Sie sind auf Seite 1von 14

Aggression and Violent Behavior 23 (2015) 2235

Contents lists available at ScienceDirect

Aggression and Violent Behavior

Bullying in South-East Asian Countries: A review

Ruthaychonnee Sittichai a,, Peter K. Smith b
Prince of Songkla University, Thailand
Goldsmiths, University of London, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: Bullying among young people is a major concern for educators and health professionals. It has been mostly stud-
Received 12 February 2015 ied in Western countries, but with a signicant amount of research in the Asian Pacic Rim countries of Japan,
Received in revised form 27 May 2015 South Korea, Mainland China and Hong Kong, which suggests that it can exhibit different characteristics in
Accepted 17 June 2015
other cultural contexts. Research in this area in South-East Asian countries has been relatively neglected, but
Available online 3 July 2015
has begun to appear in recent years. Here, we review studies on bullying in the 10 ASEAN countries. We summa-
rize the nature and main ndings of these studies, and comment on similarities and differences with studies in
Bully Western and Asian Pacic Rim societies. Finally we make suggestions for future research which will enhance
Cyberbully comparability, respecting cultural differences but moving toward a more effective comparative analysis.
ASEAN 2015 Elsevier Ltd. All rights reserved.
South-east Asia


1. Studies in Western and Asian Pacic Rim societies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

2. Comparative neglect of studies in South-East Asian countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3. Literature search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4. Data presented in Table 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.1. Sample size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2. Date of study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.3. Language and linguistic terms used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.4. Denitions of bullying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.5. Measures used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.6. Main ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5. Two comparative data sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.1. Comparative data from GSHS (Indonesia, Myanmar, Thailand, Philippines, Malaysia) . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2. Comparative data from TIMMS (Indonesia, Malaysia, Philippines, Singapore, Thailand) . . . . . . . . . . . . . . . . . . . . . . . . . . 27
6. Research on bullying by country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.1. Countries with no local research: Brunei Darussalam, Lao PDR, Myanmar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.2. Countries with one extended report: Cambodia and Vietnam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.2.1. Cambodia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.2.2. Vietnam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.3. Countries with limited local research: Indonesia, Philippines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.3.1. Indonesia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.3.2. Philippines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.4. Countries with more extensive local research: Malaysia, Singapore, Thailand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.4.1. Malaysia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.4.2. Singapore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6.4.3. Thailand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
7. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
7.1. Limitations of reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Corresponding author at: Prince of Songkla University, Muang, Pattani 94000, Thailand.
E-mail address: (R. Sittichai).
1359-1789/ 2015 Elsevier Ltd. All rights reserved.
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 23

7.2. Prevalence of bullying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

7.3. Cyberbullying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
7.4. Gender differences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
7.5. Other ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7.6. A call for more qualitative research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7.7. Moving toward explanations for cultural differences in bullying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7.8. Neglect of some important variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7.9. Relevance for interventions to reduce bullying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
8. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Website references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Bullying, or the systematic abuse of power, can be physical, verbal, or In Japan, Kanetsuna and Smith (2002); Kanetsuna, Smith, and
relational, and direct (face-to-face) or indirect. Bullying is generally con- Morita (2006) compared ijime (the term closest to bullying in
sidered as a subset of aggression, distinguished by the criteria of repeti- Japanese) and bullying in England. They found some signicant differ-
tion, and imbalance of power (Olweus, 1999). In the last decade ences. In Japan, pupils reported ijime as most likely to come from pupils
especially, cyberbullying has emerged through the use of modern com- that they knew well, of similar age and often within the classroom; in
munication technologies (Kowalski, Giumetti, Schroeder, & Lattanner, England, pupils reported bullying as often coming from pupils they did
2014), with some debate as to whether the same dening criteria not know well, often older, and often in the playground.
apply (Bauman, Cross, & Walker, 2013). A somewhat similar pattern of difference comes from studies of
Bullying has been extensively studied in Western countries. Howev- wang-ta in South Korea, by Koo, Kwak, and Smith (2008) and Lee,
er a signicant amount of research in the Asian Pacic Rim countries Smith, and Monks (2011, 2012). Wang-ta also seems to occur between
suggests that it can exhibit different characteristics in other cultural pupils who know each other (e.g. former friends). Within a classroom
contexts, and points to cultural and educational aspects of interest in ex- context this can mean the whole class shunning one pupil. It can take
planations of this. Research in the 10 ASEAN countries of South-East the form of even more severe social exclusion, where another term
Asian countries is comparatively sparse, but is starting to appear. Here, jun-ta refers to the whole school labeling the victim and shunning that
we review published studies on bullying in these countries, including person.
their characteristics and methodological features. We summarize the Japan, and South Korea, are considered as more collectivist societies
main ndings of these studies, comment on similarities and differences than most Western industrialized countries (Hofstede, Hofstede, &
with studies in Western and Pacic Rim societies, and note some limita- Minkov, 2010); a more collectivistic culture may imply a greater possi-
tions of the research to date. We conclude with suggestions for how fu- bility of concerted whole-group (e.g. whole-class) norms emerging,
ture research can respect cultural differences but move toward a more which could at times be aggressivethus the possibility of severe
effective comparative analysis. whole-class aggression and shunning of a victim. In addition, Japan
and South Korea may be considered more hierarchical, scoring high on
the Hofstede (1980) power distance index; there is more respect for
1. Studies in Western and Asian Pacic Rim societies older persons, including older pupils, such that (ab)use of power by
older persons will be more likely to be seen as legitimate and not as bul-
Most of the now extensive research on bullying and cyberbullying lying (Hofstede, 1980; Smith, Bond, & Kagitibasi, 2005).
has been carried out in Western countries. For example, the Handbook There have also been a number of studies of bullying in Mainland
of Bullying in Schools: An International Perspective (Jimerson, Swearer, China, and Hong Kong. Schwartz, Chang, and Farver (2001) examined
& Espelage, 2010) has 41 chapters, but only 2 represent perspectives peer group victimization in a Chinese primary school. The correlates of
outside Europe, North America and Australia (1 comparative, 1 on victimization, such as poorer academic functioning, submissive-
Japan). withdrawn behavior, and aggression, suggested considerable similarity
Prevalence gures from Western countries are available from the in the processes underlying peer group victimization across Chinese and
Health Behaviour in School-aged Children (HBSC) surveys, which col- Western cultural settings (p.520); and subsequently, in a similar study
lect data from 11-, 13- and 15-year olds from nationally representative in Hong Kong schools, Tom, Schwartz, Chang, Farver, and Xu (2010)
samples, every 4 years. The reports on bullying are based on a single vic- stated that they were able to replicate the behavioral and academic
tim item and a single bully item, adapted from the Olweus Bully/Victim correlates of peer victimization from past research conducted in the
Questionnaire (Solberg & Olweus, 2003); a widely-used denition of West and Mainland China with children in Hong Kong (p.35); although
bullying is given (mentioning repetition, and imbalance of power). Vic- they noted that forms of victimization that involve social exclusion and
tim or bully rates are calculated from at least two or three times in the causing harm to relationships may be particularly relevant in Chinese
past couple of months or more (so, ignoring it only happened once or children's peer groups (p.35). Wong, Lok, Lo, and Ma (2008) also re-
twice). The 2005/2006 survey gives average rates (over 40 countries) ported on frequency and correlates of bullying in Hong Kong primary
of 10.7% for bullying others, and 12.6% for being bullied (victims) schools; one notable aspect in their report was an appreciable frequency
(Craig et al., 2009). The 2009/2010 survey gives average rates (over 38 of extortion as well as the more familiar forms of bullying from West-
countries) of 10.3% for bullying others, and 11.3% for being bullied ern studies.
(Currie et al., 2012). The countries surveyed are mostly European, but In summary, studies of bullying in the Asian Pacic Rim countries
also include the United States, Canada, Russian Federation, and suggest that there may be important differences from western countries
Ukraine; no data from Asian countries is included. in terms of who does the bullying (friends in the same class, or relative
The phenomenon of bullying does appear to have some broad simi- strangers), where it happens (classroom, playground), and types of bul-
larities across these Western cultures, but there are differences as well lying (severity of social exclusion, prevalence of extortion). In addition,
as similarities with the forms it takes in Asian Pacic Rim societies; sub- denitions of bullying-like phenomena show linguistic variation, and
stantial research has taken place in Japan, South Korea, Mainland China may be inuenced by what is viewed as legitimate or illegitimate
and Hong Kong (see Smith, Kwak, & Toda, in press). power relations. Possible explanatory factors for these differences may
24 R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235

be cultural (for example, the individualism-collectivism and power and Google, and also contacted researchers in the ASEAN countries to
distance index dimensions of Hofstede et al., 2010), and/or they may request more information and studies. Here we review published arti-
be inuenced by schooling variables (for example, prevalence of home- cles and reports, and accessible master's or doctoral theses; we do not
room class teaching in secondary schools, which is higher in Japan than include a number of newspaper reports where there is no academic or
many western countries; Kanetsuna et al., 2006). refereed publication to give fuller details. We include any study which
is purportedly on bullying or cyberbullying, or in the latter case on inter-
2. Comparative neglect of studies in South-East Asian countries net or online harassment; we do not include articles on aggression or
physical ghting which make no claim to be about bullying. We review
In his doctoral thesis on bullying in schools in Vietnam, Horton studies published (or with an abstract) in English, but mention signi-
(2011, p.15) wrote that There has until now been almost no research cant other-language material in the text. We summarize the main stud-
about school bullying in Vietnam , or indeed in any Southeast Asian ies by country in Table 2. In the text we give more details of the ndings
context. Nevertheless, although neglected in relative terms, recent of each study.
years have started to see a signicant number of such publications in
South-East Asian countries. Here, we review the literature on bullying 4. Data presented in Table 2
and cyberbullying in the 10 ASEAN countries: Brunei Darussalam,
Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Philippines, 4.1. Sample size
Singapore, Thailand, and Vietnam. These countries are more collectivist
than the predominately individualistic cultures of Western Europe, We give the number and age of participants. In the text we give any
North America, and Australasia (Hofstede, 1980; Hofstede et al., relevant details of recruitment, for example number of schools sampled,
2010), and they are high on the power distance index; but they differ and regional area.
from Japan, South Korea, China and Hong Kong in being less industrial-
ized, and many having large Muslim (Indonesia, Malaysia) or Catholic 4.2. Date of study
(Philippines) as well as Buddhist populations.
The South-East Asian countries together comprise a population of We indicate the year or date at which the data was collected, if given
over 622 million people, or about 16% of the world's total population. (which can differ substantially from the year of publication). Many
Table 1 shows (as of 2012) the population and ofcial languages of studies do not report this, but it is importanta point made generally
each country in 2012, and percent internet penetration (relevant for by Hofstede et al. (2010, p.338), and specically for bullying by Smith
consideration of cyberbullying); internet penetration is very high in (2010), in the light of historical changes in the denition of bullying,
Brunei, Singapore, and Malaysia, and also above the world average of forms of bullying, changing educational practices, and the impact of
27.5% in Vietnam, Thailand and the Philippines. It also shows GDP per large-scale anti-bullying campaigns or interventions.
capita, which we refer to later in terms of country differences in re-
search activity. 4.3. Language and linguistic terms used
Previously, Sittichai and Smith (2013) reviewed studies on bullying
and cyberbullying in Thailand, commenting on progress made and is- This is important since many languages do not have a linguistic
sues arising. Here, we aim to review the studies on bullying carried term that closely corresponds to the English term bullying. In the
out in all 10 ASEAN countries; to compare with ndings in Western Latin languages, terms such as violenza or prepotenza (in Italian)
and Pacic Rim cultures; and to offer constructive criticism for how have broader meanings than bullying, as they include aggressive
further research in these countries may be taken forward in a way acts which do not necessarily have repetition or imbalance of
that respects cultural differences but may better enable cross-cultural power (Fonzi et al., 1999). Smith, Cowie, Olafsson, Liefooghe, and
comparisons. in collaboration with 17 additional authors (2002) used a cartoon-
based method to study the meaning of terms similar to bullying, in
various languages; the only South-East Asian country included was
3. Literature search Thailand, where tum raai (thamri) was the closest in meaning to
bullying, with klang (klang) having a meaning more related to ver-
A literature search was conducted for publications on bullying (in- bal teasing, and nisai mai dee a more general term referring to
cluding cyberbullying) in these 10 countries, up to 2014. We searched nasty or aggressive behavior. In Table 2 we indicate whether the lan-
through Science Direct, PsycINFO, Springer Link, DAO, Google Scholar guage used in the study is explicitly stated.

