Passive-Aggressive Behavior and Leadership Styles in Organizations

Abstract

In this phenomenological study, 13 experts were asked about passive-aggressive (PA) behaviors in the workplace, specifically, whether leadership styles (autocratic, transactional, and transformational) can predict them. Participants were asked to consider the occurrence of PA behaviors in typically healthy working individuals' (rather than disordered individuals') responses to leadership styles and organizational events. Of the eight themes that emerged from the analyzed interviews, three hold particular significance to the workplace. First, a majority of the participants viewed PA behaviors in organizations as a combination of exogenous and endogenous factors. Second, most interviewees agreed that specific types of change in organizations contribute to PA behaviors. Third, most participants viewed the autocratic leadership style as a predictor of PA behaviors.

Full text

130
Journal of Leadership &
Organizational Studies
Volume 14 Number 2
November 2007 130-142
©2007 Baker College
10.1177/1071791907308044
http://jlos.sagepub.com
hosted at
http://online.sagepub.com
Passive-Aggressive Behavior and
Leadership Styles in Organizations
Nora J. Johnson
Wynnewood, PA
Thomas Klee
Chestnut Hill College
In this phenomenological study, 13 experts were asked about passive-aggressive (PA) behaviors in the workplace,
specifically, whether leadership styles (autocratic, transactional, and transformational) can predict them. Participants
were asked to consider the occurrence of PA behaviors in typically healthy working individuals’(rather than disordered
individuals’) responses to leadership styles and organizational events. Of the eight themes that emerged from the
analyzed interviews, three hold particular significance to the workplace. First, a majority of the participants viewed PA
behaviors in organizations as a combination of exogenous and endogenous factors. Second, most interviewees agreed
that specific types of change in organizations contribute to PA behaviors. Third, most participants viewed the autocratic
leadership style as a predictor of PA behaviors.
Keywords: passive-aggresive; leadership styles; workplace; organizations; predictors
Organizational leadership styles have a clear impact
on organizational culture, including employee
behavior (Bennis, 1987; Graham, 1995), organiza-
tional “rigidity and stagnation” (Ashforth & Lee,
1990), and corporate image (Kacmar & Baron, 1999).
Graham (1995) suggested that specific leadership
styles motivate employees differently and encourage
specific sets of responses. For example, autocratic or
coercive styles yield employee responses distinct from
those resulting from transactional or task-oriented
styles (Argyris, 1985) as well as those resulting from
transformational (Burns, 1978) or servant (Greenleaf,
1977) leadership styles. Some leadership styles or cer-
tain aspects of them add significant costs to an organi-
zation’s bottom line through decreased productivity,
lower employee morale, broken deadlines, and mis-
communication (Stohl & Cheney, 2001). In addition to
adding cost to an organization’s product or service,
certain employee responses diminish the workplace
climate (Tagiuri & Litwin, 1968).
In this exploratory, qualitative study, we attempted
to obtain and integrate expert opinions about the pre-
dictive qualities (if any) of organizational leadership
styles for passive-aggressive (PA) behaviors in the
workplace. Among the central research questions asked
were, Is the autocratic leadership style an antecedent
to this cluster of behaviors in typically productive
employees? and Does the transformational style of
leading predict fewer incidences of PA behaviors in
the workplace?
Definitions
Following are definitions of key concepts discussed
in this article: autocratic, transactional, and transfor-
mational leadership styles and PA behaviors.
Leadership styles are defined as “a pattern of
emphases, indexed by the frequency or intensity
of specific leadership behaviors or attitudes, which a
leader places on the different leadership functions”
(Casimir, 2001). The three leadership styles used in
this research are autocratic, transactional, and trans-
formational. Although categorization is restricting, it
is also convenient. Instead of falling in distinct cate-
gories, leadership styles exist on a continuum. This
continuum reflects the range of styles, not the fre-
quency with which they are exhibited. Most leaders
use combinations of skill types depending on the situa-
tion, context, and the styles of those within their spheres
of influence. Nonetheless, each exhibits a predominant
style. It is those dominating styles that we addressed
in this research and used for the sake of comparison.
The autocratic style falls at one end of the leadership
style continuum. It is sometimes depicted as coercive

Johnson, Klee / Passive-Aggressive Behavior 131
in leadership literature, with coercion running “counter
to working with followers to achieve a common goal”
(Northouse, 2001, p. 8). This style aims to engender
obedience in those working for an organization to
comply with and conform to the directives of the
leader. The autocratic style does not provide ethical
inspection, review, or input by subordinates and ulti-
mately provokes negative subordinate reaction
(Greenleaf, 1978).
With a task-oriented style, a transactional leader
occupies the middle area of the continuum. Leaders
with this style largely react to the performance of their
employees and reward for compliance to their direc-
tives, a factor Northouse (2001) called “contingent
reward” (p. 140). Transactional leaders do not address
employees’ needs, motivations, or development. In gen-
eral, their focus does not include intangibles, such as
goodwill, because the influence of similar intangibles
may be subtle, future oriented, or not easily calculated.
Organizational positions are defined according to tasks
to be fulfilled, and employees are evaluated on the
same. Workaholic patterns are modeled and rewarded
(Graham, 1995, p. 47). Transactional leaders “focus on
the exchanges that occur between leaders and their
followers” (Northouse, 2001, p. 132) and use “corrective
criticism, negative feedback, and negative reinforce-
ment” (p. 140) actively and passively.
The transformational (Bass, 1985; Burns, 1978)
style occupies the opposite end of the leadership con-
tinuum from the autocratic style. Its characteristics
overlap the participative and servant (Greenleaf, 1977)
styles but can be distinguished by its motivating power.
Within the continuum of styles, a transformational
leader may swing from “charismatic” to “individual-
ized consideration” (Northouse, 2001). We focused on
the latter, which shows a leader approaching employ-
ees in a caring way, coaching each to develop his or
her capabilities and to grow intellectually and profes-
sionally. The transformational leadership style over-
laps with the servant leadership style in its attempts to
share knowledge and power and to recognize the
“have-nots...as equal stakeholders in the life of the
organization” (Northouse, 2001, p. 257).
The following three widely accepted constructs
explain PA behaviors: the Diagnostic and Statistical
Manual of Mental Disorders (4th ed.) (American
Psychiatric Association, 1994) categorical construct
(previously defined), Millon and Davis’s (1996) multi-
dimensional traits, and McCrae and Costa’s (1987)
five-factor taxonomic trait model. First, the Diagnostic
and Statistical Manual of Mental Disorders requires
that a person with PA (negativistic) personality disorder
exhibit four or more of the following seven criteria
beginning in “early adulthood and in a variety of
contexts” (American Psychiatric Association, 1994,
pp. 734-735):
1. passively resists fulfilling routine social and occu-
pational tasks;
2. complains of being misunderstood and unappreci-
ated by others;
3. is sullen and argumentative;
4. unreasonably criticizes and scorns authority;
5. expresses envy and resentment toward those appar-
ently more fortunate;
6. voices exaggerated and persistent complaints of
personal misfortune; and
7. alternates between hostile defiance and contrition.
Second, Millon and Davis (1996) proposed that
their negativistic descriptor is broader, encompassing
the “total pattern” (Millon, Davis, Millon, Escovar, &
Meagher, 2000, p. 472). They specified four types
of negativists: vacillating (which adds borderline
components to PA behavior), discontented (which adds
depressive components to PA behavior), abrasive
(which adds sadistic components to PA behavior), and
circuitous (which adds passive-dependent components
to PA behavior).
Third, the five-factor model has assumed many
forms since its inception in 1932 by McDougall and its
subsequent validating factor analysis by Thurstone
(Digman, 2002, p. 17). By 1994, Costa and Widiger
characterized the current five factors (or dimensions) as
extraversion or surgency, agreeableness, conscientious-
ness, emotional stability or neuroticism, and openness
to experience or intellect (Digman, 2002).
For the purposes of this research, which did not focus
on the personality disorder construct, the PA cluster
includes the following behaviors that convey aggressive
feelings through passive means: verbal indirectness, ver-
bal passivity, indirect and physically passive behaviors,
action avoidance, blame avoidance, change avoidance,
resistance, “passing the buck, playing dumb, over-
conforming, depersonalizing, smoothing and stretching,
stalling, playing safe, justifying, scapegoating, misrep-
resenting, escalating commitment, resisting change,
protecting turf” (Ashforth & Lee, 1990, p. 621),
obstructionism, passive deceit (Wetzler & Morey, 1999),
and “negativism” (Millon, 1993, p. 78).
Review of Literature
Most extant literature about organizational passive-
aggressiveness details the effects it has on workplaces,

