varsling
Whistleblowing in Organizations:
An Examination of Correlates of
Whistleblowing Intentions,
Actions, and Retaliation
Jessica R. Mesmer-Magnus
Chockalingam Viswesvaran
ABSTRACT. Whistleblowing on organizational
wrongdoing is becoming increasingly
prevalent. What
aspects of the person, the context, and the
transgression
relate to whistleblowing intentions and to
actual whistleblowing
on corporate wrongdoing? Which aspects
relate to retaliation against whistleblowers?
Can we draw
conclusions about the whistleblowing process
by assessing
whistleblowing intentions? Meta-analytic
examination
of 193 correlations obtained from 26 samples
(N = 18,781) reveals
differences in the correlates of
whistleblowing intentions and actions.
Stronger relationships
were found between personal, contextual, and
wrongdoing characteristics and whistleblowing
intent
than with actual whistleblowing. Retaliation
might best
be predicted using contextual variables.
Implications for
research and practice are discussed.
KEY WORDS: retaliation, retaliatory acts,
organizational
wrongdoing, organizational justice,
whistleblowing,
whistleblower characteristics, whistleblowing
intentions
Incidents of organizational wrongdoing are
increasingly making headlines (e.g., fraud,
corruption,
and other unethical acts in organizations
like
Enron, WorldCom, Anderson, and Tyco). While
once thought to be rare, broad coverage of
recent
corporate wrongdoing has led to a widely-held
belief
that scandal may be uncovered in virtually
every
business or organization (Anand et al.,
2004). With
greater frequency, reports of wrongdoing are
made
by members close to the inner workings of the
organization (e.g., by employees, board
members or
internal auditors), rather than by external
auditing
agencies. These individuals, often referred
to as
whistle-blowers, risk retaliation both by
their organization
(e.g., via job loss, demotion, decreased
quality of working conditions) and by the
public
(e.g., character assassinations, accusations
of being
merely ‘‘sour grapes’’, spies, or
‘‘squealers’’) in their
efforts to expose perceived immoral or
illegal acts
(Jubb, 1999; Near and Miceli, 1985).
In this manuscript, we examine potential
predictors
and correlates of whistleblowing behavior
and of retaliation against whistleblowers. Specifically,
we employ meta-analytic methodology to
examine the personal and contextual
correlates of
Chockalingam Viswesvaran (Ph.D. University
of Iowa) is a
Professor of Psychology at
His research interests include business
ethics, personnel selection,
and human resource management. He has
published in
Journal of Applied Psychology, Organizational
Behavior and Human Decision Processes, and
Psychological Bulletin. He has served on 5 editorial boards
and as an Associate Editor of the International Journal of
Selection and Assessment. He is an elected fellow of the
Society for Industrial and Organizational
Psychology as well
as Divisions 5 (Measurement) and 14
(Industrial-Organizational
Psychology) of the American Psychological
Association.
Jessica Mesmer-Magnus (Ph.D. 2005,
University) is an Assistant Professor of
Management with the
family conflict/balance, organizational
training program design,
whistleblowing/counterproductive behavior,
and business
ethics. She has published in the Journal of Vocational
Behavior, Journal of Business Ethics and Journal of
Labor Research. She is certified as a Senior Professional in
Human Resources and has worked as a Human
Resources
Manager for a US-based national consulting
firm.
Journal of Business Ethics (2005) 62: 277–297
_ Springer 2005
DOI 10.1007/s10551-005-0849-1
whistleblowing intentions and behaviors, as
well as
the correlates of retaliation against
whistleblowers
in an effort to improve our understanding of
the
whistleblowing process. Our main purpose in
conducting this study is to guide future
research in
efforts to better predict and understand whistleblowing
and retaliation, and to inform practice as
to the aspects of the work environment that
may
foster the potential for effective
whistleblowing. In
the following review, we will (1) provide an
overview of the whistleblowing construct, (2)
review
potential correlates of whistleblowing and of
retaliation against whistleblowers, and (3)
draw a
distinction between an employee’s intent to
blow
the whistle and actually blowing it. We hope
to
inform research as to the feasibility of
drawing
conclusions about whistleblowing from data
based
on intentions rather than actual behaviors.
Whistleblowing defined
Whistleblowing is ‘‘the disclosure by
organization
members (former or current) of illegal,
immoral, or
illegitimate practices under the control of
their
employers, to persons or organizations that
may be
able to effect action.’’ (Near and Miceli,
1985, p. 4).
While whistleblowers typically have both
internal
and external reporting channels available to
report
organizational transgressions, research
suggests that
nearly all whistleblowers initially attempt
to report
wrongdoing via internal channels before
utilizing (or
in lieu of) external channels (Miceli and
Near, 1992,
2002). Even though whistleblowing via
internal
channels is less threatening to an organization
(as
compared with external reporting which
threatens
public scrutiny or legal intervention; Miceli
et al.,
1991a), whistleblowing within an organization
is not
often welcomed. Rather, whistleblower reports
of
wrongdoing are frequently buried or ignored
(Miceli
et al., 1991b). In this scenario, the
whistleblower is
unsuccessful in stopping organizational
wrongdoing,
and worse, is placed in a position to
experience
negative consequences to their action.
Burying or
ignoring reports and retaliation against
whistleblowers
are more likely when whistleblowing is
perceived by top management to represent a
questioning of or challenge to the
organization’s
authority structure (e.g., Miceli and Near,
2002).
Whistleblowing research
Reviews of whistleblowing research have
identified
two relatively robust foci of examination, in
which
researchers have sought to identify the
conditions
under which whistleblowing intentions are
formed
and action taken, and retaliation occurs
(Ellis and
Arieli, 1999; Miceli and Near, 2002). Studies
of the
antecedents, correlates, and consequences of
the
decision to blow the whistle on corporate
wrongdoing,
as well as those focused on modeling its
process,
have typically drawn conceptual distinctions
between
those variables related to the whistleblower
(i.e.,
whistleblower characteristics), the context
in which
the whistleblowing occurs (i.e., contextual
variables),
and aspects of the wrongdoing and wrong-doer
(e.g.,
King. 1997; Miceli et al., 1991a; Miceli and
Near,
1985, 1988; Miceli et al., 1991b). Research
exploring
the role of these variables in the
whistleblowing process
is typically approached in one of two ways:
(1)
conducting surveys of actual whistleblowers,
or (2)
using scenarios, interviews, or survey-based
methods
to ascertain when an observer of
organizational
wrongdoing will be likely to blow the whistle
or will
report the intention to make a claim.
Specifically,
while some authors have been successful in
gaining
access to actual whistleblowers, others have
resorted
to drawing conclusions about the
whistleblowing
process using reports of whistleblowing
intentions. In
the first case, actual whistleblowers have
observed a
transgression, made the decision to report
it, and then
actually followed through in making a claim.
‘‘Likely’’
or ‘‘intended’’ whistleblowers, on the other
hand,
have been given information about a
wrongdoing
(whether real or fictitious) and have
reported either
‘‘how likely’’ they would be to blow the
whistle or
whether they intended blow the whistle at
some future
time. The obvious difference between these
approaches
is that intended whistleblowers have not
actually followed through with blowing the
whistle.
Researchers have justified using data gained
from
intended rather than actual whistleblowers by
(1)
citing the difficulty of carrying out
investigations
into unethical conduct in actual
organizations (e.g.,
Chiu, 2003), (2) suggesting that actual
whistleblowers
censor the information they provide to
investigators due to the perception that data
gathered
in actual organizations precludes their
confidentiality
or anonymity (e.g., Sims and Keenan,
278 Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
1998), (3) illuminating the difficulty of
locating
actual whistleblowers for questioning, or (4)
citing
the inherently flawed nature of such data
(e.g., selfreports
of past events). While we believe these
concerns to be valid, we also believe that
conclusions
drawn based on testimonies of actual
whistleblowers
will be different than those from likely
whistleblowers. The real question, however,
is how
different will these conclusions be.
Specifically, are
the correlates of actual whistleblowing
similar enough
to those of likely/intended whistleblowing to
warrant firm conclusions about the
whistleblowing
process? Or, given the nature of
whistleblowing as a
behavior (rather than an attitude or
intention), is
the ‘‘distance" between whistleblowing
intent and
actual whistleblowing (e.g., deciding to blow
the
whistle, accessing the internal or external
channel,
making a claim of wrongdoing, supplying the
necessary evidence) so great as to void
direct
comparison? Research on the
intention–behavior
relationship suggests only a weak
relationship
between the intention to act and the actual
performance
of a behavior (Miller and Grush, 1988).
In fact, measures of behavioral intentions
have accounted
for, at most, 10% of the variance in overt
behaviors (e.g., Mitchell, 1974). Further,
the
intention–behavior relationship is thought to
be
even weaker for some behaviors (i.e., those
behaviors that may be governed more heavily
by
subjective norms and external support, like
whistleblowing;
e.g., Miller and Grush, 1988; Mitchell,
1974). According to Bagozzi’s (1992)
Volitional
Model of Goal Directed Behaviors, behavioral
intention and overt action are
‘‘separated" by
extensive psychological, motivational, and
implementation
processes (i.e., planning, monitoring
activities, guidance and control,
psychological
commitment, effort). This suggests there may
be
multiple phases involved in moving from an
awareness of an illegitimate practice, to the
decision
to blow the whistle, to actually making a
claim.
Correlates of whistleblowing
Below we review research resulting from both
reports
of whistleblowing intent and whistleblowing
action.
Researchers have examined the same sets of
variables
in studies of intended and actual
whistleblowers (i.e.,
relating to aspects of the whistleblower,
context, and
wrongdoing potentially predictive of
whistleblowing
intention/action). A qualitative review of the
whistleblowing
process yields relatively consistent findings
resulting from each approach. However, a
cursory
examination of their results suggests
stronger relationships
may exist between key variables and
whistleblowing
intentions than whistleblowing actions.
Given that whistleblowing behavior may be
heavily
governed by subjective norms and external
supports, it
stands to reason that the relationship
between intention
and action may be even lower for
whistleblowing
than other actions.Using a meta-analytic
approach, we
examine correlates of whistleblowing intent
and actual
whistleblowing separately so as to lend a
greater
understanding of their common relation to
predictors
and correlates. We hope to identify the
relationship
between whistleblowing intent and
whistleblowing
action, so that researchmaybe better informed
as to the
applicability of data regarding
whistleblowing intentions
to conclusions about whistleblowing actions.