4.4. Denitions of bullying

Table 1
Bullying is usually dened as being an aggressive, intentional act
Population (in millions), ofcial language(s), internet penetration (percent), and per
capita GDP (dollars) in ASEAN countries. or behavior that is carried out by a group or an individual repeatedly
and over time against a victim who cannot easily defend him or her-
Country Population Ofcial languages Internet GDP $ ppp
self (Olweus, 1999). It is thus usually distinguished from categories
Brunei 0.41 Malay 78.0 75,222 such as aggression, or physical ghting, by the specic criteria of rep-
Cambodia 14.95 Khmer 4.4 2845 etition and power imbalance. However denitions, and the use of
Indonesia 248.65 Indonesian 22.1 9102
these criteria, vary in different studies. Some studies in Table 2 (in-
Laos 6.59 Lao 9.0 4335
Malaysia 29.12 Bahasa Melayu 60.7 22,104 cluding the GSHS surveys) give a denition of bullying, based on
Myanmar 54.58 Burmese 1.0 3989 the OBVQ (Solberg & Olweus, 2003). This provides a degree of stan-
Philippines 103.78 Filipino Spanish 32.4 6171 dardization, though there are issues about how much the denition
Singapore 5.35 English, Chinese, Malay, Tamil 75.0 75,949
is followed by respondents, or how much it corresponds to their
Thailand 67.09 Thai 30.0 13,605
Vietnam 91.52 Vietnamese 33.9 5001 own understanding of the term. Some studies use the term bullying
(or equivalent) without a denition. Some use behavior-based mea-
Sources: Internet World Stats, updated June 30, 2012.
sures, for example the Peer Relations Questionnaire (Rigby & Slee,
GDP per capita for 2012. 1993), with items such as I get called names by others; behavior- based studies avoid the issues around imposing a denition, but
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 25

Table 2
Summary of published studies of bullying in South-East Asian countries.

Authors Sample Date of Language Denition Measures Findings


Miles, 2004 1300 aged 1215 years Probably Not stated No denition of bullying Focus groups and self-report Prevalence, how to help bullies and
2004 questionnaire (victim) victims

Fleming & 2694, 1216 years 200306 Not stated Olweus-type denition Self-report questionnaire (victim) Prevalence, associations with mental
Jacobsen, 2010 (GSHS data) health
Lai, Ye, & Chang, 5542 eighth graders 2003 Not stated Behavior based, 5 types Self-report questionnaire (victim) Prevalence, associations with school
2008 (TIMMS data) attitudes, learning, home conditions
Muryani & 152, 1120 years Not stated Not stated Behavior based Peer Relations Questionnaire Prevalence, associations with age and
Phongpat, 2014 gender
Akbar, Huang, & 245, grades 13 Not stated Not stated Behavior based, 49 Self-report questionnaire of Factor structure and reliability of
Anwar, 2014 items cyberbullying cyberbullying scale

Lai et al., 2008 5287 eighth graders 2003 Not stated Behavior based, 5 types Self-report questionnaire (victim) Prevalence, associations with school
(TIMMS data) attitudes, learning, home conditions
Yaakub, Haron, & 3816 in secondary Not stated Not stated No denition of Survey of 5 types of bullying Outcomes of a version of the
Goh, 2010 school bullying Olweus Bullying
Prevention Programme
Wan, Tan, Nik, 410 sixth graders AprilJune Bahasa No denition of Self-report questionnaire Prevalence, association with family
Tuti, Syamsul, 2006 Malaysia bullying (bully, victim) factors, academic performance
Aniza, & (Malay)
Zasmani, 2009
Wan et al., 2010 410 sixth graders, aged 2006 Bahasa No denition of Self-report questionnaire (bully, Association with ADHD
12 years Malaysia bullying victim) plus reports on ADHD
Yahaya, Ramli, 480 of secondary school Not stated Not stated No denition of Self-report questionnaire (victim) Prevalence, location, reasons for
Hashim, age, plus 80 teachers bullying bullying, views on prevention
Ibrahim, &
Rahman, 2009
Uba, Yaacob, & 242 aged 1316 years Not stated Not stated Behavior based PRQ, Child Depression Inventory Prevalence, associations with family
Juhari, 2010 factors and depression
Wan, Nik, Hatta, 410 aged 12 years Not stated Not stated Behavior based Self-report questionnaire Prevalence, associations with
Marhani, & (Malaysian Bullying Questionnaire), demographic aspects, academic
Shamsul, 2014 ADHD, CBC performance, ADHD symptoms,

Lai et al., 2008 6840 eighth graders 2003 Not stated Behavior based, 5 types Self-report questionnaire (victim) Prevalence, associations with school
(TIMMS data) attitudes, learning, home conditions
Rudatsikira, 7338 aged around 15 2003 Not stated Behavior based (GSHS Self-report questionnaire (victim) Prevalence, association with
Mataya, Siziya, years data analysis) physical ghting
& Muula, 2008
Fleming & 7338 aged around 15 2003 Not stated Olweus-type denition Self-report questionnaire (victim) Prevalence, association with
Jacobsen, 2010 years (GSHS data) internalizing and externalizing factors
Wilson, Dunlavy, 8659 aged 1216 years 200608 Not stated Olweus-type denition Self-report questionnaire (victim) Prevalence, forms, associations with
& Berchtold, (GSHS data) age, gender, truancy, poverty

Lai et al., 2008 6009 eighth graders 2003 Not stated Behavior based, 5 types Self-report questionnaire (victim) Prevalence, associations with school
(TIMMS data) attitudes, learning, home conditions
Koh & Tan, 2008 786 primary pupils c.102006 English Behavior-based plus Self-report questionnaire (bully, Prevalence, forms, consequences,
years, 513 secondary (secondary), use of term bullying, victim) sources of support
pupils c.15 years 2007 not dened
Ong & Elliott, 400 + 200 adults 2529 2009 English Denition given Retrospective self-report Prevalence at school, forms,
2010 years questionnaire (victim) through associations with mental well-being,
structured interview; psychometric suggestions for prevention
Ang, Ong, Lim, & 809 (463 aged 913; Not stated English Power differential Self-report questionnaire (bully), Associations between narcissism,
Lim, 2010 346 aged 1216 years) Narcissistic Personality approval-of-aggression and
Questionnaire, Normative Beliefs bullying, test mediational model
about Aggression Scale
Chew, 2010 3488, 1317 years 2006 English Denition discussed Self-report questionnaire (cyber Prevalence, forms, gender, coping
victim) strategies
Ng, 2010 4000+, 716 years 2006 English Behavior based PRQ, DINO-Map Forms, location, feelings, results of
Ang & Goh, 2010 396 aged 1218 years Not stated English Behavior-based, 9 Self-report questionnaire (cyberbully), Prevalence of cyberbullying,
items Basic empathy scale associations with empathy
Ang, Tan, & 336 aged c.14 years in Not stated English Behavior-based, 9 Self-report questionnaire (cyberbully), Associations between narcissism,
Mansor, 2011 Singapore, 374 aged items Narcissistic Personality Questionnaire, approval-of-aggression and
c.15 years in Malaysia Normative Beliefs about aggression cyberbullying, test mediational model

(continued on next page)

26 R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235

Table 2 (continued)

Authors Sample Date of Language Denition Measures Findings

Pozzoli, Ang, & 430 aged c.12 years Not stated Not stated Behavior-based Self-reported active and passive Associations of attitudes to bullying
Gini, 2012 (and 601 Italian c.12 bystanders (3 items each), attitudes and perceived peer pressure to
years) to bullying, perceived peer pressure active and passive bystanding
Lwin, Li, & Ang, 537 aged 1219 years Not stated Not stated Mainly behavior-based Self-reports on behavioral intention Associations of predictor variables
2012 to protect against online with behavioral intention to test
harassment, and four predictor model
Khiat, 2012 75 (essay) + 64 Not stated Not stated Not givengrounded Reective essay; semi-structured Codings of various aspects; includes
(interview), ages not theory approach interview typologies of bullies, bystanders,
stated and victims
Ang, Huan, & 332 aged c.14 years in Not stated English Behavior-based, 9 Self-report questionnaire Associations between reactive and
Florell, 2013 Singapore, 425 aged items (cyberbully), Reactive-Proactive proactive-aggression and
c.13 years in U.S. Aggression Questionnaire cyberbullying, test moderator
Kwan & Skoric, 1,597 aged c.1317 2011 English Behavior-based Self-report questionnaires Associations between cyber and
2013 years (cyberbully, cybervictim; school school bully and victim rates and
bully, victim), Facebook use Facebook usage