132 Journal of Leadership & Organizational Studies
company productivity, and coworkers (Kantor, 1997;
Tracey, 2004). Although some researchers contend
that workplace incongruity can be productive (Stohl
& Cheney, 2001), others (Katz & Kahn, 1966) claim
that it precipitates employee “burnout and stress”
(pp. 191-192). The prevalent bottom-up view of the
effect of an employee’s PA behavior on the organiza-
tion ignores the antecedents of the PA cluster and its
effects on the workplace (Baron & Neuman, 1998;
Ferris et al., 1996; Geddes & Baron, 1997, Kacmar &
Baron, 1999).
Although little research focuses on organizational
antecedents (e.g., leadership styles) that may foster
PA behaviors in workers, a few theorists see forces
operating top down from the workplace climate to
employees (Baron & Neuman, 1998; Kantor, 1997)
and have proposed characteristics of organizational
climates that precipitate or predict PA behavior
(Ashforth & Lee, 1990; Hoffmann, 1994).
First, Ferris et al. (1996) tested a model that linked
employees’ perceptions of organizational politics with
various predictors, moderating effects, and mediating
effects, using the results from 822 questionnaires
administered to nonacademic employees of a large
university. This quantitative research negatively corre-
lated bureaucratic and heavily layered organizations
with employees’ perceptions, senses of control, and
understanding. The response rate was low (29%), but
the study offered a good cross-sampling of race,
gender, job positions, and salary grades. The authors
performed ttests and multiple regressions with their
data from six measures, correlating organizational
centralization positively with organization politics, the
formalization of policies and procedures negatively
with organization politics, hierarchical level nega-
tively with politics, advancement opportunity nega-
tively with politics, female gender negatively with
politics, and age negatively with politics.
Fedor, Ferris, Harrell-Cook, and Russ (1998)
reviewed and assessed literature on the multidimen-
sionality of perceptions of organizational politics. Their
purpose was twofold: to assess the one-, three-, and
five-factor models that had been proposed to that date
(including the previous study by Ferris et al., 1996) and
to examine the predictability of organizational and per-
sonal factors in political perceptions (pp. 1760-1761).
This study differed from others in its inclusion of data
from 282 organizations through questionnaires sent to
975 members of the Society for Human Resource
Management, each from a different organization
(p. 1773). Fedor et al. found that the five-factor model
of perceptions underlying organizational politics pro-
duced the best fit (p. 1785). The five dimensions under-
lying employees’ perceptions of organizational politics
reflected how employees perceived (a) the presence of
coercive others, (b) that how one dealt with uncertainty
rather than hard work led to advancement in the orga-
nization, (c) the distortion of information that bene-
fitted a few individuals, (d) that supervisors protected
and enhanced their images through veiled comments, and
(e) that policies regulating pay and promotion were not
clear (p. 1785).
The focus of a study by O’Connor and Morrison
(2001) was to determine and describe employee pre-
dictors to organizational politics. The authors used
seven measures to evaluate 501 surveys returned by
the Royal Canadian Mounted Police and then evalu-
ated the measures using factor analysis and the χ2/df
ratio. The results showed that 52% of the variance in
respondents’ perceptions of organizational politics
could be predicted by organizational climate, formal-
ization, internal and external loci of control, and
Machiavellianism. The authors substantiated their
hypothesis that workers’ perceptions of organiza-
tional politics were negatively correlated with situa-
tional variables, such as job autonomy, level in the
hierarchy, departmental formalization, and favorable
organizational climate. In addition, workers’ percep-
tions of organizational politics were positively corre-
lated with “dispositional” variables, such as external
loci of control (being at the mercy of others) and
Machiavellianism (“cynical beliefs about human
nature, morality, and the acceptability of various
manipulative tactics to satisfy one’s goals”; p. 304).
Kiewitz, Hochwarter, Ferris, and Castro (2002)
examined employees’ perceptions of organizational
politics, with a similar focus to that of O’Connor and
Morrison (2001). However, Kiewitz et al. surveyed
131 restaurant employees to determine whether the
psychological climate (the perceptions and interpreta-
tions of the work environment) of the organization
moderated the relationship between politics and work
outcomes in the organization. Kiewitz et al. defined
psychological climate as composed of six elements:
“supportive management, role clarity, contribution,
recognition, self-expression, and challenge” (p. 1189).
The authors’ confirmatory factor analysis strongly
supported one of their hypotheses, that employee
commitment is predicted by the interaction between
political perceptions and each of the psychological
climate elements except challenge (p. 1200). That is,
when an employee perceives strong organizational