Whistleblower characteristics
A variety of personal characteristics related
to the
decision to engage in whistleblowing have
been
examined: whistleblower demographics (i.e.,
age,
sex, level of education, level of job held,
etc.),
personality variables (i.e., locus of
control), morality
(i.e., ethical judgment), and other
characteristics
(i.e., job performance, organizational
commitment,
role responsibility, approval of
whistleblowing; e.g.,
Brief and Motowidlo, 1986; Near and Miceli,
1996).
While results seem to differ slightly across
studies,
whistleblowers (as compared with inactive
observers)
tend to have good job performance, to be more
highly educated, to hold higher-level or
supervisory
positions, to score higher on tests of moral
reasoning,
and to value whistleblowing in the face of
unethical
behavior (e.g., Brabeck, 1984; Miceli and
Near,
1984; Near and Miceli, 1996; Sims and Keenan,
1998). Also, it appears that whistleblowers
are more
likely (than inactive observers) to report a
rolerelated
responsibility or obligation to blow the
whistle
(e.g., Near and Miceli, 1996). Age and
organizational
tenure as predictors of whistleblowing have
yielded
mixed results (e.g., Near and Miceli, 1996;
Sims and
Keenan, 1998).
Whistleblowing and Retaliation 279
Empirical findings regarding the personal
correlates
of whistleblowing intentions and actions are
typically interpreted through the lens of
social–psychological theories of behavior
(cf. Casal
and Zalkind, 1995; Ellis and Arieli, 1999;
Miceli and
Near, 1994; 2002; Near and Miceli, 1995). For
example, Hollander’s (1958) idiosyncrasy
model
suggests that those individuals who are
considered to
be the best contributors to group and
organizational
objectives (i.e., have high job performance)
will be
given greater leeway to deviate from group or
organizational norms. Specifically, employees
with
good job performance will be more likely to
have
accumulated idiosyncrasy credits
(interpersonal
bonuses) within the organization, thus
providing
some latitude to report wrongdoing without
suffering
retaliation, and more importantly, to effect
the
desired change (e.g., Miceli et al., 1991a).
Similarly,
Pheffer and Salancik’s (1978) Resource
Dependence
Theory posits that when one party possesses
resources
upon which another is dependent, that party
will be
more powerful. Within the whistleblowing
context,
an individual with more experience, tenure,
and
better job performance is more valuable to an
organization,
thus giving them some leverage to report
misdeeds (e.g., Miceli and Near, 2002).
Perceived
leverage may increase whistleblowing
potential and
action. Theories of power relationships
(e.g., French
and Raven, 1959) suggest that individuals
gain (and
exercise) various bases of power by
possessing valued
and not easily replaceable characteristics
(e.g., desired
or unique skills, good job performance, tenure
or
position status, credibility). Thus,
whistleblowers
with better job performance and tenure are
more
likely to be successful in persuading
organizational
actors to stop the undesirable acts, also
increasing the
potential they will report (or intend to report)
a
transgression (e.g., Miceli and Near, 1994).
Organizational employees have three options
to
address an unsatisfactory situation faced
within an
organization: (1) to exit the organization,
(2) voice
discontent (i.e., blow the whistle), or (3)
remain
silent. Employees with greater tenure are
more invested
in the organization and may prefer voice to
exit. This is also congruent with predictions
from
theories of power in organizations, where
employees
with greater tenure may have greater power to
effect
change, and therefore may prefer voice to
exit or
silence (e.g., French and Raven, 1959). In a
similar
vein, individuals demonstrating higher
organizational
commitment are more invested in staying with
the
organization, therefore are more likely to
blow the
whistle rather than exit the organization
(particularly
when the prospect of continued wrongdoing is
uncomfortable or unacceptable). Thus,
theoretical
explanations based on available idiosyncrasy
credits,
power relations, voice-exit choices, and
control
theory are congruent with empirical findings
that
older, high performing, more committed and
more
experienced employees are more likely to
report
wrongdoing in and by organizations.
Contextual variables
Compared with the personal characteristics of
whistleblowers, contextual variables seem to
explain
more variance in an individual’s decision to
blow the
whistle (e.g., Miceli et al., 1991b; Miceli
and Near,
1984; Near and Miceli, 1996). Such contextual
variables may include supervisor and coworker
support, organizational climate, threat of
retaliation,
and size of organization. Research suggests
that
perceptions of supervisor or top management
support
are instrumental in the decision to blow the
whistle, and in the determination of which
reporting
channel will be employed (i.e., internal vs.
external
to the organization; e.g., Keenan, 2000; Sims
and
Keenan, 1998). For instance, predictions from
Social
Exchange Theory (Thibaut and Kelley, 1959)
would
suggest that when there is a high level of
supervisor
and co-worker support, norms of reciprocity
will
develop that channel an individual to use
internal
means to effect behavior change (before
making the
charges public).
Other contextual variables are also worth
pursuing.
For example, whistleblowing seems to be
greater in organizations that value
whistleblowing
and in those in which the whistleblower
perceives a
high congruence between personal and
organizational
values (Berry, 2004; Near and Miceli, 1996).
This is in accordance with Enz’s (1988)
theory of
value congruence. Similarly, potential
whistleblowers
who perceive a threat of retaliation (by the
organization, immediate supervisors, or
co-workers)
are much less likely to blow the whistle than
those
who do not perceive a retaliatory climate
(Keenan,
1995; King, 1999; Near and Miceli, 1996).
280 Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
Characteristics of wrong-doing
Evidence suggests that characteristics of the
wrongdoing
(i.e., perceived severity of the wrongdoing,
evidence of wrongdoing) and/or characteristics
of
the wrong-doer (i.e., likeability of or
closeness to the
wrongdoer) may have significant implications
in the
decision to blow the whistle. Wrongdoing type
and
perceived severity has been found to be
moderately
positively related to whistleblowing (Miceli
and
Near, 1985; Near and Miceli, 1996). There is
also a
tradition of research in whistleblowing that
addresses
whether the wrongdoing is sufficiently costly
to
warrant pursuit. Thus, it has been suggested
that
potential whistleblowers look for
‘‘materiality’’ (as
auditors do) before embarking on any actions.
Similarly, organization members seem to be
more
likely to blow the whistle on organization
wrongdoing
when they have convincing evidence of the
wrongdoing, and when the transgression personally
affects them (Miceli and Near, 1985). The
latter
finding parallels the research on the
‘‘bystander
effect’’ in explaining prosocial behavior:
due to a
diffusion of responsibility, individuals are
less likely
to help when several others are present
(Latane and
Darley, 1970). On the other hand, to the
extent
strong norms of reciprocity develop and
social support
is high between organizational members,
wrongdoing that harms the organization and/or
co-workers is more likely to be reported
(especially
using internal channels). Interestingly, in
organizations
that depend upon the continuation of the
wrongdoing, whistleblowers are more likely to
select
external reporting channels rather than
attempt to
make an internal claim (Miceli and Near,
1985).
This relationship appears to be especially
strong
when the whistleblower is fearful of
retaliation by
the organization, supervisors, or coworkers
(Miceli
and Near, 1985).
Retaliation against whistleblowers
Once an organization member has blown the
whistle
on an organizational wrongdoing, management
may
make two types of decisions: (1) whether to
disregard
the claim or take appropriate action, and (2)
whether to reward or retaliate against the
whistleblower
(Near and Miceli, 1986). It is assumed that a
whistleblower’s experiences (perceived or
actual,
reward or retaliation) following a
whistleblowing
event will have strong effects on others’
willingness
and likelihood to blow the whistle in the
future
(Casal and Zalkind, 1995; Miceli and Near,
1992).
Logically, organizational response to
whistleblower
action depends in part upon whether
management
agrees with the merit of the claim and with
the
whistleblower’s obligation to take action
(Parmerlee
et al., 1982). Under circumstances where an
organization is dependent upon the
continuation of
the wrongdoing or when they are not dependent
upon the whistleblower (e.g., Resource
Dependence
Theory; Pheffer and Salancik, 1978), the
organization is more likely to retaliate
against the
whistleblower and continue the wrongdoing
(Near
and Miceli, 1986).
Retaliation may take many forms, ranging from
attempted coercion of the whistleblower to
withdraw
accusations of wrongdoing to the outright
exclusion of the whistleblower from the
organization
(e.g.,
Parmerlee et al., 1982). Other
retaliatory
acts may include organizational steps taken
to
undermine the complaint process, isolation of
the
whistleblower, character defamation,
imposition of
hardship or disgrace upon the whistleblower,
exclusion from meetings, elimination of
perquisites,
and other forms of discrimination or
harassment
(e.g.,
Parmerlee et al., 1982). Retaliatory
acts may be
motivated by the organization’s desire to (1)
silence
the whistleblower completely, (2) prevent a
full
public knowledge of the complaint, (3) discredit
the
whistleblower, and/or (4) discourage other
potential
whistleblowers from taking action (Miceli and
Near,
1994; Parmerlee et al., 1982).
Retaliation is not always initiated by
organizational
top management. Rather, isolated acts of
retaliation may be initiated by the
whistleblower’s
supervisor or coworkers with or without
(formal or
informal) sanctioning by top management.
Supervisors
may be motivated to retaliate against
whistleblowers
for a variety of reasons, but they
frequently do so out of fear that a
whistleblowing
claim signals their inability to maintain
order and
compliance within their departments, or the
fear that
valid complaints will result in the
restriction or
cessation of their own operations or
influence
(O’Day, 1972; Parmerlee et al., 1982).
Whistleblowing and Retaliation 281
Correlates of retaliation against
whistleblowers
Predictors or correlates of retaliation
against whistleblowers
fall into one of four broad categories: (1)
characteristics of the whistleblower, (2)
actions taken
by the whistleblower in reporting
organizational
wrongdoing, (3) situational or environmental
variables
related to the organization, and (4)
characteristics
of the wrongdoing or wrongdoer.
Characteristics of the whistleblower
Characteristics of the whistleblower examined
in
relation to retaliation include whistleblower
age,
education level, job level, role
responsibility, and
value congruence with the organization. While
demographic characteristics of whistleblowers
are
thought to be less predictive of retaliation
than are
contextual variables (Miceli and Near, 2002),
research
suggests that individuals who blow the
whistle because
it is their job to do so (e.g., audit or role
responsibility) are less likely to be
retaliated against
and are more likely to be successful in
stopping the
transgression (e.g., Casal and Zalkind, 1995;
Miceli
and Near, 2002). Further, Parmerlee and
colleagues
(1982) found preliminary evidence that older
whistleblowers are more likely to be
retaliated against
than are younger whistleblowers.
