Acta Pædiatrica ISSN 0803–5253

REGULAR ARTICLE

Poor performance in physical education – a risk factor for bully

victimization
A case–control study
Susanne Bejerot ([email protected])1, Johan Edgar2, Mats B Humble1,3
1.Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
2.Faculty of Health Sciences, Linköping University, Linköping, Sweden
3.Department of Neuroscience, Psychiatry, Uppsala University, Uppsala, Sweden

Keywords ABSTRACT
Attention-deficit disorder with hyperactivity, Autistic
Aim: Poor social skills are a risk factor for becoming bullied, which could explain why
disorder, Bullying, Motor skills, Peer victimization
this frequently occurs to children with autism spectrum disorders (ASD) and attention-
Correspondence
deficit hyperactivity disorder (ADHD). Poor social skills tend to coexist with clumsiness.
Susanne Bejerot, Ph.D., M.D., KogNUS, VUB-team,
Northern Stockholm Psychiatry, St Göran Hospital, According to a pilot study, poor performance in physical education (PE) was correlated
SE-112 81 Stockholm, Sweden. with bully victimization.
Tel: +46 (0)8 672 24 19 | Methods: Sixty-nine healthy university students reported performance in PE and
Fax: +46 (0)8 656 70 00 |
Email: [email protected] bully victimization in childhood. In addition, the participants responded to questionnaires
for ADHD and ASDs to assess personality traits related to increased risk for bully
Received
13 July 2010; revised 12 August 2010; victimization.
accepted 13 September 2010. Results: Below average performance in PE was a risk factor of being bullied in school
DOI:10.1111/j.1651-2227.2010.02016.x with an odds ratio of 3.6 [95% confidence interval: 1.23–10.5; p = 0.017]. Strong
correlations between poor performance in PE and long duration of victimization
(p = 0.007) and poor performance in PE and high frequency of victimization (p = 0.008)
were found. Autistic traits were related to performance below average in PE.
Conclusion: Poor motor skills are a strong risk factor for becoming bullied. Prevention
programmes that identify, protect and empower the clumsy children could be an important step
to avoid bullying of the most vulnerable children.

Any child could become bullied, but it occurs more often This may explain why children with attention-deficit
toward children who are fragile. Being bullied is associated hyperactivity disorder (ADHD) or autism spectrum disor-
with significantly poorer ratings of physical, social and emo- der (ASD) often are bullied (6,7). Both ADHD and ASD
tional functioning in adolescents (1) and has been suggested are neuropsychiatric conditions with an early onset, and
as a risk indicator for various mental disorders in adoles- both disorders are related to poor motor skills, i.e. these
cence (2). Bully victimization in childhood is related to children are often viewed as clumsy (8,9,10). It has been
long-term unemployment in adulthood (3). suggested that impaired motor development by itself is a
A lack of social skills has been associated with victims risk factor for later poor cognitive performance and
of bullying (4,5). Non-assertive behaviour, provocative behaviour problems (11,12). Interestingly, abnormal
behaviour (including fighting back) and withdrawal ⁄ movement patterns in infancy may be the first signs of a
solitary behaviour are indicative of poor social skills (5). disturbed brain function in ASD (13).
Parents to children with developmental coordination dis-
order (DCD) reported that their offspring often were bullied
Abbreviations (14). In our clinical practice, we have observed a strong
ADHD, Attention-deficit hyperactivity disorder; ASD, Autism association between motor clumsiness and bully victimiza-
spectrum disorder; ASRS, Adult ADHD self-report scale; CI, tion among adult psychiatric patients (data on file). To
Confidence interval; DCD, Developmental coordination disor- investigate whether this would stand true in the normal
der; HAGS, High-functioning autism ⁄ asperger syndrome global population, we conducted a pilot study. In a large group of
scale; PE, Physical education; RAADS-R, Ritvo autism and asper- adults, there was a strong association between poor perfor-
ger diagnostic scale revised.
mance in PE in school PE and bully victimization, as

ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 1

Bullying and performance in physical education Bejerot et al.