support, he or she perceives lower levels of destruc-
tive organizational politics and vice versa.
Parker et al. (2003) performed a meta-analysis of
121 independent samples from 94 studies of employ-
ees’ perceptions of their work “outcomes, such as
employee attitudes, psychological well-being, moti-
vation, and performance” (p. 389). Because the
authors found no quantitative analyses of the rela-
tionships between individual perceptions and work
outcomes or between individual perceptions and
organizational work outcomes, they used structural
equation modeling to examine competing theories.
The results revealed that individual psychological
well-being has the strongest relationship with one’s
perceptions of job and leader, whereas individual
work attitudes related most strongly to psychological
climate dimensions of workers’ leaders, work groups,
and organizations (p. 405).
Fox and Spector (1999) assessed the antecedents of
counterproductive work behaviors. The authors used
four scales and one survey to study 186 employees
from eight corporations. Their data analysis involved
structural equation modeling and zero-order correla-
tional analysis to evaluate three antecedents: situation,
disposition, and affect. They found a positive relation-
ship between employees’ perceptions of situational
constraints and counterproductive behavior (pp. 923-
924). Although PA behavior was not named as one of
the counterproductive behaviors, these data pointed to
related behaviors (employee and organizational
aggression) as consequences of situational constraints.
Fox and Spector also found relationships “between
behavioral responses and frustration (r =0.35,
p <0.001)” and between situational constraints and
behavioral responses (r =0.36, p <0.001)” (p. 926).
Method
Design of the Study
Because this study’s participants and design were
significantly different from those of extant literature,
it was appropriately a quantitative, phenomenological
study whose results are meant to lay the groundwork
for subsequent research about leadership styles and
PA behavior in organizations. The method consisted
of the lead researcher conducting individual inter-
views using semistructured questions with each con-
senting participant.
Pilot interviews were conducted with three Cau-
casian men, each of whom had a PhD and consulted
for organizations. The first pilot interview was con-
ducted with a clinical psychologist who had been an
organizational vice president, had consulted for orga-
nizations for over 20 years, had taught various psy-
chology courses in academic graduate programs, and
had published widely in the field of organizational
dynamics. He suggested no changes to the interview
questions. The second interviewee was a Caucasian
man who had a doctorate in organizational psychol-
ogy, directed a consulting firm, and had written arti-
cles about leadership and human resources. This
interviewee suggested that the researcher clarify her
definitions of frequently used terms (e.g., passive-
aggressiveness) with each participant before the inter-
view. The third pilot interview was conducted with
the principal of a research consulting organization.
Also a Caucasian man with over 15 years of experi-
ence in business and academia, this interviewee had a
doctorate in social theory. His suggestion for change
was similar to that of the second interviewee:
Introduce and discuss the protocol’s key terms and
definitions to participants before the interviews.
Participants
A nonrandom, purposive sampling (Heppner &
Heppner, 2004) of 13 qualified individuals was con-
ducted. Initial interview participants were obtained
from Internet searches of professional organizations
and universities. All participants were initially con-
tacted by telephone. The researcher then met with all
who agreed to the interview process. At the end of
each interview, the investigator asked participants if
they could recommend similarly qualified and poten-
tially interested colleagues for the same interview.
Through this nonparametric snowball sampling tech-
nique, the remaining participants were obtained. The
snowball sampling technique is used when partici-
pants with specific qualifications are difficult to find.
Data Collection
The insights and narratives of the participants were
collected individually using semistructured, face-to-
face interviews ranging from 45 to 75 minutes in
length. All interviews used the same 13 questions
(listed below), inquiring about the participants’ know-
ledge of PA behavior in organizations, their experience
of it in the workplace, and whether and/or how they
saw a relationship between PA behavior and leader-
ship styles. The depth and detail of a participant’s
responses determined whether the interviewer
prompted the participant for more information with
Johnson, Klee / Passive-Aggressive Behavior 133

134 Journal of Leadership & Organizational Studies
preformed follow-up questions. Whenever an intervie-
wee’s responses warranted prompting, the interviewer
asked one or more of the italicized questions following
each main question:
1. Do you have experience with the three styles
depicted in the handout? If so, please describe that
experience. Are there other leadership styles that you
see more often or that you think are more prevalent?
If so, what are they? Why do you perceive them as
more prevalent or dominant?
2. Have you witnessed, experienced, or worked with
passive-aggressive behavior in workplaces? If so,
please describe your experience in detail, using
examples if possible. Would you characterize your
experience with passive-aggressive behavior as
minimal, moderate, or extensive?
3. Does your perception of passive-aggressive behavior
agree with those in the handout? If so or if not, in
what ways? What behaviors in the handout would
you add, eliminate, or change? Can you say why?
4. Do you think passive-aggressive behavior is exhib-
ited more often under certain circumstances than
others? If so, please describe those circumstances
in detail. Are you aware of passive-aggressive
behavior as being manifested more frequently at a
specific organizational level (e.g., employee, man-
ager, leader, officer)? If so, please specify the level.
5. What do you perceive as the causes of passive-
aggressive behavior in the workplace? How have
you arrived at that judgment? In your organiza-
tional experience, what perceptions brought you to
those causes?
6. What are the effects or costs of passive-aggressive
behavior to the workplace? To the recipients of the
behavior? How have you seen this manifested in
the organization? Were the perpetrators and recip-
ients aware of these effects or costs?
7. What are the effects or costs of passive-aggressive
behavior to the perpetrator? How have you experi-
enced this in the organization? Can you give
examples?
8. What have you seen as some advantages of passive-
aggressive behavior in the workplace to the perpe-
trator? If you saw some, how were these advantages
used or expressed? Can you cite examples?
9. Do you suspect there are antecedents or predictors of
passive-aggressive behavior in the workplace? If so,
please describe them in detail. In your experience
in organizational change consulting, how have you
determined these factors to be antecedents or predic-
tors? Are they components of leadership constructs?
10. Do you think any of the following factors contribute
to passive-aggressive responses in the workplace? If
so, how? Could you explain why you feel any of the
following are factors? (a) Number and/or position
of women in the workplace; (b) change in commu-
nications or way messages and directives are con-
veyed within organizations; (c) change in personal
communication styles; (d) change in number of
workers in the workplace; (e) change in organiza-
tional structure (formal or informal); (f) change in
group processes (e.g., team work, group communi-
cation, etc.); (g) layoffs, firings, consolidations,
takeovers, etc.; (h) ageism; (i) interracial relations;
(j) sexism; and (k) anything else.
11. Do you see a connection between leadership style
and passive-aggressive behavior in the workplace?
If so, what is that connection and how does it work?
If so, did you perceive this connection before our
interview or as a result of these questions? If not,
please explain why not.
12. Could certain leadership styles predict passive-
aggressive behavior? If so, what style(s) do you
think predict(s) PA behavior and how? If not, what
do you perceive to be antecedents to passive-
aggressive behavior? If you do not recognize
antecedents to PA behavior, do you see all PA
behavior as endogenous?
13. Do leadership styles cause, provoke, or encourage
PA behavior? If so, please explain why you believe
this is true. If so, do you differentiate between
styles being antecedents and predictors? If you do
not see leadership styles as fostering PA behavior,
please explain your view.
After the interviews, each interviewee was assigned
a number, which was used in lieu of the participant’s
name. Replacing the participant’s name with a number
ensured confidentiality for the interviewees and
increased the analyst’s impartiality during data sort-
ing, definition, and analysis. The audiotaped inter-
views were transcribed by the researcher and sent to
each participant for review of accuracy. In addition,
the responses were evaluated and reviewed for dis-
crepancies or incompleteness.
“Debriefing” involved sending the participants
copies of their transcribed interviews and requesting
their feedback or any needed changes. After the data
were analyzed and summarized, the findings were
sent to the participants with a request for their feedback
if they saw a need. Participants were also requested
to supply explanations for any suggested changes.
Detailed field notes were maintained to maximize
reliability of the data. Each audiotaped session was
actively searched for discrepant data because poten-
tial bias is built into the interview process, but none
was found.