Interestingly, their
results also suggest that whistleblowers that
are valuable
to their organization (e.g., due to age,
experience,
education, job level) are more likely to be
retaliated
against as compared to less valuable
whistleblowers.
Perhaps, for older individuals and those at
higher job
levels and with more experience, greater
organizational
loyalty is expected. When such individuals
blow the whistle, other organizational
membersmay
feel a greater sense of betrayal, thus paving
the way
for more retaliatory behaviors. This is
especially true
when external channels are employed to report
violations.
Norms of reciprocity and notions of perceived
justice violations (however misguided) appear
to
predict retaliation. Theories of power also
suggest that
whistleblowers at higher job levels, who are
expected
to enforce the power structure, upon
violating this
mandate are more likely to suffer
retaliation. On the
other hand, individuals at lower levels of
the organizational
structure may have lesser power, thus being
easy
targets for retaliation. Finally, evidence
suggests that
whistleblowerswhose values regarding right
andwrong
are not congruent with those of the
organization, are
more likely to be retaliated against (Miceli
and Near,
1994), presumably because organization top
management
does not deem the wrongdoing to be as
severe as is perceived by thewhistleblower,
thus casting
doubt on themerit of the whistleblower’s
complaint.
Actions taken by the whistleblower
Some researchers have examined whether
specific
actions taken by a whistleblower influence
the degree
to which they are retaliated against (e.g.,
whether the whistleblower used an internal or
external channel to report wrongdoing,
whether the
whistleblower attempted to remain anonymous
during the whistleblowing process, how
successful
the whistleblower was in ultimately curbing
the
organizational wrongdoing, and even whether
others in the organization ignored
wrongdoing).
Indeed, research suggests that when
whistleblowers
report wrongdoing via external channels, they
are
more likely to receive retaliation, and such
retaliation
is likely to be more severe than when
internal
channels are utilized (Near and Miceli,
1986).
Starting with Weber (1947) and Barnard
(1938),
organizations have relied on the concept of
legitimate
authority. Thus, the use of external channels
is more likely to elicit retaliatory
behaviors, as it
violates the power and authority structure
present
within the organization. Similarly,
whistleblowers
who unsuccessfully attempted to remain
anonymous
during the whistleblowing process were more
likely
to be retaliated against (Miceli and Near,
1994).
Inconsistent results have been reported
regarding
the effectiveness of the whistleblower in
curbing
wrongdoing and experience of retaliation.
Specifically,
Miceli and Near (2002) report that effective
whistleblowers are less likely to experience
retaliation
unless the supervisor or top management
simultaneously
curbed the wrongdoing and retaliated
simply for blowing the whistle. This suggests
that
even when the transgression is agreed to be
harmful
and is subsequently stopped, some
organizations are
particularly sensitive to violations of the
authority
structure via whistleblowing. This seems to
be
282 Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
especially likely when whistleblowers have
utilized
external channels.
Contextual variables
Context variables examined in relation to
retaliation
include top management, supervisor, and
coworker
support, as well as organizational climate for
whistleblowing.
While lack of support from supervisors
and top management is likely predictive of
retaliation
against whistleblowers, coworker support does
not appear to be related to perceived or
experienced
retaliation (Near and Miceli, 1986). This finding
is
likely due to the low occurrence of
retaliation initiated
by coworkers (e.g., Miceli and Near, 1994).
However, this finding may be indicative of
organizational
norms for treatment of whistleblowers.
Specifically, in organizations where
whistleblowing
is not sanctioned, coworkers are likely less
willing to
offer support or protection from retaliation
to a
whistleblower.
Characteristics of the wrongdoing
Lastly, in addition to the other predictors
of retaliation,
researchers have examined aspects of the
wrongdoing that may influence retaliation
(e.g.,
frequency, severity, and evidence of
wrongdoing).
Logic suggests that when wrongdoing in an
organization
is widespread or when the organization is
dependent upon the continuation of
wrongdoing,
the organization’s top management would be
more
likely to lash out at individuals who blow
the whistle
(Casal and Zalkind, 1995). However, Near and
Miceli (1986) found that a whistleblower’s
report of
(a) multiple incidents of wrongdoing, (b)
multiple
individuals
involved in the wrongdoing, or (c)
multiple
sources of evidence, appear unrelated to
retaliation.
This could be explained in terms of the
effort
needed to retaliate against a whistleblower
with
strong evidence of wrongdoing, particularly
when
the wrongdoing is widespread and impacts
multiple
individuals (Parmerlee et al., 1982).
In
sum, our first purpose in conducting this research
was
to investigate the antecedents and correlates
of
whistleblowing, and to determine whether
these
were different for whistleblowing intent versus
actual
behavior. Our second purpose in conducting
the
present study was to examine likely antecedents
and
correlates of retaliation against whistleblowers,
and
to determine the relative importance of each
variable
in predicting retaliation. To address these
questions,
meta-analytic cumulation of the extant
literature
was employed to summarize and integrate
findings
from individual studies.
Method
Database
One
hundred and ninety-three correlations from 26
samples
reported in 21 articles (total N = 18,781)
examining
whistleblowing (including intent to blow
the
whistle, likelihood of blowing the whistle, and
actual
whistleblowing, both via internal and external
channels)
and retaliation against whistleblowers were
included
in this meta-analysis. To ensure a comprehensive
search,
these studies were located using
the
following strategies: (1) conducting a computerized
search
of the PsycInfo (1887 to present) and
ABI
Inform (1971 to present) databases, using
appropriate
keywords and phrases (e.g., whistleblowing,
blowing
the whistle, organizational wrongdoing,
retaliation, dissent, counterproductive
behavior, corruption),
(2)
conducting a manual search of references cited in
studies
included in this meta-analysis and cited in this
manuscript,
(3) snowballing references cited in
recently
published reviews of the whistleblowing
literature
(e.g., Gundlach et al., 2003; Miceli et al.,
1991b;
Near and Miceli, 1995, 1996), and (4)
soliciting
relevant, but as yet unpublished, research
from
authors at the 2004 and 2005 meetings of the
Society
of Industrial Organizational Psychology.
These
methods yielded an initial 67 journal
articles,
dissertations, and theses with the potential to
provide
usable data. An attempt was made to obtain
all
studies identified. When a manuscript was not
available
via interlibrary loan, a request for a copy
was
sent directly to the study’s primary author.
Studies
were included only if they reported a correlation
(or
an effect size which could be converted
to
a correlation using appropriate conversion formulas)
between
one of the several predictors and
whistleblowing
(intent/likelihood or behavior) or
instances
of retaliation. Studies that examined
Whistleblowing
and Retaliation 283
constructs
different from those of interest (e.g.,
perceptions
of or reactions to unethical behaviors/
incidences
other than those of actual or intended
whistleblowing)
or those that reported only anecdotal
data,
regression coefficients, or other effect
sizes
not representative of a clean relationship
between
whistleblowing intent/action and a relevant
correlate,
were not code-able. When authors
reported
separate correlations for different subgroups
(e.g.,
males and females), samples, administrations (as
in
a longitudinal study), or measures of the same
construct,
those correlations were examined separately.
The
studies included in this meta-analysis are
listed
in the references prefixed with an asterisk. The
studies
included in this meta-analysis, for the most
part,
utilized largely male samples of working adults
(average
ages between mid-1920’s and late-1940’s)
who
have (or could have) witnessed and reported (or
considered
reporting) organizational wrongdoing
(i.e.,
internal auditors, military and government
employees,
accountants). Experimental, scenariobased,
and
survey-based designs were utilized.
Coding
procedure
The
first author made an initial independent effort to
code
the 21 articles selected for inclusion in this
study.
A random subset of these articles was coded
by
the second author in an effort to determine coder
reliability.
Inter-coder agreement was very high,
likely
due to the objective nature of the data coded.
In
reference to whistleblowing intentions and
actions,
the variables coded included those related
to
(1) the characteristics of a whistleblower (e.g.,
demographics,
like age, sex, education, tenure, and
job
level, and other characteristics, like ethical
judgment,
job satisfaction, job performance, role
responsibility
to blow the whistle, and approval of
whistleblowing),
(2) the context in which whistleblowing
takes
place (e.g., organizational climate for
whistleblowing,
fear or threat of retaliation against
whistleblowers,
organizational size, and supervisor
and
co-worker support), and (3) the characteristics of
the
wrongdoing or the wrong-doer (e.g., the seriousness
of
the wrongdoing, the amount of evidence
of
the wrongdoing possessed by the whistleblower,
whether
the whistleblower was effective in stopping
the
wrongdoing, and the degree to which the
whistleblower
works near or closely with the
wrong-doer).
These correlates were coded with respect
to
whether the data was collected from intended
or
likely whistleblowers or from actual
whistleblowers.
Further, the data were coded with
respect
to whether the whistleblower used (or intended
to
use) an internal or external reporting
channel.
If the reporting channel was not specified
or
was mixed, this was coded separately. While we
had
hoped to be able to examine correlations between
whistleblowing
correlates and whistleblowing
intentions
and behaviors by the type of reporting
channel
employed (intended), there was an insufficient
number
of studies available to facilitate this
specificity.
We chose instead to examine the relationships
of
whistleblowing correlates with either
intentions
or actions, regardless of reporting channel.
In
reference to correlates of retaliation against
whistleblowers,
the variables coded included those
related
to the (1) characteristics of a whistleblower
(e.g.,
like age, education, job level, role responsibility
to
blow the whistle, perceived alternative job
opportunities,
and degree to which the whistleblower
perceived
value congruence with the organization),
(2)
actions taken by the whistleblower
in
reporting an organizational wrongdoing (e.g.,
utilizing
an external channel, attempting to remain
anonymous
during the whistleblowing process, and
success
in stopping the wrongdoing), (3) contextual
variables
(e.g., supervisor and co-worker support),
and
(4) characteristics of the transgression (e.g.,
frequency
of wrongdoing in the organization, the
severity
of the transgression, and the amount of
evidence
about the wrongdoing possessed by the
whistleblower).
Analysis
The
meta-analytic methods outlined by Hunter and
Schmidt
(2004) were employed to analyze this data.