Laeheem, Kuning, 1440, 710 and 1013 20052006 Not stated Denition (similar to Self-report questionnaire (bully) Prevalence, location, associations
McNeil, & years that of aggression), but with entertainment, family abuse
Besag, 2008 behavior-based
Laeheem, Kuning, 1440, 710 and 1013 20052006 Not stated Denition (similar to Self-report questionnaire (bully) Prevalence, associations with school
& McNeil, 2009 years that of aggression), but type, entertainment family abuse
Yodprang, 244, 1219 years 2006 Not stated No denition, partly Self-report questionnaire (bully, Prevalence (bully), associations
Kuning, & behavior-based victim) with school type, family abuse
McNeil, 2009
Buttabote, 2011 102 ADHD patients, 20092010 Thai Denition as in OBQ Self-report questionnaire (OBQ) and Association of bully and victim with
1018 years Rosenberg self-esteem self-esteem
Musikaphan, 2000 students, 14 key Not stated Thai Not stated Self-report survey of cyberbullying Prevalence of being cyberbullied,
2009 informants (plus 100 behavior, and interviews attitudes
students, 12 key
informants in Japan)
Songsiri & 1200, c.1416 years Not stated Not stated Not stated Self-report survey of cyberbullying Types, direct or indirect experiences
Musikaphan, behavior
Laeheem & 1592 2010 Not stated Behavior based, 28 Self-report questionnaire of risk of Discriminative power and reliability
Sungkharat, items exposure to bullying of scale
Pengpid & 2758, 1315 years 2008 Not stated Olweus-type denition Self-report questionnaire (victim) Prevalence, forms, correlates
Peltzer, 2013 (GSHS data)
Laeheem, 2013a 15 each bullies, victims, Not stated Not stated Not givengrounded In-depth interviews Codings of types of bullying, where
teachers, parents in theory approach takes place, attitudes and feelings
primary school
Wilson et al., 6120 aged 1216 years 200608 Not stated Olweus-type denition Self-report questionnaire (victim) Prevalence, forms, associations with
2013 (GSHS data) age, gender, truancy, poverty
Tudkuea & 480, c.1318 years Not stated Not stated Behavior based, 24 Self-report questionnaire of Factor structure and reliability of
Laeheem, 2014 items cyberbullying cyberbullying scale
Sittichai, 2014 1183, 1417 years 2012 Not stated Olweus-type denition Self-report questionnaire (victim of Prevalence, associations with
traditional and cyber bullying) gender, province, grade, religion,
parent education
LGBT-friendly 67 focus groups, 56 20122013 Thai Behavior-based Focus groups, interviews, self-report Understandings of bullying,
Thailand?, interviews, 2070 survey questionnaire prevalence based on LGBT, forms,
2014 aged 1320. bystanders, coping, impact

Horton, 2011, 906 aged 11 to 15 years 20072008 Vietnamese No denition given Self-report questionnaire (bully, Prevalence, nature of bat nat,
2012 victim), interviews, participant disciplinary power and role of
observation teachers

run the risk of assessing the broader category of aggression than of be viewed or passed on many times by others; since this could often
bullying (Furlong, Sharkey, Felix, Tanigawa, & Green, 2010; Ybarra, be foreseen by the perpetrator, it might be legitimate to count a single
Boyd, Korchmaros, & Oppenheim, 2012). In Table 2 we indicate perpetrator act as cyberbullying. Regarding the imbalance of power cri-
whether a denition or behavior based approach was used. terion, the usual signs of this for traditional bullying (physical strength,
Particular denitional issues arise for cyberbullying. One denition social status, or numbers of bullies) do not so obviously apply, especially
which keeps the standard criteria is that cyberbullying is an aggressive, if the perpetrator withholds their identity; however anonymity in itself
intentional act carried out by a group or individual, using mobile phones may indicate an imbalance of powerthe perpetrator knows the victim,
or the internet, repeatedly and over time against a victim who cannot but not vice versa; and if there is not anonymity and the victim does know
easily defend him or herself (Smith et al., 2008). There is debate who the perpetrator is, then the traditional criteria may still be relevant
about whether it is appropriate to just carry over the denition in this (Menesini et al., 2012). Some researchers prefer a behavior based
way. Regarding the repetition criterion, a single perpetrator act may approach to avoid these denitional controversies; nevertheless much
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 27

research has used the term cyberbullying. Prevalence rates can vary wide- whereas physical bullying is relatively low. In Myanmar, sexual bullying
ly depending on the denitional approach used in cyberbullying (Smith, is particularly low, but being bullied because of religion is relatively
2012). high. Exclusion is low in Thailand. A surprisingly high proportion
of Indonesian victims (one-fth) say it was in other ways. The
4.5. Measures used Philippines data on forms varies considerably between 2003 and 2007.
However all the data sets have a high proportion (13% or more) of miss-
Although self-report surveys have been most commonly used to ing responses, especially high for Philippines 2007 (21%), which may
study bullying, other common quantitative methods include question- indicate some problems in data collection or comprehension.
naires to teachers or parents, and peer nominations. Less frequently, GSHS data from 2003 to 2006 was used by Fleming and Jacobsen
more qualitative or ethnographic approaches have been used, such as (2010) in an analysis of bulling rates in 19 countries. The only South-
interviews, focus groups, and participant or non-participant observa- East Asian country included was the Philippines, for which they found
tion. In Table 2 we outline the main measures used to study bullying, associations of victim rates with younger age, feeling sad/hopeless, suf-
and measures for other important variables associated with bullying. fering from loneliness and insomnia, having suicidal ideation, and also
smoking, drinking alcohol and using drugs. The 2003 Philippines data
4.6. Main ndings was also analyzed further by Rudatsikira et al. (2008). Being a victim
of bullying was found to be related to involvement in physical ghting.
The main types of ndings are summarized in Table 2, with details in There were no gender differences in either variable. Summary GSHS
the text below. We start with two comparative data sets; this is followed data for the Philippines is also available for 2011, for the percentage of
by a review of studies carried out by country; here we group countries students who were bullied on one or more days during the past
in terms of the volume of research carried out. 30 days. This was 47.7% (46.9 males, 48.4 females).
GSHS data from 2006 to 2008 was used by Wilson et al. (2013) in an
5. Two comparative data sets analysis of bullying rates in 15 countries, including Philippines and
Thailand. They found associations of victim rates (over all 15 countries)
5.1. Comparative data from GSHS (Indonesia, Myanmar, Thailand, with male gender, often being hungry, and rates of truancy from school.
Philippines, Malaysia) Pengpid and Peltzer (2013) used the 2008 data from Thailand, and
found victim rates higher in males, and associated with physical ght-
The Global School Health Survey (GSHS), as part of a large and ing, psychosocial distress, being physically inactive (boys only), truancy
regular inventory on health-related behaviors, provides data on two (boys only), and lack of parental bonding (girls only).
questions about being a victim of bullying (frequency, forms) from a
wide range of countries. An Olweus-type denition is given rst (men- 5.2. Comparative data from TIMMS (Indonesia, Malaysia, Philippines,
tioning repetition and imbalance of power as criteria for bullying), then Singapore, Thailand)
the two questions, referring to the past 30 days, and 7 response options
for forms of bullying. Table 3 gives a summary of the ndings (from the The Trends in International Mathematics and Science Study
GSHS website) for Indonesia (2007), Myanmar (2007), Thailand (2008), (TIMMS) survey began in 1995 and is assessed every 4 years, with
and the Philippines (2003, 2007). There is also summary data only from some 50006000 eighth grade students in each country. Primarily
the Philippines (2011), and also for Malaysia (2012), for the question on about mathematics achievement, it also includes a section on school
how many days you were bullied during the past 30 days; this is men- safety, including questions on experiences of being bullied in the past
tioned under the country headings. month. No denition of bullying is given, so the questions may be tap-
Regarding frequency, Table 3 shows this to be clearly substantially ping some aggressive acts as well as bullying more strictly dened,
higher in Indonesia and then the Philippines, than in Myanmar, or yielding relatively large prevalence rates.
Thailand. Regarding forms, there is some country variation in ranking. The 2003 questionnaire asked if any of ve forms of bullying behav-
Even bearing in mind overall frequency of being bullied, being made ior had happened during the last month. Lai et al. (2008) reported an
fun of because of how my body or face looks is high in Indonesia, analysis of this data for 10 countries, including Indonesia, Malaysia,

Table 3
Comparative results from GSHS surveys in Indonesia, Myanmar, Thailand and Philippines. Percentages of pupils responding, by country.

Indonesia Myanmar Thailand Philippines Philippines

Year of survey 2007 2007 2008 2003 2007

Total sample N = 2694 N = 2438 N = 2386 N = 6268 N = 4496

Missing data N = 422 N = 368 N = 381 N = 1070 N = 1161

During the past 30 days, on how many days were you bullied?
0 days 50.3 79.3 72.2 64.3 54.4
1 or 2 days 29.6 15.7 17.5 23.5 24.3
3 to 5 days 9.9 2.8 4.9 7.8 9.7
6 to 9 days 4.4 1.3 2.5 2.2 3.9
10+ days 5.8 1.0 2.9 2.1 7.7
During the past 30 days, how were you bullied most often?
I was hit, kicked, pushed, shoved around, or locked indoors 4.3 5.3 6.3 10.0 6.2
I was made fun of because of how my body or face looks 8.7 2.2 3.1 1.8 5.5
I was made fun of with sexual jokes, comments, or gestures 4.9 0.5 6.9 2.9 8.1
I was made fun of because of my race or color 6.0 3.9 2.1 3.8 4.1
I was left out of activities on purpose or completely ignored 4.5 1.7 0.7 2.8 5.6
I was made fun of because of my religion 0.7 2.2 0.8 1.4 1.7
I was bullied in some other way 20.6 4.9 7.9 13.1 15.0

10+ days sums 1019 days, 2029 days, and all 30 days, in the original data sets.
28 R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235