reported in retrospect (15). However, the study was limited Thereafter, 44 of the participants came in for a 30-min’
by the fact that the participants were anonymous audiences medical examination, when assessments with the Neurolog-
on lectures about psychiatric disorders, and their education ical Evaluation Scale (NES) (19) and the High-Functioning
level was not determined. Considering that learning disabil- Autism ⁄ Asperger Syndrome Global Scale (HAGS) (20)
ity has been associated with an increased risk of bully vic- were completed. NES was used to measure neurological soft
timization (and bullying) (16), cognitive abilities should signs, and HAGS was used to rule out prominent autistic
preferably be at least normal in the study population. traits in the participants of this study.
Accordingly, this study aims to explore the relationship Throughout the study, bullying was defined as being
between a history of poor motor skills in childhood and either physical or verbal, including spreading rumours and
bully victimization among high-performing individuals social exclusion, according to the revised definition given by
without diagnoses of ASD, ADHD, DCD or other psychiat- Olweus (21): ‘a student is being bullied when another stu-
ric disorders. Thus, only healthy high-performing univer- dent, or a group of students, say or do nasty and unpleasant
sity students were included to decrease confounding by things to him or her. It is also bullying when a student is
low intelligence (associated with poor motor skills) and teased repeatedly in a way he or she does not like or when
psychiatric disorders (associated with bully victimization). they are deliberately left out of things. But it is not bullying
However, we also wanted to investigate whether subthresh- when two students of about the same strength or power
old traits related to ASD or ADHD would increase the argue or fight. It is also not bullying when the teasing is
risk of bully victimization in this highly achieving group. done in a friendly and playful way.’

Measures
METHOD Questionnaire regarding self-conception, executive prob-
Participants lems, motor and academic skills and bully victimization
University students, recruited from three different campuses This questionnaire comprises the questions that were previ-
in Sweden, were included in this study. They were enrolled ously included in the pilot study (15), but a more detailed
through advertisements on university notice boards, and description of the definition of bullying according to Olw-
both bully victims and those who never had been bullied eus (21) was amended. Moreover, two questions about
were invited to participate. The study was briefly described hoarding were added to this study. Hoarding behaviours
as intended to increase knowledge about bullying, whereas tend to develop gradually and are prevalent in adults with
motor skills were not mentioned. In the advertisement, they neuropsychiatric disorders, obsessive-compulsive disorder
were also informed about inclusion criteria (age between 19 and schizophrenia. Thus, hoarding can be viewed as an
and 29 years) and exclusion criteria (any psychiatric diag- indicator of non-diagnosed psychiatric dysfunction. As we
nosis in child- or adulthood). The prerequisite was to anon- wanted to investigate subtle psychiatric traits associated
ymously fill in a web-based questionnaire. Optionally, they with poor social skills, probing for hoarding seemed rele-
could, thereafter, go through a brief medical examination. vant.
The participants were offered a cup of coffee and a lottery In addition to the questions that covered demographics,
ticket for their participation in the study. We were heading the following questions and response options were pre-
for 50 subjects to be seen in person, but only 44 of the 69 sented to the participants: (i) ‘Were you exposed to physical
subjects that met the inclusion criteria and had filled in the violence, or were you bullied and excluded in school (age
web questionnaire during the recruitment period came in 10–15)?’ (Never bullied; Exposed to physical violence;
for a physical assessment. Exposed to psychological violence; Exposed to both), (ii)
‘How often were you bullied in school?’ (Never; Once or
Procedures twice; 2–3 times a month; Weekly; Several times a week),
Following ethical approval from the Ethics Review Board, (iii) ‘If you were bullied in school, for how long did it last?’
Karolinska Institutet, Stockholm and written informed con- (Never bullied; A few weeks; A couple of months; For sev-
sent, the subjects filled out a web-based questionnaire set eral years; Throughout primary school), (iv) ‘Did you bully
comprising questions concerning demographic data, the other children in school? (Very much; A little; No), (v) Do
questionnaire concerning self-conception etc. (described you feel different from others? (Yes; No; I don’t know) (vi)
elsewhere), the Adult ADHD Self Report Scale (ASRS) (17) Do others have difficulties to understand you? (Yes; No; I
and the Ritvo Autism and Asperger Diagnostic Scale don’t know), (vii) Do you think it is difficult to know
Revised (RAADS-R) (18). The administration of these scales whether it is OK to pass on information that was told to
was not monitored, but each web entry of answer sets was you? (Yes; No; I don’t know), (viii) Were you physically
viewed for inconsistencies and timed. It took approximately stronger than other children between the ages 7 and 12?
40 min to respond to the questionnaire set. Codes were (Yes; No; Average; I don’t know), (ix) Were you regarded as
used to preserve confidentiality and entered together with academically talented in school? (Yes; No; Average), (x)
the responses into Open Office Calc spreadsheets. Informa- Were you regarded as talented in PE (i.e. motor smoothness,
tion about confidentiality was given to and signed by each well coordinated, good ball skills) (Yes; No; Average), (xi)
subject on the web site. The complete web questionnaire is Do you have difficulties to throw things away because you
available from the corresponding author on request. think you might need them in the future? (Very much; A

