The Leadership Quarterly: Rachel E. Frieder, Wayne A. Hochwarter, Philip S. Deortentiis

The Leadership Quarterly 26 (2015) 821837
Contents lists available at ScienceDirect
The Leadership Quarterly

journal homepage: www.elsevier.com/locate/leaqua
Attenuating the negative effects of abusive supervision: The role

of proactive voice behavior and resource management ability
Rachel E. Frieder a,, Wayne A. Hochwarter b, Philip S. DeOrtentiis b
a
b
Department of Management, Strome College of Business, Old Dominion University, 2171 Constant Hall, Norfolk, VA 23529, USA
Department of Management, College of Business, Florida State University, 821 Academic Way, P.O. Box 3061110, Tallahassee, FL 32306-1110, USA
a r t i c l e
i n f o
Article history:
Received 23 April 2014
Received in revised form 2 June 2015
Accepted 4 June 2015
Available online 22 June 2015
Handling Editor: Leanne Atwater
Keywords:
Abusive supervision
Resource management ability
Voice
a b s t r a c t
While a large portion of the abusive supervision literature has examined the negative consequences
of such perceived mistreatment, little research has examined individual-level characteristics
capable of helping victims survive under such conditions. The purpose of this two-sample study,
therefore, is to examine the factors that attenuate the negative affective and behavioral reactions
stemming from perceived abusive supervision. Supported by recent extensions of the Job DemandControl model (JDC; Karasek, 1979; Meier, Semmer, Elfering, & Jacobshagen, 2008), we
suggest that individuals who exhibit proactive voice behaviors and perceive that they are better
able to manage their resources will experience less dissatisfaction, emotional exhaustion, turnover
intentions, and reductions in work effort when faced with perceived supervisory abuse than those
not demonstrating proactive voice and incapable of managing their resources. Cross-sample ndings demonstrated support for our hypotheses. Implications for theory and practice, strengths
and limitations, and avenues for additional research are discussed.
2015 Elsevier Inc. All rights reserved.
Introduction
Over 60 years ago, Hughes (1950, p. 321) noted that others at work do the most to make our life sweet or sour. Not surprisingly,
considerable research seeks to understand how contextual factors make our lives sweet (e.g., perceived supervisor support;
Eisenberger, Stinglhamber, Vandenberghe, Sucharski, & Rhoades, 2002) or sour (e.g., abusive supervision; Tepper, 2000). With
regard to the latter, abusive supervision is dened as subordinates' perceptions of the extent to which supervisors engage in the
sustained displays of hostile verbal and nonverbal behaviors, excluding physical contact (Tepper, 2000, p. 178). The deleterious
effects of perceived abusive supervision on employee health and well-being have been widely documented. For example, research
indicates such perceived abuse impacts an estimated 14% of individuals annually and costs organizations exponentially more in
terms absenteeism, health, and reduced performance (Tepper, 2000, 2007). Unfortunately, the negative effects of perceived abusive
supervision should be felt long into the foreseeable future as researchers afrm that incidents of nonviolent workplace aggression are
gradually increasing (Burton & Hoobler, 2006; Burton, Mitchell, & Lee, 2005).
Despite its well-established theoretical foundation and over a decade of scholarly inquiry, the abusive supervision literature has
primarily emphasized outcomes. Less attention has been directed towards understanding the factors capable of mitigating the harmful effects of abusive supervision. The research that does exist suggests that individual differences such as social adaptability (Mackey,
Ellen, Hochwarter, & Ferris, 2013), emotional intelligence (Hu, 2012), and perceived job mobility (Wei & Si, 2013) are capable of
attenuating relationships between abusive supervision and work outcomes. While these studies lack a single, cohesive explanation
for such mitigating effects, factors related to control represent a common theme.
Corresponding author.
E-mail addresses: rfrieder@odu.edu (R.E. Frieder), whochwarter@cob.fsu.edu (W.A. Hochwarter), psd11@my.fsu.edu (P.S. DeOrtentiis).
http://dx.doi.org/10.1016/j.leaqua.2015.06.001
1048-9843/ 2015 Elsevier Inc. All rights reserved.
822
R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837
The purpose of this study is to examine additional factors that offset the negative affective and behavioral reactions stemming from
abusive supervision. Specically, our study focuses on differences that emphasize one's ability to directly exert inuence in threatladen situations. Supported by recent extensions of the Job DemandControl model (JDC; Karasek, 1979; Meier, Semmer, Elfering,
& Jacobshagen, 2008), we suggest that individuals who report high levels of proactive voice behavior and resource management ability will experience fewer negative reactions (i.e., dissatisfaction, emotional exhaustion, turnover intentions, reduced work effort)
when faced with perceived supervisory abuse than those less adept and less vocal.
In doing so, we contribute to the organizational sciences literature in a number of ways. While past research has identied a handful
of individual difference factors (e.g., conscientiousness, Nandkeolyar, Shaffer, Li, Ekkirala, & Bagger, 2014; emotional intelligence, Hu,
2012; social adaptability, Mackey et al., 2013) capable of mitigating the negative effects of perceived supervisory abuse, it is unclear
what this array of factors has in common that allows individuals to survive (and sometimes even thrive) under conditions of supervisory mistreatment. Thus, this study contributes to the abusive supervision literature in that we utilize the Job DemandControl model as
a synthesizing framework by which we may better understand why certain individual difference factors may help individuals ward off
the negative effects of perceived mistreatment. The factors capable of attenuating the negative effects of abusive supervision are
especially important to identify given that abusive supervision reects a perception, and therefore it may not be realistic to eradicate
such sensitivities entirely from employees' cognitions.
Secondly, we examine a demand (i.e., abusive supervision) and form of decision latitude (i.e., proactive voice behavior) different
from that which is originally specied in the JDC model. In keeping with recent extensions of the model (e.g., Meier et al., 2008;
Schaubroeck & Merritt, 1997), we also examine an individual difference in control beliefs (i.e., perceived resource management ability) as a factor capable of inhibiting the negative effects of abusive supervision. As such, this study represents both a supplement and
extension of the original specications of the Job DemandControl model to recognize alternate forms of job demands and decision
latitude, as well as an individual difference in control beliefs.
Theoretical foundations and hypothesis development
The Job DemandControl model
The Job DemandControl (JDC) model or decision latitude model positions strain reactions as the result of the combined effects
of workplace demands and the amount of discretion (i.e., decision latitude) available for employees to manage (Karasek, 1979). The
JDC model is reinforced by two primary assertions. First, increases in job demands engender increased job strain. Second, individuals
with the requisite decision latitude needed to cope with high job demands experience heightened psychological well-being and
motivation to perform, while individuals faced with high demands and low amounts of decision latitude experience reductions in
psychological health and well-being. In other words, job control serves to buffer the negative effects of job demands on strain.
While the premises of the JDC model are intuitively appealing, empirical support for the JDC model, particularly for the buffering
hypothesis, has been largely equivocal (e.g., de Jonge & Kompier, 1997; Ganster & Fusilier, 1989; Perrew & Ganster, 1989; Van der
Doef & Maes, 1999; cf. de Lange, Taris, Kompier, Houtman, & Bongers, 2003).
It has been suggested that this lack of empirical support is due to the overly simplistic nature of the JDC model as well as the
narrow classication of both demands (i.e., time pressure, work overload, role conict) and control (i.e., autonomy, decision latitude)
(Bakker & Demerouti, 2007). Nonetheless, even Karasek (1979) acknowledged that other possible demands and sources of control
(e.g., job-related personal conict, fear of unemployment, p. 291) were likely to inuence strain reactions. Other researchers have
suggested that the buffering hypothesis may only hold true when examining specic outcomes (e.g., Dwyer & Ganster, 1991) or
individuals working within certain occupations (e.g., de Jonge, Dollard, Dormann, Le Blanc, & Houtman, 2000).
In contrast, others have found support for the JDC model, but only when individuals demonstrated certain dispositional characteristics (e.g., internal locus of control, self-efcacy; Meier et al., 2008; Schaubroeck & Merritt, 1997) or were privy to external resources (e.g., social support; Daniels & Guppy, 1994; Johnson & Hall, 1988). In support, Meier et al. (2008) concluded that support
for the JDC model's buffering hypothesis was tenuous unless individuals possessed certain individual characteristics that allowed
them to better, or more fully, exercise their control options.
Building on these supportive ndings, we utilize the extended JDC framework to examine the interactive relationship among
abusive supervision, proactive voice, and resource management ability on theoretically and practically relevant outcomes. Specically,
the three-way interactions of these constructs on job satisfaction, turnover intentions, emotional exhaustion and work effort are
examined in accordance with prior abusive supervision research (Wu & Hu, 2009). These constructs, as well as their combined multiplicative effect on each outcome, are discussed below.
Abusive supervision
Abusive supervision reects a subordinate's subjectively-held perception. As such, while subordinates' accounts of abusive supervision are not always indicative of their supervisor's objective treatment (Martinko, Harvey, Brees, & Mackey, 2013), subordinates will
respond to abuse in ways indicative of their perceptions.1 We suggest that abusive supervision represents the classic job demand,
where job demands are dened as those physical, psychological, social, or organizational aspects of the job that require sustained
1
It is important to emphasize that the proceeding discussion deals entirely with subordinates' perceptions of abusive supervision; while some supervisors do engage
in objectively abusive behavior, such objective abusive supervision is not the focus of this manuscript.
823
physical and/or psychological (cognitive and emotional) effort or skills and are therefore associated with certain physiological and/or psychological costs (Bakker & Demerouti, 2007, p. 312, italics added for emphasis). By denition, abusive supervision is a perception of
unremitting mistreatment on the part of the supervisor. Therefore it should require individuals' sustained efforts to properly cope.
These efforts come at a documented cost, be that physical, psychological, or both. For instance, perceptions of abusive supervision
have been positively associated with emotional exhaustion (e.g., Aryee, Sun, Chen, & Debrah, 2008) and dissatisfaction (Tepper, Duffy,
Hoobler, & Ensley, 2004), reecting inherent psychological costs. Moreover, research also indicates that individuals respond to
perceptions of abusive supervision by way of exerting physical efforts towards counterproductive (e.g., Bowling & Michel, 2011) or
deviant behaviors (e.g., Mitchell & Ambrose, 2007) or by reducing task (e.g., Hoobler & Hu, 2013) or contextual performance
(e.g., Rafferty & Restubog, 2011). Taken together, the deleterious effects of perceived abusive supervision on employees' affective
and behavioral functioning are well-documented and provide evidence in favor of its classication as a job demand requiring a sufcient coping response. Nonetheless, while much of the research fueled by Tepper's (2000) seminal piece has identied the outcomes
with which abusive supervision has been associated, little research has been conducted to examine the factors capable of attenuating
the negative effects of perceived supervisory abuse.
Research examining individual differences in these contexts has focused largely on control-related remedies. For instance, research has found that the positive relationships between perceived abusive supervision and problem drinking (Bamberger &
Bacharach, 2006) and perceived abusive supervision and dysfunctional resistance (Tepper, Duffy, & Shaw, 2001) were attenuated
by heightened conscientiousness and agreeableness. Similarly, Nandkeolyar et al. (2014) found that conscientiousness lessened the
effects of abuse on job performance. As an explanation, research indicates that individuals low in conscientiousness and agreeableness
are more impulsive (Costa, McCrae, & Dye, 1991), have antagonistic and argumentative tendencies (Costa & McCrae, 1992), and tend
to utilize avoidant as opposed to active coping strategies (Nandkeolyar et al., 2014). Conversely, individuals high in conscientiousness
and agreeableness are considered more controlled than their non-conscientious, disagreeable peers. Similarly, Harvey, Stoner,