Philippines and Singapore. Being made fun of/called names was the (by 60%), followed by emotional support to the bullied child (12%),
most frequent form in all countries, followed by being hit or hurt dealing with bullies through local authorities (11%), and bullying pre-
(Indonesia), having things stolen (Malaysia, Singapore), made to do vention/protection (10%). When asked how the bullies can be helped,
things they didn't want to do (Philippines), and being left out of activi- the most frequent choice was again to educate/advise them not to
ties (usually less frequent). Prevalence rates for all ve behaviors were harm others (72%), followed by protecting other children from harm
noticeably higher in the Philippines (range 3059%) than Singapore (20%).
(range 1237%), Indonesia (range 1336%) or Malaysia (range 730%).
Males generally experienced such behaviors more than females (with 6.2.2. Vietnam
the exception of left out in Singapore). Risk factors included immigrant Horton (2011, 2012) published a doctoral thesis and a book based on
status in Indonesia and Philippines; and having many books in the the topic of bullying in Vietnamese schools. This was based on ethno-
home, in Indonesia. Protective factors included higher parental educa- graphic eld work in two lower secondary schools in Haiphong.
tion in the Philippines, and having learning tools (e.g. computers) at Although primarily using participant observation and group and indi-
home, in Singapore and Malaysia. Generally, experiencing these vidual interviews with students and teachers, some questionnaire sur-
behaviors was associated with more negative attitudes to school, and vey data from pupils in grades 6 to 9 is reported. This data (Horton,
there were signicant but small negative correlations with academic 2011, pp.7678) asked about experiences of bat nat, a Vietnamese
achievement. word close to bullying. Taking a frequency criterion of more than occa-
The 2011 data set included six types of behavior (adding a catego- sionally, then 11.9% of boys and 6.8% of girls reported being victims of
ry of someone spread lies about me). Mullis, Martin, Foy, and Arora bat nat, and 10.0% of boys and 5.9% of girls were perpetrators. Besides
(2012) provide prevalence tables for both 4th grade and 8th grade the gender differences, victim rates increased with age from 6th grade
students, but the 2011 data is only available for Singapore and to 9th grade; perpetrator rates also increased with age for boys, but
Thailand of the South-East Asian countries. The prevalence data for peaked in 7th grade for girls.
these two countries is shown in Table 4, compared to the interna- Horton's ethnographic approach also gives much insight into the na-
tional average (for 50 countries at 4th grade, 42 countries at 8th ture of bat nat, which he denes as being made to do something you
grade). Prevalence is slightly higher in Singapore but much higher would not otherwise do. This naturally leads to a discussion of power re-
in Thailand than the international average; and is generally less for lationships, and the author adopts a Foucauldian approach to power;
older students. power is not held but is exercised in strategic situations. For example
one child may make another child go to the canteen to get them a
6. Research on bullying by country snack; this means the rst child (bully) saves time and avoids possible
attacks in an unsupervised canteen setting. This in turn leads Horton
6.1. Countries with no local research: Brunei Darussalam, Lao PDR, Myanmar to question whether a bully is necessarily intentionally wishing to
hurt the victim, rather than responding to the institutional power struc-
No studies were located in Brunei Darussalam, or Lao PDR. Apart ture in the school. He similarly examines the actions of teachers, ways in
from the comparative data available from GSHS (2007), we did not which pupils may bully teachers and teachers may bully pupils, and
nd any publications on bullying in Myanmar. interprets this in relation to power relations within the immediate con-
text of the school and in the wider society.
6.2. Countries with one extended report: Cambodia and Vietnam
6.3. Countries with limited local research: Indonesia, Philippines
6.2.1. Cambodia
Miles (2004) compiled a report on violence against and by children 6.3.1. Indonesia
in Cambodia, which included a section on bullying. Some focus group As noted above, comparative data for Indonesia is available from
research was followed by a survey of pupils drawn from 130 schools GSHS, and TIMMS (Lai et al., 2008). Muryani and Phongpat (2014) car-
(10 from each school) across all the provinces of Cambodia. Some 57% ried out a survey in Jakarta, and reported prevalence of bullying as 75.7%
said that they had heard of children being bullied by other children at and victim as 96.1%; these extraordinarily high gures appear to be be-
school, and around 73% said they had seen other children teased or cause of a very lenient criterion used for dening bullying, from the PRQ.
mistreated. Altogether 37% (slightly more boys than girls) said they Contrary to most international ndings, bully rates were lowest in mid-
had ever been bullied by other children because they were different. adolescence, and there was no gender difference.
Nearly half described it as an extremely serious issue. Poverty was Akbar et al. (2014) report on the development of a cyberbullying
seen as a main reason for being bullied (by 40%), followed by disability scale, testing it on students at a private school in Aceh province. The
(25%), with much less frequent choice of gender (8%), ethnicity (6%) or scale was based on Willard (2007) with seven subscales of aming, ha-
religious belief (5%). When asked how children who have been bullied rassment, denigration, impersonation, outing & trickery, exclusion, and
can be helped, by far the most frequent choice was to educate the bullies cyberstalking. Of an initial 49 items, 26 were retained to nalize an in-
strument with satisfactory factor loadings and internal reliability.
Table 4
Rahayu (2012) (main article in Indonesian) reports on a survey on
Comparative results from TIMMS 2011 survey in Singapore and Thailand compare to av- cyberbullying in secondary schools in three provinces, but with very lit-
erage from all countries surveyed. Percentages of pupils responding for perceptions of be- tle detail on ndings in the English abstract. Not included in Table 2 is a
ing bullied at school. conference report by Margono, Yi, and Raikundalia (2014), which collat-
Almost never Bullied about Bullied about ed words used for bullying in Indonesian, from an analysis of Indonesian
bullied monthly weekly Twitter feeds.
4th grade
Singapore 39 38 23 6.3.2. Philippines
Thailand 17 35 48 As noted above, comparative data is available for the Philippines
International average 48 32 20 from GSHS (Fleming & Jacobsen, 2010; Rudatsikira et al., 2008; Wilson
8th grade et al., 2013) and TIMMS (Lai et al., 2008).
Singapore 52 36 12 Not included in Table 2 is a discussion article by Ancho and Park
Thailand 30 43 27 (2013) on the importance of tackling school violence, including bullying
International average 59 29 12
and cyberbullying; and a conference report by Gonzales (2014) which
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 29

summarizes views on cyberbullying from eight key professional coping strategies, from which it appears that pupils see a role for both
informants. teachers and students in stopping bullying. There were some different
perceptions about the prevalence of bullying between students and
6.4. Countries with more extensive local research: Malaysia, Singapore, teachers; teachers reported the bullying prevalence rate at a lower
Thailand level, than students. However teachers agreed with students in seeing
verbal abuse as the most common type, and the classroom as the most
6.4.1. Malaysia common location, and they too saw both teachers and students as hav-
The GSHS 2012 survey reports summary data for Malaysia, for the ing important roles in prevention. Teacher's views on intervention strat-
percentage of students who were bullied on one or more days during egies were also obtained, with discipline school board and updating
the past 30 days. This was 20.9% for 1315 year olds (males 24.0, fe- school regulations getting the highest rating.
males 17.8), and 12.5% for 1617 year olds (males 12.9, females 12.0). Uba (2009) carried out a Master's thesis on the relationship between
Apart from this, the comparative survey with TIMMS data by Lai et al. bullying, victimization, prosocial behavior and depression among teen-
(2008), and the inclusion of a Malaysian sample in a Singapore-based agers in Selangor. The main ndings are published in Uba et al. (2010).
study by Ang et al. (2011), there are several specically Malaysian Teenagers from 5 secondary schools in Selangor were given the PRQ to
studies. assess bullying, victimization and pro-social behavior, and the Children
Nora Yaakub and colleagues carried out some surveys of bullying in Depression Inventory to measure depression. Pupils were dichotomized
elementary school children, and in secondary school children, which are into bully or non-bully groups, which constituted 49.2% and 50.8% re-
not published but are available through a website ( spectively, probably using a lenient criterion including students who
pers/bully/bully.asp) or from the rst author. Building on these surveys, carried out one or two aggressive behaviors once in a while. The main
Yaakub et al. (2010) describe ndings from an intervention study, ndings were a higher rate of bullying in males than females, no gen-
funded by UNICEF. They worked with three experimental and three der differences in depression, but a correlation of around 0.3 between
control schools, in the Federal Territory of Kuala Lumpur; in each set bully scores and depression. The longer report in the thesis (Uba,
of three, one was a girls school, one a boys school, and one was co- 2009) also reports a correlation of around 0.4 between victim scores
educational. A pre and post 24-item survey asked about ve types of and depression, with victimization being the only unique predictor of
bullying: physical, verbal, relational, signal (staring, jutting out tongue), depression (p.xvii). Pro-social behavior had a positive correlation
and extortion. In the experimental schools, rates of bullying decreased with victimization only.
signicantly in the girls school; but it increased in the other two, as it In summary, there are several studies on prevalence (but using
did in the control schools. Only brief details are given in the published different instruments), and correlates including ADHD, one study on
report, but the authors suggest that the success in the girls school was coping strategies, and one on an intervention procedure.
because of better implementation of the program. A Manual for teachers
in Malay language was also produced as a result of this program 6.4.2. Singapore
(Yaakub, Haron, & Jusoh, 2009). Apart from the comparative data reported by Lai et al. (2008), there
In three related studies, Wan and colleagues (2009, 2010, 2014) are a number of studies based in Singapore.
gave the Malaysian Bullying Questionnaire (developed by Yaakub and Koh and Tan (2008), in a report to the Singapore Children's Society,
colleagues) to pupils from 7 primary schools in the Federal Territory carried out a survey of pupils of primary and secondary school age; they
of Kuala Lumpur. Pupils rated themselves on a 4-point frequency recruited a random stratied sample, and pupils were interviewed
scale, with bully, victim or bully-victim dened in terms of bullying (given the questionnaire) at home. Their comprehensive report repro-
others or being bullied, or both, three or more times during the last 4 duces the questionnaires used, which asked about 15 forms of bullying
weeks. By this denition, prevalence rates were bully 2.4%, victim behaviors in primary, and 17 in secondary, and how often these had
41.2%, and bully-victim 17.6%. More boys than girls were involved as been experienced in the last year. Overall 21% of primary pupils and
bullies and bully-victims. As there were only 10 children counted as 25% of secondary pupils were victims of at least one form of bullying;
bullies, the statistical associations with family variables and academic in primary school only, the rate was higher in boys than girls. For
performance may not be reliable; and the non-involved category was doing these things to others (bullying), the rates were 5% in primary
not used as a comparison. In the second report, on the same sample, and 10% in secondary (more boys in both sectors). Of the forms of bul-
the data were additionally analyzed in relation to self, teacher and par- lying, verbal was most frequent, followed by relational items, then
ent reports of Attention-Decit/Hyperactivity Disorder (ADHD) symp- physical forms; cyberbullying was least frequent. When asked what
toms. ADHD symptoms were self-reported more by bullies and bully- they did or how they felt after being bullied, the most common response
victims, but parent-reported more for victims, with no signicant asso- was feeling angry, followed by feeling sad; nearly one-third also
ciations for teacher reports. In the third report, apparently on the same mentioned retaliating against the bully; other fairly common responses
sample, associations were reported with demographic aspects (not sig- included wanting to be alone, loss of appetite, problems in concentra-
nicant), academic achievement (lowest in bully-victims), and ADHD tion at school, or sleeping at night. When asked if they told anyone,
(mixed ndings depending on informant); conduct problems were and if this helped, in primary schools 60% told parents, 35% told no-
highest, and internalizing symptoms lowest, in those involved in bully- one, 31% told teachers, and 8% told peers; teachers were rated most ef-
ing others. fective. In secondary schools, 81% told peers, 46% told teachers, 40% told
Yahaya et al. (2009) investigated the prevalence and types of being parents, 17% told no-one; teachers were rated more effective than par-
bullied at 8 secondary schools in Batu Pahat, Johor, using an adaptation ents or peers, but less so than in primary school.
of questionnaires such as the PRQ. Taking a criterion of once or twice a Ong and Elliott (2010), in another report to the Singapore Children's
month or more, 10.8% of students reported they had been bullied in Society, used a retrospective questionnaire with young adults. Bullying
school this year. Students described verbal bullying as more frequent was dened as repeated and intentional attempts by others to hurt
than physical bullying. It is stated in the Abstract (although not in the you or to cause distress to your daily life. When asked if they had
main text) that there was no signicant difference in prevalence be- been bullied during their primary or secondary school days, one-
tween male and female students. Being bullied was most frequent in quarter of an initial random stratied sample of 400 said yes (with no
the classroom, followed by recess time, on the way from school, and gender difference). A further 200 bullied respondents were then
then on the way to school. Students were also asked for their percep- added to the sample to equate bullied and non-bullied groups. Verbal
tions of reasons for the bullying: the most frequent responses were be- and relational bullying was recalled most often, followed by physical
cause they hurt my feelings and to revenge. They also gave views on bullying. There was no gender difference overall in being a victim, but
30 R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235

