1007/s12144-015-9363-5

Coping with Abusive Supervision: the Neutralizing Effects

of Perceived Organizational Support and Political Skill
on Employees’ Burnout
Xiaoyan Li 1 & Jing Qian 2 & Zhuo Rachel Han 3 & Zhuyun Jin 3

Published online: 2 August 2015

# Springer Science+Business Media New York 2015

Abstract While research into the antecedents of burnout has relationship was weaker for people with high, rather than low-
steadily grown, the relationship between abusive supervision er level of political skill.
and burnout remains largely unknown. In addition, we know
little about the contingencies under which abusive supervision Keywords Burnout . Abusive supervision . Perceived
may be related to employees’ burnout. This study aims to organizational support . Political skill
examine the contingency side of the abusive supervision-
burnout relationship by addressing the exploratory question
of whether perceived organizational support and individual The burnout syndrome refers to three distinct psychological
differences in political skill play moderating roles in the abu- responses to chronic work stress in which employees feel
sive supervision-burnout relationship. The present article de- emotionally “spent” or depleted of emotional resources (emo-
veloped a model and tested it with data from a sample of 248 tional exhaustion), display a detached attitude towards others
supervisor–subordinate dyads. We hypothesized and found (depersonalization), and experience a low sense of efficacy at
that (1) Abusive supervision was positively associated with work (reduced professional efficacy) (Maslach and Jack-
burnout; (2) the positive relationship between abusive super- son 1986). It has been consistently linked with physiolog-
vision and burnout was moderated by the employees’ per- ical and affective outcomes (Burke and Greenglass 1995;
ceived organizational support in such a way that the relation- Cherniss 1992; Lee and Ashforth 1988) as well as with
ship was weaker for employees who perceive higher rather organizational consequences such as reduced level of per-
than lower organizational support; (3) the positive relationship formance, negative work attitudes and increased turnover,
between abusive supervision and burnout was moderated by absenteeism, and organizational health care costs (Camer-
the employees’ political skill in such a way that the on et al. 1994; Jackson et al. 1986; Lee and Ashforth
1996; Wolpin et al. 1991; Wright and Bonett 1997;
Wright and Cropanzano 1998). In light of this, it is not
* Jing Qian surprising that scholars and practitioners alike have
[email protected] sought to identify the antecedents of burnout in order to
Xiaoyan Li effectively prevent and reduce it in the work context (e.g.,
[email protected] Halbesleben and Buckley 2004). Various personal charac-
Zhuo Rachel Han teristics such as extraversion and agreeableness as well as
[email protected] contextual factors such as job characteristics of perceived
Zhuyun Jin control and autonomy at work have been identified as the
[email protected] antecedents of burnout (Alarcon 2011).
1
School of Business, Jiangxi Normal University, Nanchang, China A recent study conducted by Chen and colleagues (2015)
2
Business School, Beijing Normal University, Beijing 100875, China suggested that leadership style is an important factor in the
3
Beijing Key Laboratory of Applied Experimental Psychology, service industry. Given the dominant role of supervisors in
School of Psychology, Beijing Normal University, Beijing 100875, the workplace, one key factor that may have substantial im-
China pact on employees’ burnout level is the supervisors’ behaviors
78 Curr Psychol (2016) 35:77–82