Data Analysis
The analytic process was based on evaluation of
the data and iterative reviews, sorting, and catego-
rizations that characterize the phenomenological
approach. Because literature on the predictor effects
of top-down flow from leadership style to employee
PA behavior does not exist, the phenomenological
approach lays the groundwork for defining, sorting,
and populating categories, which will eventually con-
tribute to the formulation of a theory. With themes
established, future qualitative and empirical research
will ensue.
The data were analyzed in six steps, similar to
those proposed by Colaizzi in 1978 (Heppner &
Heppner, 2004, pp. 174-175), and the coding proce-
dure for data analysis was specified by Heppner and
Heppner (2004, p. 152).
The first step involved the researcher’s reading
each interview several times to gain an overall sense
of each interviewee’s opinions and ideas. The second
step involved analyzing the raw data further by identi-
fying and extracting key responses from each partici-
pant (e.g., “I think when someone changes his or her
personal way of communicating, it could be a factor in
contributing to PA behavior”) for all questions in the
interview. The third step involved collecting and list-
ing the key responses (by participant number) under
each interview question. In this new list, key words
and phrases within each response were highlighted.
In the fourth step, a Microsoft Excel spreadsheet
was created for each interview question. Each key
word and/or phrase became a data unit, for example,
Question 10b (“change in personal communications”).
Each data unit was then recorded in its own row, with
adjacent columns noting the page number of a partici-
pant’s interview on which the citation (data unit) was
located. Whenever follow-up questions (those in italics
after the main questions) were asked, responses were
included in the spreadsheet with the main question,
even if they were not answered by all respondents. A
total of 346 data units were derived from this first set
of Excel spreadsheets that categorized all data units
within all questions.
Step 5 involved identifying “emergent themes.”
When a question yielded agreeing responses from a
majority of participants (7 of 13), it was classified as an
emergent theme and moved to a second, separate set of
Excel spreadsheets that tracked the themes, associated
data units, and participant responses (Heppner &
Heppner, 2004, pp. 316-318). If fewer than 7 intervie-
wees expressed similar thoughts on an issue, it was not
considered an emerging theme. Each theme was
labeled as succinctly and accurately as possible to
maintain the interviewees’ ideas while showing their
participation in a common idea. Eight themes and 246
data units survived from the original data. The data
units from Questions 1, 2, and 3 were eliminated from
the second group of Excel spreadsheets (themes and
data units) because they were “warm-up” questions
that established participants’ experience with PA
behavior and the three leadership styles. Themes were
grouped by closeness in idea or separated because of
the importance the participants placed on different
words within a question. To illustrate, the second
theme (PA origins and causes and its three subcate-
gories) included two questions: Question 5 and a
follow-up to Question 12. And some questions in the
interview yielded two or more themes. For example,
Question 6 yielded two main themes (“PA effects,
costs to workplace” and “PA effects, costs to recipi-
ents”), each with multiple, associated data units.
In the sixth and final step, the theme results were
sent to all interviewees. The study participants were
asked to review them for congruence with their own
experiences and suggest any changes, additions, or
deletions. None were found.
Validity and Reliability
To maintain research validity, the interviewees’
words were used verbatim while creating data units and
themes during analysis. Exact quotations were tracked
and used to reduce the possibility of contamination.
Because this phenomenological study used the
snowball technique of sampling, reliability was upper-
most in the researcher’s mind. To maximize reliability,
interviewees who possessed understanding of the
concepts in question and considerable experience in
dealing with them were sought. Both academic and
experiential knowledge were important requirements.
Investigator’s Biases
Having experienced and observed PA behavior from
so many angles and in so many contexts, the researcher
is biased against its destructive effects in any organiza-
tional setting. She is also aware that destructive PA
behavior is different from the occasional PA responses
every human being may use intermittently. This inves-
tigator has seen work environments and groups deteri-
orate and become unbalanced under the inappropriate
manipulation of PA leaders, managers, and employees.
To counteract her bias, the investigator made a
commitment to not engage in agreeing or disagreeing
Johnson, Klee / Passive-Aggressive Behavior 135

