Content uploaded by Roger Persson
Author content
All content in this area was uploaded by Roger Persson
Content may be subject to copyright.
ORIGINAL PAPER
Personality trait scores among occupationally active bullied
persons and witnesses to bullying
Roger Persson Æ Annie Hogh Æ A
˚
se-Marie Hansen Æ
Catarina Nordander Æ Kerstina Ohlsson Æ Istvan Balogh Æ
Kai O
¨
sterberg Æ Palle Ørbæk
Published online: 22 May 2009
Ó Springer Science+Business Media, LLC 2009
Abstract There is little research on the personalities and
emotional stability of persons who report being bullied or
witnessing bullying at work. Men and women (N = 247)
from 19 to 64 years of age and in manual labour completed
a questionnaire concerning the psychosocial work climate,
bullying and personality. Three groups were defined: bul-
lied (N = 14), witnesses (N = 31) and non-bullied
respondents (N = 202). The Swedish universities Scale of
Personality was used to assess personality traits related to
the three major dimensions of neuroticism, extraversion,
and aggressiveness. Bullied persons had higher scores on
all six scales within the neuroticism dimension as well as
higher irritability (aggressiveness dimension) and impul-
siveness scores (extraversion dimension), when compared
with their non-bullied work colleagues. To conclude, bul-
lied persons display a self-image that is dominated by
mistrust and embitterment as well as irritability and
impulsiveness. Accordingly, when dealing with bullying at
work, the need for ego-supportive actions should be con-
sidered in conjunction with more organisational, collective
oriented action.
Keywords Aggressiveness Anxiety Extraversion
Harassment Job content questionnaire
Work environment
Introduction
Within the occupational health literature there is increasing
interest in the relationship between individual personality
traits and workplace bullying (Einarsen 2000). During the
course of this development, personality traits, or personal
dispositions, have primarily been studied as an antecedent
to becoming a victim of harassment or bullying, as a factor
that moderates the bullying process, or as a factor associ-
ated with the perpetrators (Djurkovic et al. 2005; Einarsen
et al. 1990; Hansen et al. 2006; Mikkelsen and Einarsen
2002; Seigne et al. 2007).
Since being bullied at work implies systematic exposure
to repeated negative acts (e.g. name-calling, threats,
physical aggression and social isolation), it is also con-
ceivable that enduring harassment, via traditional learning
mechanisms (i.e. classical and operant conditioning as well
as modelling), change the individual and consequently
affects the individual’s self-presentation in personality trait
questionnaires. However, because no study yet has
attempted to disentangle these questions in a longitudinal
design, it remains undetermined whether certain personal-
ity traits should be regarded as antecedents or conse-
quences of worksite bullying (Einarsen 2000).
Another unanswered question is whether witnesses to
bullying show deviations in personality trait questionnaires.
In contrast to bullied persons and perpetrators, witnesses
may choose sides as well as degree of engagement in the
conflict. In theory, witnesses may therefore, not be a very
homogenous group. However, and in view of the main
R. Persson (&) A
˚
se-MarieHansen P. Ørbæk
National Research Centre for the Working Environment, Lersø
Parkalle
´
105, 2100 Copenhagen, Denmark
e-mail: rpe@nrcwe.dk
A. Hogh
Department of Psychology, University of Copenhagen,
Copenhagen, Denmark
C. Nordander K. Ohlsson I. Balogh K. O
¨
sterberg
Division of Occupational and Environmental Medicine, Lund
University, Lund, Sweden
123
Motiv Emot (2009) 33:387–399
DOI 10.1007/s11031-009-9132-6
stream negative attitude towards bullying, the mere
acknowledgment of being a witness implies to some extent
that one dislikes bullying. In addition we have, in a previous
Swedish study, observed that witnesses displayed lower
trait anxiety scores than bullied persons (Hansen et al.
2006). Hence, it seems warranted to treat witnesses as a
group in their own right and to examine the possibility that
witnesses may be characterized by certain personality traits.
From a theoretical perspective personality traits are
thought to be relatively stable characteristics that account
for individual consistency in behaviour, feelings and
thinking across a wide range of situations. Personality traits
are often contrasted with states, i.e. short-lasting emotional
and cognitive reactions that unfold in a specific situation. It
is commonly assumed that traits and states are correlated.
Accordingly, a person who has a propensity for certain
consistent behaviour, feelings and thinking is also more
likely to display signs of these behaviours, thinking and
feelings in a given situation. Empirically, personality trait
research relies heavily on the use of questionnaires which
at their best offer an economic, conceptually accurate,
quick and easy sampling of the mixture of behaviours and
situations that are necessary to be aware of when drawing
inferences about a person’s personality (Persson 2002).
The majority of studies on personality and bullying
seem to have been conducted among adolescents (i.e.
persons under 18 years) and therefore, in a school context
(Bjorkqvist et al. 1992; Roland 2002). Occasionally, older
students in contact with the labor market have also
attracted the researchers interest (Parkins et al. 2006).
However, the few studies dealing with personality traits
and worksite bullying have mostly involved white-collar
workers (e.g. teachers) and other workers employed in the
public sector. Very often, the focus has also been on
individual traits, for example aggression or anxiety
(Bjorkqvist et al. 1994; Einarsen et al. 1996; Hansen et al.
2006; Moreno-Jime
´
nez et al. 2007). Comprehensive per-
sonality trait inventories have more infrequently been
employed to describe bullied persons’ self-image (Matt-
hiesen and Einarsen 2001).
One exception is an American study which used the
Minnesota Multiphasic Personality Inventory-2 (MMPI-2)
to characterize 47 harassed and 82 non-harassed workers
with psychological problems who were awaiting the out-
come of their compensation claims (Gandolfo 1995). Only
the paranoia scale score was found to be elevated among
bullied persons when compared with non-harassed persons.
However, when compared to norm data both harassed and
non-harassed persons showed elevations above 60 T-scores
on the following scales: Infrequency (F-scale), Hypo-
chondriasis (Hs), Depression (D), Hysteria (H), Paranoia
(Pa), Psychastenia (Pt) and Schizophrenia (Sc) (Gandolfo
1995). Interestingly this overall profile configuration was
essentially repeated in a Norwegian study, in which the
MMPI-2 was applied to 85 members of two support asso-
ciations for bullied people. It was observed that bullied
persons had higher scores on six of the ten clinical scales
when compared to American norm data (Matthiesen and
Einarsen 2001). Almost the identical pattern of scores was
observed in a recent Italian study (Girardi et al. 2007). In
this case the MMPI-2 was applied on 146 bullied workers
from various occupations who had been identified via
interviews at a workers union and subsequently referred to
an occupational health clinic.
In another Norwegian study, differences between 72
bullied and 72 non-bullied persons were observed on four of
five personality scale scores, as assessed with the Interna-
tional Personality Item Pool (IPIP) (Glasø et al. 2007).