2 ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica

Bejerot et al. Bullying and performance in physical education

little; No; I don’t know), (xii) Do you hoard papers or items research team, three were removed as the logged time
to an extent that becomes a problem for yourself or others? of entries was too short (<5 min), and three were
(Very much so; A little; No; I don’t know). excluded because of age that did not meet the inclusion
criteria. In addition, another 20 response sets were
Neuropsychiatric rating scales incomplete and subsequently removed. No subjects were
The ASRS comprises 18 items and includes two subscales excluded for any other reasons. In all, 69 participants
measuring inattention and hyperactivity ⁄ impulsivity (17), provided complete answers and were included in the
and scores above 34 suggest ADHD. The Ritvo Autism study. Forty-four persons had the medical examination,
and Asperger Diagnostic Scale Revised (RAADS-R) whereas 25 simply responded to the web-based ques-
consists of 80 questions, where scores above 66 suggest tionnaire from their own computer.
ASD (22). This scale covers areas of social interaction, lan-
guage, sensory motor function and circumscribed interests Demographics
(18). Twenty-six participants studied medicine or nursing, seven-
teen studied technology, 11 studied social sciences, and
Soft neurological signs and autistic traits nine studied teaching (missing data in six subjects). Half the
The NES (19) was created to give a standardized approach group had a girl ⁄ boyfriend or was married. Forty-three sub-
to neurological findings in patients with schizophrenia but jects had at least one parent with university education. The
has also been used in other psychiatric populations where vast majority lived independent from their parents. See
deviances can be expected. The scale consists of 26 assess- Table 1.
ments and measures integrative sensory dysfunction, motor There were no differences in age, gender, academic per-
in-coordination and impaired sequencing of complex motor formance in school and type of university studies, or par-
acts, short-term memory, frontal release signs and abnor- ents’ educational level between bully victims and those who
malities in eye movements. The HAGS provides a global never had experienced being bullied.
score, based on the clinician’s observation, of social and
emotional reciprocity, social competence, rigidity, speech Academic and motor performance
and language, body posture, gestures, facial expression, eye The vast majority (n = 51) reported to have performed
contact and functional impairment (20). above average academically in school (between the age of
10 and 15 years), while only seven reported to have per-
Statistical analysis formed below. Thirty-five reported to have performed above
Those who reported having been bullied at least 2–3 times a average in PE, thirteen reported adequate performance,
month during at least a couple of months were defined as while 21 reported to have performed below average in their
‘bully victims’ in this study according to Olweus’ definition class. See Table 2.
(21). Those who had been only briefly bullied (i.e. only once
or twice or for a period shorter than a couple of months) History of bullying
were not regarded as ‘bully victims’ as the bullying did not Twenty-five subjects (36%) reported to have been bullied at
last. However, they were included in the analyses of bully least 2–3 times a month during at least a couple of months
duration and frequency. and thus defined as ‘bully victims’. In addition, another fif-
Categorical data were studied by means of Pearson chi- teen subjects were only briefly bullied, i.e. only once or
square tests. For cross-tabulation analyses, the Fisher exact twice, or for a period shorter than a couple of months. Three
test was used. Odds ratios with a 95% confidence interval reported that they had been bullied throughout compulsory
(95% CI) were calculated for dichotomous outcome vari- school, twenty had been bullied for at least several years,
ables (i.e. performance below average in PE or not, and a nine were bullied for several months, and eight were bullied
history of being bullied or not). Continuous data were anal- for a few weeks. Seventy-five per cent of this bullied group
ysed using t-tests or Mann–Whitney U-test. Spearman’s cor- reported having been subjected to psychological violence,
relation test was used for non-parametric analysis. The 2% to physical violence and 23% to both. Twenty-eight per
frequencies and duration of bullying were transformed into cent of the bully victims had been bullying others, compared
a 5-point Likert scale: (Never = 0; Once or twice = 1; 2–3 to 27% in the non-bullied group.
times a month = 2; Weekly = 3; Several times a week = 4) There was a strong positive correlation between duration
and (Never bullied = 0; A few weeks = 1; A couple of of being bullied and frequency of being bullied (r = 0.89,
months = 2; For several years = 3; Throughout primary p < 0.00001), and this relationship remained highly signifi-
school = 4). The significance level was set to p < 0.05. cant when the 29 subjects who never had experienced being
STATISTICA release 7 was used for all analyses. bullied were excluded from the calculation (v2 = 26.6;
df = 9, p = 0.0016).