and practice. Abusive supervision, a form of nonphysical ag- political skill as those who are keenly aware of their social
gression, is a reality of today’s organization and, much of the context, are able to accurately interpret others’ behaviors and
extant research has, not surprisingly, focused on the influence motives, and to manage social interactions. Politically skillful
of abusive supervision on employees’ stress-related outcomes individuals have better abilities in interpersonal interactions
(e.g., Keashly et al. 1997; Tepper 2000, 2007). Defined as and view these interactions as opportunities, rather than
“subordinates’ perceptions of the extent to which leaders en- threats (Perrewé et al. 2000; Ferris et al. 2005, 2007). Al-
gage in the sustained display of hostile verbal and nonverbal though employees with political skills may still experience
behaviors, excluding physical contact” (Tepper 2000, p. 178), abusive supervision, possessing more political skills can help
abusive supervision has been empirically linked with em- them manage their supervisors’ behaviors and may potentially
ployees’ burnout (e.g., Carlson et al. 2012). influence the supervisors’ interactions with them. As a result,
More recently, leadership researchers have adopted an in- the positive relationship is weaker for employees with high
teractional approach in arguing that leadership or managerial political skill.
practice’s influence on subordinates may be contingent upon In the present study, we hypothesize that (1) abusive super-
certain contextual and personal factors (e.g., Qian et al. 2012; vision is positively related to employees’ burnout; (2) em-
Wang et al. 2005). This approach suggests that to fully ployees’ perceived organizational support moderates the pos-
understand the relationship between abusive supervision itive relationship between abusive supervision and burnout in
and subordinates’ burnout, one also needs to look at the such a way that the relationship will be weaker for employees
potential moderators to see a more complete picture. In who perceive higher rather than lower organizational support
this study, we examine one example of a contextual mod- and (3) employees’ political skill moderates the positive rela-
erator (i.e., perceived support from the organization) as tionship between abusive supervision and burnout in such a
well as one example of an individual moderator (i.e., fol- way that the relationship is weaker for people with high, rather
lowers’ political skill) for the relationship between abu- than lower level of political skill.
sive supervision and burnout. By doing so, our research
seeks to provide a more complete picture about the influ-
ence of abusive supervision on burnout level. Methods
Randall et al. (1999) defined perceived organizational sup-
port (POS) as employees’ perceptions about the extent to Sample and Procedure
which an organization values them and cares about their wel-
fare and well-being (Eisenberger et al. 1986; Loi et al. 2006). We collected data in person from a hotel group located in a
Research has suggested that POS serves as “a social- major city in China. The supervisor questionnaires were dis-
emotional resource for employees” when the need for support tributed to 64 supervisors at a company meeting held for su-
is salient (Armeli et al. 1998, p. 289). For example, Armeli pervisors. This was one of the four meetings during that
et al. (1998) found that POS can alleviate police officers’ need month, all of which were designed for training purpose to limit
for social-emotional support. In the present study, we suggest the size of each meeting. The participants for each meeting
that POS could buffer the harmful effect of abusive supervi- were randomly selected by computer. With the permission of
sion on employees’ burnout. This could ease the resource and the human resources department, we obtained the names of
emotion constraints suffered by those under abusive supervi- their subordinates and randomly selected 5 subordinates for
sion, and the positive relationship between abusive supervi- each supervisor from the direct name list via a computer pro-
sion and burnout could therefore be weakened if the POS is gram. With the help of the human resources department, we
higher. administered separate questionnaires to 64 supervisors and
Unsurprisingly, it is observed in the workplace that not all 320 of their immediate subordinates. Each questionnaire was
employees who have been abused by their supervisors expe- assigned an identification number so the responses of the sub-
rience the same level of burnout, highlighting the fact that the ordinates could be matched with the evaluations of their im-
relationship between abusive supervision and burnout may be mediate supervisors. Subordinates were asked to provide in-
influenced by employees’ individual differences. In this study, formation about their demographic variables, perceived abu-
we explore the moderating role of employees’ political skill sive supervision, burnout, and perceived organizational sup-
on the relationship between abusive supervision and burnout. port, while their supervisors were asked to rate the subordi-
Political skill refers to “the ability to effectively understand nates’ political skill.
others at work, and to use such knowledge to influence others To ensure confidentiality, the respondents were instructed
to act in ways that enhance one’s personal or organizational to complete the questionnaires, seal them in the return enve-
objectives” (Ferris, Treadway, et al. 2005: 127). Research by lope, and deliver them at a company meeting for all employees
Ferris and colleagues (e.g., Ferris et al. 2002; Ferris, (both supervisory and non-supervisory) 2 weeks later. Specif-
Treadway, et al. 2005) has described individuals possessing ically, they were instructed to put the finished questionnaires
Curr Psychol (2016) 35:77–82 79