Although it was concluded that personality is an important
factor in understanding the bullying phenomenon, the
authors also emphasized that it was only a smaller fraction of
the bullied persons that tended to be more emotionally
instable, less extravert, less agreeable, less conscientious,
and less open to experience. Similar to the Norwegian
MMPI-2 study, the bullied participants were recruited via
two support associations for bullied persons; the reference
group comprised demographically matched part-time stu-
dents, who were recruited from various locations in Norway.
Taken together, since all participants in Gandolfo’s
study and many of the participants in the Norwegian
MMPI-2 study had left work (due to sick leave, unem-
ployment, or disability pension), it is very difficult to
conclude whether the observed high degree of emotional
imbalance is a result of worksite bullying, or merely a
secondary reaction to disabling social circumstances. Also
the unclear compositions of the study samples in the Italian
MMPI-2 study and the Norwegian IPIP study makes it
difficult to rule out secondary reactions.
However, a few studies from Germany and Ireland seem
to indicate that the increased degree of emotional imbal-
ance observed among bullied persons is also observable
among bullied persons in active work, also when they are
compared with other persons in active work. For example,
in a recent German study of primarily white-collar workers
(i.e. administration, office and health care jobs), which
used a German version of the NEO-PI (i.e. NEO-Fu
¨
nf
Faktoren Inventar), positive relations were observed
between bullying and neuroticism and openness to expe-
rience (Rammsayer et al. 2006). Scores relating to the other
three dimensions, i.e. extraversion, agreeableness and
conscientiousness were, however, unrelated to bullying.
Moreover, in one Irish study that used the ICES Personality
Inventory (i.e. Independence, Contentiousness, Extraver-
sion, Stability, and Social Desirability), and which, similar
to the NEO-PI, adheres to the well-known Big-Five theory,
it was observed that bullied persons tended to be less
388 Motiv Emot (2009) 33:387–399
123
extravert, less stable, less independent, more conscientious,
and to give more socially desirable answers than non-bul-
lied workers (Coyne et al. 2000).
On the whole, however, it is clear that the documenta-
tion regarding the degree to which bullied people in active
work show signs of being habitually, emotionally imbal-
anced, is limited. Accordingly, the foundation for under-
standing a bullied person’s self-image and subsequent
repercussions in terms of health and social behaviors is
weak. There is, therefore, a need for further systematic
investigations concerning bullying and personality traits at
the worksite (Rammsayer et al. 2006).
Against this background, we decided to examine the
answers from a set of questions that were originally com-
piled to record the psychosocial work climate at two man-
ufacturing companies, where the vast majority of men and
women performed identical work tasks and worked on the
same job rotation schedule (Nordander et al. 2007). Since
questions about bullying and witnessing bullying were
included together with a complete personality inventory, it
was not only possible to study the relation between per-
sonality traits and bullying in people in active work but also
to extend previous research to include witnesses.
The main objective was to examine and describe whe-
ther bullied persons and witnesses to bullying showed
deviations on the Swedish universities Scale of Personality
(SSP) (Gustavsson et al. 2000), when compared with their
non-bullied work colleagues. The SSP, which is an updated
and modified version of the Karolinska Scales of Person-
ality (KSP) inventory (Schalling et al. 1987), covers traits
relating to three major personality dimensions, i.e. neu-
roticism, extraversion and aggressiveness. In view of pre-
vious findings, it was expected that bullied persons would
show signs of more emotional imbalance, particularly as
regards neuroticism related traits.
In addition to examine personality traits, the original aim
of the study to assess the psychosocial climate opened up for
a possibility to explore other insufficiently illuminated
questions as regards work related bulling. One such question
concern how bullied persons and witnesses to bullying
report psychosocial exposures at work in terms of demands,
control and support, as expressed in the Job Content Ques-
tionnaire (JCQ) (Karasek et al. 1998). Even if the organi-
zational impact on bullying has been a recurrent topic in
previous research, the extant knowledge is relatively sparse
and needs to be supplemented (Hauge et al. 2007). Hence, to
improve the possibilities for estimating and describing the
degree and nature of mismatch between work and those who
are bullied, or witness to bullying first hand, we also decided
to look at the scores from the Job Content Questionnaire.
This allowed us to gauge how bullied persons and witnesses
to bullying judged their own performance at work and the
extent to which they perceived work to impact on family life.
Recent studies has observed that bullied persons perceive
lower levels of social support from coworkers and supervi-
sors and that people witnessing bullying perceive the work
environment to be almost as stressful as people who are
bullied (Hansen et al. 2006; Hauge et al. 2007). Accordingly,
we expected bullied persons and witnesses to report more
adverse circumstances at work. Another set of questions in
the questionnaire dealt with performance at work and work
family conflicts. Since these issues in this context are linked
with even sparser empirical observations than the JCQ, but
are of relevance for understanding the situation of bullied
persons and witnesses to bullying, it was also decided to
explore these questions.
Methods
Participants
In total 247 employees from two worksites participated. In
one worksite, employees manufactured rubber-sealings of
various dimensions (N = 124; 48 men/76 women). At the
other worksite, employees manufactured break regulators
(N = 123; 56 men/67 women). All participants gave their
written informed consent to participate in the study, and the
study protocol was approved by the Ethics Committee of
Lund University.
The workers in the rubber manufacturing plant produced
rubber-sealings of various dimensions. This entailed that the
workers operated vulcanization machines, which typically
should be opened and emptied once every minute. The
removal of the rubber-sealings was quite physically
demanding, also involving a risk of getting burnt. This had a
negative effect on working postures. The workers in the
mechanical assembly plant produced brake-regulators.
There were six workstations and the work line involved the
mounting of empty brake-regulators (with a weight about 2–
3 kg) on a fixture, fitting additional details by hand, as well
as using pneumatic screwdrivers and riveting hammers. The
final step involved wiping and removing the completed
regulators (weight 3–4 kg). The cycle time at each of the six
work stations was about 25 s. At both workplaces, a con-
siderable part of the workday involved solitary work.
The identification of participants
The participants were first identified in conjunction with a
project concerning repetitive work and manual labour,
aiming to study differences between men and women with
identical work tasks (Nordander et al. 2007
). In a first step,
all currently employed production workers at both work-
sites, including workers on long-term sick leave, were
invited to participate. However, workers on leave, or with
Motiv Emot (2009) 33:387–399 389
123
\3 months seniority in their present job, were not inclu-
ded. One female in active work, and five male and two
female workers who were on long-term sick leave, could
not be reached or refrained from participation. All in all,
277 workers were included 105 males (mean age 36 years,
SD 11) and 172 females (mean age 42 years, SD 9). All
underwent a physical examination and a questionnaire-
based interview with focus on health complaints. In the
second step, and several months later, these 277 men and
women were asked to respond to a questionnaire on the
psychosocial environment (including the questions about
worksite bullying and personality traits), activities outside
work, and smoking. Dropouts were either on long-term
parental leave, or had left their employment at the partic-
ular department, while two further participants refrained
from participation (N = 264). Moreover, eleven partici-
pants were on long-term sick leave and were not included,
and six participants did not respond to the bullying ques-
tions that were used to identify subgroups (N = 247).