RESULTS Bullying related to motor performance

Attrition Below average performance in PE, when compared to aver-
In total, 98 entries were made in the web-based ques- age or normal, was a risk factor for being bullied in school
tionnaire. Of these, three were trial runs made by the with an odds ratio of 3.59 (95% CI = 1.23–10.5) (v2 = 5.71;

ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 3

Bullying and performance in physical education Bejerot et al.

Table 1 Characteristics of the participants

Bully victim n = 25 Not bully victim n = 44 Pearson chi square df p

Gender, m ⁄ f (male ⁄ female) n (%) 5 (20) ⁄ 20 (45.5) 20 (80) ⁄ 24 (54.5) 4.47 1 0.03
Age, years (range, SDS) 24 (20–29 ± 2.3) 24 (19–29 ± 2.9)
Civil status, single n (%) 15 (60) 19 (43) 1.80 1 0.18
Type of education ⁄ Faculty*
Health (n) 12 14 2.69 3 0.44
Technology (n) 6 11
Philosophy (n) 2 9
Education (n) 3 6
University education in parent n (%) 13 (52) 30 (68) 1.78 1 0.18
Living independent from parents n (%) 25 (100) 38 (86) 3.73 1 0.05

SDS, standard deviation score; n, number.

*Missing data in 6 subjects.

Table 2 Cognitive and gross motor signs reported by participants

Bully victim n = 25 Not bully victim n = 44 Pearson
Questions* n (%) n (%) Chi-square df p

Achievements in theoretical education between age 17 (68) ⁄ 3 (12) 34 (77) ⁄ 4 (9) 0.25 1 0.62
10 and 15 years
Above average ⁄ below average
Achievements in physical education between age 10 9 (36) ⁄ 12 (48) 26 (59) ⁄ 9 (20) 5.53 1 0.019
and 15 years
Above average ⁄ below average
Do you think it is difficult to know whether it is OK to 8 (32) ⁄ 15 (60) 5 (11) ⁄ 38 (86) 5.08 1 0.024
pass on information that was told to you?
yes ⁄ no
Do you feel different from others? 20 (80) ⁄ 5 (20) 19 (43) ⁄ 22 (50) 7.28 1 0.007
yes ⁄ no
Do others have difficulties to understand you? 18 (72) ⁄ 5 (20) 9 (20) ⁄ 32 (73) 19.16 1 0.000012
yes ⁄ no
Do you have difficulties to throw things away because 14 (58) ⁄ 8 (33) 23 (54) ⁄ 18 (42) 0.34 1 0.56
you think you might need them in the future?
yes ⁄ no
Do you hoard papers or items to an extent that 2 (8) ⁄ 21 (88) 3 (7) ⁄ 41 (93) 0.08 1 0.78
becomes a problem for yourself or others?
yes ⁄ no

*Subjects responding ‘I don’t know’ or ‘neither’ are excluded from the analyses, while the per cent figure is based on all participants.

df = 1, p = 0.017). Among those who reported below aver- information given by others compared to the non-bullied
age PE performance, 57% were bully victims, compared to group. See Table 2.
26% among those who reported above average
motor performance (v2 = 5.53; df = 1, p = 0.019). There Neuropsychiatric measures and hoarding
was a strong correlation between poor PE performance and The ADHD symptoms were nearly identical in the two
a longer duration of victimization (r = )0.32, p = 0.007), groups; the bully victims received a medium score of 27 on
and between poor PE performance and frequency of the ASRS, compared to 26 in the non-bullied group. Autistic
being bullied (r = )0.31, p = 0.008). Reported muscle traits according to RAADS-R also did not differ between the
strength in childhood was equal in the bully and non-bullied bullied and non-bullied group; however, the bullied group
group. had numerically higher scores: the bully victims received
medium scores of 38 and the non-bullied group 26, respec-
Self-perception tively. Poor performance in PE was related to higher scores
Subjects who had experienced being bullied reported that on RAADS-R (Mann–Whitney U = 335, p = 0.03) and inat-
they felt different from others and sensed that other people tention according to the ASRS (Mann–Whitney U = 351,
have difficulties to understand them. Also, they reported p = 0.04). None of the 44 participants seen in person was
more problems in knowing whether it’s OK to pass on considered to have clear-cut autistic traits according to the