136 Journal of Leadership & Organizational Studies
with the participants. To help maintain objectivity, the
interviewer responded to interviewees’ statements
with neutral comments, such as “I understand what
you’re saying” or “I see your point of view.” These
types of responses served to restrain the interviewer’s
judgment or opinions while allowing the respondents
to feel understood. The researcher also worked to
counteract her bias while culling and categorizing
data during the analysis phase by first questioning her
choice of each data unit by looking at whether it
answered the interview question, not whether it rep-
resented her point of view. Second, the investigator
reviewed each data unit as an isolated response so
as not to overlook it as a redundant response by any
participant or not to exclude a response because it
conflicted with a prior view expressed by the same
participant. In other words, the researcher did not
exclude responses because they appeared contradic-
tory to previous views expressed by a participant.
Results
During the study interviews, 13 participants shared
their ideas and feelings about PA behavior in the
workplace, three leadership styles (autocratic, transac-
tional, and transformational), and the possibility of a
relationship between leadership styles and PA behav-
ior. Participants in this research had worked profes-
sionally for an average of 21.2 years (range =14 to 35
years). All were based in and around the Philadelphia
area, with many traveling extensively within the
United States to work with clients. At the time of
interviewing, most participants were self-employed as
management or organizational consultants. Of the
exceptions to the consultant rule, one was the director
of program development in a trade school. Another
was an attorney who owned an investment firm and
chaired a reform commission of a major city school
district. The only participant without a graduate degree
managed a methadone clinic. Participant characteristics
are listed in Table 1.
Without exception, all participants expressed having
had experience with autocratic, transactional, and trans-
formational leadership styles that were defined in the
handouts given by the interviewer to the participants.
Two participants stated that autocratic leadership was
most prevalent. Five participants viewed transactional
leadership as the most prevalent in current workplaces.
One participant said that a combination of styles was
prevalent, and one participant did not know. Two par-
ticipants reported that all styles were prevalent.
The second interview question inquired about the
participants’ experience in witnessing or working with
PA behavior in the workplace. All participants reported
having had either moderate or extensive experience
with PA behavior in the workplace. From these inter-
views, 346 original data units of meaningful words and
phrases were derived.
During the coalescing of themes, associated data
units were reduced to 246. Table 2 details the eight
emergent themes (with subcategories), the numbers
of participants citing the themes, and the number of total
data units within each theme.
Discussion
The following eight themes or findings build on
one another in a stepwise fashion.
Theme 1: Circumstances Under Which PA
Occurs Most Often
The findings of this first theme reflect those of
Ashforth and Lee’s (1990) study, which linked organi-
zational stressors with bureaucratic forces and individ-
ual reactions. Their meta-analysis focused on “threat”
and “powerlessness” and how they combine with “for-
malization” and “specialization” to yield defensive reac-
tions, such as “over-conformance,” “playing dumb,”
“stalling,” and “resisting change” (p. 624). This study’s
participants also suggested a correlation between threat
and powerlessness and the costs of PA behavior.
For example, respondents noted circumstances such
as when a person “doesn’t have the power” or a power
struggle is likely to invoke PA behavior. Respondents
also cited threatening circumstances, such as “working
Table 1
Description of Participants
Variable n
Highest degree earned
BA 1 (8%)
MA or MS 2 (15%)
PhD 7 (54%)
EdD 1 (8%)
PsyD 1 (8%)
JD 1 (8%)
Sex
Female 4 (31%)
Male 9 (69%)
Race
African American 1 (8%)
Caucasian 12 (92%)

in a state of confusion” or “pressure,” that would
increase the likelihood of PA behaviors being exhib-
ited. Through these examples, the study participants
concurred with Ashforth and Lee’s (1990) findings that
organizational stressors (e.g., “mixed messages” and
“organizational change”) are the fuel that ignite mate-
rials such as “bureaucratic forces” and “individual
reactions” to create a bonfire of PA behavior. Both
studies agree that PA behaviors erupt when several
components are in place. Without stressors such as
“mergers” and “autocratic workplaces,” PA behaviors
would appear less frequently. Similarly, without the
formalization and specialization of bureaucratic orga-
nizations, another component in the formula would be
missing, decreasing the incidence of PA behavior.
Theme 2: Causes of PA Behavior
in the Workplace
Ferris et al. (1996) proposed that employees’ per-
ceptions of organizational politics are linked to two
moderating effects: employee control and under-
standing of organizational politics. When queried
about causes of PA behavior, participants in this study
extended this link by citing “loss of control” as an
exogenous cause of PA behavior. One also suggested
that feeling out of control would moderate one’s view
of organization politics.
Kantor (1997) proposed that an individual’s
endogenous qualities can merge with exogenous fac-
tors (e.g., hostile work environments) to inhibit pro-
ductive functioning. Twelve respondents agreed that
PA behavior is a mixture of endogenous and exoge-
nous origins. Only 1 respondent voiced strong feel-
ings about the two sources of PA behavior as distinct
origins. In agreeing that PA behavior is a combination
of endogenous and exogenous behavior, 12 intervie-
wees acknowledged that both types had to be present
for a person to behave passive-aggressively. As such,
these respondents support Kantor’s notion that the
interplay between endogenous qualities or traits and
exogenous factors can bear on and alter organiza-
tional climates, morale, and productivity.
To a manager or any troubleshooter within an
organization who must resolve problems, knowing
whether behavior is largely endogenous or exogenous
can point to the source of a problem and its solution.
For example, if an employee’s behavior is reported as
blocking the progress of a project, the person in
charge must figure out where the problem originates,
sometimes without much cooperation from the
involved players. One way to find out is to consult
psychological assessment results (if previously
administered) to determine whether the person had
been assessed as PA. If the behavior were determined
to be atypical for the individual, the project leader
could use approaches with the individual that proba-
bly would not work with someone who routinely
behaved passive-aggressively.
Theme 3: Effects or Costs of PA Behavior
to the Workplace and Recipients
The relationship between stress and PA behavior in
the workplace was suggested both overtly and indi-
rectly by several participants in this study supporting
Kantor’s (1997, 2002) theory that stress in the work-
place contributes to certain PA behaviors. Specifically,
participants cited costs of PA behavior as “lack of
Johnson, Klee / Passive-Aggressive Behavior 137
Table 2
Themes
Participants Data Units
Citing Within
Theme Theme Theme
Circumstances when PA
behavior occurs 12 17
PA behavior origins/causes
Endogenous 7 10
Exogenous 10 20
Not all endogenous/both 10 4
PA behavior costs, deleterious
effects
To workplace 13 28
To recipients 11 16
PA behavior costs, advantages
to perpetrators
Costs to perpetrators 13 21
Advantages to perpetrators 12 29
PA behavior awareness 10 6
PA behavior antecedents, predictors
Antecedents 3 4
Predictors 11 23
Contributing factors
Not number and position of women 7 4
Change in organizational
communication 13 1
Change in personal communication 13 1
Change in number of workers 7 4
Change in organizational structure 13 1
Change in group processes 12 2
Layoffs, firings, takeovers 12 2
Ageism 8 3
Interracial relations 8 5
Sexism 11 2
Other 14 11
Styles connect to, predict, encourage,
cause PA behavior 41 22
Note: PA =passive-aggressive.