in a designated box located at the front of the meeting venue. Eisenberger 2008), the six-item version was used in this study.
Text messages were sent to the participants the day before the A sample item is “[Organization name] values my contribu-
meeting and they were instructed to put the finished question- tion to its well-being.” The reliability estimate for the scale
naires in a designated box located at the front of the meeting was .91.
venue.
Of the questionnaires distributed, 57 supervisor and 248 Political Skill The Political Skill Inventory of 18 items devel-
subordinate questionnaires were returned, representing re- oped by Ferris et al. (2005) was used in a translated version.
sponse rates of 89.06 and 77.5 % respectively. All data were Sample items included: “He/she spends a lot of time and effort
entered into SPSS 22 for data analysis and the MODPROBE networking with others”; “It is easy for him/her to develop
macro, an aid for probing single-degree-of-freedom interac- good rapport with most people”; and “He/she is able to com-
tions in OLS (ordinary least squares) and logistic regression, municate easily and effectively with others.” The reliability
were used to probe moderation models. The average number estimate for the scale was .83.
in each workgroup was 4.35. Subordinate respondents were
predominantly male (62.1 %), reported an average age of Control Variables We controlled for the participants’ age,
32.58 years (SD=8.28), an average organizational tenure of gender, education, and company tenure. Age, education, and
6.31 years (SD = 3.99), and an average team tenure of company tenure were measured by number of years. Gender
4.18 years (SD=2.21). was coded 0 for “female” and 1 for “male.”

Measures Data Analytic Strategies

To ensure measure equivalence between the Chinese and En- First, preliminary analyses evaluating the descriptive statistics
glish versions, the translation and back-translation method and correlations among study variables, and possible group
(Brislin 1990) was applied. With the exception of the burnout differences in study variables based on demographic
scale, we used a 5-point response scale ranging from “strongly characteristics were performed. Next, the two moderation
disagree” (1) to “strongly agree” (5) for all items. For burnout, models, with perceived organizational support and political
the seven-point response scale ranging from 0 (never) to 6 skill as moderators on the relations between abusive
(daily) was applied. supervision and job burnout were tested using SPSS
MODPROBE macro, developed by Hayes and Matthes
Abusive Supervision We measured abusive supervision with (2009) for estimating the single-degree-of-freedom interac-
a 15-item scale developed by Tepper (2000) which has previ- tions in Ordinary Least Square (OLS) and logistic regression.
ously been used in a Chinese context (e.g., Aryee et al. 2007; Preacher et al. (2007) demonstrated that moderation is
Liu et al. 2010). Sample items are “My supervisor tells me my established when the independent variable and moderator sig-
thoughts or feelings are stupid” and “My supervisor expresses nificantly interact and the bootstrapped confidence interval
anger at me when he/she is mad for another reason.” The does not contain zero.
scale’s alpha reliability in this study is .73.

Burnout The Maslach Burnout Inventory – General survey Results

(16 items) was used to measure subordinates’ burnout
(Maslach and Jackson 1986). This instrument includes ex- Preliminary Analyses
haustion (five items; e.g., “I feel used up at the end of a work
day”), cynicism (five items; e.g., “I doubt the significance of See Table 1 for mean scores, standard deviations and
my work”) and professional efficacy (six items; e.g., “I can Pearson’s correlations among study variables (i.e., abusive
effectively solve the problems that arise in my work”). High supervision, job burnout, organizational support, and political
scores on exhaustion and cynicism and low scores on profes- skill). All study variables are normally distributed and the
sional efficacy are indicative of burnout. The reliability esti- relations among variables appear linear. Abusive supervision
mate for the scale was .71. and job burnout significantly and positively correlated. There
was no significant correlation between organizational support
Perceived Organizational Support Prior research has con- and other study variables (abusive supervision, job burnout
sistently shown the high internal reliability and unidimension- and political skill). And there was no significant correlation
ality of the Survey of Perceived Organizational Support between political skill and other study variables (abusive su-
(SPOS; Eisenberger et al. 1986; Eisenberger et al. 1990; pervision, job burnout and organizational support). Initial
Shore and Tetrick 1991) in both its full and shortened ver- analyses also examined participants’ age, gender, education,
sions. Following previous research (e.g., Eder and and company tenure differences on all variables. Pearson’s
80 Curr Psychol (2016) 35:77–82