The
three outcomes investigated are (1) intent to
blow
the whistle, (2) whistle-blowing behavior, and
(3)
retaliation experienced. Each set of bivariate
correlations
(between a predictor and outcome)
were
analyzed separately. The sample-size weighted
mean
observed correlation and the sample-size
weighted
standard deviation of the observed correlations
were
computed. These two estimated
284
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
parameters
provide a distribution of the observed
effect
sizes. Unfortunately, insufficient information
was
available to correct for unreliability in the
measures.
This concern is mitigated to some extent
by
the ‘‘objective’’ nature of some of the predictors
included
(e.g., gender, tenure). To the extent statistical
artifacts
like unreliability and range restriction
in
the measures affect the results, our reported
findings
underestimate the ‘‘true’’ magnitude of the
correlations
and overestimate the ‘‘true’’ variability
across
studies. Note, however, that our analyses are
conservative
in that true correlations will actually be
higher
and we are more likely to consider potential
moderators
(even when their effects are weak).
The
sampling error variance associated with the
mean
observed correlation was computed (Hunter
and
Schmidt, 2004) and subtracted from the
observed
variance. The resulting residual variance
was
used to compute the confidence intervals
around
the observed mean. The percent of observed
variance
attributable to sampling error was also
computed.
Usually, if 60% or more of the observed
variance
is attributable to sampling error, one can
assume
that variance across studies due to uncorrected
artifacts
like unreliability and range restriction
in
the measures correlated can explain the remaining
variability.
Such a finding suggests that the relationship
generalizes
or transports across contexts.
Thus,
our analyses could be construed to be a barebones
meta-analysis
(Hunter and Schmidt, 2004).
We
report for each meta-analysis, the number of
correlations
included (k), the total sample size
across
all
estimates (N), the sample-size weighted
mean
observed
correlation (r), the sample-size weighted
standard
deviation (SDr), the sampling error
standard
deviation,
the residual standard deviation (resSD),
the
percent of observed variance attributable to
sampling
error, and the 95% confidence interval
around
the sample-size weighted mean observed
correlation.
For studies with a mean observed correlation
of
greater than 0.05, we conducted a file
drawer
analysis (Hunter and Schmidt, 2004), and
provide
a ‘‘file drawer k’’, which represents the
number
of missing studies averaging null results
required
to reduce our reported sample-size weighted
mean
observed correlation to 0.05.
For
the most part, we conducted meta-analyses
only
when a minimum of three studies reporting a
relationship
between whistleblowing intent/action
and
a relevant correlate were available. However,
given
that one of the purposes of this study was to
compare
conclusions drawn from reports of whistleblowing
intentions
with those drawn from reports
of
whistleblowing actions, in cases where fewer than
three
studies were available examining a ‘‘comparison’’
relationship
of interest (i.e., a sufficient number
of
studies were available to meta-analyze the relationship
between
a correlate and either whistleblowing
intention
or whistleblowing action, and less
than
three were available for the other relevant metaanalysis),
we
report the results of the other relationship
for
purposes of description and comparison.
Results
The
results of the meta-analysis of correlations
between
whistleblower characteristics and whistleblowing
intentions
(Intent WB) and actions (Actual
WB)
are presented in Table I. Results indicate that
older
employees are more likely to intend to blow the
whistle
than are younger employees (r = 0.19).
Unfortunately,
no studies examining age and actual
whistleblowing
were located, so we cannot be sure
whether
this relationship translates to actual whistleblowers.
Sex
and tenure appear to be related to actual
whistleblowing
(r = 0.13 and 0.10, respectively), but
not
to whistleblowing intent. Specifically, females
and
more tenured employees appear to be slightly
more
likely to actually blow the whistle. Job level
appears
to have a similar relationship with whistleblowing
intentions
(r = 0.10) as with actual whistleblowing.
These
results support the contention that
older
employees with greater tenure and at higher
levels
are more likely to have the commitment and
power
to employ voice rather than exit mechanisms.
More
interesting than the demographic characteristics
of
whistleblowers were the results related
to
other whistleblower characteristics and intent or
actual
whistleblowing. Specifically, ethical judgment,
while
moderately related to the intent to blow the
whistle
(r = 0.45), is not related to actual whistleblowing
(r
= )0.08). Similarly, while
role responsibility,
approval
of whistleblowing, and perceptions
that
blowing the whistle would be in one’s best
interests,
appear to be predictive of the intent to
blow
the whistle (r = 0.15, 0.44, and 0.31,
respectively),
they
do not appear to be related to actual
Whistleblowing
and Retaliation 285
whistleblowing
behavior (r = 0.06, 0.06, and 0.04).
Conversely,
while job satisfaction and job performance
relate
to actual whistleblowing (r = 0.19 and
0.11,
respectively), they are unrelated to whistleblowing
intentions
(r = )0.01
and 0.05, respectively).
Finally,
organizational commitment does not
appear
to be related to either whistleblowing intentions
or
actual whistleblowing. It should be noted,
that
while our analysis suggests that role responsibility
is
unrelated to actual whistleblowing behavior, it is
possible
that this relationship may be moderated by
the
whistleblowing channel used. Specifically, the
data
included in our analysis suggest a small, but
positive
correlation may exist between role responsibility
and
the use of an internal reporting channel,
while
no (or a negative) relationship may exist with
the
use of an external channel.
The
results of the meta-analytic examination of
correlations
between contextual aspects of the
whistleblowing
situation and whistleblowing intentions
(Intent
WB) and actions (Actual WB) are
presented
in Table II. Organizational climate for
TABLE
I
Correlations
between Whistleblower Characteristics and Whistleblowing Intentions and Actions
Meta-analysis
k N r SDr SESD resSD
%SEV 95%CI File drawer k
Whistleblower
demographics
Action
Age
Intent WB 3 1279 0.19 0.1902 0.0457 0.1847 5.77 )0.17/0.55
9
Actual
WB – – – – – – – – –
Sex
Intent WB 2 1131 )0.05 0.0866 0.0414 0.0760
22.93 )0.20/0.09 –
Actual
WB 4 1707 0.13 0.0368 0.0460 0 100 0.13/0.13 7
Education
Intent WB 3 1667 0.02 0.0867 0.0418 0.0759 23.32 )0.13/0.17
–
Actual
WB 6 7851 0.02 0.0619 0.0275 0.0555 19.76 )0.09/0.12
–
Tenure
Intent WB 3 1563 0.02 0.0651 0.0436 0.0483 44.93 )0.07/0.12
–
Actual
WB 2 1383 0.10 0.0100 0.0376 0 100 0.10/0.10 2
Job
level Intent WB 4 1938 0.10 0.1164 0.0439 0.1078 14.22 )0.11/0.31
4
Actual
WB 8 9200 0.08 0.0320 0.0293 0.0130 83.56 0.05/0.10 5
Other
characteristics Action
Ethical
judgment Intent WB 4 1147 0.45 0.0707 0.0472 0.0527 44.49 0.35/0.56 32
Actual
WB 2 320 )0.08 0.1669 0.0695 0.1518
17.32 )0.38/0.22 2
Job
satisfaction Intent WB 2 838 )0.01 0.0503 0.0487
0.0128 93.58 )0.03/0.02 –
Actual
WB 2 1164 0.19 0.0300 0.0399 0 100 0.19/0.19 6
Organizational
commitment
Intent
WB 8 2170 0.03 0.0637 0.0603 0.0202 89.89 )0.01/0.07
–
Actual
WB – – – – – – – – –
Job
performance Intent WB 2 1315 0.05 0 0.0389 0 – 0.05/0.05 –
Actual
WB 2 1401 0.11 0.0156 0.0373 0 100 0.11/0.11 3
Role
responsibility Intent WB 4 1494 0.15 0.1556 0.0474 0.1482 9.28 )0.14/0.44
8
Actual
WB 6 7762 0.06 0.1048 0.0274 0.1011 6.85 )0.14/0.26
2
Approve
of
whistleblowing
Intent
WB 6 2143 0.44 0.1406 0.0394 0.1349 7.86 0.18/0.71 47
Actual
WB 2 5514 0.06 0.0200 0.0190 0.0064 89.92 0.05/0.07 1
Notes: k
= number of studies included in the meta-analysis; N
= total number of participants across studies included in the
meta-analysis;
r = sample-size weighted mean observed
correlation; SDr = tandard deviation of
sample-size weighted
mean
observed correlation; SESD = standard error of the
standard deviation; resSD = residual standard deviation;
%SEV
= percent variance due to sampling error; 95% CI = 95% confidence interval
formed around the sample-size
weighted
mean observed correlation (the number before the slash indicates the lower
bound and the number after the
slash
indicates the upper bound); File drawer k indicates
the number of missing studies averaging null results required to
reduce
the mean observed correlation to 0.05. Intent WB indicates participants
indicated they intended or were likely to
blow
the whistle on a wrongdoing; Actual WB indicates participants actually blew the
whistle on a wrongdoing.
286
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
whistleblowing
is more strongly related to whistleblowing
intentions
than actual whistleblowing, but
the
direction of the relationship is consistent
(r
= 0.28 and 0.10, respectively). Of greater interest
is
that threat of retaliation is negatively related to the
intent
to blow the whistle, but appears to be unrelated
to
actual whistleblowing behaviors (r = )0.27
and
0.04, respectively). Also intriguing is the finding
that
supervisor support is positively related to
whistleblowing
intent (r = 0.28), but negatively
related
to actual whistleblowing (r = )0.12),
suggesting
that
supervisor support may facilitate one’s
decision
to blow the whistle, but work to inhibit
behaviors
required to actually blow it.
The
results of the meta-analysis of correlations
between
characteristics of the wrongdoing and of
the
wrong-doer and whistleblowing intentions
(Intent
WB) and actions (Actual WB) are presented
in
Table III. We found similar small relationships
between
seriousness of the wrongdoing and whistleblowing
intentions
and actions (r = 0.16 and 0.13,
respectively).