4 ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica

Bejerot et al. Bullying and performance in physical education

HAGS. There were no significant differences on the HAGS they were more often uncertain whether it is OK to
score, the NES score or hoarding behaviours between the pass on information, compared to the non-bullied sub-
bully victims and the other subjects, nor between those with jects. The first two of these statements were also
poor performance versus adequate performance in PE. strongly related to the total RAADS-R scores (Mann–
Whitney U = 334, p = 0.006; Mann–Whitney U = 333,
p = 0.0006; Mann–Whitney U = 301, p = 0.33). Accord-
DISCUSSION ing to our clinical experience, patients with ASD often
This study further supports the results from the previous endorse these three statements, which we have per-
pilot study, in which a strong correlation between bully vic- ceived as problems with mentalizing, i.e. poor abilities
timization and poor performance in PE was shown. The to make inferences about mental states in others. How-
present study, thus, suggests that poor motor skills predes- ever, this way of reasoning could likewise be a result of
tine the person to a more than threefold risk of becoming the bullying trauma with loss of confidence in others.
bullied, in subjects with a presumably high cognitive func- Albeit hoarding symptoms did not differ between the
tion and without any history of psychiatric illness. This bullied and the non-bullied group; these behaviours
would also imply that the risk of being a bully victim is fur- strongly correlated to both the ASRS and RAADS-R
ther increased in persons with impairments of both motor total scores (Mann–Whitney U = 292, p = 0.0009;
and cognitive skills. We also found that the children that Mann–Whitney U = 269, p = 0.02).
were persistently bullied are the very same children that Furthermore, we found an association between subtle
were frequently bullied in school. personality deviancies (as measured with rating scales for
It is highly unlikely that bullying can result in poor motor identifying ADHD and ASD traits) and poor motor skills
skills. However, there are several possible reasons for bul- among our participants. This could be expected, as motor
lied children to dislike PE other than poor motor skills: bul- clumsiness is common in childhood onset neuropsychiatric
lying frequently occurs in the locker room, and overweight disorders, and these traits are dimensionally distributed in
children are at increased risk of being bullied (23). Unfortu- the normal population. Possibly, the lack of statistical asso-
nately, we did not measure weight in this study; however, ciation between these personality traits and bully victimiza-
only two of the 44 participants seen in person impressed as tion could be attributed to a type II error, i.e. explained by
being overweight, and none of these had been bullied. This our small study sample, and that subtle problems with
does not support the strong finding by Lumeng (23) that motor skills are a more sensitive marker. Alternatively, poor
showed the odds of being bullied of 1.6 for children with motor skills do indeed exert an independent effect on the
overweight. This discrepancy could possibly be explained by propensity of becoming bullied.
a hesitance, because of embarrassment in young persons Evolutionary mechanisms may explain why individuals
with overweight, to participate in a study that includes a with poor social skills are bullied. Humans are ‘social
physical examination. The presumed embarrassment to animals’, i.e. our brains contain evolutionarily prepared
undress in front of others and poor self-esteem may hold a behavioural programmes related to social functions.
child back in the PE class and cause poor achievements and Such programs have presumably evolved through selec-
low grades. tion mechanisms, implying that those with inadequate
In this study, however, we did not inquire about the PE social skills are deselected. Bullying might perceivably
grades; instead, we specifically asked about coordination, represent one way in which this deselection is exerted.
smoothness and ball skills, i.e. gross motor function. The bully victims will be viewed as less attractive and
We do not believe that the poor motor skills per se consequently become less successful in finding a mate in
are necessarily the cause of the bully victimization. adulthood (26). These mechanisms for explaining bully-
Although good motor skills are associated with higher ing are currently neglected. Political correctness tells us
social status in adolescents, this truth applies mostly to that the risk of becoming bullied in childhood is a ran-
men (24), whereas most of the bully victims in this dom event, but everyone who has gone through school
study were women. Another possibility is that poor knows that this is not the full truth.
motor skills are closely related to poor social skills in We acknowledge the ethical dilemma in pointing out pre-
general, as has been shown in children with severe sumed biological markers, such as poor motor skills, as an
motor skills problem, i.e. DCD (25). Poor social skills important predictor of the risk of being bullied in child-
could be perceived as being different e.g. ‘weird’, ‘annoy- hood. Our conclusions could be criticized for being deter-
ing’ or ‘unpleasant’ in the eyes of other children. ministic and for ‘blaming the victim’. On the other hand,
Accordingly, we suggest that subtle signs of poor gross our findings imply that prevention programmes against bul-
motor skills may indicate an increased risk of also hav- lying could gain from focusing on children at risk if the
ing subtle signs of poor social skills and, consequently, proper risk factors are identified. Prevention programmes
running an increased risk of becoming bullied. This view that identify and empower the clumsy children could be an
is supported by differences in responses between the important step to avoid severe victimization of the most vul-
bully victims and the non-bullied participants. Those nerable children and protect them from developing a sense
who had been bullied sensed that others have difficulties of inferiority and shame. Individualized motor skill training
to understand them, they felt ‘different’ from others and could also be an option worth investigating; presumably,

ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 5

Bullying and performance in physical education Bejerot et al.

this could increase coordination and strength and indirectly ACKNOWLEDGEMENTS

increase self-assertiveness. The importance of emphasizing Thanks to Sead Omerov for assistance with the web-based
and stimulating adolescents’ strengths and talents, particu- questionnaire.
larly for children and adolescents with weaknesses in other
areas, has been underscored by others (24).
There are several limitations in this study. First, the par- CONFLICT OF INTERESTS
ticipants were all self-referred and may not fully correspond None.
to bully victims in the general population; moreover, only
64% were seen in person, and significantly more women
References
than men were bully victims. Second, the school charts that
could validate the PE and academic performances were not 1. Frisén A, Bjarnelind S. Health-related quality of life and bully-
retrieved; in other words, we relied on the information pro- ing in adolescence. Acta Paediatr 2010; 99: 597–603.
2. Kaltiala-Heino R, Rimpela M, Rantanen P, Rimpela A. Bullying
vided by the participants. However, there were no obvious
at school-an indicator of adolescents at risk for mental disor-
reasons why they would provide us with false information, ders. J Adolesc 2000; 23: 661–74.
as they had nothing to gain from doing so. Third, gross 3. Varhama LM, Björkqvist K. Relation between school bullying
motor function was not measured; we only measured so- during adolescence and subsequent long-term unemployment
called neurological soft signs with the NES. This instrument in adulthood in a Finnish sample. Psychol Rep 2005; 96:
is in our view too rough to detect subtle motor problems in 269–72.
healthy young adults, regardless of whether they had been 4. Olweus D. Bullying at school: what we know and what we can
do. Oxford: Blackwell Publishers, 1993.
clumsy or not in childhood. Also, motor problems in child-
5. Fox CL, Boulton MJ. The social skills problems of victims of
hood tend to diminish in adulthood because of maturation bullying: self, peer and teacher perceptions. Br J Educ Psychol
of the involved brain areas. 2005; 75(Pt 2): 313–28.
In conclusion, the strong correlation between bully vic- 6. Holmberg K, Hjern A. Bullying and attention-deficit- hyperac-
timization and reported poor motor skills in childhood fur- tivity disorder in 10-year-olds in a Swedish community.
ther supports the results from our pilot study in which 27% Dev Med Child Neurol 2008; 50: 134–8.
of 1043 predominantly middle-aged low-educated health 7. Little L. Middle-class mothers’ perceptions of peer and sibling
victimization among children with Asperger’s syndrome and
care workers, attending 1-day conferences about psychiatric
nonverbal learning disorders. Issues Compr Pediatr Nurs 2002;
disorders from seven different districts in Sweden, acknowl- 25: 43–57.
edged being poor performers in PE. These individuals were 8. Gillberg C, Kadesjö B. Why bother about clumsiness? The
more often bullied than those who reported superior perfor- implications of having developmental coordination disorder
mance (p < 0.000001) (15). In the present study, these find- (DCD). Neural Plast 2003; 10: 59–68.
ings were extended to subjects with high education level, 9. Dewey D, Cantell M, Crawford SG. Motor and gestural perfor-
mance in children with autism spectrum disorders, develop-
supporting the generalizability of the association. Prospec-
mental coordination disorder, and ⁄ or attention deficit
tive studies of the relationship between motor function, per- hyperactivity disorder. J Int Neuropsychol Soc 2007; 13:
sonality traits and bully victimization are warranted. 246–56.
10. Kopp S, Beckung E, Gillberg C. Developmental coordination
disorder and other motor control problems in girls with autism
SIGNIFICANT OUTCOMES spectrum disorder and ⁄ or attention-deficit ⁄ hyperactivity disor-
der. Res Dev Disabil 2010; 31: 350–61.
11. Diamond A. Close interrelation of motor development and
• Performing below average in physical education
cognitive development and of the cerebellum and prefrontal
(PE) was associated with a more than threefold cortex. Child Dev 2000; 71: 44–56.
risk of being bullied in school. This was reported in 12. Piek JP, Dawson L, Smith LM, Gasson N. The role of early fine
retrospect by healthy and cognitively high-performing and gross motor development on later motor and cognitive
young adults. ability. Hum Mov Sci 2008; 27: 668–81.
• A strong association between long duration and high 13. Teitelbaum O, Benton T, Shah PK, Prince A, Kelly JL, Teitel-
baum P. Eshkol-Wachman movement notation in diagnosis:
frequency of victimization was found, and both were
the early detection of Asperger’s syndrome. Proc Natl Acad Sci
associated with poor performance in PE in school.
U S A 2004; 101: 11909–14.
• A feeling of ‘being different’ resides into adulthood in 14. Stephenson EA, Chesson RA. Always the guiding hand’: par-
childhood bully victims. ents’ accounts of the long-term implications of developmental
co-ordination disorder for their children and families. Child
Care Health Dev 2008; 34: 335–43.
LIMITATIONS 15. Bejerot S, Humble M. Relevance of motor skill problems
in victims of bullying (letter). Pediatrics 2007;
120: 1226–7.
• The participants were self-referred 16. Kaukiainen A, Salmivalli C, Lagerspetz K, Tamminen M,
• The information relies on retrospective, biographical Vauras M, Mäki H, et al. Learning difficulties, social intelli-
reports gence, and self-concept: connections to bully-victim problems.
• Not all participants were seen in person Scand J Psychol 2002; 43: 269–78.