Identification of bullied persons and victims
The items used to identify bullied persons and witnesses to
bullying were derived from the General Nordic Question-
naire for Psychological and Social Factors at Work (QPS-
Nordic) (Dallner et al. 2000). The QPS-Nordic provides a
definition to bullying that is listed ahead of the questions
‘‘ Bullying (harassment, offending somebody) is a problem
at some workplaces and for some employees. To label
something bullying, the offensive behavior has to occur
repeatedly over a period of time and the exposed person
has to experience difficulties defending himself/herself. The
behavior is not bullying if two parties of approximately
equal ‘‘strength’’ are in conflict or the incident is an iso-
lated event’’. The bullying question read ‘‘Have you been
subjected to bullying or harassment at work during the past
6 months?’’ and the witness question read ‘‘Have you
noticed anyone being subjected to bullying or harassment
at your workplace during the past 6 months?’’. For both
questions the response categories read ‘‘yes’’ or ‘‘no’’.
Based on the 247 participants’ responses three mutually
exclusive subgroups were formed: Non-Bullied Referents
(N = 202), Witnesses (N = 31), and Bullied (N = 14). Of
the 14 persons comprising the bullied group, eleven also
acknowledged having witnessed bullying. Demographic
information for each subgroup is presented in Table 1.
Self-rated health, mental well-being and life style factors
The General Health Questionnaire-12 (GHQ-12) (Goldberg
and Williams 1988) was used to evaluate the group’s
current mental health status. In addition, two single item
indicators were used to measure the person’s perception of
current and past (last 12-months) general health status. This
comparison showed a clear gradient in mean GHQ-12
scores where the bullied persons showed signs of having
the poorest mental health, followed by witnesses and non-
bullied referents (Table 1). With respect to the general
health indicators, only bullied persons perceived their
current and past health in worse terms. The above-men-
tioned differences between groups were all statistically
significant (p \ 0.005). There were no differences between
the three groups as regards smoking and alcohol, coffee
and tea drinking habits.
Measures
Swedish universities scales of personality
The Swedish universities Scales of Personality (SSP) was
used to assess personality traits (Gustavsson et al. 2000).
SSP is a revised version of the Karolinska Scales of Per-
sonality (KSP) (Schalling et al. 1987). SSP consists of 13
seven-item scales, each covering one personality dimen-
sion. The 13 scales can be grouped into three main per-
sonality dimensions according to a three-factor solution:
Neuroticism, Aggressiveness, and Extraversion. Each item
is given as a statement and answered on a four-point scale
(1–4), ranging from ‘does not apply at all’ to ‘applies
completely’. The formulation of the questions are very
general and context free, and it is emphasized that the
person should respond in a way that best reflects their
habitual actions or feelings to a given statement. The three-
factor structure of the SSP corresponds to well-known
personality theories, particularly the model behind the
Eysenck Personality Inventory (EPI) (Eysenck and Ey-
senck 1985). The SSP dimensions of Neuroticism and
Extraversion is similar to the EPI dimensions with the same
labels, and the ‘‘Aggressiveness’’ dimension has some
overlap with the ‘‘Psychoticism’’ factor of the EPI. Scale
descriptions, factor belonging and internal consistency data
are presented in Table 2.
The job content questionnaire
The Swedish version of the Job Content Questionnaire
(JCQ) was used to assess how the participants perceived
the work environment in terms of psychological job
demands, job control and job support (Karasek et al. 1998).
Psychological job demands and job control was measured
with nine items each (Chronbach’s a = 0.71 and 0.79,
respectively) whereas Job support was measured with eight
items (Chronbach’s a = 0.70). Following JCQ theory, the
job control dimension was also broken down into the sub-
dimensions of decision authority (three items; Chronbach’s
a = 0.52) and skill discretion (six items, Chronbach’s
390 Motiv Emot (2009) 33:387–399
123
a = 0.76). Also the support dimension was broken down to
distinguish between support from managers (four items;
Chronbach’s a = 0.76) and support from co-workers (four
items Chronbach’s a = 0.77). In the JCQ each item is
given as a statement indicating degree of agreement. The
psychological job demands and job control items were
answered on a four-point scale (1–4). Support items were
answered on a five-point scale (0–4). The mean score for
each dimension and sub-dimension were used as the out-
come measure.
Personal satisfaction with job performance
Four single items derived from the General Nordic
Questionnaire for Psychological and Social Factors at
Work (QPS-Nordic; Dallner et al. 2000; QPS-items No.
66–69) were used to assess how well the participant’s
perceived themselves to master four different aspects of
work. The first item asked ‘‘Are you satisfied with the
quality of your work?’’. The second item asked ‘‘Are you
satisfied with the quantity of work you are performing’’.
Table 1 Demographics and background information
Referents (N = 202) Witness (N = 31) Bullied (N = 14)
Age [years; M, (SD)] 39 (11) 37 (11) 43 (11)
Height [cm; M, (SD)] 170 (13) 170 (10) 168 (8)
Weight [kg; M, (SD)] 73 (17) 73 (15) 76 (16)
Seniority [years in this job; M, (SD)] 11 (9) 7 (5) 11 (9)
Sickness absence [number of times the last 12-months; M, (SD)] 2.17 (0.88) 2.19 (0.75) 2.43 (1.28)
GHQ-12 mean score [range 1–4; M, (SD)]
a
1.85 (0.37) 2.01 (0.57) 2.18 (0.58)
Perceived current general health [range 1–5; M, (SD)]
a
2.48 (0.89) 2.32 (0.83) 3.07 (1.21)
Perceived general health the last 12-months [range 1–5; M, (SD)]
a
2.36 (0.82) 2.39 (0.76) 3.00 (1.04)
Women/Men (%) 55/45 65/35 79/21
Company (%)
Rubber manufacturing 51 48 36
Mechanical assembly 49 52 64
Employment conditions (%)
Employed 92 93 67
Temporary, known end date 2 3 8
Project, known end date 1 0 8
Project, no known end date 1 0 0
Other 4 3 17
Perceived risk for unemployment within the next 12 months (%)
Great risk 9 21 29
Some risk 23 28 29
Small risk 33 28 29
No risk 32 21 14
Not relevant 3 3 0
Social status (%)
Single 20 17 36
Together with other adult without children 18 27 21
Together with other adult with children 49 47 36
Together with children without other adult 8 3 0
Other 5 7 7
Do you have any chronic disease, trouble after an accident,
a handicap or any other weakness? (%)
Yes 24 23 43
No 76 77 57
Age, height, weight and seniority are rounded to nearest full figure; General health questions: 1 very good, 5 very poor. GHQ-12 General health
questionnaire-12; Percentages are rounded to the nearest full figure and may therefore not always add up 100%
M mean, SD standard deviation
a
Lower scores are better
Motiv Emot (2009) 33:387–399 391
123
The third item asked ‘‘Are you satisfied with your ability
to solve problems in your work’’. The fourth item asked
‘‘Are you satisfied with your ability to maintain a good
relationship with your fellow co-workers?’’. All items
were responded to on a five-point scale: 1 = very rarely
or never, 2 = seldom, 3 = sometimes, 4 = often,
5 = very often or always. The items were dichotomized
according to the substantive meaning of response alter-
natives (0 = very rarely to sometimes, 1 = often to very
often/always).