Importantly, the observed correlations
used
to compute the relationship between severity of
the
transgression and whistleblowing intent varied
widely
in magnitude, potentially signaling the presence
of
a moderator. Of interest is the moderate
effect
size found for the relationship between
closeness
to the wrong-doer and intent to blow the
whistle
(r = 0.45); this finding suggests that the
closer
the potential whistle-blower is to the wrongdoer
(whether
interpersonally or with respect to
organizational
structure), the greater their intention
to
blow the whistle on the transgression. Unfortunately,
similar
data was not available for actual
TABLE
II
Correlations
between contextual variables and whistleblowing intentions and actions
Meta-analysis
k N r SDr SESD resSD
%SEV 95%CI File drawer k
Contextual
variables Action
Organizational
climate
for
whistleblowing
Intent
WB 8 2947 0.28 0.1738 0.0457 0.1677 6.92 )0.05/0.61
37
Actual
WB 5 6960 0.10 0.0375 0.0263 0.0266 49.45 0.05/0.15 5
Threat
of retaliation Intent WB 7 2849 )0.27
0.1292 0.0459 0.1208 12.61 )0.51/-0.03 31
Actual
WB 8 11974 0.04 0.0907 0.0251 0.0872 7.65 )0.13/0.21
–
Supervisor
support Intent WB 4 1443 0.28 0.0530 0.0484 0.0216 83.41 0.23/0.32 19
Actual
WB 2 789 )0.12 0.0088 0.0495 0 100 )0.12/)0.12
3
Organizational
size Intent WB 1 725 )0.10 – – – – – 1
Actual
WB 5 5803 0.09 0.0783 0.0286 0.0729 13.34 )0.05/0.23
4
See
Note to Table I.
TABLE
III
Correlations
between characteristics of the wrongdoing/wrongdoer and whistleblowing
intentions and actions
Meta-analysis
k N r SDr SESD resSD
%SEV 95% CI File drawer k
Wrongdoing/wrongdoer
Action
Seriousness
of
wrongdoing
Intent
WB 7 3207 0.16 0.2642 0.0413 0.2609 2.44 )0.35/0.67
16
Actual
WB 6 6460 0.13 0.0348 0.0299 0.0180 73.41 0.10/0.17 10
Evidence
of
wrongdoing
Intent
WB 1 636 )0.06 – – – – – 1
Actual
WB 3 4759 0.06 0.0079 0.0250 0 100 0.06/0.06 1
Closeness
to
wrongdoer
Intent
WB 4 965 0.45 0.1768 0.0527 0.1687 8.88 0.12/0.78 32
Actual
WB – – – – – – – – –
Effective
in stopping
wrongdoing
Intent
WB – – – – – – – – –
Actual
WB 3 1714 )0.07 0.0741 0.0410 0.0617
30.65 )0.19/0.05 2
See
Note to Table I.
Whistleblowing
and Retaliation 287
whistleblowing,
so we do not know if this relationship
will
generalize. Also noteworthy is the small
negative
correlation found between actual whistleblowing
and
the whistleblower’s success in stopping
the
wrongdoing (r = )0.07).
This finding suggests
that
whistleblowers are not frequently successful in
their
efforts to curb organizational wrongdoing. It is
worth
noting, however, that all correlations included
in
this computation reflect the success of whistleblowers
who
utilized an external reporting channel.
Lastly,
displayed in Table IV are the results of the
meta-analysis
of correlates of retaliation against
whistleblowers.
Information is arranged according to
the
four types of correlates examined (1) characteristics
of
the whistleblower, (2) actions taken by the
whistleblower,
(3) contextual variables, and (4)
characteristics
of the wrongdoing. Education, job
level,
and the role-prescribed responsibility to blow
the
whistle appear to be relatively unrelated to
retaliation
against whistleblowers (r = 0.04, )0.07,
and
)0.07, respectively). While small relationships,
it
is
important to note that the correlation between
retaliation
and role responsibility is in the desired
direction.
Also, while based on only one study each,
it
appears that older whistleblowers are more likely
to
be retaliated against, as are those with greater
perceived
value congruence with the organization.
This
second finding may indicate a misperception as
to
the organization’s values regarding whistleblowing
and/or
wrongdoing.
Not
surprisingly, whistleblowers who utilize an
external
reporting channel are more likely to be
retaliated
against (r = 0.17). However, those whistleblowers
who
are effective in stopping the transgression
are
less likely to experience retaliation for
blowing
the whistle (r = )0.20),
as are those who
produce
convincing evidence (r = )0.25),
and those
who
enjoy support from their supervisor (r = )0.39).
More
intriguing, however, was our finding that
blowing
the whistle on serious transgressions or
those
that frequently occur in the organization, are
more
likely to be met with retaliation, than are
infrequent
or less severe wrongdoings (r = 0.30 and
0.13,
respectively).
Discussion
Our
purpose in conducting this research was to
examine
the antecedents and correlates of whistleblowing
and
retaliation against whistleblowers so
that
we might better understand their role as predictors
and
within the whistleblowing process. A
secondary
goal was to evaluate the feasibility of
drawing
conclusions about whistleblowing utilizing
data
gathered on whistleblowing intentions rather
than
whistleblowing actions. Overall, our results
suggest
that the predictors of the intent to blow the
whistle
are not the same as those of actual whistleblowing.
Further,
many of the variables typically
measured
in studies of whistleblowing are stronger
TABLE
IV
Correlations
of retaliation against a whistleblower
Meta-analysis
k N r SDr SESD resSD
%SEV 95%CI File drawer k
Whistleblower
characteristics
Education
5 2405 0.04 0.0329 0.0452 0 100 0.04/0.04 –
Job
level 6 2624 )0.07 0.0606 0.0471 0.0381
60.41 )0.14/0.01 3
Role
responsibility 4 2314 )0.07 0.0730 0.0408 0.0605
31.27 )0.18/0.05 2
Whistleblower
actions
Used
external channel 4 2337 0.17 0.0871 0.0391 0.0778 20.12 0.02/0.33 10
Success
in stopping wrongdoing 5 2410 )0.20 0.0782 0.0425
0.0656 29.63 )0.33/)0.07
15
Contextual
variables
Supervisor
support 3 923 )0.39 0.0027 0.0482 0 100 )0.39/)0.39
21
Co-worker
support 3 1194 0.03 0.0155 0.0501 0 100 0.03/0.03 –
Characteristics
of the wrongdoing
Frequency
of wrongdoing 2 855 0.30 0.0788 0.0448 0.0648 32.30 0.18/0.43 10
Seriousness
of wrongdoing 6 3046 0.13 0.0503 0.0431 0.0259 73.58 0.08/0.19 10
See
Note to Table I.
288
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
correlates
of whistleblowing intent than of whistleblowing
action.
Correlates
of whistleblowing
We
organized the presentation of the results of our
meta-analyses
of whistleblowing (intent and action)
according
to characteristics of the whistleblower,
work
context, wrongdoing and wrongdoer, as we
felt
that such variables would provide unique
information
regarding the whistleblowing process
(Gundlach
et al., 2003; Miceli and Near, 1988; Near
and
Miceli, 1995, 1996). Further, we anticipated
differences
in both the direction and strength of
these
relationships according to these dimensions
(e.g.,
Dozier and Miceli, 1985; Miceli et al., 1991b;
Near
and Miceli, 1995). We expected that demographics
and
other characteristics of the whistleblower
would
have stronger correlations with the
intent
to blow the whistle, while contextual variables
and
aspects of the transgression would have stronger
implications
for actual whistleblowing.
Whistleblower
characteristics
With
regard to demographic characteristics, our
results
suggest that whistleblowing intent might best
be
predicted from whistleblower age and the level of
the
job held by the intended whistleblower. Similarly,
actual
whistleblowing is likely to be predicted
by
job level and organizational tenure but also the
sex
of the whistleblower (e.g., Miceli et al., 1991b).
While
we found no studies which examined age of
whistleblower
with actual whistleblowing, one
would
think that older employees are more likely to
hold
positions at higher levels within the organizational
hierarchy
and also have greater tenure (this is
particularly
true for tenure-based promotion systems
commonly
found in government organizations).
Therefore,
we would expect that older employees
are
also more likely to follow through with blowing
the
whistle. Also likely is that older workers are
more
secure in their value systems than are younger
workers,
and therefore are more willing to defend
these
values throughout the lengthy whistleblowing
process
and in the face of retaliation (Chiu, 2003).
Likewise,
more senior and higher-level employees
are
more likely to have access organizational members
with
the ability to exact change within the
organization
(Keenan, 2000). Consistent with control
theory
and theories of power, organizational
members
with greater tenure and of higher job levels
usually
have stronger power bases (as compared with
their
younger and newer counterparts), and therefore
may
be more confident in their abilities to exact
change
within the organization via whistleblowing.
In
other words, these individuals may be more
certain
that the risks they take in blowing the whistle
will
not be in vain. Lastly, employees higher in the
organizational
hierarchy and those that have been
with
the organization for longer are likely to have a
greater
understanding of, appreciation for, and
commitment
to the espoused values of the organization,
and
thus a greater desire to protect this culture
(e.g.,
Berry, 2004; Schein, 1996).
We
were surprised to find that ethical judgment,
approval
of whistleblowing, and a perception that
blowing
the whistle is in one’s best interests were
related
to whistleblowing intent but not to actual
whistleblowing.
One would have thought that
morality
or ethicality, in particular, would be predictive
of
whistleblowing (e.g., Brabeck, 1984; Brief
and
Motowildo, 1986; Miceli et al., 1991b). Certainly,
we
anticipated that workers would be more
likely
to blow the whistle when they perceive that it
is
in their best interests to do so (e.g., Miceli and
Near,
1985). Perhaps, however, the influence of
contextual
factors on the decision to blow the
whistle
(e.g., fear of retaliation, perceived costs
versus
benefits, severity of the transgression, perceived
support
from supervisors and coworkers, etc.)
are
so great as to preclude whistleblowing on the
basis
of personal need or desire. Or, perhaps, the
commitment
to the organization is so great that the
desire
to meet the needs of the greater workgroup is
stronger
than that of meeting individual needs.
While
this potential is appealing, we were disheartened
to
find no correlation between organizational
commitment
and whistleblowing intent (e.g.,
Somers
and Casal, 1994). We expected to find that
greater
organizational commitment would yield a
desire
to exercise voice rather than either silence or
exit
strategies. It should be noted, however, that we
were
unable to locate any studies that reported a
correlation
between actual whistleblowing and
organizational
commitment. As such, we cannot be
sure
that there is not a relationship between this
variable
and the decision to blow the whistle.
Whistleblowing
and Retaliation 289
Of
particular interest is that job satisfaction and job
performance
were positively related to whistleblowing
action
but not to whistleblowing intent.
This
finding makes sense in light of various propositions
of
resource dependence and power theories
(e.g.,
Miceli et al., 1991b). Specifically, while one’s
job
performance may not enter into the decision to
blow
the whistle on an organizational wrongdoing,
when
one considers the likelihood that one might be
able
to effect change by blowing the whistle and/or
when
one considers the likelihood one will suffer
retaliation
by organizational members, it becomes
likely
that the intended whistleblower will consider
their
relative value to and power within the organization
prior
to taking action (e.g., Miceli and Near,
1994).