6
Table 1 Means, standard deviations, and correlations among study
variables
5
Mean SD 1 2 3

1. Abusive supervision 2.87 0.34 4

2. Job burnout 4.27 0.54 0.23**

Burnout
Low organizational
3. Organizational support 3.91 0.81 0.06 0.10 3 support
4. Political skill 3.28 0.68 0.08 −0.03 0.07 High orgniazational
support
2
* p<.01, ** p<.001

correlational test demonstrated that age was negatively asso-

ciated with abusive supervision, r (248)=−.13, p=.041, and 0
positively associated with political skill, r (248)=.14, p=.024. Low High
abusive abusive
There was no significant difference of years of company ten-
Fig. 1 Interaction between abusive supervision and perceived
ure on any study variable. Independent samples t-tests dem-
organization support on job burnout
onstrated no gender difference on any study variable. One-
Way ANOVA tests showed no education or tenure differences
on any study variable.
The first moderation analysis examined whether perceived CI=(−0.1697, 0.3595). High and low perceived organization-
organizational support moderated links between abusive su- al support groups were defined by means.
pervision and job burnout. Applying the MODPROBE Similarly, the second moderation analysis examined wheth-
macros for moderation analysis, the conditional effect of per- er political skill moderated the relations between abusive su-
ceived organizational support was estimated at values of one pervision and job burnout. Applying the MODPROBE macros
standard deviation below the mean, the mean, and one stan- for moderation analysis, the conditional effect of political skill
dard deviation above the mean. Demographic variables (i.e., was estimated at values of one standard deviation below the
participants’ age, gender, education, and year of company mean, the mean, and one standard deviation above the mean.
tenure) were also entered in the model as controlled variables. Demographic variables were again controlled in the model.
Results showed that perceived organizational support moder- Results showed that political skill moderated the links between
ated the relations between abusive supervision and job burn- abusive mentoring and job burnout (Table 2). Specifically, as
out (Table 2). As shown in Fig. 1, the conditional effect esti- shown in Fig. 2, the conditional effect estimates indicated that
mates indicated that the interaction between abusive supervi- the interaction between abusive supervision and political skill
sion and job burnout was such that abusive supervision and was such that abusive supervision and burnout was positively
job burnout was positively associated for low organization and significantly associated for low political skill group,
support group, b=.66, p<.000, 95 % CI=(.3760, .9415), b=.62, p=.000, 95 % CI=(.3461, .8967), whereas the associ-
whereas the association was positive but not significant for ation was positive but non-significant for high political skill
high organizational support group, b=.48, p=.481, 95 % group, b=.11, p=.419, 95 % CI=(−0.1648, 0.3945). High and
low political skill groups were defined by means.
Table 2 Results of ordinary least square regression analyses

Moderator b (SE) t R2 F 6

Perceived organizational 0.09 8.01** 5

support (POS)
Abusive supervision 0.37 (0.10) 4.85** 4
(AS)
Burnout

Job burnout (JB) 0.04 (0.04) 1.05 3 Low Political Skill

AS*POS −0.35 (0.12) −2.85* High Political Skill

2
Political skill (PS) 0.08 6.98**
Abusive supervision 0.37 (0.10) 3.75** 1
(AS)
Job burnout (JB) −0.03 (0.05) −0.66 0
Low Abusive High Abusive
AS*PS −0.37 (0.15) −2.51*
Fig. 2 Interaction between abusive supervision and political skill on job
* p<.05, ** p<.01 burnout
Curr Psychol (2016) 35:77–82 81