Work and family conflict: Spillover from work to home life
Three items, derived from previous studies, measured
negative spillover from work to family (Chandola et al.
2004; Netemeyer et al. 1996). The first item read ‘‘Would
you say your job takes so much energy you don’t feel up to
doing things that need attention at home’’. The second item
read ‘‘Would you say your job reduce the time you can
spend with the family’’. The third item read ‘‘Would you
say problems at work make you irritable at home’’. Each
Table 2 Overview of the Swedish universities scales of personality (SSP) (descriptions after Gustavsson et al. 2000) and the job content
questionnaire (JCQ)
Description. Examples of people with high scores No.
items
Chronbachs’s
a
Pearson
correlation
with age
Point bi-serial
correlation
with gender
SSP dimension and scale
Neuroticism
Somatic trait anxiety Autonomic disturbances, restless, tense 7 0.67 -0.04 -0.16*
Psychic trait anxiety Worrying, anticipating, lacking-self confidence 7 0.78 -0.02 -0.21**
Stress susceptibility Easily fatigued, feeling uneasy when urged to speed up 7 0.63 0.12 -0.14*
Lack of assertiveness Lacks ability to speak up and to be self-assertive in
social situations
7 0.77 0.08 -0.06
Embitterment Unsatisfied, blaming and envying others 7 0.74 -0.13* 0.02
Mistrust Suspicious, distrusting peoples motives 7 0.79 0.05 -0.05
Aggressiveness
Social desirability
a
Socially conforming, friendly, helpful 7 0.63 -0.05 -0.13*
Trait irritability Irritable, lacking patience 7 0.78 -0.13 0.01
Verbal trait aggression Getting into arguments, berating people when annoyed 7 0.68 -0.25** 0.11
Physical trait aggression Getting into fights, starts fights, hits back 7 0.84 -0.22** 0.24**
Extraversion
Impulsiveness Acting on the spur of the moment, non-planning,
impulsive
7 0.59 -0.18** 0.02
Adventure seeking Avoiding routine, need for change and action 7 0.78 -0.26** 0.01
Detachment
a
Avoiding involvement in others, withdrawn, ‘schizoid’ 7 0.65 0.18** 0.02
JCQ dimension and scale
Psychological job demands
Total job demand score Perceiving time pressure and a need to work fast, hard
and concentrated
9 0.71 -0.05 -0.04
Job control
Total job control score Perceiving the possibility for personal development in
work and to exert influence over work
9 0.79 0.16* 0.06
Skill discretion score Perceiving the work to be variable, creative and flexible 6 0.76 0.20* 0.00
Decision authority score Perceiving the possibility to exert influence over work 3 0.52 0.08 0.11
Job support
Total job support score Perceiving support from managers and co-workers 8 0.70 -0.06 -0.08
Support from managers Perceiving that the manger listens and facilitates work
and cooperation
4 0.76 0.04 -0.09
Support from co-workers Perceiving that the co-workers are friendly, interested,
and facilitates work
4 0.77 -0.14* -0.02
In the point bi-serial correlations, women are coded 0 and men 1
* p \ 0.05; ** p \ 0.01
a
Subscale with negative loading on factor
392 Motiv Emot (2009) 33:387–399
123
item was responded to on a three point scale indicating
degree of agreement: 0 = not at all, 1 = to some degree,
2 = to a large degree. The items were dichotomized
according the substantive meaning of response alternatives
(0 = not at all to some degree, 1 = to a large degree).
Procedure and questionnaire evaluation
The questionnaire was administered during work hours on
location at the work sites and under supervision of research
personnel. Participants were moved from the production
area and seated in small groups (typically between 5 and 15
persons at the time) in a comfortable office environment.
The participants were coached to read the instructions
carefully and to take their time and give individual
answers. It was emphasized that there were no right or
wrong answers, and that they should be as honest as pos-
sible. The examiners had no information about the partic-
ipant’s concurrent health status or whether a specific
person was bullied or not. Communication between par-
ticipants, other than communication concerning factual
matters such as sharing of information concerning work
schedules and vacations, were politely discouraged. Most
participants (94%) completed the questionnaire within half
an hour to 2 h time. The evaluation showed that the
questionnaire was well approved of and that most partici-
pants were in a good mood when completing it. Apart from
the fact that a relatively larger proportion of bullied persons
reported being in a poor mood when completing the
questionnaire (Pearson v
2
-test, p = 0.009; 8, 10 and 36%,
respectively), there were no systematic differences between
referents, witnesses and bullied persons.
Results
The statistical software SPSS 15.0 was used for all analyses
and p values\0.05 were considered statistically significant
(two-tailed tests).
Swedish universities scales of personality
The statistical analyses showed ANOVA group differ-
ences primarily on the scales within the Neuroticism
dimension (Table 3). Pair-wise post-hoc analyses with
the Least Significant Difference procedure (LSD)
showed that bullied persons had elevated scores on the
somatic trait anxiety, psychic trait anxiety, stress sus-
ceptibility, embitterment and mistrust scale scores when
compared with the referent group (all p values \0.027).
When bullied persons were compared with witnesses the
pattern, with the exception of the somatic trait anxiety
score, was repeated. As regards personality trait scores
in the aggressiveness dimension, ANOVA group differ-
ences were only found for trait irritability (p = 0.005).
Pair-wise post-hoc analyses with the LSD procedure
showed that bullied persons had elevated scores when
compared to witnesses and referents (p = 0.002). As
regards scores in the Extraversion dimension, ANOVA
group differences were only found for the impulsiveness
score (p = 0.001). Pair-wise post-hoc analyses with the
LSD procedure showed that both bullied persons and
witnesses had elevated scores when compared to the
referents (p values \0.010). The results including effects
size estimates (partial eta squares, g
p
2
)andpresentation
in T-scores are presented in Table 4. Adjusting the
results for age and gender resulted in only very minor
changes in the effect estimates and p values (data not
shown).
The job content questionnaire
The statistical analysesshowed noANOVA group differences
on any of the three JCQ dimensions: psychological demands
(p = 0.096), job control (p = 0.267), and job support
(p = 0.464; Table 3). Similarly there were no ANOVA group
differences on any of the sub-dimensions. Adjusting the
results for age and gender resulted in only very minor changes
in the effect estimates and p values (data not shown).