Hochwarter, and Kacmar (2007) found that high positive affect, high ingratiatory individuals experienced less job tension and emotional exhaustion under conditions of high perceived supervisory mistreatment.
More recently, Hu (2012) found that emotionally intelligent individuals experienced less psychological distress under high perceived supervisory abuse conditions than their less astute peers. Finally, scholars argued that socially adaptable individuals would
be better able to leverage resources to adjust to the demands of abusive supervision (e.g., less strain and exert greater effort)
(Mackey et al., 2013). In sum, these studies support the important role of both control options (e.g., ingratiation) and individual
differences regarding one's (actual or perceived) ability to exercise control options (e.g., social adaptability, conscientiousness) in
mitigating the negative effects of abusive supervision. In what follows, we draw from the Job DemandControl model to argue for
the importance of proactive voice behavior as an additional factor indicative of individuals' control options or decision latitude.
Employee proactive voice behavior
Many conceptualizations of employee proactive voice behavior have been articulated since the concept was rst described by
Hirschman (1970). Nonetheless, a recent review of the voice literature (Morrison, 2011) found three commonalties across the various
conceptualizations. Specically, voice behavior is 1) a discretionary act of 2) verbal expression which is 3) constructive in nature such
that it is intended to bring about positive improvements to one's work experiences (Morrison, 2011). As such, proactive voice behavior refers to employees' discretionary communication of ideas, suggestions, concerns, or opinions about work-related issues with the
intent to improve organizational or unit functioning (Morrison, 2011; p. 375). Proactive voice behavior differs from other forms of
citizenship behaviors in that voice is intentionally change-oriented and necessarily challenging (Van Dyne, Cummings, & McLean
Parks, 1995). Moreover, voice behavior is motivated by individuals' desires to inuence the treatment they receive from others
(Fuller, Barnett, Hester, Relyea, & Frey, 2007).
For the purposes of this research, proactive voice behavior is conceptualized as a type of job control (i.e., decision latitude) in that it
can be (and is) used to bring about constructive change to aspects of one's work. In support, Daniels, Boocock, Glover, Hartley, and
Holland (2009) conceptualized job control as the extent to which workers change aspects of their work to solve problems. As
such, voice behavior works either to change the abused individual's situation or, even if such behavior is unsuccessful at changing
the situation, it is helpful in achieving catharsis (e.g., van Prooijen, Van Den Bos, & Wilke, 2007). In support, individuals who engage
in voice behaviors seem to benet from reduced strain and demonstrate heightened levels of performance (Ng & Feldman, 2012;
Whiting, Podsakoff, & Pierce, 2008).
Moreover, organizational justice research suggests that voice plays an instrumental role in process (Lind & Tyler, 1988) and
outcome control (Leventhal, 1980) such that individuals who engage in voice will be more in control of both what they receive
and what befalls them (Lind, Kanfer, & Earley, 1990). Taken together, voice behavior is consistent with Karasek's (1979) notion of decision latitude (i.e., discretion) in that individuals engaged in voice behavior are thought to exert considerable inuence and affect
change on the environment in which they operate and/or the demands with which they are faced. Thus, according to the basic tenets
of the JDC model, individuals should be less negatively affected by perceived supervisory abuse (i.e., a job demand) to the extent that
they report engaging in voice behavior (i.e., decision latitude; discretion).
However, it is unlikely that all employees perceiving abuse will exercise voice behaviors, especially if they feel that doing so will
threaten their current or future resource reservoirs. In support, meta-analytic research suggests that engaging in voice behavior taxes
individuals' resources and thus, when individuals are already confronted with excessive workplace demands, they are less likely to
use resources to engage in such constructive voicing (Ng & Feldman, 2012). This suggests that voice behavior alone may not be sufcient to buffer the negative effects of perceived supervisory abuse. Hence, in keeping with more recent extensions of the JDC model
824
(Meier et al., 2008), we hypothesize that the strain-buffering effects of voice behavior on perceived supervisory abuse hold only for
certain individuals, namely those more capable of managing their resources.
Job resources reect psychological, social, physical, and/or organizational features of one's job that contribute to individuals'
growth and development, aid in goal accomplishment, and/or assist individuals in reducing job demands and their associated costs
(Bakker & Demerouti, 2007; de Jonge & Dormann, 2006; Hobfoll, 2001). According to conservation of resources (COR) theory
(Hobfoll, 1989), individuals are motivated to protect and acquire resources and are threatened by the actual or potential loss of
their resources. Furthermore, individuals can anticipate threats and subsequently minimize their negative effects by acquiring and
protecting resources (Aspinwall & Taylor, 1997; Bolino, Valcea, & Harvey, 2010). In this manner, resources act as buffers that serve
to separate individuals from the sources of stress and strain (Grandey & Cropanzano, 1999; Hobfoll & Leiberman, 1987; Sumer,
Karanchi, Berument, & Gunes, 2005).
Nonetheless, in recognition that the objective quantities of resources individuals possess are less valuable than one's ability to appropriately utilize and protect one's resources (Diener, Suh, Lucas, & Smith, 1999), Hochwarter and colleagues advanced the notion of
resource management ability (Hochwarter, Laird, & Brouer, 2008; Hochwarter, Perrew, Meurs, & Kacmar, 2007; Laird, Brouer, &
Hochwarter, 2005). Resource management ability, or RMA, refers to one's ability to maintain and mobilize one's resources; individuals
who are able to manage their resources are equipped to protect and acquire resources that include access to equipment, assistance,
exibility, and control over the pace of, and exertion towards, one's work (Hochwarter et al., 2007). While RMA is grounded in
COR, it differs in that COR captures whether individuals have resources whereas RMA captures how one manages the resources at
their disposal (Halbesleben, Neveu, Paustian-Underdahl, & Westman, 2014).
Research indicates that individuals with greater access to, or availability of, resources are better able to regulate their behavior in
response to work demands (Muraven & Baumeister, 2000) and have an increased likelihood of coping successfully with the demands
with which they are faced (Janssen, De Jonge, & Bakker, 1999). As such, resource management ability represents an individual difference with documented positive effects on individuals' health and well-being especially when faced with stressful demands. In support, research suggests that individuals who perceived themselves as better able to manage their resources experienced more job
satisfaction and less tension when being held accountable by others (i.e., a job demand) (Zellars, Hochwarter, Lanivich, Perrew, &
Ferris, 2011). Similarly, when individuals faced with work-induced guilt also perceived they were able to manage their resources,
they experienced heightened levels of job and life satisfaction as compared to their peers who perceived themselves as unable to manage their resources; in the latter case, individuals experiencing work-induced guilt but unable to manage their resources experienced
reductions in both life and job satisfaction (Hochwarter et al., 2007).
Further, research has indicated that the ability to dictate the pace of one's work tasks (King, Winett, & Lovett, 1986) and control the
amount of exertion at work (Grandey & Cropanzano, 1999) results in anxiety reduction across both traumatic and non-traumatic
work environments (Monnier, Cameron, Hobfoll, & Gribble, 2002). Coping with (perceived) abusive supervision requires sufcient
psychological efforts and consequently comes at a documented psychological and physiological cost (Tepper, 2007). However,
those with greater perceived RMA should be less negatively affected by such demands. Specically, individuals high in RMA are
able to control how their resources are employed; moreover, such individuals should arguably be less dependent on their supervisors
for current or future resource disbursements. Thus, perceived RMA is a valuable individual difference that should provide insight into
when voice behavior can more fully attenuate the negative effects of perceived supervisory abuse.
The interactive effects of resource management ability and proactive voice behavior
We suggest that under conditions of high perceived abusive supervision, employees who report high levels of voice behavior and
better perceived resource management abilities will demonstrate more auspicious affective (i.e., job satisfaction, emotional exhaustion)
and behavioral reactions (i.e., turnover intentions, work effort) in response to perceived mistreatment than they would otherwise
should they report low RMA and low levels of voice behavior. The choice of dependent variables was driven by past abuse and stress
research as well as the prominence of these variables in the organizational sciences literature.
Job satisfaction represents an individual's positive or negative attitude towards one's job (Brief, 1998; Weiss, 2002) and has been
one of the most studied outcomes of abusive supervision in particular (cf. Tepper, 2007) and the organizational sciences literature in
general (Iaffaldano & Muchinsky, 1985). Emotional exhaustion (i.e., a loss of concern, interest, and spirit characterized by feelings of
fatigue, frustration, and wearing out; Maslach & Jackson, 1981) has documented a positive association with perceptions of abusive
supervision (e.g., Chi & Liang, 2013; Wu & Hu, 2009). Turnover intentions refer to an individual's subjective probability for leaving
his or her organization in the near future (Mowday, Koberg, & McArthur, 1984); not surprisingly, individuals' perceptions of abusive
supervision have been positively associated with their intentions to turnover (Tepper et al., 2009). Finally, research has indicated that
work effort (i.e., the amount of resources that individuals choose to expend on work-related behaviors; Yeo & Neal, 2004) is negatively
associated with perceived abusive supervision (Harris, Kacmar, & Zivnuska, 2007).
We posit that when perceived supervisory abuse is high, voice behavior will tax individuals' current resources and future resource
reservoirs at a time when individuals already face excessive demands; thus voice behavior may be insufcient in reducing the
demands of perceived supervisory abuse. Individuals who perceive that they are better able to manage their resources, however,
are capable of protecting and acquiring resources at their disposal. Consequently, we hypothesize that those individuals high in
RMA and high in proactive voice should be the least negatively affected by perceived supervisory abuse because they possess both
825
the personal control options (voice) and control beliefs (resource management ability) capable of attenuating the demands of
perceived supervisory abuse.
In support, research suggests that individuals employ a cognitive costbenet analysis before engaging in voice behaviors (Dutton,
Ashford, O'neill, Hayes, & Wierba, 1997) such that individuals should refrain from speaking up when doing so is considered personally
costly (e.g., in terms of reduced career mobility) (Detert & Burris, 2007). In this regard, resource management ability reduces the likelihood that speaking up would be personally costly. Thus, individuals with high RMA should be able to protect their current and future
resource reserves (Hochwarter et al., 2007) and should have an increased likelihood of coping success (Janssen et al., 1999).
The JDC model suggests that strain results from the combination of high demands and low decision latitude (i.e., discretion).
Thus, individuals who do not exercise their control options by speaking up to constructively affect change within their work environments should not be able to effectively cope with the demands (i.e., abuse) with which they are faced. Further compounding this effect, such individuals who are also low in RMA are likely to expend high amounts of their resources in trying to cope with the demands
with which they are faced (i.e., perceived abuse) and due to their low perceived RMA, should have difculty managing and recouping
their resource expenditures. Taken together, the combination of lack of control opportunities (voice) and diminished control abilities
(RMA) is hypothesized to result in the maximal amount of strain under conditions of high demands (perceived abusive supervision).
Finally, when considering opposing levels of voice and RMA (i.e., high voicelow RMA, low voicehigh RMA), the relationships of
high supervisory abuse perceptions with the affective and behavioral outcomes of interest are more difcult to anticipate. On one
hand, voice behavior is effective in achieving catharsis (i.e., reduced strain) (van Prooijen et al., 2007), but engaging in voice requires
the expenditure of resources (Ng & Feldman, 2012); thus, voice behavior may tax the very resources that individuals low in RMA are