harmony”; “culture of negativity”; “frustration, tension”;
“dysfunction”; “doubt, fear, anxiety”; “change”;
“lay-offs and firings”; and “feeling attacked.” All are
specific examples of stress-inducing conditions. In fact,
stress is such an omnipresent concept and topic of dis-
cussion that most of the negative effects offered by the
study’s participants could easily fall under the rubric
of stress. The presence of these factors should precipitate
an investigation into the cause of the stress.
All 13 participants cited unequivocally negative
repercussions of PA behavior on productivity, inter-
personal relations, and organizational climate. The
combined readiness with which respondents cited
examples and the stress on negative consequences
revealed a common, disturbed reaction to the memo-
ries of PA behavior in the workplace, despite or
because of the respondents’ understanding of them
and their dynamics. In other words, the participants
viewed PA behavior in the workplace as serious and
sometimes dire: “career-stopper,” “good people will
leave,” “loss of morale,” and “reduces creativity.”
Once PA behaviors find a safe place to reside, they
are not easily removed. Nothing short of a thorough
examination of the causes and factors that allow them
to thrive and an active and inclusive resolution will
expel them from the environment. PA behaviors are
also resistant to change because they have a large
emotional component; the endurance of emotions is
no mystery. A leader who finds himself or herself
consistently confronting this behavior cluster must
be persistent, creative, and systemic in planning and
enacting a resolution. Unfortunately, few managers
have the resources of time and money to follow
through. By not rooting out their causes, the dynamics
of this cluster are prone to multiply and thrive as they
foster more of their own. When non-PA individuals
watch the behaviors multiply, they grow demoralized
and lose hope. Soon, the spread of PA behaviors is
systemic and very costly to eradicate.
Theme 4: Costs and Advantages
to the Perpetrator
This study offers anecdotal confirmation to
Geddes and Baron’s (1997) survey data, which found
that the most common reactions to negative feedback
were passive forms of aggression. Similarly, Fedor
and Maslyn (2002) proposed that the type of power a
manager or supervisor is perceived to have influences
employees’ perceptions negatively or positively about
the organization’s politics. Specifically, employees
who received negative feedback from bosses whom
they viewed as having expert and referent types of
power took more corrective actions to change their
behavior and sought further feedback. In these cases,
employees perceived a lower degree of organizational
politics playing a role in their feedback. Conversely,
employees who received negative feedback from
bosses whom they viewed as having coercive power
avoided further feedback from their bosses and did
not seek to make suggested feedback changes to their
behavior. They saw organizational politics as playing
a more commanding role in their work environment
(p. 279). Although Fedor and Maslyn addressed
neither PA behavior nor leadership styles per se, they
linked perceived management power styles and
employee responses.
Participants also cited employees’ responses to
unbalanced and unconstructive feedback by bosses,
including the avoidance of employee responses, such
as “anger at the boss,” “backed off the challenge,”
“not productive,” exhibiting “parental transference,”
“costs to personal development,” and “slow[ing]
things down.” These examples support the contention
that autocratic, directive (i.e., noninclusive), or coer-
cive leadership styles do not encourage or aid the
change they are intended to initiate.
The implication to workplaces is that negative mes-
sages usually beget negative results. Managers and
leaders (and others in powerful positions) should be
trained to communicate in congruent verbal and
nonverbal ways and to deliver difficult messages in a
noninflammatory way. Unfortunately, congruent
communication is assumed to be mastered by anyone
who secures a position of responsibility. Nothing could
be further from the truth. Delivering bad news in a
palatable way is a honed skill requiring forethought
and delicacy. Organizations are shortsighted in under-
estimating this ability or ignoring its value altogether.
Theme 5: Awareness of PA Behavior
A perpetrator’s and recipient’s awareness of his or
her PA behavior is another area in which no previous
literature exists. Nonetheless, this issue is important
because it addresses the notion that PA behavior is often
an automatic response without concern for its rippling
consequences in the workplace. In addition, it illus-
trates the other side of the coin: Sometimes, PA behav-
iors are used intentionally, as volunteered by three
study participants, two of whom admitted to using it
with purpose themselves. One has used it to control
other’s responses, and one used it in response to a
boss’s aggressive and PA behavior. When PA behaviors
138 Journal of Leadership & Organizational Studies

Johnson, Klee / Passive-Aggressive Behavior 139
are used intentionally, a perpetrator can control the
degree, timing, and extent of the behaviors, thereby
using them in a prudent manner to make a point or to
cope with a perpetually difficult person. In the latter
case, using PA behaviors may reflect an individual’s
efforts to hold onto a job in a hostile environment or to
protect himself or herself.
This study’s participants overwhelmingly opined that
the PA behavior dynamic is not in the awareness of most
perpetrators or recipients, even though its effects are
strongly felt. Only one respondent said the individuals
who behave passive-aggressively are always aware of
what they are doing. This finding suggests that behav-
iors used are not necessarily used consciously. It may
also suggest that behaviors that evoke strong reactions or
emotions may not be consciously analyzed or realized.
In other words, perhaps “loaded” behavior such as the
PA cluster is not responded to logically because we are
busy reacting to it emotionally. All the more reason that
workplaces would benefit from investing energy in eval-
uating employees before hiring them and tracking their
behavior in a consistent and constructive manner (e.g.,
using baseline and follow-up standardized measures).
Theme 6: Antecedents or
Predictors of PA Behavior
Because PA behavior has been so little studied in the
workplace, no antecedents or predictors of it have been
substantiated through research. Only Fedor et al. (1998)
have suggested that further research about organiza-
tional antecedents to job dissatisfaction and perceptions
of political injustice in the workplace may offer a pre-
dictive link between coercive (autocratic) leadership
and dysfunctional behavior in the workplace.
In this theme, participants specifically cited “per-
formance reviews” as predictors of PA behavior,
coinciding with Geddes and Baron’s (1997) survey.
Although the authors’ survey mentioned “negative
feedback,” which encompasses several forms of com-
munication, performance reviews are the formal
mode of feedback that often elicits negative reaction
from employees. The five subthemes of antecedents
or predictors derived from the participants’ responses
showed great range: psychological assessments or
instruments, personality style and traits, specific PA
behaviors (e.g., “no follow-through”), organizational
culture or events, and autocratic behavior. This wide
range of antecedents and predictors indicates two
things. First, antecedents of PA behavior could be
global or specific, inherent or external, intentional or
uncontrolled. Second, predictors or antecedents are
viewed as relatively nebulous and varied. When asked
the question, most respondents cited nonspecific fac-
tors, such as “market down-turn” or “culture of the
organization.” The difficulty this study’s participants
had in honing in on specific antecedents may reflect
investigators’ reticence to research this area because
it is so situationally dependent.
Theme 7: Contributing Factors
Although most of the factors included in this ques-
tion were offered by participants prior to this question,
the investigator had no way of knowing that prior to the
interviews. This question was meant to ensure that par-
ticipants addressed these factors so often mentioned in
the literature as pressing organizational factors. Only 1
of the 10 factors posed to the participants was rejected
as a contributing factor to PA behavior. That rejected
factor was “number and position of women” in the
workplace. Interestingly, “yeas” and “nays” were not
split exactly on lines of participant gender, so we can
only conjecture that a complex mixture of attitudes
entered the decisions of the respondents. In contrast,
sexism was cited as a contributing factor. Without
extensive further inquiry, any possible contradictions
between these votes cannot be determined.
Six factors revolving around change were all consid-
ered contributing factors to PA behavior: change in
organizational communication, change in personal
communication, change in number of workers, change
in organizational structure, change in group processes,
and change in employee status (i.e., firings, lay-offs,
mergers). Of these six factors, five were unanimous or
near unanimous in respondent agreement. The only
factor with a wider split was “change in number of
workers,” which could be grouped with the three
remaining demographics considered contributing fac-
tors (ageism, interracial relations, and sexism) that
also had less dramatic majorities. This finding sug-
gests the participants’ certainty that change in any
process invokes PA behavior. However, demographic
change (though still considered a contributing factor) is
less clear-cut. Demographics such as age, sex, and race
are often emotionally charged subjects with compli-
cated and mixed components. The emotional involve-
ment or attitude toward these factors could account for
the less than unanimous votes.
Two findings within this theme are supported by
Baron and Neuman’s (1998) survey, which found that
most workplace aggression “does not involve direct,