Specifically, resource dependence theory
suggests
that individuals with better job performance
are
more valuable to the organization, suggesting the
organization
officials would be more likely to want to
correct
the transgression than risk losing the whistleblower
(e.g.,
Miceli et al., 1991b). Similarly,
power
theory would predict that high performing
whistleblowers
would be in a better position to exact
change
through voice mechanisms (cf. French and
Raven,
1959). Also, in the making the decision to
engage
in whistleblowing, an employee with better
job
performance might be more comfortable in
expressing
a dissenting opinion than their fellow
employees
who may possess fewer idiosyncrasy
credits
(due to less satisfactory job performance; e.g.,
Hollander,
1958). Future research should examine
the
potential for job performance to predict retaliation
against
whistleblowers. It would seem likely that
individuals
with higher job performance would have
accumulated
more idiosyncrasy credits and would
also
be less likely to suffer retaliation.
Curiously,
we found that role responsibility to
blow
the whistle is predictive of intent to blow the
whistle
but not of actual whistleblowing. This finding
is
unexpected, as one would expect that the organization
creates
and staffs internal ‘‘watchdog’’ positions
to
facilitate whistleblowing on unsanctioned practices
(e.g.,
Miceli et al., 1991b; Near and Miceli, 1996).
Furthermore,
in certain occupations and industries,
all
employees are expected to voice concerns of illegitimate
or
unsanctioned activities (e.g., government
employees,
military, and employees of many government
contractors).
However, it may be that these
individuals
actually follow through with blowing the
whistle,
but utilize non-traditional channels in doing
so.
While in many cases this would still qualify as
whistleblowing,
use of such reporting channels makes
the
tracking and identification of these whistleblowers
more
difficult. Alternatively, certain internal
auditors
may be chosen from a select inner circle
within
the organization and possess the ultimate goal
of
preventing damage to the organization’s public
image
rather than stopping wrongdoing. In this case,
these
individuals may opt to ignore certain transgressions
they
know to be unethical, or in some other
non-traditional
way deal with the transgression so as
to
minimize public knowledge and scrutiny. Our data
does
not preclude the potential that this relationship
(between
role responsibility and whistleblowing) is
moderated
by the whistleblowing channel utilized
(internal
versus external). Regardless, our data suggest
that
having a role-related responsibility to blow the
whistle
may not be sufficient to ensure whistleblowing
action.
Other factors may be of greater
importance,
or may play a moderating role, in the
whistleblowing
decision-making process. This is a
matter
for future research.
Contextual
aspects of whistleblowing
An
examination of the meta-analyses of contextual
variables
and whistleblowing reveal somewhat surprising
findings
for organizational climate, fear of
retaliation,
and supervisor support. As expected,
organizational
climate for whistleblowing is positively
related
to whistleblowing intent (Berry, 2004),
however,
the relationship is considerably weaker for
whistleblowing
action. This may be a function of the
‘‘distance’’
between whistleblowing intent and
action
(e.g., Bagozzi, 1992). Alternatively, this may
signal
the presence of other contextual or individual
variables
that may moderate this relationship and
may
account for the reduction in explained variance.
For
example, supervisor support, if low, may serve
to
de-motivate an employee to blow the whistle
even
when this individual is part of an organization
that
(with regard to organizational wrongdoing)
values
employee dissent (Dozier and Miceli, 1985;
Miceli
and Near, 1985). Also of interest, is that
threat
or fear of retaliation appears to greatly reduce
the
likelihood that an observer of wrongdoing will
intend
to blow the whistle, but does not impact
actual
whistleblowing. Therefore, it appears that
once
the intention to whistle-blow is formed, fear of
290
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
retaliation
for whistleblowing does not serve to
de-motivate
action. It may be that, in their decision
to
blow the whistle, employees weigh aspects of the
situation
(like their relative power to exact change,
the
severity of the transgression, the likelihood that
they
will experience retaliation for blowing the
whistle
on this particular transgression, which is also
likely
to be a function of their relative power within
the
organization and their relative balance of idiosyncrasy
credits
which will ‘‘protect’’ them from
retaliation),
and thus only decide to blow the whistle
when
they perceive their chances of receiving
retaliation
are low (e.g., Miceli and Near, 1994; Sims
and
Keenan, 1998). Also, once the decision is made
to
blow the whistle, they may be so committed to
taking
action that they are no longer able to back
out.
Researchers should explore the possibility that
the
relationship between fear of retaliation and actual
whistleblowing
is moderated by such processes and
contextual
variables, as well as by the reporting
channel
employed (Miceli and Near, 2002).
Another
unexpected finding was that for supervisor
support
and whistleblowing intent versus action (e.g.,
Miceli
et al., 1991b). Specifically, supervisor support
seems
to increase the likelihood that an observer of
wrongdoing
will intend to blow the whistle, but to
decrease
actual whistleblowing behavior (e.g., Near
and
Miceli, 1986). This suggests that employees who
enjoy
the support of their supervisors in some domains
perceive
that this support is likely to extend to a
whistleblowing
situation. As such, they may be more
likely
to intend to blow the whistle upon observing a
wrongdoing,
knowing their supervisor will likely
‘‘back
them up’’. However, given the reverse in the
direction
of this relationship for actual whistleblowing,
some
other variable or process is likely acting as a
moderator.
One possible explanation is that employees
with
supportive supervisors may feel a greater
responsibility
to ensure that their supervisors are not
adversely
affected by a whistleblowing claim. These
individuals
may choose silence to voice in an effort to
spare
the supervisor embarrassment or discipline that
may
result from a report of illegitimate, unsanctioned,
or
immoral acts occurring within their department/
division.
The
role of wrongdoing in whistleblowing
Examination
of meta-analytic results for the influence
of
characteristics of the wrongdoing and wrongdoer
on
whistleblowing yield two interesting findings.
First,
while the relationships were similar for whistleblowing
intent
and action, we were surprised by the
relatively
small correlation between severity of the
wrongdoing
and whistleblowing. We would have
thought
that a more serious transgression would merit
action
more so than a less severe transgression (e.g.,
Trevino
and Victor, 1992). In fact, severity of the
wrongdoing
explains less than 3% of the variance in
both
whistleblowing intent and action. It may be that
in
some cases employees perceive that reports of very
severe
organizational wrongdoing are more likely to
be
met with resistance and retaliation by organizational
members
(e.g., Dozier and Miceli, 1985).
Alternatively,
this finding may be indicative of an
overall
inconsistency in assessments of the severity of
various
actions. For example, it may be that individuals
with
varying value orientations may assess the
same
action as more or less severe, or even as more or
less
wrong. Additionally, this may indicate incongruence
between
the employee’s values and the
organization’s
espoused values (e.g., Miceli and Near,
1994).
Researchers might explore the possibility that
value
congruence between the organization and the
whistleblower
moderates the relationship between
transgression
severity and whistleblowing.
Finally,
we were at first somewhat amazed by the
strong
positive correlation between whistleblowing
intent
and relational closeness to the wrong-doer (i.e.,
the
closer the observer was to the wrong-doer, whether
interpersonally
or with respect to organizational
structure,
the more likely it was that he or she intended
to
blow the whistle; e.g., King, 1997). However, it
would
make sense that one is more likely to observe
wrongdoing
by a person with which one is in close
proximity.
Further, the closer one is to the wrongdoer,
the
more likely one has access to sufficient evidence
of
wrongdoing. Knowledge of wrongdoing is
logically
the first step in the whistleblowing process. It
would
be interesting, however, to determine whether
this
relationship holds for actual whistleblowing.
Unfortunately,
no studies were found that reported
this
correlation.
Intending
to blow the whistle versus actually blowing it
Our
discussion of the correlates of whistleblowing
intent
versus actual whistleblowing points to a likely
Whistleblowing
and Retaliation 291
difference
in the predictors of each. More importantly,
these
differences probably reflect the ‘‘psychological
distance’’
between behavioral intention
and
overt action (Bagozzi, 1992). Specifically,
whistleblowing
intention and action are logically
separated
by psychological, motivational, and
implementation
processes (e.g., planning, monitoring,
guidance
and control, psychological commitment,
effort;
e.g., Azjen, 1992; Bagozzi, 1992).
Therefore,
a complete model of the whistleblowing
process
requires the identification and explication of
the
entire process (from observation, to intent, to
action,
and all steps in between). Certainly, most
researchers
in this domain are interested in drawing
conclusions
about actual whistleblowing; however,
few
have designed research to allow such enumeration
of
the process. We suspect that this is a
byproduct
of the difficulty in gaining longitudinal
data
and data from actual whistleblowers (e.g., Chiu,
2003;
Sims and Keenan, 1998).
Of
the studies included in this meta-analysis, only
two
examined whistleblowing from the perspective
of
both the intended and actual whistleblower;
therefore
only two reported a correlation between
intending
to blow the whistle and actual whistleblowing
behavior.
The sample-size weighted, mean
observed
correlation between whistleblowing intent
and
actual whistleblowing was only 0.05 (total
N
= 789, k = 2). While we can not be
sure of the
true
relationship between intent and action, this is
consistent
with research in other domains that has
reported
a lower than expected relationship (e.g.,
Ajzen,
1992). In sum, generalizing from intent to
actual
whistleblowing may not be a wise strategy.
Granted,
actual behavior is more difficult to study,
especially
in a sensitive area such as this. However, it
is
still imperative that such studies are conducted and
published.
Correlates
of retaliation against whistleblowers
We
examined four potential types of correlates of
retaliation
against whistleblowers. The strongest
correlates
of retaliation against whistleblowers were
found
for characteristics of the wrongdoing
(frequency
and severity of wrongdoing), whistleblower
actions
during the whistleblowing process
(use
of an external reporting channel and success in
stopping
the wrongdoing), and contextual aspects,
namely
supervisor support. It is interesting that positive
correlations
surfaced between retaliation and
both
the severity and the frequency of the wrongdoing.
This
finding likely reflects a greater threat to
the
organization’s future performance by whistleblowers
who
report information on a severe or
frequent
organizational wrongdoing (e.g., Miceli
and
Near, 2002). This may be particularly strong in
instances
where an external reporting channel is
utilized,
as the risk of public scrutiny and legal
intervention
increases (Miceli et al., 1991b). In
addition,
in organizations where wrongdoing is
frequent,
it is more likely that a cultural norm
actively
operates to continue and support the
transgressions
(e.g., Schein, 1996). In such cases, a
whistleblower
who reports wrongdoing (regardless
of
the channel employed) is actively violating this
norm.