not good at managing. In other words, when faced with the demands of perceived abusive supervision, the benets of voice behavior
may be offset by low levels of perceived resource management ability.
On the other hand, prolonged exposure to perceived supervisory mistreatment without the requisite decision latitude option
needed to cope should be especially detrimental to individuals' physical and psychological resource reservoirs; nonetheless, individuals with high RMA should be insulated and protected from resource loss spirals (Hochwarter et al., 2007) associated with such perceived mistreatment. In other words, the benets of high RMA may offset the disadvantages of low levels of voice (i.e., decision
latitude) when faced with perceived abusive supervision. Therefore, regardless of the combination of high RMAlow voice or low
RMAhigh voice, we do not anticipate that there will be signicant differences in affective or behavioral functioning under high perceived supervisory abuse conditions. Instead, we anticipate that high levels of voice behavior and RMA will attenuate strain stemming
from high perceived supervisory abuse whereas low levels of voice behavior and RMA are hypothesized to amplify strain stemming
from high perceived supervisory abuse. More formally, we offer the following hypotheses:
Hypothesis 1. The perceived abusive supervisionjob satisfaction relationship is jointly moderated by voice behavior and resource
management ability (RMA). The negative relationship between abusive supervision and job satisfaction is strongest (weakest)
when an individual's voice behavior and RMA are low (high).
Hypothesis 2. The perceived abusive supervisionturnover intentions relationship is jointly moderated by voice behavior and
resource management ability (RMA). The positive relationship between abusive supervision and turnover intentions is strongest
(weakest) when an individual's voice behavior and RMA are low (high).
Hypothesis 3. The perceived abusive supervisionemotional exhaustion relationship is jointly moderated by voice behavior and
resource management ability (RMA). The positive relationship between abusive supervision and emotional exhaustion is strongest
(weakest) when an individual's voice behavior and RMA are low (high).
Hypothesis 4. The perceived abusive supervisionwork effort relationship is jointly moderated by voice behavior and resource
management ability (RMA). The negative relationship between abusive supervision and work effort is strongest (weakest)
when an individual's voice behavior and RMA are low (high).
Method
Participants and procedures
Sample 1
Consistent with prior studies (Meier & Spector, 2013), students in introductory business classes were asked to distribute surveys to
be completed by individuals working more than 35 hours per week in exchange for (minimal) extra course-credit. When completed,
respondents were instructed to return the survey to the student in a sealed envelope. A total of 146 surveys were returned (170 were
initially distributed). Three surveys were not included due to incomplete/illegible responses or by indicating that they did not have a
supervisor; individuals without a supervisor would be unable to provide meaningful ratings on the focal variable, perceived abusive
supervision. Therefore the nal sample consisted of 143 respondents. The majority of the sample reported administrative or professional occupations. Respondents averaged 41 years of age while males comprised 52% of the sample. Consistent with similar studies
(e.g., Byrne & Hochwarter, 2006; Harvey et al., 2007; Hochwarter & Treadway, 2003), respondents provided their daytime telephone
number on a page separated from the survey upon return. A total of 15 respondents were selected at random and asked generally
826
innocuous questions about their level of participation. One number was disconnected; however, the others indicated sufcient levels
of involvement.
Scholars advocate for this data collection strategy (Hochwarter, 2014), acknowledging that information gathered is often consistent
with other, more well-recognized methods. In support, Wheeler, Shanine, Leon, and Whitman (2014) conrmed few observational
correlational differences between student- and non-student-recruited samples. Lastly, Demerouti and Rispens (2014) argued that
student-recruited samples have merit in that they may increase external validity (p. 39).
Sample 2
Surveys were distributed to all 216 white-collar medical personnel of a hospital in the Western United States (no direct providers
of health care were included in the research). The hospital had seminars conducted at the end of each month that focused on hospitalrelated topics. At the end of two consecutive seminars, employees were given time to complete surveys and return them to a locked
container near the exit of the presentation room. Independent variables were collected during the rst administration, while dependent variables were assessed one month later; data were matched based on a unique respondent identier (i.e., last three digits of the
social security number). Due to non-participation and missing data, the nal sample consisted of 202 respondents. Neither presentation directly addressed issues contained on the survey. The majority of the sample indicated administrative or clerical occupations,
averaged 44 years of age, 9 years of organizational tenure and were 58% male.
Measures
Unless otherwise noted, the measures described below utilized a seven-point response format ranging from 1 (strongly disagree)
to 7 (strongly agree).
Abusive supervision
Abusive supervision was measured with Tepper's (2000) 15-item scale. Doesn't give me credit for jobs requiring a lot of effort
and Tells me my thoughts or feelings are stupid represent sample items (Sample 1: = .88; Sample 2: = .91).
Resource management ability was measured using six items developed by Hochwarter et al. (2007). I am able to pace myself at
work when things get hectic and I can change my behavior at work to make sure that I don't run on an empty tank represent scale
items (Sample 1: = .84; Sample 2: = .87).
Employee voice behaviors
Employee voice behavior was measured using Van Dyne and LePine's (1998) six-item scale. I speak up and encourage others in
the group to get involved in issues that affect the group and I develop and make recommendations to my supervisor concerning
issues that affect my work represent scale items (Sample 1: = .79; Sample 2: = .84).
Job satisfaction
We used a ve-item scale developed by Brayeld and Rothe (1951) to measure job satisfaction (Sample 1: = .89; Sample 2:
= .87). Most days I am enthusiastic about my work and Each day of work seems like it will never end (reversed scored)
represent scale items.
Turnover intentions
Turnover intentions were measured using a six-item scale (Mobley, Horner, & Hollingsworth, 1978; Mowday et al., 1984) (Sample
1: = .76; Sample 2: = .91). I plan to look for a new job in the future and I intend to quit my job represent sample items. A vepoint response format was used (1 = strongly disagree to 5 = strongly agree).
Emotional exhaustion (Sample 2 only)
We used a nine-item scale developed by Maslach and Jackson (1981) to measure exhaustion on the job (Sample 2: = .85). I feel
emotionally drained from my work and I feel used up at the end of the work day represent scale items. A ve-point response format
was used (1 = strongly disagree to 5 = strongly agree).
Work effort (Sample 2 only)
Work effort was measured using a ve-item work effort scale developed by Brown and Leigh (1996) (Sample 2: = .90). When I
work, I really exert myself to the fullest and When there's a job to be done, I devote all of my energy to getting it done represent
scale items.
Data analyses
Across both samples, we conducted multiple moderated hierarchical regression analyses to determine the contribution of the
three-way interaction term above that of the main effects and each of the two-way interaction terms (Cohen, Cohen, West, &
Aiken, 2003). The main effects for abusive supervision, resource management ability, and proactive voice behavior were entered in
827
Step 1, followed by each of the combinations of two-way interactions between the three independent variables in Step 2. Finally, the
three-way interaction of abusive supervision, resource management ability, and proactive voice behavior was included in Step 3.
Results
The means, standard deviations, and correlations for all variables in each sample can be found in Table 1. The correlations between
the dependent variables across both samples were modest (highest: r = .48 between job satisfaction and turnover intentions) and
consistent with previous research in regard to magnitude and direction. Further, conrmatory factor analyses revealed that the dependent variables were related but sufciently unique across both samples. According to the one of the most stringent tests of discriminant validity (i.e., Fornell & Larcker, 1981), evidence of discriminant validity is present when the average variance extracted
(AVE) for each dependent variable exceeds the shared variance between each pair of dependent variables. For Sample 1, the AVE
for job satisfaction (.49) was greater than the shared variance (.40) between the two constructs while the AVE for turnover intentions
(.32) did not exceed the shared variance between the two constructs. For Sample 2, the AVE values for job satisfaction (.50), turnover
intentions (.42), exhaustion (.31), and effort (.58) all were greater than the shared variance between each pair of constructs, which
ranged from .01 to .42. Across both samples, therefore, results provided support for the discriminant validity of all constructs with
the exception of turnover intentions in Sample 1. The Sample 1 measure of turnover intentions did, however, demonstrate evidence
of discriminant validity based on Jreskog's (1971) more lenient constrained phi (i.e., x and free) test. Thus, on the bases of conceptual and empirical grounds, we are condent that the inclusion of four different dependent variables was both warranted and
meaningful.
Complete hierarchical moderated regression results can be found in Table 2. As a check, collinearity diagnostics were examined to
identify potential method effects. Specically, variance ination factor (VIF) scores, which measure the extent to which collinearity
among the predictors in each step affects the precision of a regression model, were calculated. VIF scores less than 5 typically are considered acceptable (Hair, Anderson, Tatham, & Black, 1998). Nonetheless, VIF scores did not exceed 1.50 for any model in either sample indicating that the prevalence of multicollinearity was minimal (O'Brien, 2007).
Regression results: Sample 1
The interactive effects of perceived abusive supervision, voice behavior, and resource management ability were hypothesized to
predict job satisfaction and turnover intentions in Sample 1. Regression results provided initial support for the existence of such an
interaction; specically, in the nal step of regression analyses, the three-way interaction term predicted job satisfaction ( = .16,
R2 = .03, p b .01) and turnover intentions ( = .13, R2 = .02, p b .01).
Post hoc tests
We plotted combinations of proactive voice behavior and resource management ability at one standard deviation above and below
the mean across the range of abusive supervision scores in order to determine the form of the signicant interactions (Aiken & West,
1991; Stone & Hollenbeck, 1989; Stone-Romero & Liakhovitski, 2002). Figs. 1 and 2 illustrate these interactive relationships. Signicant abusive supervisionjob satisfaction relationships were conrmed for low voice behaviorhigh resource management ability
(slope: = .25, p b .01), low voice behaviorlow resource management ability (slope: = .17, p b .05), and high voice behaviorlow resource management ability groups (slope: = .14, p b .05). Contrarily, the abusive supervisionjob satisfaction relationship remained unchanged in the high voice behaviorhigh resource management ability setting (slope: = .02, ns).
Moreover, consistent with comparable studies (e.g., Takeuchi, Chen, & Cheung, 2012), pairwise slope tests were conducted in efforts to determine whether the slopes corresponding to the various combinations of high and low levels of the moderator variables
differed signicantly from one another. Results of these tests are summarized in Table 3. Results indicated that the slope
Table 1
Means, standard deviations, and intercorrelations among study variables.
Variable
1. Abusive supervision
2. Voice behavior
3. Resource management ability
4. Job satisfaction
5. Turnover intentions
6. Emotional exhaustion
7. Work effort
Sample 1 mean
Sample 1 SD
Sample 2 mean
Sample 2 SD
.41
.35
.39
.40
.15
.34
.45
.31
.31
.29
.17
.37
.43
.05
.41
.44
.34
.02
.31
.31
.38
.27
.41
.19
.39
.25
1.62
.60
2.91
1.21
5.28
1.20
5.45
.90
3.48
1.56
.08
5.56
.88
.12
.36
.39
4.35
1.03
5.17
1.22
Sample 1 estimates included below the diagonal; Sample 2 estimates included above the diagonal.
p b .05.
p b .01.
.48
5.29
1.21
4.93
1.13
2.81
1.63
2.58
1.12
828
Table 2
Results of regression analyses.
Job satisfaction
Step 1:
Abusive supervision (A)
Voice behavior (V)
RMA (R)
adj R2
Step 2:
AV
VR
RA
adj R2
Step 3:
AVR
adj R2
Turnover intentions
Emot. exh.
Effort
Sample 1
Sample 2
Sample 1
Sample 2
Sample 2
Sample 2
.38
.26
.20
.13
.14
.33
.40
.11
.35
.19
.21
.31
.21
.17
.15
.23
.34
.15
.35
.16
.03
.24
.05
.07
.05
.02
.07
.01
.02
.03
.01
.01
.09
.16
.08
.02
.05
.10
.09
.00
.04
.02
.26
.04
.01
.15
.01
.03
.16
.03
.08
.02
.13
.02
.08
.03
.09
.02
.08
.02
Sample 1: N = 143; Sample 2: N = 202.