Discussion resources such as counseling and well-being workshops are

available for them to seek help and assistance. In terms of the
While research into the antecedents of burnout has steadily moderating role of employees’ political skill, our study suggests
grown, the relationship between abusive supervision and that political skill could weaken the harmful influence of abu-
burnout remains largely unknown. In addition, we know little sive supervision on burnout. Hence, organizations should pref-
about the contingencies under which abusive supervision may erentially hire employees with greater political abilities while
be related to employees’ burnout. To this end, our study sug- providing training as well as mentoring programs to encourage
gests three findings. First, abusive supervision was positively employees to develop their political skills (Armstrong-Stassen
associated with burnout. Second, the positive relationship be- and Ursel 2009; Eder and Eisenberger 2008).
tween abusive supervision and burnout was moderated by the The limitations of our study also point to a possible direction
employees’ perceived organizational support in such a way for future research. One of the main weaknesses of the present
that the relationship was weaker for employees who perceive study was the use of a cross-sectional design, which does not
higher rather than lower organizational support. Third, the allow for an assessment of the causality of the relationship
positive relationship between abusive supervision and burnout examined. Additional quasi-experimental or longitudinal stud-
was moderated by the employees’ political skill in such a way ies would be useful in testing the underlying causality of the
that the relationship was weaker for people with higher, rather relationships researched in this study. Another limitation of the
than lower, level of political skill. current study relates to the generalizability of our results. The
The results of the present study provide important theoretical study was conducted in the hotel industry with 248 participants.
contributions to the literature on abusive supervision and burn- We don’t know whether these results would generalize to other
out in three ways. First, this study increases our understanding settings or other cultures. Generalizability of the present find-
of the effects of supervisors’ behaviors and managerial practice ings should therefore be examined in future studies in other
on employees’ burnout level by examining one threat to em- types of organizations and cultures. Finally, we only examined
ployees’ burnout at work: abusive supervision. In addition, our the association between individual employees’ perceptions
study provides further support for a previous study linking abu- about their supervisors’ abusive supervision and their burnout
sive supervision with burnout (Carlson et al. 2012) while testing levels at the individual level. Recent studies have demonstrated
our model on a Chinese sample. Second, our study advocated theoretical and empirical evidence that abusive supervision can
an underrepresented perspective on the contingency effect of also occur at the group level (Aryee et al. 2007; Priesemuth
antecedents’ influence on burnout. We addressed the explorato- et al. 2014). Future studies could therefore examine abusive
ry question of whether a contextual factor of perceived organi- supervision at the group level to investigate how abusive super-
zational support and an individual difference variable of polit- vision may affect the group at large.
ical skill moderate the relationship between abusive supervision
and burnout. To the best of our knowledge, no study has exam-
ined moderating effect on the relationship between abusive su- Conclusion
pervision and burnout. In response to Carlson and colleagues’
(2012) call for further research into this area, we have demon- Given the high costs and deleterious effects associated with
strated that abusive supervision and burnout was moderated by burnout, organizations often face the challenge of preventing
perceived organizational support and political skill. and reducing employees’ burnout in the workplace. Our study
There are several practical implications that can be derived addresses this challenge by suggesting that supervisors’ abu-
from our findings. First, abusive supervision is costly for both sive supervision is positively related with burnout. Our results
organizations and employees in terms of generating burnout. also highlight that this relationship is contingent upon em-
Our work suggests a need for organizations to send a clear ployees’ perceived organizational support and political skill.
message to supervisors that these hostile and harmful behaviors
will not be tolerated. The most salient findings of this study are Acknowledgments This research is supported by the national Natural
that perceived organizational support could buffer the harmful Science Foundation of China (project 71302022 & 71162008), Jiangxi
effect of abusive supervision on employees’ burnout. Thus, in Social Science Foundation (14GL11) and the Fundamental Research
Funds for the Central Universities (project SKZZY2014022,
addition to preventing abusive supervision in the work context, SKZZA2014003 & SKZZX2013032).
organizations should implement fair and equitable policies and
practices as well as building a friendly organizational work
climate for individual employees to perceive high level of sup- References
port from the organization (Tepper et al. 2008; Thau et al. 2009;
Zellars et al. 2002). For example, when abusive supervision Alarcon, G. M. (2011). A meta-analysis of burnout with job demands,
occurs and is experienced by employees, certain channels such resources, and attitudes. Journal of Vocational Behavior, 79(2),
as an employee assistance program (EAP) or other support 549–562.
82 Curr Psychol (2016) 35:77–82