Personal satisfaction with job performance
Fisher’s exact probability tests showed that a larger pro-
portion of bullied persons (50%) were unsatisfied with their
ability to maintain a good relationship with their co-
workers when compared with both witnesses (17%) and
non-bullied work colleagues (20%, p = 0.032 and 0.018,
respectively). A smaller proportion of witnesses (61%)
were satisfied with their quality of work when compared
with the non-bullied work colleagues (80%, p = 0.036)
and bullied persons (86%, p = 0.165). A larger proportion
of witnesses (87%) were satisfied as regards their ability to
solve problems in their work when compared with the non-
bullied work colleagues (66%, p = 0.021) and bullied
persons (71%, p = 0.231). Bullied persons, witnesses and
non-bullied work colleagues were equally satisfied with the
quantity of work they performed (80, 70, and 68%,
respectively, p value’s [0.7).
Work and family conflict: Spillover from work to home
life
Fisher’s exact probability tests showed that a larger pro-
portion of bullied persons (46%) reported that problems at
work made them annoyed at home when compared with
non-bullied work colleagues (7%, p \ 0.001) and
Motiv Emot (2009) 33:387–399 393
123
witnesses (13%, p = 0.046). There was no statistically
significant difference between the proportions of bullied
persons (30%), witnesses (18%) and non-bullied work
colleagues (15%) that reported that work had a large
impact on the amount of energy available to deal with
problems at home (p = 0.422 and 0.130, respectively).
There was no difference between the proportions of bullied
persons (23%), witnesses (14%) and non-bullied work
colleagues (11%) who reported that work had a large
impact on the possibilities to spend time with the family
(p = 0.196 and 0.657, respectively).
Discussion
The present study investigated whether persons in active
work and who reported being bullied, or being witnesses to
bullying, showed deviations on a personality test when
compared with their non-bullied work colleagues. In addi-
tion, we explored whether the bullied persons and the wit-
nesses to bullying perceived their work and their ability to
perform at work differently and, likewise, whether they
perceived more conflicts between work and family. By doing
this, the present study contributes with new knowledge to a
Table 3 Swedish universities scales of personality (SSP) and the job content cuestionnaire (JCQ) mean crude scores across groups
Referents
(N = 202)
Witness
(N = 31)
Bullied
(N = 14)
ANOVA
a
Post-hoc t test
Group Witness vs.
reference
Bullied vs.
reference
Wittness vs.
bullied
M SD M SD M SD p g
p
2
ppp
SSP dimension and scale
Neuroticism (1–4)
Somatic trait anxiety 1.90 0.52 2.10 0.56 2.33 0.54 0.004 0.05 0.056 0.004 0.195
Psychic trait anxiety 1.98 0.54 2.05 0.62 2.75 0.68 \0.001 0.09 0.563 \0.001 \0.001
Stress susceptibility 2.16 0.44 2.06 0.54 2.53 0.52 0.006 0.04 0.227 0.005 0.002
Lack of assertiveness 2.18 0.55 1.99 0.58 2.53 0.84 0.014 0.04 0.084 0.027 0.004
Embitterment 1.89 0.51 2.01 0.53 2.60 0.53 \0.001 0.10 0.227 \0.001 0.001
Mistrust 2.00 0.54 1.97 0.50 2.80 0.61 \0.001 0.11 0.768 \0.001 \0.001
Aggressiveness (1–4)
Social desirability
b
2.91 0.44 2.80 0.55 3.00 0.56 0.352 – – – –
Trait irritability 2.13 0.56 2.31 0.65 2.62 0.80 0.005 0.04 0.121 0.003 0.102
Verbal trait aggression 2.07 0.50 2.25 0.58 2.18 0.75 0.167 – – – –
Physical trait aggression 2.04 0.62 2.21 0.75 2.18 0.87 0.352 – – – –
Extraversion (1–4)
Impulsiveness 2.18 0.46 2.45 0.43 2.51 0.45 0.001 0.06 0.003 0.009 0.660
Adventure seeking 2.31 0.58 2.43 0.56 2.57 0.73 0.164 – – – –
Detachment
b
2.19 0.44 2.21 0.58 2.41 0.55 0.253 – – – –
JCQ dimension and scale
Psychological job demands (1–4)
Total job demand score 2.48 0.38 2.61 0.42 2.62 0.37 0.096 – – – –
Job control (1–4)
Total job control score 2.37 0.47 2.28 0.53 2.57 0.41 0.267 – – – –
Skill discretion score 2.34 0.51 2.23 0.59 2.64 0.46 0.098 – – – –
Decision authority score 2.39 0.56 2.32 0.60 2.50 0.47 0.614 – – – –
Job support (0–4)
Total job support score 1.71 0.43 1.66 0.46 1.58 0.38 0.464 – – – –
Support from managers 1.48 0.62 1.48 0.50 1.46 0.67 0.996 – – – –
Support from co-workers 1.95 0.50 1.87 0.61 1.70 0.51 0.212 – – – –
g
p
2
partial eta squared
a
For the SSP, the degrees of freedom and F value for the ANOVA F tests were in the following ranges: F (2, 237–240) = 1.05–14.42; For the
JCQ, the degrees of freedom and F value for the ANOVA F tests were in the following ranges. F (2, 240–242) = 0.004–2.371
b
Subscale with negative loading on factor
394 Motiv Emot (2009) 33:387–399
123
poorly studied area in bullying research and gives new
insights as to how to understand and characterize witnesses.
It was observed that the group of bullied persons had profiles
of elevated scores on the neuroticism scales as well as higher
trait irritability and impulsiveness scores when compared
with their non-bullied work colleagues. Witnesses to bully-
ing had higher impulsiveness scores than non-bullied but did
not otherwise markedly differ from non-bullied work col-
leagues. When repeating the statistical analyses with the
women only, the same pattern was essentially found, but the
differences were more pronounced.
Our observations of bullied persons with higher scores on
neuroticism related traits agree well with previous studies
(Gandolfo 1995; Glasø et al. 2007; Hansen et al. 2006;
Matthiesen and Einarsen 2001; Rammsayer et al. 2006).
Since the neuroticism related scales in the SSP entails
question about a lack of assertiveness, that is, an ability to
speak up and be self-assertive in social situations, the results
also harmonize with Coyne et al. (2000) and Moreno-
Jimenez et al. (2007) who both investigated assertiveness.
However, a complication with the comparison is that the
SSP, MMPI-2, ICES and NEO-PI inventories have been
developed for different purposes and with different theo-
retical point of departures. Hence to avoid flawed compari-
sons appropriate caution needs to be exercised when
comparing results across studies. For example, the higher
impulsiveness scores (belonging to the extraversion
dimension) may, at first glance, seem to contradict previous
observations regarding bullied people scoring lower than
reference groups on the extraversion dimension in the 4-
dimensional ICES Personality Inventory as used in the Irish
studies (Coyne et al. 2000, 2003). It seems, also, to contrast
with studies not showing any link between bullying and
extraversion in the 5-dimensional NEO-PI inventory as used
in the German study (Rammsayer et al. 2006). However, this
difference is resolved when reading the more detailed scale
descriptions for the ICES and SSP inventories, which sug-
gest it would be more appropriate to compare the SSP
impulsiveness score with the ICES stability (neuroticism)
score. Hence, a robust classification system with three major
dimensions is not always easily compatible with more ‘‘fine
grained’’ approaches using four or five dimensions.