According to Schein (1996), norm-violators,
particularly
in cases where a norm is widely shared,
often
face severe repercussions. Whistleblowers who
already
possess fewer idiosyncrasy credits (e.g., due
to
low job performance, low organizational value,
etc.;
Hollander, 1958), upon dissent, are likely to
face
the harshest retaliation.
As
expected, whistleblowers who possess convincing
evidence
of wrongdoing and those who are
effective
in stopping the transgression are less likely
to
be retaliated against (e.g., Miceli and Near, 2002;
Parmerlee
et al., 1982). It is logical that when the
whistleblower
possesses convincing evidence of the
transgression,
the organization would be well-served
to
halt the practice and not retaliate. Similarly, it
seems
likely that in organizations where whistleblowers
are
effective in stopping wrongdoing, the
organization
and the whistleblower were in agreement
that
the wrongdoing must be stopped (e.g., the
practice
violated the espoused values of the organization),
and
therefore it would be less likely that the
whistleblower
would experience organizationallysanctioned
retaliation
(Miceli and Near, 1994). It is
important
to note that retaliation may occur even
when
not sanctioned by the organization’s top
management
(Miceli and Near, 1994; Parmerlee
et
al., 1982). Future research should examine
sanctioned
and unsanctioned retaliation separately,
as
this would facilitate a more comprehensive
understanding
of the retaliation process.
292
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
We
were not surprised to find that whistleblowers
who
employed an external reporting channel were
more
likely to suffer retaliation, as the use of an
external
channel is likely to be interpreted as a threat
to
the organization’s structure and legitimate
authority
(e.g., Miceli and Near, 2002; Near and
Miceli,
1986; Weber, 1947). The correlation,
however,
was smaller than would be predicted by
these
theories. This is likely due to the immediate
external
scrutiny that these organizations would
suffer
as a result of the whistleblower’s use of an
external
reporting channel (Dworkin and Baucus,
1998).
Any retaliatory acts sanctioned by the
organization
are likely also to be reported through
this
external channel, thus further increasing the
potential
for public scrutiny and legal intervention
(Miceli
et al., 1991).
Our
finding of a small negative relationship
between
job level and retaliation is also noteworthy.
Power
theories and dependence-control theories
would
have predicted that individuals of higher job
level
would be in a better position to blow the
whistle
on organizational wrongdoing (Casal and
Zalkind,
1995; Near and Miceli, 1986). However,
organizational
members may feel a greater sense of
betrayal
when a member of higher status in the
organization
blows the whistle (e.g., Parmerlee et al.,
1982).
Therefore, instead of being better protected
against
retaliation, members of higher job level may
be
at somewhat higher risk. Given the small relationship
we
found, however, neither of these
explanations
fit the data. If valuable members of the
organization
are more likely to be retaliated against
(due
to feelings of betrayal, violation of power
structure,
etc.), it may be that job level is no longer
an
appropriate proxy variable for value. Specifically,
organizations
are increasingly hiring for top management
and
supervisory positions from outside the
organization,
rather than promoting from within (as
is
typical in tenure-based promotion systems). In
these
cases, individuals of high level in the organization
are
not necessarily more valuable, nor do they
necessarily
possess a relationship with the organization
as
a whole that would induce a feeling of
betrayal
after whistleblowing. Specifically, a feeling
of
betrayal implies a break of a psychological contract
made
in a dyad (i.e., between two individuals,
such
as between the whistleblower and the retaliator,
who
is perhaps a supervisor or coworker) not
between
an entire organization and the whistleblower.
Future
research should explore the potential
that
retaliation is better predicted by examining
dyadic
interactions or relationships (between whistleblower
and
likely retaliator) than by using a proxy
that
is representative of an organizational-level variable
(i.e.,
job level, value to organization, etc.; e.g.,
Casal
and Zalkind, 1995).
While
we did not have access to such data, it
would
be interesting to examine the potential for
retaliation
against intended whistleblowers. For
example,
what is the likelihood that these individuals
share
their intentions with others in the organization,
perhaps
to assess if, when, or how they should
blow
the whistle on organizational wrongdoing
(e.g.,
via social comparison processes)? Individuals
with
whom they share their intentions may be in a
position
to retaliate against the intended whistleblower
(perhaps
to prevent the individual from
taking
the charges any further), or to share his or her
intentions
with another organizational member who
is
in a position to retaliate. As this would have
implications
for actual whistleblowing, it seems
important
to understand and enumerate this process
as
well as to determine the predictors and correlates
of
retaliation against potential whistleblowers.
Practical
implications
Whistleblowing
on organizational wrongdoing has
the
potential for many positive outcomes for the
organization
(Miceli and Near, 1994; Near and
Miceli,
1995). Our results have a few implications
for
the types of actions that should be taken to improve
the
potential for actual whistleblowing in
organizations.
First, consistent with past research
findings
(e.g., Miceli et al., 1991b), our results suggest
that
employees of higher tenure and job level
are
more likely to blow the whistle. It may be that
younger
employees or those of a lower job level are
uncomfortable
with reporting wrongdoing (particularly
by
those at a higher level in the organizational
hierarchy).
Similarly, these employees may lack
sufficient
power or knowledge to blow the whistle
and
effect change (Keenan, 1995). Organizations
may
consider targeting this group of employees for
specific
training focused on (1) what types of activities
might
be considered unethical/unsanctioned,
Whistleblowing
and Retaliation 293
(2)
how reports of wrongdoing might be made, and
(3)
reemphasizing the organization’s commitment to
ethical
practices, etc. (e.g., Miceli and Near, 1994).
Perhaps
such training might be delivered during the
organizational
orientation process. However, the
organization
should be certain that these espoused
ethical
values are consistent with those that are
enacted
on a daily basis (e.g., Schein, 1996).
Incongruence
between espoused and enacted values
may
lead to feelings of confusion and even suspicion
within
the targeted audience.
Second,
we found that ethical judgment is related
to
whistleblowing intent, but not actual whistleblowing.
This
suggests that employees may be aware
when
an observed practice is questionable and
should
be reported, however, this knowledge is
insufficient
to instigate actual reporting. While this is
also
a matter for future research, organizational
officials
should consider the presence of other variables
within
their control that may work against
whistleblowing
(e.g., fear or threat of retaliation,
perception
that the costs of whistleblowing outweigh
potential
benefits, accurate perception/evaluation
of
the severity of the transgression, value
congruence
with the organization, perception that
whistleblowing
will not lead to change, etc.) Similarly,
our
research suggests that whistleblowers may
not
frequently be successful in stopping wrongdoing.
A
potential whistleblower may weigh this into his or
her
decision to actually blow the whistle.
Third,
our results suggest that role responsibility
to
blow the whistle is relatively unrelated to actual
whistleblowing.
Organizations should consider the
possibility
that the individuals in these positions
(wrongly)
believe their primary responsibility is to
‘‘save
face’’ for the organization by, at times, covering
up
transgressions, or by dealing with wrongdoing
using
other non-traditional or unsanctioned
measures.
Further, even though these employees
have
been appointed to a ‘‘watchdog’’ role, they
may
still fear they would betray their organization by
blowing
the whistle; similarly, they may still fear
unsanctioned
retaliation for whistleblowing by
coworkers
or supervisors. Top management should
ensure
that the internal reporting channels are
‘‘free
from leaks’’ and are manned by trustworthy
individuals.
Finally,
when possible, organizations may consider
publicizing
when an incident of whistleblowing has
led
to positive change. This act may improve the
likelihood
that other potential whistleblowers will be
motivated
to alert top management of wrongdoing.
It
would be wise, however, to exercise caution that
this
practice does not foster unsanctioned retaliatory
acts.
Limitations
and future research directions
One
limitation of our analyses is the relatively small
number
of studies included under each meta-analysis.
Although
the number of studies was few, the
actual
sample sizes were often in the thousands
suggesting
some generalizability of our findings.
Furthermore,
by integrating findings over three
decades
of research, this study provides a comprehensive
view
of the correlates of intended and actual
whistleblowing
as well as lending insight into the
retaliation
process. Thus, while there remains a need
for
more individual studies, this meta-analysis was
successful
in providing a snapshot of the current state
of
the field as well as in providing directions for
improving
future research.
A
second concern relates to our choice of only
published
articles in the meta-analyses. Arguments
can
be made that only significant findings are published,
and
thus by confining ourselves to the published
literature
we have overestimated the effect
sizes.
This is the classic file-drawer effect (Rosenthal,
1979).
To address this potential concern, we conducted
file-drawer
analyses and found that in many
instances
we needed several studies to reverse our
substantive
findings.
A
third limitation concerns the lack of reliability
data
available in the studies included in this metaanalysis.
All
measures are affected by measurement
error
and thus the correlates reported here are
underestimates
of true relationships. Related to
unreliability
in the measures, we found that many
studies
did not report adequate measures of construct
validity
for the measures that were employed. Similarly,
many
studies used ad-hoc measures of retaliation.
No
evidence was presented why certain
behaviors
are considered to be retaliatory. This
illuminates
the need to focus on construct validity
issues
in future. It would also be profitable if standardized
measures
are used for the main variables in
future
research.
294
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
Another
issue concerns the potential for alternative
explanations
for the differences reported
between
studies examining whistleblowing intentions
and
those exploring whistleblowing actions.
Specifically,
an anonymous reviewer raised the
important
point that since many studies on behavioral
intentions
utilize scenario-based approaches,
something
about this research methodology, other
than
the fact that intentions rather than actions are
studied,
may explain the differing results. This is always
a
concern in content domains where substantially
similar
research methodologies are consistently
employed
(i.e., whistleblowing intentions research
relies
heavily upon survey-based and scenario-based
approaches).
It is possible that the use of the scenariobased
method
may be acting as a moderator. This
concern
may best be addressed with future research
relying
on the triangulation of research methods.
A
final important consideration relates to the
samples
used by the authors of four of the studies
included
in this meta-analysis. Specifically, four
samples
were drawn from the 1980 U.S. Merit
System
Protection Board Survey. While it is possible
that
the samples overlapped to some extent, the
selection
criteria, sample sizes, and measured variables
differed
across the studies. As such, the potential
for
bias due to non-independence to impact
our
results is mitigated.