Armeli, S., Eidenberger, R., Fasolo, P., & Lynch, P. (1998). Perceived Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the
organizational support and police performance: the ating influence correlates of the three dimensions of job burnout. Journal of Applied
of socioemotional needs. Journal of Applied Psychology, 83(2), Psychology, 81(2), 123–133.
288–297. Liu, J., Kwan, H. K., Wu, L.–. Z., & Wu, W. (2010). Abusive supervision
Armstrong-Stassen, M., & Ursel, U. D. (2009). Perceived organizational and subordinate supervisor-directed deviance: the moderating role
support, career satisfaction, and the retention or older workers. of traditional values and the moderating role of revenge cognitions.
Journal of Occupational and Organizational Psychology, 82(1), Journal of Occupational and Organizational Psychology, 83(4),
201–220. 835–856.
Aryee, S. S., Chen, Z. X., Sun, L. Y., & Debrah, Y. A. (2007). Loi, R., Ngo, H. Y., & Foley, S. (2006). Linking employees’ justice
Antecedents and outcomes of abusive supervision: test of a trickle- perceptions to organizational commitment and intention to leave:
down model. Journal of Applied Psychology, 92(1), 191–201. the mediating role of perceived organization support. Journal of
Brislin, R. W. (1990). Applied cross-cultural psychology. Newbury Park: Occupational and Organizational Psychology, 79, 101–120.
Sage. Maslach, C., & Jackson, S. E. (1986). The MaslachBurnour Inventory.
Burke, R. J., & Greenglass, E. R. (1995). A longitudinal study of psy- Palo Alto: Consulting Psychologists Press.
chological burnout in teachers. Human Relation: Towards the Perrewé, P. L., Ferris, G. R., Frink, D. D., & Anthony, W. P. (2000).
Integration of the Social Sciences, 48(2), 187–202. Political skill: an antidote for workplace stressors. Academy of
Cameron, S. J., Horsburgh, M. E., & Armstrong-Stassen, M. (1994). Job Management Executive, 14, 115–123.
satisfaction, propensity to leave, and burnout in RNs and RNAs: a Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing mod-
multivariate perspective. Canadian Journal of Nursing erated mediation hypotheses: theory, methods, and prescriptions.
Administration, 7(3), 43–64. Multivariate Behavioral Research, 42, 185–227.
Carlson, D., Ferguson, M., Hunter, E., & Whitten, D. (2012). Abusive Priesemuth, M., Schminke, M., Ambrose, M. L., & Folger, R. (2014).
supervision and work-family conflict: the path through emotional Abusive supervision climate: a multiple-mediation model of its im-
labor and burnout. The Leadership Quarterly, 23, 849–859. pact on group outcomes. Academy of Management Journal, 57,
Chen, Z., Zhu, J., & Zhou, M. J. (2015). When does a servant leader fuel 1513–1534.
the service fire? A multilevel model of servant leadership, individual Qian, J., Lin, X. S., & Chen, Z. X. (2012). Authentic leadership and
group identity, group competition climate, and customer service feedback-seeking behavior: an examination of the cultural context
performance. Journal of Applied Psychology. of mediating processes in China. Journal of Management and
Cherniss, C. (1992). Long-term consequences of burnout: and explorato- Organization, 18, 286–299.
Randall, M. L., Cropanzano, R., Bormann, C. A., & Birjulin, A. (1999).
ry study. Journal of Organizational Behavior, 13(1), 1–11.
Organizational politics and organizational support as predictors of
Eder, P., & Eisenberger, R. (2008). Perceived organizational support:
work attitudes, job performance, and organizational citizenship be-
reducing the negative influence of coworker withdrawal behavior.
havior. Journal of Organizational Behavior, 20, 159–174.