Despite the difficulties of comparing across studies, the
bullied person’s higher scores on neuroticism related scales
seem to be a readily reproducible finding. Another con-
sistent observation across studies is the fairly distinct
deviations in personality trait scores between bullied and
non-bullied norm groups. For example, in all three MMPI-
2 studies previously mentioned many scale scores were in
fact above 65 T-scores, which indicate, according to
MMPI-2 standards, distinct psychological problems or
pathology (Gandolfo 1995; Girardi et al. 2007; Matthiesen
and Einarsen 2001). Victims of bullying were, moreover,
more likely to score at the lower and higher ends of the
distribution in the ICES Personality Inventory (Coyne et al.
2000). In this case, the mean score of the bullied group
typically diverged from the non-bullied group mean score
with at least one standard deviation (T-score of at least 60).
Table 4 Swedish universities scales of personality (SSP) T-scores in the groups of reference persons, witnesses and bullied
SSP dimension and scale Referents (N = 202) Witness (N = 31) Bullied (N = 14)
Mean Median 25–75th perc. Mean Median 25–75th perc. Mean Median 25–75th perc.
Neuroticism
Somatic trait anxiety 49.09 48 43–56 52.84 51 47–57 56.99 56 53–65
Psychic trait anxiety 49.10 49 41–55 50.20 49 44–55 62.18 62 52–73
Stress susceptibility 49.84 49 43–56 47.55 46 40–59 57.64 59 49–65
Lack of assertiveness 50.07 49 44–57 46.77 47 42–49 56.13 61 43–65
Embitterment 48.96 48 43–56 51.26 51 43–56 62.17 60 54–73
Mistrust 49.26 49 42–54 48.71 50 39–55 63.20 64 54–72
Aggressiveness
Social desirability
a
50.17 52 46–55 47.86 49 41–55 52.17 51 42–65
Trait irritability 49.15 49 45–54 52.19 49 46–60 57.44 61 49–69
Verbal trait aggression 49.44 51 43–56 52.98 51 47–60 51.69 47 41–65
Physical trait aggression 49.56 49 42–55 52.08 52 40–60 51.73 50 38–61
Extraversion
Impulsiveness 48.90 48 42–54 54.64 54 49–59 56.04 56 47–67
Adventure seeking 49.47 49 44–56 51.66 52 47–57 53.98 55 43–62
Detachment
a
49.69 49 42–58 50.08 52 42–57 54.27 53 45–61
a
Subscale with negative loading on factor
Motiv Emot (2009) 33:387–399 395
123
Similar distinct deviations in personality trait scores were
also observed in our study. The largest differences between
groups were observed on the mistrust scale (median T-
score 64) for which the bullied groups’ lower quartile score
(25th perc.) was similar to the highest quartile score (75th
perc.) in the group of non-bullied work colleagues. Hence,
irrespective of personality trait inventory, the magnitude of
difference between bullied persons and non-bullied persons
reports seems to be quite large.
Perceptions of job demands, job control and job support
Contrary to what we expected, bullied persons, witnesses to
bullying and their non-bullied work colleagues reported
fairly similar levels of psychological job demands, job
control and job support. As with Hauge et al. (2007) there
was a non-statistically significant tendency for witnesses
and bullied persons to report higher job demands than non-
bullied work colleagues. Regarding job control scores, it
was observed that witnesses had the lowest score of all
three groups on the subscale skill discretion. A lower score
on the skill discretion scale indicate a tendency to perceive
work as less varied, creative and flexible. In contrast to our
previous study (Hansen et al. 2006) there were no differ-
ences as regards job support scores. In fact all three groups
reported fairly low job support. This possibly reflects the
machine paced work that did not necessitate constant
supervision or dialog with leaders and work colleagues. In
short, being bullied or being a witness to bullying seem not
to have a large impact on the perception of job demands,
job control, and job support or vice versa.
Perceptions of job performance, ability to keep good
social relations with co-workers and conflict between
work and family
That a larger proportion of bullied persons were less sat-
isfied with their ability to maintain good relationships with
co-workers and that they more often experienced problems
at work leading to irritability at home, underlines that being
bullied is not an isolated experience and some bullied
persons have concerns that extend beyond the work con-
text. No similar pattern was observed among the witnesses
who were, instead, less satisfied with the quality of their
work. That proportionally more witnesses were less satis-
fied with the quality of the work they performed seems,
however, to conform with their tendency to report lower
skill discretion. Keeping this in mind and considering that
witnesses also reported that they were more satisfied with
their ability to solve work related problems than both
bullied persons and non-bullied work colleagues, it seems
reasonable to suggest that witnesses may be more inclined
to see both problems and solutions. This characteristic,
united with their extravert personalities, may possibly
facilitate the detection of work place bullying.
Differential responses and the meaning of higher
neuroticism scores
Despite the marked group differences one must not over-
look that there is still an overlap between the three group’s
distributions of personality traits scores. Not all bullied
persons show elevated personality trait scores, nor did they
all express that they were unsatisfied with the ability to
maintain good relationships with co-workers, or felt that
problems at work made them irritable at home. Similarly,
not all witnesses displayed high impulsiveness scores, were
unsatisfied with their quality of work, or claimed that they
were good at solving work related problems. Obviously,
this raises the question of what way personality traits, or
personal dispositions, contribute to the understanding of
the bullying phenomenon. That some bullied persons seem
to be more emotionally imbalanced could indicate that they
are sensitive, more vulnerable individuals. Indeed, this
view is partly supported by our background data, which
showed that bullied persons reported poorer health and
more often acknowledged chronic diseases, trouble after
accidents, handicap or some other weakness. Whether this
implies that the bullied persons were more sensitive from
the start or that they had developed their sensitivity during
the course of the bullying, can not be determined with the
present study design.
However, focusing on social dynamics and the ‘‘here-
and-now’’ impact of personality traits instead, the present
pattern of personality trait differences opens up for some
interesting interpretations. When considering the generally
higher scores on neuroticism related traits, it seems con-
ceivable that bullied persons might be more likely to
interpret acts that they are exposed to as negative and ego
threatening (Beck et al. 1985). If this is the case then the
bullied person’s anxiety laden personality and negative
cognitions might constitute a barrier for attempts that aim
to neutralize the bullying process, by altering the tone in
the social interaction. Moreover, knowing that behaviour
may be evoked by certain behaviours and features other
group members display (e.g. scents, sounds, colours and
movements, etc.), it also seems conceivable that person-
ality might impact on the social dynamics in a more pas-
sive way (Tinbergen 1951). People who are more bitter,
mistrusting, irritable and impulsive are, perhaps, preferred
targets due to their tendencies to be more emotionally
reactive. In other words, the bullied person’s personality
and associated behaviours could be expected to trigger the
perpetrator to conduct negative acts which are subse-
quently rewarded in terms of receiving an emotional
reaction.