Note: RMA = resource management ability; Emot. exh. = emotional exhaustion; results were robust even when controlling for demographic (i.e., age, gender, and
organizational tenure) and personality variables (i.e., Big 5), work engagement, and the squared main effect terms. Therefore, only the most parsimonious results
are reported.
p b .05.
p b .01.
corresponding to the high voicehigh RMA line predicting job satisfaction was signicantly different from all the other slopes; no
other signicant differences between slopes were detected.
As Fig. 2 suggests, signicant abusive supervisionturnover intention relationships were conrmed for low voice behaviorlow resource management ability (slope: = .23, p b .01), high voice behaviorlow resource management ability (slope: = .23, p b .05),
and low voice behaviorhigh resource management ability groups (slope: = .19, p b .05). Conversely, the abusive supervisionturnover relationship remained unchanged in high voice behaviorhigh resource management ability (slope: = .05, ns) settings.
Pairwise slope tests (Table 3) revealed that the slope corresponding to the high voicehigh RMA line predicting turnover intentions
was signicantly different from all of the other slopes; no other signicant differences between slopes were detected.
Regression results: Sample 2

In Sample 2, we assessed the interactive effects of perceived abusive supervision, voice behavior, and resource management ability
on job satisfaction, turnover intentions, emotional exhaustion, and work effort. Complete regression results can be found in Table 2.
After entering the main effects as well as each of the two-way interactions, the abusive supervision voice behavior resource management ability interaction term explained incremental variance in job satisfaction ( = .08, R2 = .02, p b .01), turnover intentions
Job Satisfaction
6
(1) High Voice, High
Resources
(2) High Voice, Low

Resources
(3) Low Voice, High

Resources
(4) Low Voice, Low

Resources
Low Abuse
High Abuse
Fig. 1. Three-way interaction of abusive supervision voice behavior resource management ability on job satisfaction (Sample 1).
829
Turnover Intentions
6
Resources
(2) High Voice, Low

Resources
(3) Low Voice, High

Resources
(4) Low Voice, Low

Resources
1
Low Abuse
High Abuse
Fig. 2. Three-way interaction of abusive supervision voice behavior resource management ability on turnover intentions (Sample 1).
( = .08, R2 = .03, p b .01), emotional exhaustion ( = .09, R2 = .02, p b .01), and work effort ( = .08, R2 = .02, p b .05) in
the last step of regression analyses.
Post hoc tests

Figs. 36 illustrate the documented interactive effects of abusive supervision, voice behavior, and resource management ability on
each outcome in Sample 2. Signicant abusive supervisionjob satisfaction relationships were conrmed for low voice behaviorlow
resource management ability (slope: = .31, p b .01), high voice behaviorlow resource management ability (slope: = .27,
p b .01), and low voice behaviorhigh resource management ability (slope: = .16, p b .05) groups. Conversely, the abusive supervisionjob satisfaction relationship remained unchanged in high voice behaviorhigh resource management ability settings (slope:
= .02, ns). Moreover, the slope corresponding to the high voicehigh RMA line predicting job satisfaction was signicantly different than the slopes corresponding to all other combinations of voice and RMA (see Table 3). Taken together, the results across both
samples support Hypothesis 1.
Moreover, signicant abusive supervisionturnover intention relationships were conrmed for low voice behaviorlow resource
management ability (slope: = .26, p b .01) and high voice behaviorlow resource management ability groups (slope: = .24,
p b .01). Conversely, the relationships remained unchanged in high voice behaviorhigh resource management ability (slope:
= .04, ns) and low voice behaviorhigh resource management ability (slope: = .02, ns) settings. Pairwise comparisons revealed
that the slope corresponding to the high voicehigh RMA line predicting turnover intentions was signicantly different than the
slopes corresponding to high voicelow RMA and low voicelow RMA effects but was not signicantly different from the slope corresponding to the low voicehigh RMA line. Taken together, the results across both samples provide fairly consistent support
Hypothesis 2.
Table 3
Pairwise tests of slope differences.a
Highlow
Combination
Sample 1
Satisfaction
Sample 2
Satisfaction
Sample 1
Turnover intent
Sample 2
Turnover intent
Sample 2
Exhaustion
Sample 2
Work effort
1 and 2
1 and 3
1 and 4
2 and 3
2 and 4
3 and 4
Sig.
Sig.
Sig.
Nonsig.
Nonsig.
Nonsig.
Sig.
Sig.
Sig.
Nonsig.
Nonsig.
Nonsig.
Sig.
Sig.
Sig.
Nonsig.
Nonsig.
Nonsig.
Sig.
Nonsig.
Sig.
Sig.
Nonsig.
Sig.
Sig.
Sig.
Sig.
Nonsig.
Sig.
Nonsig.
Sig.
Sig.
Sig.
Nonsig.
Nonsig.
Nonsig.
Note. Sig. indicates a signicant slope difference was detected whereas nonsig. indicates no slope difference was detected.
1 = high voicehigh RMA
2 = high voicelow RMA
3 = low voicehigh RMA
4 = low voicelow RMA
a
Due to power limitations, an alpha level of .10 was used.
830
Job Satisfaction
6
5
4
3
2
1
0
Fig. 3. Three-way interaction of abusive supervision voice behavior resource management ability on job satisfaction (Sample 2).
Signicant abusive supervisionemotional exhaustion relationships were conrmed for low voice behaviorlow resource management ability (slope: = .31, p b .01), low voice behaviorhigh low resource management ability (slope: = .25, p b .01), high
voice behaviorlow resource management ability (slope: = .17, p b .05), and high voice behaviorhigh low resource management
ability groups (slope: = .13, p b .05). Moreover, the slope corresponding to the high voicehigh RMA line predicting emotional
exhaustion was signicantly different than the slopes corresponding to all other combinations of voice and RMA (see Table 3).
Thus, support was found for Hypothesis 3 as abusive supervision was most strongly (weakly) related to emotional exhaustion
under conditions of low (high) voice behavior and low (high) resource management ability.
Finally, signicant abusive supervisionwork effort relationships were conrmed for low voice behaviorlow resource management ability (slope: = .22, p b .01), low voice behaviorhigh resource management ability (slope: = .18, p b .05), and
high voice behaviorlow resource management ability groups (slope: = .16, p b .01). The abusive supervisionwork effort relationship remained unchanged in the high voice behaviorhigh resource management ability setting (slope: = .02, ns). Moreover,
Turnover Intentions
6
5
4
3
2
1
0
Fig. 4. Three-way interaction of abusive supervision voice behavior resource management ability on turnover intentions (Sample 2).
831
Emotional Exhaustion
6
5
4
3
2
1
0
Fig. 5. Three-way interaction of abusive supervision voice behavior resource management ability on emotional exhaustion (Sample 2).
the slope corresponding to the high voicehigh RMA line predicting work effort was signicantly different than the slopes corresponding to all other combinations of voice and RMA (see Table 3). Thus, Hypothesis 4 was supported.
Discussion
The primary purpose of this study was to examine individual-level factors capable of attenuating the deleterious effects of abusive
supervision. Building off of recent extensions (Meier et al., 2008) of the Job DemandControl model (Karasek, 1979), we examined the
joint interactive effects of proactive voice behavior (i.e., a form of decision latitude or control option) and perceived resource management ability (i.e., an individual difference in control beliefs) in relation to perceived abusive supervision (i.e., a job demand). Individuals who perceived control options available to them (i.e., voice) and indicated strong individual beliefs concerning their control
options (i.e., RMA) (Meier et al., 2008) were hypothesized to experience greater affective and behavioral well-being in response to
the demands of perceived abusive supervision than those who reported fewer voice behaviors and/or reported lesser perceived abilities to manage their resources.
6
Resources
Work Effort
(2) High Voice, Low