physical assaults” but “typically encompasses rela-
tively subtle (i.e., covert) forms of harm-doing behav-
ior” (p. 446) and results from recent changes, such as
downsizing or increased workforce diversity. Both
are factors cited in this theme as contributing PA
behavior. In addition, cost cutting and organizational
change were cited by Baron and Neuman as two fac-
tors that related to verbal aggression and obstruction-
ism (a PA quality).
Theme 8: Leadership Styles Are Connected
to, Predict, Encourage, or Cause PA
Behavior
The study’s eighth theme supports Graham’s (1995)
link between leadership style and a cluster of behaviors
evoked by each style. Graham focused on three leader-
ship styles (autocratic or coercive leadership, institu-
tional leadership, and transforming or servant
leadership) that correlate closely with the styles posed
in this study (autocratic, transactional, and transforma-
tional). Graham proposed that each style evokes a set
of behaviors: (a) Autocratic leadership appeals to fol-
lowers’ self-interests and seeks dependable task per-
formance; (b) institutional leadership (roughly
equivalent to transactional leadership) evokes work
group collaboration; and (c) transforming (or transfor-
mational) leadership seeks “constructive participation
in organizational governance” (p. 4). The participants
in this study repeatedly supported Graham’s sugges-
tion that leadership styles evoke specific sets of behav-
iors. Twelve of the 13 respondents in the current study
saw leadership styles as connected to and predictive of
PA behavior. Ten were convinced that the autocratic
style predicts PA behavior more frequently than the
other two styles, and 8 participants agreed that leader-
ship styles in general encourage PA behavior.
However, only four participants were willing to
say that the autocratic style causes PA behavior.
Participants emphasized their choices of verbs, usually
differentiating among their choices of encourages,
causes, and predicts and specifying why they may
have said “yes” to one and “no” to others. As the type
of link progressed from “connected to” to “causes,”
fewer respondents agreed, suggesting that all did not
see a causal relationship but did see a connection of a
less directional sequence.
This finding suggests that the potential links (of
whatever nature) between leadership styles and PA
behavior in the workplace should be taken seriously
and its consequences thoroughly scrutinized. The fact
that three participants did not see a causal relationship
between any specific leadership style and PA behavior
is not worrisome. Considering the average participant
education and training, this response is to be expected,
because few individuals familiar with the scientific
method would venture such a connection without evi-
dence to support it. The fact that almost all partici-
pants saw the connection as predictive is significant
and should be applied to management training, orga-
nizational consulting ventures, and academic venues.
The reason this link is often not addressed is because
of the difficulty in persuading autocratic leaders that
their style can be ineffective. The cycle perpetuates
itself: The traits that support an autocratic style are
also the traits that need to be restructured to reduce
unwanted behaviors in the workplace.
Implications for Practice
and Further Research
The discussion and research begun in this study open
the door for several arenas of future research and prac-
tice. To definitively establish a connection between
leadership style and PA behavior, research both in the
workplace and outside the workplace is needed.
Empirical research with generalizable samples
would go far to establish the existence of a connection
and the type of connection (e.g., predictive, causal).
The incidence of PA behavior in the workplace, the
circumstances under which it occurs most often, and
the forms it takes could all be further understood by
pursuing the study of these two components in the
workplace.
Leadership institutes and management organiza-
tions may be interested in sponsoring further research
to improve the productivity and morale of workplaces.
Better understanding the relationship between leader-
ship styles and PA behavior (or other dysfunctional
behaviors) would also benefit organizations such as
the American Management Association and countless
other business supportive associations that exist to
exchange information and improve communication
and processes in the workplace. Every year, new
assessment instruments enter the marketplace for use
by consultants and organizations to evaluate individu-
als at work or before hire, the attitudes at play in an
organization, and the quality of interfacing between
and among business groups. Putting them to work
could result in their own improvement and obtain
valuable workplace data.
Investigation into other counterproductive behavior in
the workplace might also open exploration into worth-
while areas. For example, although violence (physical
140 Journal of Leadership & Organizational Studies