396 Motiv Emot (2009) 33:387–399
123
Bullying and health
Although not of primary interest in the present paper, it is
noteworthy that the bullied group reported higher levels of
current mental distress, and perceived their current, and
past health (referring to the last 12-months), to be worse
than their non-bullied work colleagues. With respect to the
higher trait anxiety scores and current mental distress
scores (i.e. GHQ-12 scores) this observation fits well with
our previous study which used the identical bullying defi-
nition and focused on how bullying was related to mental
health outcomes and cortisol secretion (Hansen et al.
2006). The results also fit well with a Spanish study that
reported associations between bullying as measured by the
Work Place Bullying Questionnaire and the 28-item ver-
sion of GHQ (Moreno-Jime
´
nez et al. 2007). Since high trait
anxiety implies a personal inclination to express negative
affect it is not very strange that a larger proportion of the
bullied persons reported poorer mood when completing the
questionnaire. As mentioned in the introduction, person-
ality traits and corresponding short-term emotional and
affective states are expected to be correlated. The way to
disentangle traits from states in questionnaires is primarily
achieved by phrasing the questions so as to refer to either
typical and habitual behaviours, feelings and thinking (i.e.
traits); or, to phrase the questions so as to refer to, current
or right-here-and-now behaviour, feelings and thinking (i.e.
states).
Methodological considerations
Firstly, since the aim was to describe and estimate the
degree of differences between well-defined groups of par-
ticipants, it is clear that the cross-sectional design is ade-
quate. However, a more in-depth analysis of the effect of
bullying on personality or vice versa will need to be studied
in a longitudinal design.
Secondly, the identification procedure of bullied people
imposes some important interpretative limitations. Because
the QPS-Nordic questions did not ask for duration, or
provision of any detailed information regarding frequency
or intensity, we do not know whether the participants have
been exposed to bullying for 3, 6, 12, 24 months or longer
and neither do we know the emotional intensity of the
experience as perceived by the bullied individuals. Simi-
larly, we did not ask the participants whether they them-
selves had bullied someone or facilitated bullying.
Thirdly, a concern is the small study sample and
unbalanced group sizes, factors which theoretically would
make it more difficult to statistically reject the stated null-
hypothesis (Campbell et al. 1995). Due to the large dif-
ferences in mean scores between groups this turned out to
be less of a problem, and it may be noted that the observed
proportion of bullied fairly accurately mirrored the base
rate of bullying in the population, which often is estimated
to be between 4 and 8%. Yet, the small sample makes it
impossible to confirm previous interesting observations
suggesting a possible clustering of personality traits among
bullied persons (Glasø et al. 2007; Matthiesen and Einarsen
2001). At the same time, the underlying message that not
all bullied persons show personality deviations is also
evident in the present study. In view of the small study
sample it is recommended that caution be exercised when
interpreting the results. Nonetheless, the fact remains that
both statistically significant results and non-statistically
significant results seem to follow a rather coherent pattern
that fits with theory increase the trustworthiness of the
results.
Finally, a related concern is that the many statistical
comparisons seem to call for a downwards adjustment of
the p value before declaring statistical significance in order
to avoid making type I errors. However, a downwards
adjustment of the p value will lead to an increase in type II
errors, i.e. negatively affect the possibilities of finding true
effects. Due to the explorative nature of the present study
we decided on a liberal approach. Accordingly it is rec-
ommended to temper the enthusiasm over single signifi-
cances and instead focus on the overall pattern of results
and to think over the potential for reproducibility. From
this perspective it is, however, somewhat consoling, as
mentioned above, that the results seem to follow a rather
coherent pattern that entails at least partial reproduction of
previous findings.
Conclusion
In conclusion, bullied persons’ responses to the SSP
inventory indicate that this group has a self-image that is
dominated by worry, mistrust and embitterment, as well as
irritability and impulsiveness. In addition, bullied persons
feel that work causes problems at home and they more
often doubt their own ability to maintain good relationships
with co-workers. Witnesses describe themselves as more
impulsive and sceptical toward their own work perfor-
mance, but seem content with regard to their ability to
solve problems at work. From a practical perspective, the
results underscore the point that workplace practitioners,
consultants and managers who deal with bullied persons
should be prepared to encounter people who are emotion-
ally imbalanced and have a negative self-image. Both
might pose a barrier for action. Accordingly, when dealing
with bullying at work, the need for ego-supportive actions
should be considered in conjunction with more organisa-
tional, collective oriented action.
Motiv Emot (2009) 33:387–399 397
123
Acknowledgments This study was supported by grants from the
Swedish Medical Research Council, the Swedish Council for Work
Life and Social Research, AFA insurance, the Medical Faculty of
Lund University, and the County Councils of Southern Sweden. Ms.
Lothy Granquist and Ms. Anita Ohlsson, are acknowledged for their
efforts in collecting data and other assistance. Secretary Bodil Holst
and senior researcher Bryan Cleal provided instructive lingual
remarks. The cooperation of participants is also gratefully
acknowledged.
References
Beck, A. T., Emery, G., & Greenberg, R. L. (1985). Anxiety disorders
and phobias. A cognitive perspective. USA: Basic Books.
Bjorkqvist, K., Lagerspetz, K. M. J., & Kaukiainen, A. (1992). Do
girls manipulate and boys fight—developmental-trends in regard
to direct and indirect aggression. Aggressive Behavior, 18,
117–127. doi:10.1002/1098-2337(1992)18:2\117::AID-AB248
0180205[3.0.CO;2-3.
Bjorkqvist, K., Osterman, K., & Hjeltback, M. (1994). Aggression
among university employees. Aggressive Behavior, 20, 173–184.
doi:10.1002/1098-2337(1994)20:3\173::AID-AB2480200304[
3.0.CO;2-D.
Campbell, M. J., Julious, S. A., & Altman, D. G. (1995). Estimating
sample sizes for binary, ordered categorical, and continuous
outcomes in two group comparisons. BMJ (Clinical Research
Ed.), 311, 1145–1148.
Chandola, T., Martikainen, P., Bartley, M., Lahelma, E., Marmot, M.,
Michikazu, S., et al. (2004). Does conflict between home and
work explain the effect of multiple roles on mental health? A
comparative study of Finland, Japan, and the UK. International
Journal of Epidemiology, 33, 884–893. doi:10.1093/ije/dyh155.
Coyne, I., Seigne, E., & Randall, P. (2000). Predicting workplace
victim status from personality. European Journal of Work and
Organizational Psychology, 9, 335–349. doi:10.1080/13594
3200417957.
Coyne, I., Smith-Lee Chong, P., Seigne, E., & Randall, P. (2003). Self
and peer nominations of bullying: An analysis of incident rates,
individual differences, and perceptions of the working environ-
ment. European Journal of Work and Organizational Psychol-
ogy, 12, 209–228. doi:10.1080/13594320344000101.