Resources
(3) Low Voice, High

Resources
(4) Low Voice, Low

Resources
1
Low Supervisor
Abuse
High Supervisor
Abuse
Fig. 6. Three-way interaction of abusive supervision voice behavior resource management ability on work effort (Sample 2).
832
The results provided support for the corresponding hypotheses. Moreover, abusive supervision demonstrated a harmful effect on
job satisfaction, turnover intentions, and emotional exhaustion that was consistent with past literature (e.g., Aryee et al., 2008; Tepper
et al., 2004; Tepper et al., 2009). Voice was positively and signicantly related to satisfaction and effort and signicantly negatively
related to emotional exhaustion; thus, irrespective of whether or not voice behavior actually changes supervisory mistreatment,
being provided with the opportunity to speak has value in itself (van Prooijen et al., 2007: p. 1288). Additionally, voice demonstrated
a signicant negative effect on turnover intentions in Sample 1, which is not entirely surprising given that voice was originally conceptualized as an alternative to turnover (Hirschman, 1970). Finally, individuals' resource management ability was positively and
signicantly related to job satisfaction and negatively and signicantly related to turnover intentions and emotional exhaustion.
Most notably, we found support for the hypothesized three-way interaction of abusive supervision resource management ability
proactive voice behavior across all of the dependent variables in both samples. Moreover, the forms of the interactions were as hypothesized. Specically, when high levels of abusive supervision were perceived, individuals who reported high resource management ability and high voice behavior demonstrated the most auspicious levels of affective well-being and behavioral functioning.
On the other hand, satisfaction (Sample 2 only) and effort were at their lowest levels and turnover intentions and emotional exhaustion at their highest levels when individuals perceived abusive supervision but reported low levels of resource management ability
and low levels of proactive voice behaviors.
Moreover, while some of the patterns detected did not demonstrate consistent effects across the two samples, there was consistency regarding the effects of high RMA and high voice as compared to all other combinations of voice and resource management ability. Specically, across all outcomes in both samples, the slopes corresponding to the high resource management ability and high voice
condition were signicantly different from the slopes corresponding to the low RMA and low voice condition. Thus, results seemed to
indicate that the combination of high RMA and high voice reects an ideal combination especially when faced with the demands of
perceived abusive supervision. Conversely, low RMA and low voice reects the most subpar combination. In fact, level of perceived
abuse seemed irrelevant to outcomes when RMA and voice were both high.
As expected, neither combination of opposing levels of voice and RMA (i.e., high voicelow RMA, low voicehigh RMA) seemed
better than the other. More specically, in the majority of instances, we did not detect signicant differences in affective or behavioral
functioning for either the high RMAlow voice or low RMAhigh voice combinations. In fact, in all but one instance (i.e., turnover intentions, Sample 1), the highlow and lowhigh lines were not signicantly different from one another. This was not surprising given
that opposing levels of voice and RMA were hypothesized to offset one another (i.e., the benets of high RMA would be offset the
disadvantages of low voice; the benets of high voice would be offset by the disadvantages of low RMA). Thus, as hypothesized, it
appears that neither combination of high voice and low RMA or low voice and high RMA is more effective in ameliorating strain outcomes stemming from perceived supervisory abuse; instead, strain amelioration was consistently evidenced only when individuals
indicated both high levels of voice behaviors and high perceived resource management ability.
Perhaps most tellingly, the results for the most part demonstrated a lack of support for the buffering hypothesis originally advanced by Karasek's (1979) specication of the Job DemandControl model. While increases in perceived abusive supervision
(i.e., demands) were associated with increased strain (i.e., increase dissatisfaction, turnover intentions, emotional exhaustion), the
two-way interaction of perceived abusive supervision and voice behavior (i.e., demand control interaction) was not signicant in
any of the analyses. Thus, we found that job control/decision latitude alone did not result in strain reduction when individuals
faced high job demands. Taken together, these results provide further support for the recent extensions of the JDC model (Meier
et al., 2008; Schaubroeck & Merritt, 1997); specically, decision latitude buffers the job demandstrain relationship only to the extent
that individuals have opportunities for decision latitude and indicate beliefs conducive to exercising their control options. While decision latitude/control in isolation was insufcient in attenuating strain stemming from perceived supervisory abuse, its combination
with favorable control beliefs resulted in the least affective and behavioral strain. Thus, adopting the extended JDC framework may
prove especially benecial when examining additional factors that may serve to attenuate strain stemming from perceived supervisory abuse.
Strengths and limitations
The ndings of this paper should be interpreted in light of a number of notable strengths and limitations. Specically, a notable
strength includes our ability to constructively replicate and extend study results across two samples of working adults. Moreover,
in Sample 2, data were collected on independent and dependent variables at different points in time; by doing so, we tried to curtail
the likelihood of results plagued by common method bias.
In terms of limitations, the main limitation of our study was the reliance on entirely self-report measures. Nonetheless, many of the
phenomena under investigation are more properly and fully understood utilizing self-report measures. For instance, abusive supervision reects individuals' perceptions concerning the way they are treated by their supervisor largely irrespective of the objective reality whereas resource management ability, as we conceptualize it, is an individually held perception of one's own abilities rather
than a reection of an objectively observed skill set. Thus, we feel the use of self-reported data is justiable. In addition, we incorporated a number of Podsakoff, MacKenzie, and Podsakoff's (2012) recommendations into the survey instrument (e.g., balancing positive and negative scale items) and research design efforts (specically, the use of time-separated data in Sample 2). Taken together,
we feel that the use of these numerous methodological precautions as well as our ability to replicate the signicant three-way interaction across two diverse datasets minimizes the likelihood that ndings result from common method bias.
Additionally, two issues pertaining to the level of analysis exist as notable limitations of this manuscript. Specically, while it is
entirely reasonable to assume that perceptions of abusive supervision may exist at the individual, team or group, and organizational
833
levels, this manuscript focused entirely on individually experienced perceptions of abusive supervision. This is in keeping with the
original conceptualization of abusive supervision as an inherently perceptual phenomenon (Tepper, 2000). Relatedly, with regard
to Sample 2, the research design prevented us from determining whether, or the extent to which, multiple subordinates responding
to our survey shared the same supervisor. Nonetheless, the results from our student-recruited sample consisting of employees from
numerous companies (thus precluding the possibility of nesting) as well as the results from the sample comprised of employees from
one company were consistent. Therefore, while future research may benet from examining inter-individual or between-level differences in abuse perceptions, we feel condent from both conceptual (i.e., perceptual, subjective, and independent nature of the abusive
supervision construct) and empirical grounds (i.e., consistent results and patterns across the 2 diverse samples) that our results are
truly reective of individual-level phenomena.
Moreover, results suggest that the three-way interaction terms only explained an additional 2 to 3% of variance in the criterion of
interest, which could be seen as an additional limitation of this study. However, while this amount of incremental explained variance
might seem small, this increment is comparable to that typically expected and found (i.e., 13%) for moderator effects in eld studies
(Champoux & Peters, 1987; Chaplin, 1991). Also, scholars have recently discussed a concern regarding the use of static analyses for
testing dynamic conceptual models (e.g., Pitariu & Ployhart, 2010; Ployhart & Vandenberg, 2010). The measures used to operationalize
perceived abusive supervision, resource management ability, and proactive voice behavior captured individuals' perceptions of each
construct at the moment of measurement. Thus, despite that these constructs are likely dynamic and varied over time, a notable limitation of this study was that hypotheses focused solely on the interactive relationship between these constructs at a specic point in
time; therefore, issues such as causality and potential for feedback loops could not be addressed within the current study. Finally,
pairwise tests of slope differences need to be interpreted cautiously given the inuence that this strategy has on statistical power
(Dawson & Richter, 2006).
Directions for future research
Abusive supervision is a complex construct with intricate relationships to many important work-related outcomes; therefore,
there exist a number of fruitful avenues for future research. First, research should explicitly examine the mechanisms through
which voice and perceived RMA impact reactions to abusive supervision. It is possible that voice behavior serves to restore interactional justice whereas resource management ability serves to reduce the employee's dependence on their supervisor, thus reducing
the supervisor's power. While we relied on theory to support our interactional hypotheses, future research should explicitly measure
the mechanisms through which the three-way interaction of abusive supervision proactive voice behavior RMA inuences
affective well-being and behavioral functioning.
Moreover, future research should continue to examine factors that reduce the negative impact of perceived supervisory mistreatment. Other forms of control (e.g., job security, autonomy, physical distance) and other individual differences in control beliefs