males were higher in recollections of being physically bullied or threat- beliefs moderating the relationship between narcissistic exploitativeness
ened. Some 38% of bullied respondents said they had sought help, most and bullying; these ndings held separately for both samples.
often from school and family, then peers; some 62% said that this help Pozzoli et al. (2012) examined how individual attitudes to bullying,
had made things better, and about 10% that it made things worse. Seek- and perceived peer pressure for intervention against bullying, related to
ing help was also the most frequently suggested way of dealing with self-reported defending behavior. The same procedure was carried out
bullying, followed by having friends around, action by the school or with samples of primary and middle school pupils in Singapore, and
teachers, avoiding/ignoring, education, standing up to the bully. How- Italy; the number of schools is not reported. Both individual attitudes
ever in terms of how they had dealt with an incident successfully, and peer pressure predicted active and passive bystanding in predicted
avoiding/ignoring was the most frequent response. Compared to non- directions; however perceived peer pressure had a stronger inuence in
bullied respondents, those who had been bullied scored lower on tests Singapore (seen as a more collectivist culture), whereas individual atti-
of self-esteem and well-being and higher on depression. tudes had less inuence than in Italy (seen as a more individualistic
A book edited by Ng and Rigby (2010) contains a range of chapters on culture).
bullying experiences and anti-bullying methods, with two primarily em- Lwin et al. (2012) used Protection Motivation Theory as a model for
pirical chapters. As well as the PRQ, Ng (2010) used a DINO-Map as a pic- looking at predictors of behavioral intention to protect against online
torial way of assessing younger children's experiences of being bullied, harassment. Their sample was drawn from four high schools. Partici-
and how they felt about it. As well as the playground and classroom, a pants were briefed on what online harassment constitutes prior to be-
common venue for bullying was found to be the canteen. Name calling ginning the surveys (p.36). Incidental to the main aims of the article,
and being teased and laughed at were the most common forms, followed 53% of the sample reported being a victim of online harassment at
by physical forms. A range of negative emotions were felt by victims, in- least once (and of these victims, 23% had experienced it more than
cluding anger as well as sadness and a feeling of unfairness. For a majority once). Predictor variables measured were perceived severity, suscepti-
of children who told someone about it, things got better, although only a bility, response efcacy, and self efcacy. These were found to correlate
minority of teachers were seen as usually interested in trying to stop it. with behavioral intention, with the exception of susceptibility. Self ef-
Chew (2010), working in eight secondary schools, reported that 75% cacy had a stronger effect for younger adolescents, and high self efcacy
had never been cyberbullied, 16% occasionally, 5% fairly often and 4% was more important for behavioral intention in males than females.
very often. Of eight forms, SMS and Instant Messaging were most fre- Ang et al. (2011) used the cyberbullying questionnaire of Ang and
quent. Rates of being cyberbullied were slightly higher at home than Goh (2010) together with a ReactiveProactive aggression question-
at school. Rates were higher for girls overall, and especially for rumor naire, comparing samples in Singapore and the USA. Prevalence rates
spreading and being teased, with boys higher for being called names. of cyberbullying were 17.9% in the US sample and 16.4% in Singapore;
Victims were most likely to tell friends (60%), mother (39%) or father rates were signicantly correlated with both types of aggression, in
(27%) with teachers least (24%); while 37% said they had tried to both countries; regression analyses showed this to be driven by proac-
cyber bully back. tive aggression. Nationality and gender were not found to be modera-
A series of studies by Ang and Goh (2010), Ang et al. (2010, 2011, tors of the relationship.
2013), Lwin et al. (2012) and Pozzoli et al. (2012), published in interna- Kwan and Skoric (2013) worked in two secondary schools, examin-
tional journals, are directed not so much toward prevalence and nature ing cyberbullying occurring on Facebook, and using scales adapted from
of bullying, but to testing out various models which might predict the North American studies, plus an ofine bullying scale adapted from Ng
relationship between individual psychological characteristics, and (2010); 59.4% of Facebook users reported experiencing at least one
roles in bullying or as a bystander. Ang et al. (2010) worked with pupils form of bullying in the last year (most common was receiving nasty
from one elementary and one middle school. They used a 7-item messages), and 56.9% of engaging in at least one form (most commonly
behavior-based bullying questionnaire to measure bullying others; the blocking someone); intensity of Facebook use was a signicant predic-
power difference between perpetrator and victim was explained prior tor of victimization, and risky Facebook use of bullying. Boys were
to administration. Prevalence rates are not reported; the researchers more involved as bullies or victims, and there were signicant associa-
were interested in testing mediational and moderating models between tions between Facebook bullying and school bullying.
the three variables of narcissistic exploitativeness, normative beliefs Khiat (2012) reported a more qualitative study, with second and
about aggression, and bullying. All three variables correlated positively, third year students at one polytechnic. Some were asked to write an
and support was found for approval-of-aggression beliefs moderating essay reecting on any event related to bullying during their time at
the relationship between narcissistic exploitativeness and bullying. the polytechnic; based on analysis of this, a semi-structured interview
Ang and Goh (2010) worked with adolescents from one middle was carried out with further students. Grounded theory was used to
school and one high school, and examined the association between af- analyze the data. Findings are discussed in terms of the contexts for bul-
fective empathy, cognitive empathy, and gender on cyberbullying lying, aims of bullying, bullying tactics (this includes a typology of
among adolescents. A 9-item cyberbullying behavior scale was used, bullies as Manipulating, Entertaining, Reprising, Excluding), bystanders
validated using EFA and CFA, with a 5-point frequency scale. Frequent (typology of tutor bystanders as Apathetic, Focussed or Intervening;
cyberbullying of others (weekly or more over the last year) was report- pupil bystanders as Reassured, Guilty, Collaborating and Hindering), re-
ed by 3.7% of boys and 0.9% girls; infrequent cyberbullying, which could actions toward bullying (typology of victims as Indifferent, Resigned,
be once or more this year but less than weekly, was reported by another and Proactive), and ways to reduce bullying.
19.9% of boys and 14.2% of girls. As main effects, both low cognitive em- In summary, there is a quite extensive series of studies in Singapore,
pathy and low affective empathy predicted cyberbullying; and this was covering prevalence, correlates, coping strategies including narcissism,
clearly so for boys with no interaction between the two kinds of empa- normative beliefs, empathy, and self efcacy, and a qualitative study.
thy. However for girls, the level of cognitive empathy had no effect if Six studies included cyberbullying.
they had high affective empathy.
Ang et al. (2011) used the cyberbullying questionnaire of Ang and 6.4.3. Thailand
Goh (2010) and the design of Ang et al. (2010) with samples of students Apart from the comparative data from GSHS reported by Pengpid
from Singapore and Malaysia; the number of schools is not reported. and Peltzer (2013) and Wilson et al. (2013), and the TIMMS data from
Prevalence rates are not reported, although the cyberbullying score was 2011 (see Table 4), there are several publications on bullying in
higher in Malaysia than Singapore. The three variables of narcissistic Thailand. A survey by Tapanya (2006) is available only in Thai and is
exploitativeness, normative beliefs about aggression, and cyberbullying thus not in Table 2. Pupils from grades 4 to 9 were given the OBVQ.
correlated positively, and support was found for approval-of-aggression For all forms of bullying, boys were bullied more than girls, especially
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 31