6 ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica

Bejerot et al. Bullying and performance in physical education

17. Kessler RC, Adler L, Ames M, Demler O, Faraone S, Hiripi 22. Ritvo RA, Ritvo ER, Ritvo MJ, Guthrie D. A Scale to Assist the
E, et al. The World Health Organization Adult ADHD Diagnosis of Autism Spectrum Disorders in Adults: (RAADS-
Self-Report Scale (ASRS): a short screening scale for use R) An International Multi-Center Standardization Study. Sym-
in the general population. Psychol Med 2005; 35: 245– posium presented at IMFAR, Philadelphia, Pennsylvania, USA,
56. May 21st, 2010
18. Ritvo RA, Ritvo ER, Guthrie D, Yuwiler A, Ritvo MJ, Weisb- 23. Lumeng JC, Forrest P, Appugliese DP, Kaciroti N, Corwyn RF,
ender L. A scale to assist the diagnosis of autism and Asperger’s Bradley RH. Weight status as a predictor of being bullied in
disorder in adults (RAADS): a pilot study. J Autism Dev third through sixth grades. Pediatrics 2010; 125: 1301–7.
Disord 2008; 38: 213–23. 24. Oldehinkel A, Rosmalen JGM, Veenstra R, Dijkstra JK, Ormel
19. Buchanan RW, Heinrichs DW. The Neurological Evaluation J. Being admired or being liked: classroom social status and
Scale (NES): a structured instrument for the assessment of depressive problems in early adolescent girls and boys.
neurological signs in schizophrenia. Psychiatry Res 1989; J Abnorm Child Psychol 2007; 35: 417–27.
27: 335–50. 25. Kanioglou A, Tsorbatzoudis H, Barkoukis V. Socialization and
20. Bejerot S, Nylander L, Lindström E. Autistic traits in obsessive- behavioral problems of elementary school pupils with develop-
compulsive disorder. Nord J Psychiatry 2001; 55: 169–76. mental coordination disorder. Percept Mot Skills 2005; 101:
21. Olweus D. The revised Olweus bully ⁄ victim questionnaire. 163–73.
N-5015 Bergen, Norway: Mimeo, HEMIL, University of 26. Gilmartin BG. Peer group antecedents of severe love-shyness
Bergen, 1996 in males. J Pers 1987; 55: 467–89.

ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 7