(e.g., self-efcacy, locus of control) could impact the degree to which abusive supervision is considered threatening. Furthermore, research focusing on the appraisal of abusive supervision may be useful as much of the literature has been focused on how individuals
cope with or react to abusive supervision after they have assumedly gone through the appraisal process. When conceptualizing abusive supervision as a source of stress, it important to understand the complete process in order to fully understand and determine its
effects. Thus, according to Lazarus and Folkman (1984), individuals might react differentially to abuse depending on whether they feel
abuse is taxing to them personally (primary appraisal) and if they have the resources needed to cope (secondary appraisal).
Additionally, from the outset, we have emphasized subordinates' perceptions of abusive supervision. Nonetheless, it would be interesting to examine the extent to which subordinates' perceptions correspond to objective accounts of supervisory mistreatment.
Some research has already established that individual differences in core-self evaluations (Wu & Hu, 2009) and attribution style
(Martinko, Harvey, Sikora, & Douglas, 2011) explain meaningful variance in supervisory abuse perceptions; however, to what extent
do such individual differences accurately predict objective abuse conditions? Moreover, it would be quite informative to determine
whether certain subordinate and/or supervisor characteristics amplify or attenuate the differences between subordinates' perceptions
and objective reality.
Are there situations in which supervisors are perceptually and objectively perceived as abusive regardless of the subordinate? For
instance, it is possible that the behavior of narcissistic or Machiavellian supervisors is perceived accurately (i.e., perceptions
correspond with objective reality) regardless of subordinates' individual differences; in essence, such overtly abusive behavior
would reect a strong situation in which all subordinates, regardless of their individual differences, would perceive (and correspondingly, react) similarly (Mischel, 1977; cf. Meyer, Dalal, & Hermida, 2010). Contrarily, it is possible that the behavior of an authentic
supervisor is perceived inaccurately (i.e., perceptions deviate from objective reality) based on subordinates' individual differences
(e.g., trait or organizational cynicism, trait hostility, neuroticism). For instance, authentic leaders may be so true to themselves
that they spontaneously demonstrate impassioned language that could be interpreted as yelling. In these instances, withinsubordinate individual differences might predict whether such ambiguous supervisory behavior is interpreted as abusive (Chan
& McAllister, 2014). Researchers are encouraged to examine these possibilities regarding the consistency between perceptions and
objective supervisory abuse.
Further, we argued that resource management ability represents an individual difference that would attenuate strain stemming
from perceived abusive supervision when combined with control options/decision latitude (e.g., proactive voice behavior). While
RMA represents a perceived ability to maintain, mobilize, protect, and acquire benecial resources (Hochwarter et al., 2007), the
construct does not capture the actual amount of resources available for use. In other words, while individuals may perceive that
they have ample RMA, they might nd themselves in a resource-depleted state (e.g., due to uncontrollable economic and/or
834
organizational factors). Unfortunately, a notable limitation of our manuscript is that we were unable to determine the amount of resources available to study participants. As argued throughout the manuscript, though, individuals with fewer available resources
(i.e., fewer control options) should be more negatively affected by demands (e.g., abusive supervision) if they perceive themselves
as having substandard control beliefs (e.g., low RMA). Future research should test this assumption and explore the possibility that individuals' varying levels of resources may inuence the amount of threat received from abusive supervision.
Finally, as mentioned previously, a notable limitation of this study is the inability to test hypotheses concerning the dynamic relationships among the constructs of interest. Therefore, future research employing longitudinal or experimental designs is needed
to address whether the relationships of these constructs change over time. For instance, it is possible that the strain attenuating effects
of RMA and voice result in heightened beliefs regarding RMA and increased instances of engaging in voice in the future as a function of
individuals' previously successful strain-reduction efforts; nonetheless, the current study design precluded the investigation of such
cyclical relationships.
Contributions
The present two-sample investigation extends research on abusive supervision and the Job DemandControl (JDC) model
(Karasek, 1979; Meier et al., 2008). Specically, this is one of a few studies to examine the individual-level factors capable of attenuating the negative effects of abusive supervision. Nonetheless, given that abusive supervision is an inherently perceptual phenomenon,
it is important to examine those individual differences that enable individuals to concurrently perceive supervisory mistreatment and
maintain adequate levels of psychological and behavioral functioning. Despite the absence of a consistent theoretical explanation, the
studies that have examined the mitigating effects of subordinates' individual difference factors on the relationship between abusive
supervision and work outcomes (e.g., Hu, 2012; Mackey et al., 2013; Nandkeolyar et al., 2014) all seem to have identied factors
pertaining to control-related remedies. Thus, not only does this manuscript add to the limited body of research that examines factors
capable of mitigating the negative effects of abusive supervision, this manuscript provides a unifying framework (i.e., the extended
JDC model) by which researchers are able to classify and explain why certain individual differences enable subordinates to survive,
and sometimes even thrive, under conditions of perceived supervisory mistreatment.
Further, our study provides further support for recent extensions of the Job DemandControl model posited by Meier and
colleagues (2008). The initial specication of the Job DemandControl model suggested that the negative effects of job demands on
individuals' health and well-being are attenuated by heightened levels of job control (Karasek, 1979). Nonetheless, support for this
buffering hypothesis has been largely equivocal. In response, Meier et al. (2008) examined how individual characteristics inuenced
individuals' abilities to exercise their decision-latitude and control; they found support for the three-way interaction of job demands,
job control, and individual differences in abilities to exercise control (e.g., locus of control). Not only does our study contribute further
strength to Meier and colleagues (2008) extension of the JDC model, but also expands the content domain to which the JDC and the
extended JDC have been applied to recognize alternative forms of demands (i.e., abusive supervision), control (i.e., voice behavior),
and an additional individual characteristic that allows certain individuals to better (or more willingly) exercise their control options
(i.e., resource management ability).
Practical implications
From a practical standpoint, organizations should adopt strong policies against all forms of interpersonal mistreatment
(e.g., bullying, abusive supervision). Moreover, aside from simply adopting and promoting high standards of considerate conduct
for all individuals within the workplace, organizations should have explicit, strict, and swift policies for punishing individuals who violate these standards. Moreover, given the current state of the economy and the resultant scarcity of job alternatives, engaging in constructive voice behaviors may exist as one of the few feasible options available to employees to positively affect changes in their work
environments. Thus, voice (combined with high RMA) might serve as an effective means to combat the negative affective and behavioral effects of abusive supervision.
Given that proactive voice behaviors demonstrated a signicant main effect on a few of the outcomes of interest, formalized HR
practices that provide means for airing grievances (e.g., HR hotline, 360 degree performance appraisal) may help reduce the negative
effects of perceptions of abusive supervision. This is especially important as abusive supervision isn't always indicative of supervisors'
objective treatment of their employees. Therefore, some subordinates (e.g., those high in negative affectivity) will perceive abuse even
in the absence of overt malevolence. Thus, providing employees an outlet to constructively voice their concerns might serve as a
means of catharsis (van Prooijen et al., 2007) and might actually serve to reduce the negative effects of perceived abuse in and of itself.
Nonetheless, simply providing employees with an outlet to voice will not be benecial if organizations fail to address the underlying issues that provoked voice behavior. More specically, research has found that voice behavior that went unheard or was ignored
(Pinder & Harlos, 2001) resulted in frustration and decreased output (Hunton, Price, & Hall, 1996). In support, a recent review found
that many of the positive outcomes associated with voice (e.g., catharsis) are actually counteracted when voice behavior is
disregarded (Bashshur & Oc, 2015). Further, research has found that for some individuals, excessively talking about negative events
with coworkers (i.e., co-rumination) tends to exacerbate negative outcomes (e.g., depression; Haggard, Robert, & Rose, 2011).
Thus, organizations should attempt to simultaneously encourage voice while providing swift means to address employee's unresolved
grievances.
Also, it appears that individuals with high RMA experience less dissatisfaction, less emotional exhaustion, and less likelihood of
turning over. Thus, organizations might consider additional avenues other than one's supervisor through which employees can solicit
835
and acquire additional resources. For instance, high performing employees may be recognized through company-wide initiatives and
may be eligible for additional training, continuing education opportunities, and so forth. Similarly, organizations should design jobs
such that subordinates have some means of actual or perceived control. Nonetheless, control by itself is not sufcient to ward off
the negative effects of abusive supervision, as simply having control does not imply that individuals will feel capable or comfortable
exercising their control options.
Conclusion
Given the preponderance of evidence linking abusive supervision to deleterious affective well-being and reduced behavioral
functioning, the present study attempts to examine individual factors capable of attenuating the harmful effects of perceived