in the lower grades (grades 4 or 5). Most of the victims reported the cyberbullied more than once per month, while in Japan, reported
bully to their friends (56%) followed by their parents (37%), teachers cases of cyber bullying were very low in frequency, with more than
(32%), and siblings (29%). half of respondents said that they have never faced or got information
Laeheem et al. (2008) reported on bullying in 5 public and 2 private about cyber-bullying. In terms of attitudes toward cyberbullying, in
primary schools (one Islamic, one Chinese) in the Pattani district of Thailand, 60% of students believed that cyberbullying was negatively
southern Thailand. The researchers dened bullying as a person's ac- avoidable behavior which should be prohibited, but some 35% believed
tions to cause physical or psychological harm on another person that cyberbullying might be an acceptable behavior; whereas in Japan,
(p.180), so measuring aggression rather than bullying. Pupils were 68% thought that it was bad behavior.
asked if they had ever harmed someone or hurt their feelings, in 16 dif- Songsiri and Musikaphan (2011) reported ndings from a survey of
ferent ways (physical and verbal, but not including cyber). Students pupils from 21 secondary schools in Bangkok. Attitudes to cyberbullying
who said yes to at least 2 out of the 16 categories were labeled as were described as negative, or mixed. Seven types of cyberbullying
bullies. Based on this criterion, 32.9% of students reported that they were reported, of which electronic messages with angry and vulgar lan-
had (ever) bullied other children. The classroom was the most frequent guage, and repeatedly sending nasty, mean and insulting messages,
location. Bullying others was signicantly associated with being older, were the most frequent. The frequency measures are difcult to inter-
with having seen family physical abuse between parents, and with pref- pret, as they included both experiencing these forms oneself, or hearing
erence for action (rather than comedy or mystery) cartoons. A report by about it occurring to one's friends. Duration of internet use and aspects
Laeheem et al. (2009), appears to be based on the same sample of stu- of family relationships (exposure to violence; family income) were
dents and data set. A different (but not completely specied) dichoto- found to relate to the cyberbullying measure, but no details of these
mizing of students into bullied others and not bullied others groups ndings are presented. This team has also produced manuscripts
was carried out, with the bullied others group now constituting written only in Thai (Musikphan, Yongchin, & Chancharoen, 2011;
20.9% of the sample. Signicant risk factors now were in public school Pokpong & Musikphan, 2010).
(rather than private), being male, non-Muslim (rather than Muslim), Tudkuea and Laeheem (2014) worked with young people in
11 years or older (rather than 8 to 10 years), had witnessed parental Songkhla province to devise a 24-item inventory on cyberbullying, cov-
physical abuse, and preference for action (to comedy or mystery) ering 5 forms. Conrmatory factor analysis supported the use of this
cartoons. instrument.
Yodprang et al. (2009) reported on physical bullying among lower Sittichai (2014) sampled students from 12 secondary schools in
secondary school students in Pattani province. The procedure appears three southern provinces using an adaptation of a U.K. questionnaire.
similar to the study on primary school students; however it only Using either strict or lenient frequency criteria, prevalence for tradition-
asked about ve types of physically aggressive behavior, but as both al victimization was 6.0% and 16.0% respectively, and 3.7% and 14.9% for
perpetrator and victim, on a yes/no basis. Questions referred to behavior cyber victimization. About half of cyber victims were also traditional
during both the preceding six months and during the previous month. victims. Victim status was associated with being male (both kinds),
On these criteria, the overall prevalence of physical bullying of others and with higher parental education (only for lenient criterion cyber
was reported as 18.5% (both during the last six months and the last victims).
month). Physically bullying others was not signicantly associated A report from Mahidol University (LGBT-friendly Thailand?, 2014)
with gender, but was associated with being older, Thai compared to used a combination of qualitative and quantitative methods with stu-
Malay or Chinese, urban school rather than rural, and witnessing phys- dents at 3 secondary schools in each of 5 provinces. The report focuses
ical violence between parents. on the experiences of LGBT students; overall, 55.&% of self-identied
Buttabote (2011) (only abstract available in English), reported on LGBT students reported being bullied within the last month because
the relationship between bullying roles and self-esteem in ADHD pa- they were LGBT, with physical, verbal, social and sexual abuse all
tients, being treated at the Child and Adolescent Psychiatric Outpatient being common. Despite a range of negative outcomes from such victim-
Unit of Ramathibodi Hospital, using a Thai version of the OBVQ and the ization, only about one-third used an active coping strategy. A lack of
Rosenberg self-esteem scale. The analysis distinguished four roles: relevant anti-bullying policies in schools, both generally and for LGBT
bully-only, victim-only, bully-victim, and not involved. The proportion students specically, was identied.
of ADHD patients who exhibited or were victims of some bullying Not included in Table 2 is a report by Laeheem (2013b) which dis-
behaviors was 52.9%, most of them being in the bully-victim group. cusses group study activities from an Islamic perspective, or Hallakah,
Victims of bullying most frequently reported being made fun of or as a way to solve problems of bullying behaviors.
teased in a hurtful way, while those who bullied others most frequently In summary, there are a number of studies on prevalence (but some
reported behaviors of hitting, kicking, pushing, and placing someone measuring aggression rather than bullying), and correlates including
under arrest. The self-esteem scores for the victim-only group and self-esteem and ADHD, and four studies on cyberbullying.
bully-victim groups were statistically signicantly lower than for the
not involved group. 7. Discussion
Laeheem and Sungkharat (only abstract available in English) gave a
screening inventory to students at 10 Islamic schools in Songkhla prov- We rst note that the distribution by country of studies on bullying
ince. The properties of the instrument were reported. Laeheem (2013a) in ASEAN countries is patchy: Singapore has the most extensive range of
gathered qualitative data at 15 Islamic private schools. Examples are studies, followed by Thailand and Malaysia. In the other countries there
given of physical, verbal, and social bullying, and of sites such as the is either one extensive report (Cambodia, Vietnam), data mainly or sole-
classroom, toilets, canteen, and school yard. Attitudes to and feelings ly from HBSC and/or TIMMS (Indonesia, Myanmar, Philippines) or no
about bullying are also described. data whatsoever (Brunei Darussalam, Lao PDR). The most research has
Four studies have produced evidence regarding cyberbullying in been done in the countries with highest GDP per capita (Table 1),
Thailand. Musikaphan (2009), in a university report (only abstract which are likely to support a greater research base. The exception to
available in English), provided comparative data on cyberbullying in this generalization is Brunei Darussalam, which is by far the smallest
Thailand, and Japan. Data collection was carried out in Thailand with country of the ten. Given the size of the country and GDP ranking 5th
2,000 students for a quantitative study, and 14 key informants for in the list (Table 1), the lack of research in Indonesia is disappointing.
qualitative purposes; and in Japan with 100 students for quantitative Secondly, the sources reviewed are variable. Some are articles
study and 12 key informants for qualitative purposes. From the quanti- published in international refereed journals. Others are national journal
tative surveys, it was found that nearly 59% of respondents had been articles, perhaps with less intensive peer review; and others are theses
32 R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235

or reports for organizations or charitable bodies. We have indicated the higher prevalence rates, for example by Wan et al. (2009) in Malaysia.
status of the sources in the text. However before seeking cultural or educational variables to explain os-
These studies are starting to make a valuable contribution, and any tensibly higher prevalence rates, we need to consider methodological is-
critical remarks below are made with the aim of constructively helping sues. In the case of Wan et al. (2009) we do not know what term was
research to progress further in an effective way, both locally and in an used for bully or victim, in Malay language, or how it was dened. In
international comparative perspective. Thailand, all the studies have uncertainty related to what Thai
word(s) were used for bullying, but the GSHS data (Table 3) suggest a
7.1. Limitations of reporting prevalence broadly in line with Western gures; however Laeheem
et al.'s (2008, 2009) estimates of prevalence are much higher; this is prob-
Of the 38 independent studies in Table 2, it can be seen that 17, or ably both because of a long time referent period (have you ever ), but
nearly half, do not give the date of data collection. This is also quite a also because their denition describes aggressive behaviors generally
common feature in reports of bullying internationally, especially by psy- rather than bullying specically. Similarly, Yodprang et al.'s (2009) high
chologists, but it is regrettable (Hofstede et al., 2010; Smith, 2010). prevalence rate of bullying others may be inated as a measure of bullying
There are short-term historical changes in the denition of bullying prevalence, because their denition describes physically aggressive be-
(for example incorporation of indirect forms in the 1990s), and also in haviors generally rather than physical bullying specically. In Indonesia,
behavior (for example the development of cyberbullying in the the report by Muryani and Phongpat (2014) produced excessively high
2000s). Furthermore if school-based intervention work started to prevalence rates due to an extremely lenient criterion.
make an impact in these countries, as it has in many Western countries The two comparative data sets, GSHS and TIMMS, do allow compar-
(Krn et al., 2011; Rigby & Smith, 2011), then prevalence rates in par- ison of prevalence across countries; but unfortunately where the coun-
ticular will exhibit historical trends. As an example of how changing tries overlap (notably Indonesia and the Philippines), the two data sets
trends in cyberbullying may affect interpretation of ndings, Ang and give contrary ndings. Using TIMMS, Lai et al. (2008) comment that
Goh (2010) found much less cyberbullying in girls than boys, but it is Students in Philippine had the highest rates of all types of bullying
unfortunate that the date of the study is not given, as some studies sug- (p.395); but the GSHS data in Table 3 show Indonesian pupils reporting
gest a greater involvement of girls in cyberbullying in the last few years higher rates of being bullied than those in the Philippines, in all re-
as social networking sites become more popular (Smith, 2014). It sponse categories! This discrepancy might be due to behavior-based
should be noted that this omission of date of data collection is also vs. denition-based measures, different linguistic terms used, the differ-
found in many western studies (Kochenderfer-Ladd & Troop-Gordon, ent times of the studies (2003 and 2007), sampling differences, or other
2010). methodological factors such as response rates. In addition, and more
Also of the 38 independent studies in Table 2, 25, or nearly two- broadly than just the terms used for bullying, the equivalence of ques-
thirds, do not specify the language used to gather data (by question- tionnaire items in different contexts needs to be considered (van der
naires or interview). Even if we can guess the language, we do not Vijver, Chasiotis, & Breugelmans, 2011). Altogether, the ndings on
know what terms were used for bullying (or related concepts), in prevalence rates in these different countries are not yet on a reliable
denition-based studies. When questions are asked about bullying, the enough basis to warrant systematic attempts to explain in terms of cul-
corresponding term in the language needs to be specied and it's equiv- tural or educational variables, but we do advance some tentative hy-
alence to bullying discussed. With the exception of 3 terms in Thai men- potheses below.
tioned in the introduction, no proper study has been made of the
equivalence of such terms, for example using the cartoon test (Smith 7.3. Cyberbullying
et al., 2002). In Malay the word buli is often used informally, but it
would be important to know if this was the term used in a research Although a few studies in Singapore and Thailand specically
study, and how it was understood. As an example, Yahaya et al. (2009, measured cyberbullying, many of the behavior-based studies assessed
p.651) asked questions such as Do you being bully in school this bullying but did not include cyberbullying items. This includes the
year?, but we do not know how bully was translated in such items. Ex- TIMMS data sets. In the denition-based studies it is possible that
ceptions to this linguistic blindness are Horton (2011), who provides a cyberbullying would be picked up by a global question (such as how
lexicon of bullying-related words in Vietnamese, and discusses the often have you bullied someone in the last term?) but some studies,
meaning of bat nat as the nearest equivalent to bullying, but with a for example Buttabote (2011), appear to have used an earlier version
meaning closer to making others do things they would not otherwise of the OBVQ without items on cyber (or electronic) bullying, which re-
wish to do; and the report on LGBT bullying in Thailand (2014). cent versions of the OBVQ include. The GSHS surveys do not explicitly
If no term for bullying is used directly, as is the case in behavior- include cyberbullying (see Table 3), although it might be picked up in
based measurement studies, then it needs to be very clear that the I was bullied in some other way.
distinguishing criteria of repetition and imbalance of power are Cyberbullying has been noticeable in many countries from at least
being taken account of. If they are not, then it is questionable that bul- 2002 onward (Rivers & Noret, 2010), and the surveys reviewed were
lying is being measured. Some studies have a lax criterion for carried out (where this was reported) from 2003 onward; given reason-
bullyingin fact some are clearly measuring aggressive behaviors ably high internet penetration in many of the countries (Table 1), clearly
and not bullying, thus leading to high prevalence gures. As noted, def- cyberbullying items should be explicitly included in all research, and
initional issues are more complicated for cyberbullying. this is a clear recommendation for future work.

7.2. Prevalence of bullying 7.4. Gender differences

Bullying-like behaviors are not infrequent in the eight countries for Where reported, gender differences are mostly in line with Western
which there is some prevalence data. The sample sizes of studies are ndings, namely that boys are more often perpetrators of bullying,
generally in the hundreds or more, and most of the studies are on pupils whereas there is much less of a gender difference in being a victim of
of secondary school age. bullying. Yodprang et al. (2009) reported no gender difference in phys-
By the HBSC standards for Western countries (c.10% for bullying ical bullying of others, which is unusual, in that almost all studies nd
others, and 11% for being bullied; Currie et al., 2012), some of the physical aggression and bullying to be more common in boys. Gender
reports in Table 2 give broadly comparable prevalence rates; for differences in cyberbullying are very variable across studies (Kowalski
example Yahaya et al. (2009) in Malaysia. Other reports give much et al., 2014; Smith, 2012; Tokunaga, 2010); the much higher rate in
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 33