Conclusion
Whistleblowing
on organizational wrongdoing has
become
increasingly publicized in recent years. Such
whistleblowing
has assisted organizations and federal
agencies
in halting practices that would otherwise
harm
employees and consumers. Given the potential
for
positive outcomes to result from whistleblowing,
it
is disheartening to learn that many whistleblowers
may
fear and even suffer retaliation (e.g., Miceli and
Near,
1996). Much research has been conducted in an
effort
to understand and enumerate the whistleblowing
process
as well as to predict retaliation against
whistleblowers.
We cumulated this literature using
meta-analytic
methodology in an effort to ‘‘take
stock’’
of where we are in the field and to provide an
orientation
for future research in whistleblowing.
Of
greatest importance is the finding that research
conducted
on whistleblowing intentions does not
easily
translate into conclusions about whistleblowers.
Rather,
to advance our understanding of this field, we
need
to explore and enumerate the processes that
occur
between the time wrongdoing is witnessed and
when
actual whistleblowing occurs.
Acknowledgement
We
wish to thank the Editor and two anonymous
reviewers
for their valuable comments and suggestions
on
an earlier version of this manuscript.
References
Anand,
V., B. E. Ashforth and M. Joshi: 2004, _Business
as
Usual: The Acceptance and Perpetuation of
Corruption
in Organizations_, Academy
of Management
Executive
18(2), 39–53.
Azjen,
I.: 1992, _The Theory of Planned Behavior_,
Organizational
Behavior and Human Decision Processes 50,
179–211.
Bagozzi,
R. P.: 1992, _The Self-Regulation of
Attitudes,
Intentions,
and Behavior_, Social
Psychology Quarterly
55(2),
178–204.
Barnard,
C. I.: 1938, The Functions of the Executive
(Harvard
University Press, Cambridge, MA).
Berry,
B.: 2004, _Organizational Culture: A Framework
and
Strategies for Facilitating Employee Whistleblowing
_, Employee
Responsibilities and Rights Journal
16(1),
1–11.
*Brabeck,
M.: 1984, _Ethical Characteristics of Whistle
Blowers_, Journal
of Research in Personality 18, 41–53.
Brief,
A. P. and S. J. Motowildo: 1986, _Prosocial
Organizational
Behaviors_, Academy of Management
Review
11(4), 710–725.
*Casal,
J. C. and S. S. Zalkind: 1995, _Consequences of
Whistle-Blowing:
A Study of the Experiences of
Management
Accountants_, Psychological
Reports 77,
795–802.
*Chiu,
R. K.: 2003, _Ethical Judgment and
Whistleblowing
Intention:
Examining the Moderating Role of
Locus
of Control_, Journal of Business
Ethics 43, 65–74.
Dozier,
J. B. and M. P. Miceli: 1985, _Potential Predictors
of
Whistle-Blowing: A Prosocial Behavior Perspective
_, Academy
of Management Review 10(4), 823–836.
Dworkin,
T. M. and M. S. Baucus: 1998, _Internal vs.
External
Whistleblowers: A Comparison of Whistleblowing
Processes_, Journal
of Business Ethics 17, 1281–
1298.
Whistleblowing
and Retaliation 295
*Ellis,
S. and S. Arieli: 1999, _Predicting Intentions to
Report
Administrative and Disciplinary Infractions:
Applying
the Reasoned Action Model_, Human
Relations
52(7),
947–967.
Enz,
C. A.: 1988, _The Role of Value Congruity
in Intra-
Organizational
Power_, Administrative Science
Quarterly
33
284–304.
French,
J. R. P., Jr. and B. H. Raven: 1959, _The
Bases
of
Social Power_, in D. Cartwright (ed.), Studies
in
Social
Power (Institute for Social Research, Ann
Arbor,
MI).
Gundlach,
M. J., S. C. Douglas and M. J. Martinko:
2003,
_The Decision to Blow the Whistle: A Social
Information
Processing Framework_, Academy
of
Management
Review 28(1), 107–123.
Hollander,
E. P.: 1958, _Conformity, Status, and
Idiosyncracy
Credits_, Psychological
Review 65(2), 117–
127.
Hunter,
J. E. and F. L. Schmidt: 2004, Methods of
Meta-analysis:
Correcting for Error and Bias in Research
Findings,
2nd edn. (Sage Publications, Inc, Thousand
Oaks,
CA).
Jubb,
P. B.: 1999, _Whistleblowing: A
Restrictive
Definition
and Interpretation_, Journal
of Business Ethics
21,
77–94.
*Keenan,
J. P.: 2000, _Blowing the Whistle on Less
Serious
Forms of Fraud: A Study of Executives and
Managers_, Employee
Responsibilities and Rights Journal
12(4),
199–217.
*Keenan,
J. P.: 1995, _Whistleblowing and the
First-Level
Manager:
Determinants of Feeling Obliged to Blow
the
Whistle_, Journal of Social
Behavior and Personality
10(3),
571–584.
King,
G., III: 1999, _The Implications of an Organization’s
Structure
on Whistleblowing_, Journal
of Business
Ethics
20, 315–326.
*King,
G., III: 1997, _The Effects of
Interpersonal
Closeness
and Issue Seriousness on Blowing the
Whistle_, The
Journal of Business Communication 34(4),
419–436.
Latane,
B. and J. Darley: 1970, The Unresponsive Bystander:
Why
Doesn’t He Help Me? (Appleton-Century-Crofts,
New
York).
*Miceli,
M. P. and J. P. Near: 2002, _What Makes
Whistle-Blowers
Effective? Three Field Studies_,
Human
Relations 55(4), 455–479.
*Miceli,
M. P. and J. P. Near: 1994, _Relationships
Among
Value Congruence, Perceived Victimization,
and
Retaliation Against Whistle-Blowers_, Journal
of
Management
20(4), 773–794.
Miceli,
M. P. and J. P. Near: 1992, Blowing the Whistle:
The
Organizational and Legal Implications for Companies
and
Employees (Lexington, New York).
*Miceli,
M. P. and J. P. Near: 1988, _Individual and
Situational
Correlates of Whistle-Blowing_, Personnel
Psychology
41, 267–281.
*Miceli,
M. P. and J. P. Near: 1985, _Characteristics of
Organizational
Climate and Perceived Wrongdoing
Associated
with Whistle-Blowing Decisions_, Personnel
Psychology
38, 525–544.
*Miceli,
M. P. and J. P. Near: 1984, _The Relationships
among
Beliefs, Organizational Position, and Whistle-
Blowing
Status: A Discriminant Analysis_, Academy
of
Management
Journal 27(4), 687–705.
*Miceli,
M. P., J. B. Dozier and J. P. Near: 1991,
_Blowing
the Whistle on Data Fudging: A Controlled
Field
Experiment_, Journal of Applied Social
Psychology
21(4),
271–295.
*Miceli,
M. P., J. P. Near and C. R. Schwenk: 1991,
_Who
Blows the Whistle and Why?_, Industrial
and
Labor
Relations Review 45(1), 113–130.
Miller,
L. E. and J. E. Grush: 1988, _Improving Predictions
in
Expectancy Theory Research: Effects of
Personality,
Expectancies, and Norms_, Academy
of
Management
Journal 31(1), 107–122.
Mitchell,
T. R.: 1974, _Expectancy Models of Job
Satisfaction,
Occupational
Preference, and Effort: A Theoretical
and
Methodological Appraisal_, Psychological
Bulletin
81, 1053–1077.
*Near,
J. P. and M. P. Miceli: 1996, _Whistle-Blowing:
Myth
and Reality_, Journal
of Management 22(3),
507–526.
Near,
J. P. and M. P. Miceli: 1995, _Effective Whistle-
Blowing_, Academy
of Management Review 20(3),
679–708.
*Near,
J. P. and M. P. Miceli: 1986, _Retaliation against
Whistle
Blowers: Predictors and Effects_, Journal
of
Applied
Psychology 71(1), 137–145.
Near,
J. P. and M. P. Miceli: 1985, _Organizational
Dissidence:
The Case of Whistle-Blowing_, Journal
of
Business
Ethics 4, 1–16.
O’Day,
R.: 1972, _Intimidation Rituals: Reactions to
Reform_, Journal
of Applied Behavioral Science 10,
373–386.
*Parmerlee,
M. A., J. P. Near and T. C. Jensen: 1982,
_Correlates
of Whistle-Blowers’ Perceptions of Organizational
Retaliation_, Administrative
Science Quarterly
27,
17–34.
Pfeffer,
J. and G. R. Salancik: 1978, The External Control
of
Organizations (Harper & Row).
Rosenthal,
R.: 1979, _The ‘‘File Drawer Problem’’ and
Tolerance
for Null Results_, Psychological
Bulletin 86,
638–641.
Schein,
E. H.: 1996, ‘Culture: The Missing Concept in
Organization
Studies’, Administrative Science Quarterly,
236.
296
Jessica R. Mesmer-Magnus and Chockalingam Viswesvaran
*Sims,
R. L. and J. P. Keenan: 1998, _Predictors of
External
Whistleblowing: Organizational and Intrapersonal
Variables_, Journal
of Business Ethics 17, 411–421.
*Somers,
M. J. and J. C. Casal: 1994, _Organizational
Commitment
and Whistle-Blowing: A Test of the
Reformer
and the Organization Man Hypothesis_,
Group
and Organization Management 19(3), 270–284.
Thibaut,
J. W. and H. H. Kelley: 1959, The Social Psychology
of
Groups (Wiley).
*Trevino,
L. K. and B. Victor: 1992, _Peer Reporting of
Unethical
Behavior: A Social Context Perspective_,
Academy
of Management Journal 35(1), 38–64.
*Victor,
B., L. K. Trevino and D. L. Shapiro: 1993, _Peer
Reporting
of Unethical Behavior: The Influence of
Justice
Evaluations and Social Context Factors_, Journal
of
Business Ethics 12, 253–263.
Weber,
M.: 1947, The Theory of Social and Economic
Organizations
(Macmillan Publishing Company, Inc.).
Department
of Management & Marketing,
University
of North Carolina at Wilmington,
601
South College Road,
Wilmington,
NC 28403, U.S.A.
E-mail:
magnusj@uncw.edu
Department
of Psychology,
Florida
International University,
University
Park Campus,
11200
SW 8th Street,
Miami,
FL 33199, U.S.A.
E-mail:
vish@fiu.edu
Whistleblowing and Retaliation 297