Journal of Management, 34(1), 55–68.
Shore, L. M., & Tetrick, L. E. (1991). A construct validity study of the
Eisenberger, R., Huntington, R., Huntchison, S., & Sowa, D. (1986).
survey of perceived organizational support. Journal of Applied
Perceived organizational support. Journal of Applied Psychology,
Psychology, 76, 637–643.
71, 500–507.
Tepper, B. J. (2000). Consequences of abusive supervision. Academy of
Eisenberger, R., Fasolo, P., & Davis-LaMastro, V. (1990). Perceived or-
Management Journal, 43(2), 178–190.
ganizational support and employee diligence, commitment, and in-
Tepper, B. (2007). Abusive supervision in work organizations: review,
novation. Journal of Applied Psychology, 75, 51–59.
synthesis, and research agenda. Journal of Management, 33(3),
Ferris, G. R., Perrewe, P. L., & Douglas, C. (2002). Social effectiveness in 261–289.
organizations: construct validity and research directions. Journal of Tepper, B. J., Henle, C. A., Lambert, L. S., Giacalone, R. A., & Duffy, M.
Leadership and Organizational Studies, 9, 49–63. K. (2008). Abusive supervision and subordinates’ organizational
Ferris, G. R., Treadway, D. C., Kolodinsky, R. W., Hochwarter, W. A., deviance. Journal of Applied Psychology, 93, 721–732.
Kacmar, C. J., Douglas, C., & Frink, D. D. (2005). Development and Thau, S., Bennett, R. J., Mitchell, M. S., & Marrs, M. B. (2009). How
validation of the political skill inventory. Journal of Management, management style moderates the relationship between abusive su-
31(1), 126–152. pervision and workplace deviance: an uncertainty management the-
Ferris, G. R., Treadway, D. C., Perrewé, P. L., Brouer, R. L., Douglas, C., ory perspective. Organizational Behavior and Human Decision
& Lux, S. (2007). Political skill in organizations. Journal of Processes, 108, 79–92.
Management, 33, 290–320. Wang, H., Law, K. S., Hackett, R. D., Wang, D. X., & Chen, Z. X. (2005).
Halbesleben, J. R. B., & Buckley, M. R. (2004). Burnout in organization- Leader-member exchange as a mediator of the relationship between
al life. Journal of Management, 30, 859–879. transformational leadership and followers’ performance and organi-
Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing zational citizenship behavior. Academy of Management Journal, 48,
interactions in OLS and logistic regression: SPSS and SAS 420–432.
implementations. Behavior Research Methods, 41, 924–936. Wolpin, J., Burke, R. J., & Greenglass, E. R. (1991). Is job satisfaction an
Jackson, S. E., Schwab, R. L., & Schuler, R. S. (1986). Toward an un- antecedent or a consequence of psychology burnout? Human
derstanding of the burnout phenomenon. Journal of Applied Relations, 44(2), 193–209.
Psychology, 71(4), 630–640. Wright, T. A., & Bonett, D. G. (1997). The contribution of burnout to
Keashly, L., Hunter, S., & Harvey, S. (1997). Abusive interactions and work performance. Journal of Organizational Behavior, 18, 491–
role state stressors: relative impact on student residence assistant 499.
stress and work attitudes. Work and Stress, 11, 175–185. Wright, T. A., & Cropanzano, R. (1998). Emotional exhaustion as a
Lee, R. T., & Ashforth, B. E. (1988). A longitudinal study of burnout predictor of job performance and voluntary turnover. Journal of
among supervisors and managers: comparisons between the Leiter Applied Psychology, 83, 486–493.
and Maslach (1988) and Golembiewski et al. (1986) models. Zellars, K. L., Tepper, B. J., & Duffy, M. K. (2002). Abusive supervision
Organizational Behavior and Human Decision Processes, 54(3), and subordinates’ organizational citizenship behavior. Journal of
369–398. Applied Psychology, 87, 1068–1076.