supervisory abuse. Underpinned by recent extensions of the DemandControl model (Meier et al., 2008), we hypothesized
that those individuals high in resource management ability (RMA) and high in proactive voice would be the least negatively
affected by perceived supervisory abuse because they possess the personal control options (i.e., voice) and control beliefs
(i.e., perceived resource management ability) capable of attenuating the demands of perceived supervisory abuse.
Results from two samples provided overwhelming support for these hypotheses, as individuals perceiving high levels of abusive
supervision reported the most auspicious levels of satisfaction, turnover intentions, emotional exhaustion and effort when they
reported high levels of proactive voice behaviors and heightened resource management abilities. Contrarily, those individuals indicating low amounts of voice behavior and low resource management abilities reported the most strain under high (perceived) abuse
conditions. Results also indicated the initial buffering hypothesis advanced by Karasek's (1979) JDC model did not hold; in other
words, decision latitude alone (i.e., voice) did not attenuate the negative effects of high demands (i.e., perceived supervisory
abuse). Therefore, investigating the simultaneous effects of abusive supervision, resource management ability, and proactive voice
behavior provided a better understanding of the complexities of effectively coping with abusive supervision. Hopefully this study
sparks a proliferation of future research that addresses how individuals can best cope with perceived supervisory abuse.
References
Aiken, L., & West, S.G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park CA: Sage.
Aryee, S., Sun, L. -Y., Chen, Z.X.G., & Debrah, Y.A. (2008). Abusive supervision and contextual performance: The mediating role of emotional exhaustion and the
moderating role of work unit structure. Management and Organization Review, 4, 393411.
Aspinwall, L.G., & Taylor, S.E. (1997). A stitch in time: Self-regulation and proactive coping. Psychological Bulletin, 121, 417436.
Bakker, A.B., & Demerouti, E. (2007). The job demandsresources model: State of the art. Journal of Managerial Psychology, 22, 309328.
Bamberger, P.A., & Bacharach, S.B. (2006). Abusive supervision and subordinate problem drinking: Taking resistance, stress and subordinate personality into account.
Human Relations, 59, 723752.
Bashshur, M.R., & Oc, B. (2015). When voice matters: A multilevel review of the impact of voice in organizations. Journal of Management, 41, 15301554.
Bolino, M., Valcea, S., & Harvey, J. (2010). Employee, manage thyself: The potentially negative implications of expecting employees to behave proactively. Journal of
Occupational and Organizational Psychology, 83, 325345.
Bowling, N.A., & Michel, J.S. (2011). Why do you treat me badly? The role of attributions regarding the cause of abuse in subordinates' responses to abusive supervision.
Work and Stress, 25, 309320.
Brayfield, A.H., & Rothe, H.F. (1951). An index of job satisfaction. Journal of Applied Psychology, 35, 307311.
Brief, A.P. (1998). Attitudes in and around organizations. Thousand Oaks, CA: Sage.
Brown, S.P., & Leigh, T.W. (1996). A new look at psychological climate and its relationship to job involvement, effort, and performance. Journal of Applied Psychology, 81,
358368.
Burton, J., & Hoobler, J. (2006). Subordinate self-esteem and abusive supervision. Journal of Managerial Issues, 18, 340355.
Burton, J., Mitchell, T., & Lee, T. (2005). The role of self-esteem and social influences in aggressive reactions to interactional injustice. Journal of Business and Psychology,
20, 131170.
Byrne, Z., & Hochwarter, W. (2006). I get by with a little help from my friends: The interaction of chronic pain and organizational support on work contribution. Journal
of Occupational Health Psychology, 11, 215227.
Champoux, J.E., & Peters, W.S. (1987). Form, effect size and power in moderated regression analysis. Journal of Occupational Psychology, 60, 243255.
Chan, M.E., & McAllister, D.J. (2014). Abusive supervision through the lens of employee state paranoia. Academy of Management Review, 39, 4466.
Chaplin, W.F. (1991). The next generation of moderator research in personality psychology. Journal of Personality, 59, 143178.
Chi, S., & Liang, S.G. (2013). When do subordinates' emotion-regulation strategies matter? Abusive supervision, subordinates' emotional exhaustion, and work
withdrawal. The Leadership Quarterly, 24, 125137.
Cohen, J., Cohen, P., West, S.G., & Aiken, L.S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. Mahwah, NJ: Lawrence Erlbaum
Associates.
Costa, P.T., Jr., & McCrae, R.R. (1992). Four ways five factors are basic. Personality and Individual Differences, 13, 653665.
Costa, P.T., Jr., McCrae, R.R., & Dye, D.A. (1991). Facet scales for agreeableness and conscientiousness: A revision of the NE0 personality inventory. Personality and
Individual Differences, 12, 887898.
Daniels, K., Boocock, G., Glover, J., Hartley, R., & Holland, J. (2009). An experience sampling study of learning, affect, and the demands control support model. Journal of
Applied Psychology, 94, 10031017.
Daniels, K., & Guppy, A. (1994). Occupational stress, social support, job control, and psychological well-being. Human Relations, 47, 15231544.
Dawson, J.F., & Richter, A.W. (2006). Probing three-way interactions in moderated multiple regression: Development and application of a slope difference test. Journal
of Applied Psychology, 91, 917926.
de Jonge, J., Dollard, M.F., Dormann, C., Le Blanc, P.M., & Houtman, I.L. (2000). The demandcontrol model: Specific demands, specific control, and well-defined groups.
International Journal of Stress Management, 7, 269287.
de Jonge, J., & Dormann, C. (2006). Stressors, resources, and strain at work: A longitudinal test of the triple-match principle. Journal of Applied Psychology, 91,
13591374.
de Jonge, J., & Kompier, M.A. (1997). A critical examination of the demandcontrolsupport model from a work psychological perspective. International Journal of Stress
Management, 4, 235258.
de Lange, A.H., Taris, T.W., Kompier, M.A.J., Houtman, I.L.D., & Bongers, P.M. (2003). The very best of the millennium: Longitudinal research and the demand
control(support) model. Journal of Occupational Health Psychology, 8, 282305.
836
Demerouti, E., & Rispens, S. (2014). Improving the image of studentrecruited samples: A commentary. Journal of Occupational and Organizational Psychology, 87,
3441.
Detert, J.R., & Burris, E.R. (2007). Leadership behavior and employee voice: Is the door really open? Academy of Management Journal, 50, 869884.
Diener, E., Suh, E.M., Lucas, R.E., & Smith, H.L. (1999). Subjective well-being: Three decades of progress. Psychological Bulletin, 125, 276302.
Dutton, J.E., Ashford, S.J., O'neill, R.M., Hayes, E., & Wierba, E.E. (1997). Reading the wind: How middle managers assess the context for selling issues to top managers.
Strategic Management Journal, 18, 407425.
Dwyer, D.J., & Ganster, D.C. (1991). The effects of job demands and control on employee attendance and satisfaction. Journal of Organizational Behavior, 12,
595608.
Eisenberger, R., Stinglhamber, F., Vandenberghe, C., Sucharski, I.L., & Rhoades, L. (2002). Perceived supervisor support: Contributions to perceived organizational
support and employee retention. Journal of Applied Psychology, 87, 565573.
Fornell, C., & Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 3950.
Fuller, J.B., Barnett, T., Hester, K., Relyea, C., & Frey, L. (2007). An exploratory examination of voice behavior from an impression management perspective. Journal of
Managerial Issues, 19, 134151.
Ganster, D.C., & Fusilier, M.R. (1989). Control in the workplace. In C.L. Cooper, & I. Robertson (Eds.), International review of industrial and organizational psychology
(pp. 235280). London: Wiley.
Grandey, A.A., & Cropanzano, R. (1999). The conservation of resources model applied to work-family conflict and strain. Journal of Vocational Behavior, 54, 350370.
Haggard, D.L., Robert, C., & Rose, A.J. (2011). Co-rumination in the workplace: Adjustment trade-offs for men and women who engage in excessive discussions of workplace problems. Journal of Business and Psychology, 26, 2740.
Hair, J., Anderson, R., Tatham, R., & Black, W. (1998). Multivariate analysis. Englewood: Prentice Hall International.
Halbesleben, J.R., Neveu, J.P., Paustian-Underdahl, S.C., & Westman, M. (2014). Getting to the COR understanding the role of resources in conservation of resources
theory. Journal of Management, 40, 13341364.
Harris, K.J., Kacmar, K.M., & Zivnuska, S. (2007). An investigation of abusive supervision as a predictor of performance and the meaning of work as a moderator of the
relationship. The Leadership Quarterly, 18, 252263.
Harvey, P., Stoner, J., Hochwarter, W., & Kacmar, C. (2007). Coping with abusive supervision: The neutralizing effects of ingratiation and positive affect on negative
employee outcomes. The Leadership Quarterly, 18, 264280.
Hirschman, A.O. (1970). Exit, voice, and loyalty: Responses to decline in firms, organizations, and states. Vol. 25, Harvard University Press.
Hobfoll, S.E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44, 513524.
Hobfoll, S.E. (2001). The influence of culture, community, and the nestedself in the stress process: Advancing conservation of resources theory. Applied Psychology, 50,
337421.
Hobfoll, S.E., & Leiberman, J.R. (1987). Personality and social resources in immediate and continued stress resistance among women. Journal of Personality and Social
Psychology, 52, 1826.
Hochwarter, W. (2014). On the merits of student-recruited sampling: Opinions a decade in the making. Journal of Occupational and Organizational Psychology, 87,
2733.
Hochwarter, W.A., Laird, M.D., & Brouer, R.L. (2008). Board up the windows: The interactive effects of hurricane-induced job stress and perceived resources on work
outcomes. Journal of Management, 34, 263289.
Hochwarter, W.A., Perrew, P.L., Meurs, J.A., & Kacmar, C. (2007). The interactive effects of work-induced guilt and ability to manage resources on job and life
satisfaction. Journal of Occupational Health Psychology, 12, 125135.
Hochwarter, W., & Treadway, D. (2003). The interactive effects of negative and positive affect on the politics perceptionsjob satisfaction relationship. Journal of
Hoobler, J.M., & Hu, J. (2013). A model of injustice, abusive supervision, and negative affect. The Leadership Quarterly, 24, 256269.
Hu, H. (2012). The influence of employee emotional intelligence on coping with supervisor abuse in a banking context. Social Behavior and Personality: An International
Journal, 40, 863875.
Hughes, E.C. (1950). Work and the self. In J.H. Rohrer, & M. Sherif (Eds.), Social psychology at the crossroads (pp. 313323). New York: Harper.
Hunton, J.E., Price, K.H., & Hall, T.W. (1996). A field experiment examining the effects of membership in voting majority and minority subgroups and the ameliorating