Johnson, Klee / Passive-Aggressive Behavior 141
and sexual) in the workplace has been extensively
explored, its emotional repercussions on other work-
ers, as well as its victims, have been less pursued.
Instances of violence in the workplace are often
explained by the perpetrator as a reaction to per-
ceived negative remarks or feedback or an uncaring
delivery. Although we are unaware of a predictive link
between PA behavior and violent behavior, on the
basis of research (Fedor & Maslyn, 2002; Geddes &
Baron, 1997; this dissertation, 2006) both can be
responses to negative messages.
Last, organizations (profit and nonprofit alike)
would serve their own future functioning by investing
in research of this nature and by opening their work-
places to study and assessment. Such transparency
would yield countless benefits including more open
dialogues about the successes and failures of communi-
cation processes, interpersonal interactions, and man-
agement styles. More important, open doors would lead
to goodwill inside and outside organizations, increased
morale in the workplace, and ultimately greater profits
for any business genuinely involved.
Summary and Conclusion
The preponderance of participant-expressed views
held that autocratic leadership would “encourage” or
“predict” more PA behavior (especially in individuals
who do not normally behave passive-aggressively) than
the transactional or transformational styles of leader-
ship. They also held that coercive, inflexible leaders
who do not listen would foster PA behaviors in the
workplace. Study participants also attributed certain
environments with qualities or events that precede (and
perhaps predict) PA behavior. These circumstances
include organizational and personal mixed messages,
excluding workers from decisions that directly affect
their work, and unplanned or unannounced change
involving the organization’s structure or culture.
The results of this study also extended or supported
many of the extant literature’s theory and research
about the effects of leadership styles in the workplace
and its contributions to PA behavior, suggesting that
behavior never occurs in a vacuum and that all actions
provoke reactions. Interviewees’ responses imply that
even the most thoughtful and evolved individual may
regress in the face of demoralizing and fluctuating cir-
cumstances. Especially in the world of work, where
individuals’ livelihoods, reputations, and personal
meanings are at risk, feeling controlled by unfair,
erratic, and unpredictable others can set a behavioral
spiral into downward motion.
Because individuals in the working world are depen-
dent on the actions and behavior of so many others, one
person’s spiral into PA behaviors can initiate a rippling
flow of negative behaviors that poison any work inter-
face, creating local to global repercussions. Dismissing
the extent of the ramifications of leadership style to PA
behavior is tempting, but the quality of business rela-
tionships ripples through organizations.
References
American Psychiatric Association. (1994). Diagnostic and statis-
tical manual of mental disorders (4th ed.). Washington, DC:
Author.
Argyris, C. (1985). Strategy, change and defensive routines.
Marshfield, MA: Pitman.
Ashforth, B. E., & Lee, R. T. (1990). Defensive behavior in
organizations: A preliminary model. Human Relations,43(7),
621-648.
Baron, R. A., & Neuman, J. H. (1998). Workplace aggression—
The iceberg beneath the tip of workplace violence: Evidence
on its forms, frequency, and targets. Public Administration
Quarterly,21(4), 446-464.
Bass, B. M. (1985). Leadership and performance beyond expec-
tations. New York: Free Press.
Bennis, W. (1987). The four competencies of leadership. School
Library Media Quarterly,15(4), 196-199.
Burns, J. M. (1978). Leadership. New York: Harper & Row.
Casimir, G. (2001). Combinative aspects of leadership style:
The ordering and temporal spacing of leadership behaviors.
The Leadership Quarterly, 12, 245-278.
Digman, J. M. (2002). Historical antecedents of the five-factor
model. In P. T. Costa, Jr., & T. A. Widiger (Eds.), Personality
disorders and the five-factor model of personality (pp. 17-22).
Washington, DC: American Psychological Association.
Fedor, D. B., Ferris, G. R., Harrell-Cook, G., & Russ, G. S.
(1998). The dimensions of politics perceptions and their orga-
nizational and individual predictors. Journal of Applied Social
Psychology,28(19), 1760-1797.
Fedor, D. B., & Maslyn, J. M. (2002). Politics and political behavior:
Where else do we go from here? In F. Yammarino & F. Dansereau
(Eds.), Research in multi-level issues, Vol. 1: The many faces of
multi-level issues (pp. 271-285). New York: Elsevier.
Ferris, G. R., Frink, D. D., Galang, M. C., Zhan, J., Kacmar, K.
M., & Howard, J. L. (1996). Perceptions of organizational
politics: Prediction, stress-related implications, and outcomes.
Human Relations,49, 233-267.
Fox, S., & Spector, P. E. (1999). A model of work frustration-
aggression. Journal of Organizational Behavior,20, 915-931.
Geddes, D., & Baron, R. A. (1997). Workplace aggression as a
consequence of negative performance feedback. Management
Communication Quarterly,10(4), 433-454.
Graham, J. W. (1995). Leadership, moral development, and citizen-
ship behavior. Business Ethics Quarterly,5, 43-54.

142 Journal of Leadership & Organizational Studies
Greenleaf, R. K. (1977). Servant leadership: A journey into the
nature of legitimate power and greatness. New York: Paulist.
Greenleaf, R. K. (1978). The leadership crisis: A message for college
and university faculty. Indianapolis, IN: The Robert K. Greenleaf
Center.
Heppner, P. P., & Heppner, M. J. (2004). Writing and publishing
your thesis, dissertation & research: A guide for students in
the helping professions. Belmont, CA: Brooks/Cole.
Hoffmann, R. M. (1994). Silent rage: Passive-aggressive behav-
ior in organizations. Dissertation Abstracts International,56
(2-B), 1138. (UMI No. 0419-4217)
Johnson, N. J., & Klee, T. (In press). Leadership styles and
passive-aggresive behavior in organizations. Dissertation
Abstracts International.
Kacmar, K. M., & Baron, R. A. (1999). Organizational politics: The
state of the field, links to related processes, and an agenda for
future research. Research in Personnel and Human Resources
Management,17, 1-39.
Kantor, M. (1997). Occupational disorders: A treatment guide for
therapists. Westport, CT: Praeger.
Kantor, M. (2002). Passive-aggression. Westport, CT: Praeger.
Kiewitz, C., Hochwarter, W. A., Ferris, G. K., and Castro, S. L.
(2002). The role of psychological climate in neutralizing the
effects of organizational politics on work outcomes. Journal
of Applied Social Psychology,32, 1189-1207.
Katz, D., & Kahn, R. L. (1966). The social psychology of organi-
zations. New York: John Wiley.
McCrae, R. R., & Costa, P. T., Jr. (1987). Validation of the
five-factor model of personality across instruments and
observers. Journal of Personality & Social Psychology,52(1),
81-90.
Millon, T. (1993). Negativistic (passive-aggressive) personality
disorder. Journal of Personality Disorders,7(1), 78-85.
Millon, T., & Davis, R. (1996). Disorders of personality: DSM-IV
and beyond. Oxford, UK: Wiley.
Millon, T., Davis, R., Millon, C., Escovar, L., & Meagher, S. (2000).
Personality disorders in modern life. New York: John Wiley.
Northouse, P. G. (2001). Leadership: Theory and practice.
Thousand Oaks, CA: Sage.
O’Connor, W. E., & Morrison, T. G. (2001). A comparison of situ-
ational and dispositional predictors of perceptions of organiza-
tional politics. Journal of Psychology,135, 301-312.
Parker, C. P., Baltes, B. B., Young, S. A., Huff, J. W., Altmann, R. A.,
Lacost, H. A., & Roberts, J. E. (2003). Relationships between
psychological climate perceptions and work outcomes: A meta-
analytical review. Journal of Organizational Behavior,24,
389-416.
Stohl, C., & Cheney, G. (2001). Participatory paradoxical prac-
tices: Communication and the dilemmas of organizational
democracy. Management Communication Quarterly,14,
349-407.
Tracey, S. J. (2004). Dialectic contradiction or double bind?
Analyzing and theorizing employee reactions to organizational
tension. Journal of Applied Communications Research,32,
119-146.
Tagiuri, R., & Litwin, G. H. (1968). Organizational climate:
Explorations of a concept. Boston: Harvard University Press.
Wetzler, S., & Morey, L. C. (1999). Passive-aggressive personality
disorder: The demise of a syndrome. Psychiatry,62(1), 49-59.
Nora J. Johnson, MBA, MS, PsyD, has integrated business,
technical, and academic experience and endeavors. After 25 years
in management and consulting, she obtained a doctorate in clini-
cal psychology, blending interests in workplace dynamics and
personality disorders. She is currently completing a post-doctoral
position and conducts workshops on organizational PA behavior.
Thomas Klee, PhD, is a licensed psychologist and associate pro-
fessor in the Department of Professional Psychology at Chestnut
Hill College in Philadelphia, PA. He teaches doctoral courses
in ethics, personality theory, and social psychology. He has been
principle investigator on numerous federal, state, and foundation
funded research grants.