boys reported by Ang and Goh (2010) is relatively unusual, although the same-age peers, since such behaviors by older peers are more likely to
overall prevalence level is comparable with Western studies (Smith, be considered legitimate. There should also be less bullying of teachers
2012). by students, a phenomenon reported in some western societies (Terry,
7.5. Other ndings The masculinefeminine dimension may also be of interest. The
six relevant ASEAN countries vary greatly on this dimension; the
A number of other ndings not related to prevalence, are also con- Philippines is highest on the masculine end (but below Japan and
cordant with Western ndings; such as verbal bullying being the most China) and Thailand is in the lowest group. In countries high on the
frequent type (many studies), victims being more likely to be depressed masculine end, males are expected to be more tough and aggressive.
(Uba et al., 2010), cyberbullies more likely to have low cognitive and af- Thus higher bullying rates could be predicted, but especially for males,
fective empathy (Ang & Goh, 2010), and perpetrators more likely to with a greater male-female difference in bullying rates. Contrary to pre-
have witnessed domestic violence (Yodprang et al., 2009). In addition dictions then, Rudatsikira et al. (2008) and Fleming and Jacobsen (2010)
the studies by Ang and Goh (2010); Ang et al. (2010), Ang et al. found non-signicant gender differences in the Philippines, whereas
(2011, 2013); Pozzoli et al. (2012); Lwin et al. (2012) make distinctive typical gender differences in victimization rates have been reported in
contributions to the international literature in their testing of individual Thailand.
and peer factors predicting involvement in bullying and cyberbullying. Further investigation of these dimensions would be worthwhile as
one explanatory framework for approaching similarities and differences
7.6. A call for more qualitative research that might be found in further research. In addition, attitudes to bullying
and cultural baselines of bullying tolerance could be assessed; one study
Apart from Horton (2011), Khiat (2012), Laeheem (2013a, 2013b), found attitudes to bullying interventions in schools much more pessi-
and the LGBT report in Thailand (2014), all the studies reviewed have mistic in Japan, than in England (Kanetsuna et al., 2006). The various ed-
been primarily or exclusively quantitative in nature. Also, many have ucational systems also need to be considered; important factors would
used versions of Western instruments, particularly the OBVQ and the include use of homeroom classes, break times and their supervision,
PRQ. While such procedures have advantages in terms of using stan- and extent of grade retention.
dardized instruments and comparability with Western ndings, it re-
mains largely unexamined whether the nature and forms of bullying- 7.8. Neglect of some important variables
like phenomena in ASEAN countries are really very similar to those in
Western countries, or whether they might be more similar to phenom- In light of research in Asian Pacic Rim countries, and hypotheses re-
ena in the Asian Pacic Rim countries. For example, is extortion an garding cultural differences, it would be informative to gather systemat-
important category of bullying (Wong et al., 2008)? Do different forms ic data on some aspects of behavior that are not routinely gathered or
of social exclusion feature more prominently in these more collectivist reported on. These include the magnitude of the gender difference in
societies (Tom et al., 2010)? To ascertain this might require more bullying others; the proportion of bullying that is social exclusion-
qualitative approaches, such as interviews, focus groups, or cartoon based; the ratio of bullies to victims, whether bullies are usually in the
methodology (Smith et al., 2002), to nd out the range of bullying-like same class as victims; whether power imposition by older pupils is
phenomena in a country; rather than using pre-determined categories considered as bullying; and the extent of pupils bullying teachers.
identical to those in Western instruments such as the PRQ.
7.9. Relevance for interventions to reduce bullying
7.7. Moving toward explanations for cultural differences in bullying
Finally, we mention the potential to use comparative information on
Once reliable data is gathered on prevalence and structural charac- bullying for purposes of intervention. School-based interventions have
teristics of bullying in the various countries, a move toward explaining been carried out with some success in Western countries (Tto &
such differences can be made. One approach would be to use the Farrington, 2011), and there has been some intervention work in
Hofstede dimensions (Hofstede et al., 2010); of the ASEAN countries, Japan, South Korea, Hong Kong and mainland China (see chapters in
these are available for Indonesia, Malaysia, Philippines, Singapore, Smith et al., in press). The only intervention study we found in the
Thailand and Vietnam. South-East Asian countries is by Yaakub et al. (2010), which had
The individualism-collectivism dimension has been invoked in pre- mixed results, but is an important start; many of the Western interven-
vious studies: Li, Wang, Wang, and Shi (2010) examined intracultural tion studies have had mixed results, and identifying the important com-
variation in Chinese adolescents, linking endorsement of collectivism ponents for success is a gradual process (Smith, Salmivalli, & Cowie,
with less use of overt and relational aggression. However it may also im- 2012; Tto & Farrington, 2011). In addition there is a Bully-Free Cam-
pact on the forms of bullying; Koo et al. (2008) argued that greater col- paign website at the Singapore Children's Society, with details of
lectivism may explain the emphasis on class-based exclusion and a advice and support.
greater ratio of bullies to victims found in South Korea and Japan. The Anti-bullying intervention efforts in the ASEAN countries will un-
six ASEAN countries for which data are available (Hofstede et al., doubtedly benet from the research already carried out, but also from
2010) are generally high on the collectivism index (about as high as further research that attempts to identify the particular features of
South Korea, China and Hong Kong, and higher than Japan and most bullying-like phenomena in each country, and how to best adapt
western countries). Of these, Indonesia has the highest collectivism existing procedures and devise new ones to be most effective in these
score; this might predict lower bullying scores (partly supported by educational and cultural contexts. Given the signicant prevalence
TIMMS but not by GSHS), and more relative emphasis on social exclu- rates in these as in other countries, and the known deleterious out-
sion (partly supported by GSHS, Table 3). Thailand is also high on collec- comes of bullying-like phenomena, this will surely be a worthwhile
tivism, and does have low social exclusion scores (Table 3). endeavor.
These six ASEAN countries are also high on the power distance (hi-
erarchy) index in Hofstede's categories (Hofstede et al., 2010), 8. Summary
Malaysia being the highest of all countries, the others on a general
level with China and Hong Kong; Thailand the lowest of the six, but Within the last decade, a number of studies of bullying have begun
still higher than South Korea and Japan (and most western countries). to appear from the ASEAN countries, potentially complementing a
This might suggest that more bullying will be reported as between much larger data base from Western and Asian Pacic Rim countries.
34 R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235