effects of postdecisional voice. Journal of Applied Psychology, 81, 806812.
Iaffaldano, M.T., & Muchinsky, P.M. (1985). Job satisfaction and job performance: A meta analysis. Psychological Bulletin, 97, 251273.
Janssen, P.P., De Jonge, J., & Bakker, A.B. (1999). Specific determinants of intrinsic work motivation, burnout and turnover intentions: A study among nurses. Journal of
Advanced Nursing, 29, 13601369.
Johnson, J.V., & Hall, E.M. (1988). Job strain, work place social support, and cardiovascular disease: A cross-sectional study of a random sample of the Swedish working
population. American Journal of Public Health, 78, 13361342.
Jreskog, K.G. (1971). Statistical analysis of sets of congeneric tests. Psychometrika, 36, 109133.
Karasek, R. (1979). Job demands, job decision latitude, and mental strain. Administrative Science Quarterly, 24, 285311.
King, A.C., Winett, R.A., & Lovett, S.B. (1986). Enhancing coping behaviors in at-risk populations: The effects of time-management instruction and social support in
women from dual-earner families. Behavior Therapy, 17, 5766.
Laird, M.D., Brouer, R., & Hochwarter, W.A. (2005, November). Hurricane stress and conservation of resources. Paper presented at the meeting of the Southern Management
Association, Charleston, SC.
Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY: Springer Publishing Company, Inc.
Leventhal, G.S. (1980). What should be done with equity theory? New approaches to the study of fairness in social relationships. In K. Gergen, M. Greenberg, & R. Willis
(Eds.), Social exchange: Advances in theory and research (pp. 2755). New York: Plenum Press.
Lind, E.A., Kanfer, R., & Earley, P.C. (1990). Voice, control, and procedural justice: Instrumental and noninstrumental concerns in fairness judgments. Journal of
Personality and Social Psychology, 59, 952959.
Lind, E.A., & Tyler, T.R. (1988). The social psychology of procedural justice. New York: Plenum Press.
Mackey, J.D., Ellen, B.P., III, Hochwarter, W.A., & Ferris, G.R. (2013). Subordinate social adaptability and the consequences of abusive supervision perceptions in two
samples. The Leadership Quarterly, 24, 732746.
Martinko, M.J., Harvey, P., Brees, J.R., & Mackey, J. (2013). A review of abusive supervision research. Journal of Organizational Behavior, 34, S120S137.
Martinko, M., Harvey, P., Sikora, D., & Douglas, S. (2011). Perceptions of abusive supervision: The role of subordinates' attribution styles. The Leadership Quarterly, 22, 751764.
Maslach, C., & Jackson, S.E. (1981). The measurement of experienced burnout. Journal of Organizational Behavior, 2, 99113.
Meier, L.L., Semmer, N.K., Elfering, A., & Jacobshagen, N. (2008). The double meaning of control: Three way interactions between internal resources, job control, and
stressors at work. Journal of Occupational Health Psychology, 13, 244.
Meier, L.L., & Spector, P.E. (2013). Reciprocal effects of work stressors and counterproductive work behavior: A five-wave longitudinal study. Journal of Applied
Meyer, R.D., Dalal, R.S., & Hermida, R. (2010). A review and synthesis of situational strength in the organizational sciences. Journal of Management, 36, 121140.
Mischel, W. (1977). The interaction of person and situation. In D. Magnusson, & N.S. Endler (Eds.), Personality at the crossroads: Current issues in interactional psychology
(pp. 333352). Hillsdale, NJ: Lawrence Erlbaum.
Mitchell, M.S., & Ambrose, M.L. (2007). Abusive supervision and workplace deviance and the moderating effects of negative reciprocity beliefs. Journal of Applied
Mobley, W.H., Horner, S.O., & Hollingsworth, A.T. (1978). An evaluation of precursors of hospital employee turnover. Journal of Applied Psychology, 63, 408414.
Monnier, J., Cameron, R.P., Hobfoll, S.E., & Gribble, J.R. (2002). The impact of resource loss and critical incidents on psychological functioning in fire-emergency
workers: A pilot study. International Journal of Stress Management, 9, 1129.
Morrison, E.W. (2011). Employee voice behavior: Integration and directions for future research. Academy of Management Annals, 5, 373412.
837
Mowday, R.T., Koberg, C.S., & McArthur, A.W. (1984). The psychology of the withdrawal process: A cross-validation test of Mobley's intermediate linkages model of
turnover in two samples. Academy of Management Journal, 27, 7994.
Muraven, M., & Baumeister, R.F. (2000). Self-regulation and depletion of limited resources: Does self-control resemble a muscle? Psychological Bulletin, 126, 247259.
Nandkeolyar, A.K., Shaffer, J.A., Li, A., Ekkirala, S., & Bagger, J. (2014). Surviving an abusive supervisor: The joint roles of conscientiousness and coping strategies. Journal
of Applied Psychology, 99, 138150.
Ng, T.W., & Feldman, D.C. (2012). Employee voice behavior: A metaanalytic test of the conservation of resources framework. Journal of Organizational Behavior, 33,
216234.
O'Brien, R. (2007). A caution regarding rules of thumb for variance inflation factors. Quality and Quantity, 41, 673690.
Perrew, P.L., & Ganster, D.C. (1989). The impact of job demands and behavioral control on experienced job stress. Journal of Organizational Behavior, 10, 213229.
Pinder, C.C., & Harlos, H.P. (2001). Employee silence: Quiescence and acquiescence as responses to perceived injustice. Research in Personnel and Human Resources
Pitariu, A.H., & Ployhart, R.E. (2010). Explaining change: Theorizing and testing dynamic mediated longitudinal relationships. Journal of Management, 36, 405429.
Ployhart, R.E., & Vandenberg, R.J. (2010). Longitudinal research: The theory, design, and analysis of change. Journal of Management, 36, 94120.
Podsakoff, P., MacKenzie, S., & Podsakoff, N. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of
Rafferty, A.E., & Restubog, S.L.D. (2011). The influence of abusive supervisors on followers' organizational citizenship behaviours: The hidden costs of abusive supervision. British Journal of Management, 22, 270285.
Schaubroeck, J., & Merritt, D.E. (1997). Divergent effects of job control on coping with work stressors: The key role of self-efficacy. Academy of Management Journal, 40,
738754.
Stone, E.F., & Hollenbeck, J.R. (1989). Clarifying some controversial issues surrounding statistical procedures for detecting moderator variables: Empirical evidence and
related matters. Journal of Applied Psychology, 74, 310.
Stone-Romero, E.F., & Liakhovitski, D. (2002). Strategies for detecting moderator variables: A review of conceptual and empirical issues. Research in Personnel and
Human Resources Management, 21, 333372.
Sumer, N., Karanchi, A., Berument, S., & Gunes, H. (2005). Personal resources, coping self-efficacy, and quake exposure as predictors of psychological distress following
the 1999 earthquakes in Turkey. Journal of Traumatic Stress, 18, 331342.
Takeuchi, R., Chen, Z., & Cheung, S. (2012). Applying uncertainty management theory to employee voice behavior: An integrative investigation. Personnel Psychology,
65, 283323.
Tepper, B.J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43, 178190.
Tepper, B.J. (2007). Abusive supervision in work organizations: Review, synthesis, and research agenda. Journal of Management, 33, 261289.
Tepper, B.J., Carr, J.C., Breaux, D.M., Geider, S., Hu, C., & Hua, W. (2009). Abusive supervision, intentions to quit, and employees' workplace deviance: A power/dependence analysis. Organizational Behavior and Human Decision Processes, 109, 156167.
Tepper, B.J., Duffy, M.K., Hoobler, J., & Ensley, M.D. (2004). Moderators of the relationships between coworkers' organizational citizenship behavior and fellow employees' attitudes. Journal of Applied Psychology, 89, 455465.
Tepper, B.J., Duffy, M.K., & Shaw, J.D. (2001). Personality moderators of the relationship between abusive supervision and subordinates' resistance. Journal of Applied
Van der Doef, M., & Maes, S. (1999). The job demandcontrol (support) model and psychological well-being: A review of 20 years of empirical research. Work and
Stress, 13, 87114.
Van Dyne, L., Cummings, L.L., & McLean Parks, J. (1995). Extra-role behaviors: In pursuit of construct and definitional clarity. In L.L. Cummings, & B.M. Staw (Eds.),
Research in organizational behavior, Vol. 17. (pp. 215285). Greenwich, CT: JAI Press.
Van Dyne, L., & LePine, J.A. (1998). Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Academy of Management Journal, 41, 108119.
van Prooijen, J.W., Van Den Bos, K., & Wilke, H.A. (2007). Procedural justice in authority relations: The strength of outcome dependence influences people's reactions to
voice. European Journal of Social Psychology, 37, 12861297.
Wei, F., & Si, S. (2013). Tit for tat? Abusive supervision and counterproductive work behaviors: The moderating effects of locus of control and perceived mobility. Asia
Pacific Journal of Management, 30, 281296.
Weiss, H.M. (2002). Deconstructing job satisfaction. Human Resource Management Review, 12, 173194.
Wheeler, A., Shanine, K., Leon, M., & Whitman, M. (2014). Studentrecruited samples in organizational research: A review, analysis, and guidelines for future research.
Journal of Occupational and Organizational Psychology, 87, 126.
Whiting, S.W., Podsakoff, P.M., & Pierce, J.R. (2008). Effects of task performance, helping, voice, and organizational loyalty on performance appraisal ratings. Journal of
Applied Psychology, 93, 125139.
Wu, T., & Hu, C. (2009). Abusive supervision and employee emotional exhaustion: Dispositional antecedents and boundaries. Group & Organization Management, 34,
143169.
Yeo, G.B., & Neal, A. (2004). A multilevel analysis of effort, practice, and performance: Effects; of ability, conscientiousness, and goal orientation. Journal of Applied
Zellars, K.L., Hochwarter, W.A., Lanivich, S.E., Perrew, P.L., & Ferris, G.R. (2011). Accountability for others, perceived resources, and well-being: Convergent restricted
nonlinear results in two samples. Journal of Occupational and Organizational Psychology, 84, 95115.

The Leadership Quarterly: Rachel E. Frieder, Wayne A. Hochwarter, Philip S. Deortentiis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Leadership Quarterly: Rachel E. Frieder, Wayne A. Hochwarter, Philip S. Deortentiis

Uploaded by

Copyright:

Available Formats

The Leadership Quarterly 26 (2015) 821837

Contents lists available at ScienceDirect

The Leadership Quarterly

Attenuating the negative effects of abusive supervision: The role

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

Sample 1: N = 143; Sample 2: N = 202.

Regression results: Sample 2

(2) High Voice, Low

(3) Low Voice, High

(4) Low Voice, Low

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

(2) High Voice, Low

(3) Low Voice, High

(4) Low Voice, Low

Post hoc tests

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

(2) High Voice, Low

(3) Low Voice, High

(4) Low Voice, Low

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

R.E. Frieder et al. / The Leadership Quarterly 26 (2015) 821837

You might also like