Dallner, M., Lindstro
¨
m, K., Elo, A.-L., Skogstad, A., Gamberale, F.,
& Hottinen, V. (2000). Anva
¨
ndarmanual fo
¨
r QPS nordic:
Fra
˚
geformula
¨
r om psykologiska och sociala faktorer i arbetsli-
vet utprovat i Danmark, Finland, Norge och Sverige. [User’s
manual for the QPS Nordic: Questionnaire on psychological and
social factors in the working life evaluated in Denmark, Finland,
Norway and Sweden.] Stockholm: Arbetslivsrapport fra
˚
n Arbet-
slivsinstitutet, 2000, 19.
Djurkovic, N., McCormack, D., & Casimir, G. (2005). Neuroticism
and the psychosomatic model of workplace bullying. Journal of
Managerial Psychology, 21, 73–88. doi:10.1108/02683940610
643224.
Einarsen, S. (2000). Harassment and bullying at work: A review of
the Scandinavian approach. Aggression and Violent Behavior, 5,
379–401. doi:10.1016/S1359-1789(98)00043-3.
Einarsen, S., Raknes, B. I., Matthiesen, S. B., & Hellesoy, O. H.
(1990). Mobbism in the workplace—incidence, manifestations,
consequences. Nordisk Psykologi, 42, 294–298.
Einarsen, S., Raknes, B. I., Matthiesen, S. B., & Hellesoy, O. H.
(1996). Bullying at work and its relationships with health
complaints—moderating effects of social support and personal-
ity. Nordisk Psykologi, 48
, 116–137.
Eysenck, H. J., & Eysenck, M. (1985). Personality and individual
differences: A natural science approach. New York: Plenum
Press.
Gandolfo, R. (1995). Mmpi-2 profiles of workers compensation
claimants who present with complaints of harassment. Journal of
Clinical Psychology, 51, 711–715. doi:10.1002/1097-4679
(199509)51:5\711::AID-JCLP2270510517[3.0.CO;2-R.
Girardi, P., Monaco, E., Prestigiacomo, C., Talamo, A., Ruberto, A.,
& Tatarelli, R. (2007). Personality and psychopathological
profiles in individuals exposed to mobbing. Violence and
Victims, 22, 172–188. doi:10.1891/088667007780477320.
Glasø, L., Matthiesen, S. B., Nielsen, M. B., & Einarsen, S. (2007).
Do targets of workplace bullying portray a general victim
personality profile? Scandinavian Journal of Psychology, 48,
313–319. doi:10.1111/j.1467-9450.2007.00554.x.
Goldberg, D., & Williams, P. (1988). User’s guide to the general
health questionnaire. Oxford: NFER-Nelson.
Gustavsson, J. P., Bergman, H., Edman, G., Ekselius, L., von
Knorring, L., & Linder, J. (2000). Swedish universities scales of
personality (SSP): Construction, internal consistency and nor-
mative data. Acta Psychiatrica Scandinavica, 102, 217–225. doi:
10.1034/j.1600-0447.2000.102003217.x.
Hansen, A. M., Hogh, A., Persson, R., Karlson, B., Garde, A. H., &
Orbaek, P. (2006). Bullying at work, health outcomes, and
physiological stress response. Journal of Psychosomatic
Research, 60, 63–72. doi:10.1016/j.jpsychores.2005.06.078.
Hauge, L. J., Skogstad, A., & Einarsen, S. (2007). Relationships
between stressful work environments and bullying: Results of a
large representative study. Work and Stress, 21, 220–242. doi:
10.1080/02678370701705810.
Karasek, R., Brisson, C., Kawakami, N., Houtman, I., Bongers, P., &
Amick, B. (1998). The job content questionnaire (JCQ): An
instrument for internationally comparative assessments of psy-
chosocial job characteristics. Journal of Occupational Health
Psychology, 3, 322–355. doi:10.1037/1076-8998.3.4.322.
Matthiesen, S. B., & Einarsen, S. (2001). MMPI-2 configurations
among victims of bullying at work. European Journal of Work
and Organizational Psychology, 10, 467–484. doi:10.1080/
13594320143000753.
Mikkelsen, E. G., & Einarsen, S. (2002). Relationships between
exposure to bullying at work and psychological and psychoso-
matic health complaints: The role of state negative affectivity
and generalized self-efficacy. Scandinavian Journal of Psychol-
ogy, 43, 397–405. doi:10.1111/1467-9450.00307.
Moreno-Jime
´
nez, B., Rodrı
´
guez-Munoz, A., Moreno, Y., & Garrosa,
E. (2007). The moderating role of assertiveness and social
anxiety in workplace bullying: Two empirical studies. Psychol-
ogy in Spain, 11, 85–94.
Netemeyer, R. G., Boles, J. S., & McMurrian, R. (1996). Develop-
ment and validation of work-family conflict and family-work
conflict scales. The Journal of Applied Psychology, 81, 400–410.
doi:10.1037/0021-9010.81.4.400
.
Nordander, C., Ohlsson, K., Balogh, I., Hansson, G. A., Axmon, A.,
Persson, R., et al. (2007). Gender differences in workers with
identical repetitive industrial tasks: Exposure and musculoskel-
etal disorders. International Archives of Occupational and
Environmental Health, 81, 939–947. doi:10.1007/s00420-007-
0286-9.
Parkins, I. S., Fishbein, H. D., & Ritchey, P. N. (2006). The influence
of personality on workplace bullying and discrimination. Journal
of Applied Social Psychology, 36, 2554–2577. doi:10.1111/
j.0021-9029.2006.00117.x.
Persson, R. (2002). Trait anxiety as a determinant of psychological
test results. Lund: Lund University.
Rammsayer, T., Stahl, J., & Schmiga, K. (2006). Basic personality
dimensions and stress-related coping strategies in victims of
398 Motiv Emot (2009) 33:387–399
123
workplace bullying. Zeitschrift fu
¨
r Personalpsychologie, 5, 41–
52. doi:10.1026/1617-6391.5.2.41.
Roland, E. (2002). Aggression, depression, and bullying others.
Aggressive Behavior, 28, 198–206. doi:10.1002/ab.90022.
Schalling, D., Asberg, M., Edman, G., & Oreland, L. (1987). Markers
for vulnerability to psychopathology—temperament traits asso-
ciated with platelet mao activity. Acta Psychiatrica Scandinav-
ica, 76, 172–182. doi:10.1111/j.1600-0447.1987.tb02881.x.
Seigne, E., Coyne, I., Randall, P., & Parker, J. (2007). Personality
traits of bullies as a contributory factor in workplace bullying:
An exploratory study. International Journal of Organization
Theory and Behavior, 10, 118–132.
Tinbergen, N. (1951). The study of instinct. Oxford: Oxford Univer-
sity Press.
Motiv Emot (2009) 33:387–399 399
123