Many studies have used western studies of traditional bullying as a Koh, C. W., & Tan, A. (2008). Bullying in Singapore schools. Research monograph no 8.
Singapore: Singapore Children's Society
reference point; but the advent of social media use, social networking stories/ResearchPublicEducationPublications/bullying.pdf and www.scribd.
and cyberbullying may also have contributed to this increase in activity. com/doc/55986047/Bullying-in-Singapore-Schools.
The studies in ASEAN countries are making a contribution both locally, Koo, H., Kwak, K., & Smith, P. K. (2008). Victimization in Korean schools: The nature, in-
cidence and distinctive features of Korean bullying or wang-ta. Journal of School
and potentially for international comparison. More qualitative research Violence, 7, 119139.
would be desirable to ascertain the validity of what are usually western- Kowalski, R. M., Giumetti, G. W., Schroeder, A. N., & Lattanner, M. R. (2014). Bullying in
based instruments. We also recommend that future quantitative studies the digital age: A critical review and meta-analysis of cyberbullying research
among youth. Psychological Bulletin, 140, 10731137.
should clearly specify denitions used, the date of data collection, and Kwan, G. C. E., & Skoric, M. M. (2013). Facebook bullying: An extension of battles in
the language used in collecting it. Cyberbullying should routinely be in- school. Computers in Human Behavior, 29, 1625.
cluded as one form of bullying. Besides prevalence, other aspects of data Laeheem, K. (2013a). Bullying behavior among primary school students in Islamic private
schools in Pattani Province. Social Sciences, 34, 500513.
would be theoretically relevant. Once reliable and comparable cross-
Laeheem, K. (2013b). Guidelines for solving bullying behavior problems among Islamic
national data is obtained, explanations based on cultural, linguistic, private school students in Songkhla province. Asian Social Science, 9, 8389.
attitudinal and educational variables can be more systematically Laeheem, K., Kuning, M., & McNeil, N. (2009). Bullying: Risk factors becoming bullies.
attempted. Asian Social Science, 5, 5057.
Laeheem, K., Kuning, M., McNeil, N., & Besag, V. E. (2008). Bullying in Pattani primary
schools in southern Thailand. Child: Care, Health and Development, 35, 178183.
References Laeheem, K., & Sungkharat, U. (2012). Development of screening inventory for students at
risk of exposure to bullying behaviour in Islamic private schools, Songkhla Province.
Akbar, J., Huang, T. -W., & Anwar, F. (2014). The development of cyberbullying scale to in- Kasetsart Journal: Social Sciences, 33, 175187 (in Thai: abstract in English).
vestigate bullies among adolescents. Proceeding of International Conference on Human- Lai, S. -F., Ye, R., & Chang, K. -P. (2008). Bullying in middle schools: An Asian-Pacic re-
ities Sciences and Education (ICHE), Kuala Lumpur ( gional study. Asia Pacic Education Review, 9, 393405.
Ancho, I. V., & Park, S. (2013). School violence in the Philippines: A study on programs and Lee, S. -H., Smith, P. K., & Monks, C. (2011). Perception of bullying-like phenomena in South
policies. Advanced Science and Technology Letters, 36, 2731. Korea: A qualitative approach from a lifespan perspective. Journal of Aggression, Conict
Ang, R. P., & Goh, D. H. (2010). Cyberbullying among adolescents: The role of affective and and Peace Research, 3, 210221.
cognitive empathy, and gender. Child Psychiatry and Human Development, 41, Lee, S. -H., Smith, P. K., & Monks, C. (2012). Meaning and usage of a term for bullying-like
387397. phenomena in South Korea: A lifespan perspective. Journal of Language and Social
Ang, R. P., Huan, V. S., & Florell, D. (2013). Understanding the relationship between pro- Psychology, 31, 342349.
active and reactive aggression, and cyberbullying across United States and LGBT-friendly Thailand? (2014). Bullying targeting secondary school students who are or
Singapore adolescent samples. Journal of Interpersonal Violence, 30, 118. are perceived to be transgender or same-sex attracted: Types, prevalence, impact, moti-
Ang, R. P., Ong, E. Y. L., Lim, J. C. Y., & Lim, E. W. (2010). From narcissistic exploitativeness vation and preventive measures in n5 provinces of Thailand. Bangkok: Mahidol Univer-
to bullying behavior: The mediating role of approval-of-aggression beliefs. Social sity, Plan International Thailand and UNESCO (
Development, 19, 721735. 002275/22758e.pdf).
Ang, R. P., Tan, K. -A., & Mansor, A. T. (2011). Normative beliefs about aggression as a me- Li, Y., Wang, M., Wang, C., & Shi, J. (2010). Individualism, collectivism, and Chinese adoles-
diator of narcissistic exploitativeness and cyberbullying. Journal of Interpersonal cents' aggression: Intracultural variations. Aggressive Behavior, 36, 187194.
Violence, 26, 26192634. Lwin, M. O., Li, B., & Ang, R. P. (2012). Stop bugging me: An examination of adolescents'
Bauman, S., Cross, D., & Walker, J. (2013). Principles of cyberbullying research: Denition, protection behavior against online harassment. Journal of Adolescence, 35, 3141.
methods, and measures. New York: Routledge. Margono, H., Yi, X., & Raikundalia, G. K. (2014). Mining Indonesian cyber bullying patterns
Buttabote, P. (2011). The relationship between bullying behavior and self-esteem in at- in social networks. Proceedings of the Thirty-Seventh Australasian Computer Science
tention decit / hyperactivity disorder patients. Journal of Psychiatric Association Conference (ASSC 2014), Auckland, New Zealand (pp. 115124).
Thailand, 56(2), 93102 (in Thai: abstract in English). Menesini, E., Nocentini, A., Palladino, B. E., Frisn, A., Berne, S., Ortega Ruiz, R., et al.
Chew, C. (2010). Prevalence of cyber bullying in Singapore. In E. Ng, & K. Rigby (Eds.), (2012). Cyberbullying denition among adolescents: A comparison across
Breaking the silence: Bullying in Singapore (pp. 6583). Singapore: Armour Publishing. six European countries. Cyberpsychology, Behavior and Social Networking, 15,
Craig, W., Harel-Fisch, Y., Fogel-Grinvald, H., Dostaler, S., Hetland, J., Simons-Morton, B., 455463.
et al. (2009). A cross-national prole of bullying and victimization among adolescents Miles, G. (2004). STOP violence against us! Summary report: A preliminary national re-
in 40 countries. International Journal of Public Health, 54(Suppl. 2), 216224. search study into the prevalence and perceptions of Cambodian children to violence
Social determinants of health and well-being among young people. Currie, C., et al. (Ed.). against and by children in Cambodia (
(2012). Health Behaviour in School-aged Children (HBSC) study: International report Mullis, L. V. S., Martin, M. O., Foy, P., & Arora, A. (2012). TIMMS 2011 International Results in
from the 2009/23010 survey. Copenhagen: WHO Regional Ofce for Europe. Mathematics. Boston College: TIMMS & PIRLS International Study Center.
Fleming, L. C., & Jacobsen, K. H. (2010). Bullying among middle-school students in low Muryani, S., & Phongpat, S. (2014). Bullying term of groups and gender in adolescents.
and middle income countries. Health Promotion International, 25, 7384. Musikaphan, W. (2009). A study of cyber-bullying in the context of Thailand and Japan.
Fonzi, A., Genta, M. L., Menesini, E., Bacchini, D., Bonino, S., & Costabile, A. (1999). Italy. In Bangkok, 2009 (in Thai; abstract in English).
P. K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R. Catalano, & P. Slee (Eds.), The nature Musikphan, W., Yongchin, S., & Chancharoen, S. (2011). A study project on management
of school bullying: A cross-national perspective (pp. 140156). London: Routledge. cyberbullying with family participation. Bangkok: Mahidol University (in Thai).
Furlong, M. J., Sharkey, J. D., Felix, E. D., Tanigawa, D., & Green, J. G. (2010). Bullying assess- Ng, E. (2010). What children say about bullying in Singapore. In E. Ng, & K. Rigby (Eds.),
ment: A call for increased precision of self-reporting procedures. In S. R. Jimerson, S. Breaking the silence: Bullying in Singapore (pp. 4163). Singapore: Armour Publishing.
M. Swearer, & D. L. Espelage (Eds.), Handbook of bullying in schools: An international Ng, E., & Rigby, K. (2010). Breaking the silence: Bullying in Singapore. Singapore: Armour
perspective (pp. 329345). New York & London: Routledge. Publishing.
Gonzales, R. H. (2014). Social media as a channel and its implications on cyber bullying. De Olweus, D. (1999). Sweden. In P. K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R. Catalano,
La Salle University, Manila: DLSU Research Congress. & P. Slee (Eds.), The nature of school bullying: A cross-national perspective (pp. 727).
GSHS survey results (2007). London: Routledge.
Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Ong, W., & Elliott, J. M. (2010). Young adult's recall of school bullying. Research monograph
Newbury Park, CA: Sage. no 9. Singapore: Singapore Children's Society (
Hofstede, G., Hofstede, G. J., & Minkov, M. (2010). Cultures and organizations: Software of stories/OurServices/monograph9.pdf).
the mind. New York: McGraw-Hill. Pengpid, S., & Peltzer, K. (2013). Bullying and its associated factors among school-aged
Horton, P. (2011). School bullying and power relations in Vietnam. PhD dissertation 541, adolescents in Thailand. The Scientic World Journal (Article ID 254083).
Linkping, Sweden: Linkping Studies in Arts and Science. Pokpong, S., & Musikphan, W. (2010). Factors affecting attitudes and physically violent
Horton, P. (2012). Bullied into it: Bullying, power and the conduct of conduct. Gloucester- behavior and cyberbullying among Thai youths. Bangkok: Mahidol University (in
shire: E&E Publishing. Thai).
Handbook of bullying in schools: An international perspective. Jimerson, S. R., Swearer, S. Pozzoli, T., Ang, R. P., & Gini, G. (2012). Bystanders' reactions to bullying: A cross-cultural
M., & Espelage, D. L. (Eds.). (2010). New York & London: Routledge. analysis of personal correlates among Italian and Singaporean students. Social
Kanetsuna, T., & Smith, P. K. (2002). Pupil insights into bullying, and coping with bullying: Development, 21, 686703.
A bi-national study in Japan and England. Journal of School Violence, 1, 529. Rahayu, F. S. (2012). Cyberbullying sebagai dampak negative penggunaan teknologi
Kanetsuna, T., Smith, P. K., & Morita, Y. (2006). Coping with bullying at school: Children's informasi. Journal of Information Systems, 8, 2231 (in Indonesian: abstract in English).
recommended strategies and attitudes to school-based interventions in England and Rigby, K., & Slee, P. (1993). The Peer Relations Questionnaire (PRQ). Adelaide: University of
Japan. Aggressive Behavior, 32, 570580. South Australia.
Krn, A., Voeten, M., Little, T., Poskiparta, E., Kaljonen, A., & Salmivalli, C. (2011). A large Rigby, K., & Smith, P. K. (2011). Is school bullying really on the rise? Social Psychology of
scale evaluation of the KiVa antibullying program: Grades 46. Child Development, 82, Education, 14, 441455.
311330. Rivers, I., & Noret, N. (2010). I h8 u: ndings from a ve-year study of text and email bul-
Khiat, H. (2012). Unveiling the intricacies of bullying: Students' perspectives in a poly- lying. British Educational Research Journal, 36, 643671.
technic in Singapore. Asian Criminology, 7, 122. Rudatsikira, E., Mataya, R. H., Siziya, S., & Muula, A. S. (2008). Association between bully-
Kochenderfer-Ladd, B., & Troop-Gordon, W. (2010). Introduction to the Special Issue: ing victimization and physical ghting among Filipino adolescents: Results from the
Contexts, causes and consequences. Merrill-Palmer Quarterly, 56, 441454. Global School-based Health Survey. Indian Journal of Pediatrics, 75, 12431247.
R. Sittichai, P.K. Smith / Aggression and Violent Behavior 23 (2015) 2235 35

Schwartz, D., Chang, L., & Farver, J. M. (2001). Correlates of victimization in Chinese Uba, I. U., Yaacob, S. N., & Juhari, R. (2010). Bullying and its' relationship with depression
children's peer groups. Developmental Psychology, 47, 520532. among teenagers. Journal of Psychology, 1, 1522.
Sittichai, R. (2014). Information technology behavior cyberbullying in Thailand: Incidence van der Vijver, F. J. R., Chasiotis, A., & Breugelmans, S. M. (2011). Fundamental questions
and predictors of victimization and cyber-victimization. Asian Social Science, 10, of cross-cultural psychology. In F. J. R. van der Vijver, A. Chasiotis, & S. M.
132140. Breugelmans (Eds.), Fundamental questions in cross-cultural psychology (pp. 934).
Sittichai, R., & Smith, P. K. (2013). Bullying and cyberbullying in Thailand: A review. Cambridge & New York: Cambridge University Press.
International Journal of Cyber Society and Education, 6, 3144. Wan, S. W. I., Nik, R. N. J., Hatta, S., Marhani, M., & Shamsul, A. S. (2014). Why do young
Smith, P. K. (2010). Victimization in different contexts: Comments on the Special Issue. adolescents bully? Experience in Malaysian schools. Comprehensive Psychiatry, 55,
Merrill-Palmer Quarterly, 56, 441454. S114S120.
Smith, P. K. (2012). Cyberbullying and cyber aggression. In S. R. Jimerson, A. B. Nickerson, Wan, S. W. I., Nik, R. N. J., Tuti, I. M. D., Shamsul, A. S., Aniza, I., & Zasmani, S. (2010). The as-
M. J. Mayer, & M. J. Furlong (Eds.), Handbook of school violence and school safety: Inter- sociation between the attention decit hyperactivity disorder (ADHD) symptoms and
national research and practice (pp. 93103). New York, NY: Routledge. bully/victim problem among Malaysian sixth-graders. ASEAN Journal of Psychiatry, 11,
Smith, P. K. (2014). Understanding school bullying: Its nature and prevention strategies. 18.
London: Sage. Wan, S. W. I., Tan, S. M. K., Nik, R. N. J., Tuti, I. M. D., Syamsul, S., Aniza, A., et al. (2009).
Smith, P. B., Bond, M. H., & Kagitibasi, C. (2005). Social behavior across cultures: Living and School bullying amongst standard six students attending primary national schools
working with others in a changing world. London: Sage. in the federal territory of Kuala Lumpur: The prevalence and associated socio demo-
Smith, P. K., Cowie, H., Olafsson, R., Liefooghe, A. P. D., & in collaboration with 17 graphic factors. Malaysian Journal of Psychiatry, 18, 512.
additional authors (2002). Denitions of bullying: A comparison of terms used, Willard, N. E. (2007). Cyberbullying and cyberthreats. Champaign, IL: Research Press.
and age and sex differences, in a 14-country international comparison. Child Wilson, M. L., Dunlavy, A. C., & Berchtold, A. (2013). Determinants for bullying victimiza-
Development, 73, 11191133. tion among 11-16-year-olds in 15 low- and middle-income countries: A multi-level
Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008). study. Social Sciences, 2, 208220.
Cyberbullying, its forms and impact in secondary school pupils. Journal of Child Wong, D. S. W., Lok, D. P. P., Lo, T. W., & Ma, S. K. (2008). School bullying among Hong
Psychology and Psychiatry, 49, 376385. Kong Chinese primary schoolchildren. Youth & Society, 40, 3554.
Smith, P. K., Kwak, K., & Toda, Y. (2015). School bullying in different cultures: Eastern and Yaakub, N. F., Haron, F., & Goh, C. L. (2010). Examining the efcacy of the Olweus preven-
western perspectives. Cambridge: Cambridge University Press (in press). tion programme in reducing bullying: The Malaysian experience. Procedia: Social and
Smith, P. K., Salmivalli, C., & Cowie, H. (2012). Effectiveness of school-based programs to Behavioral Sciences, 5, 595598.
reduce bullying: A commentary. Journal of Experimental Criminology, 8, 433441. Yaakub, N. F., Haron, F., & Jusoh, A. J. (2009). Pencegahan BULI di sekolah: Manual guru [Bul-
Solberg, M., & Olweus, D. (2003). Prevalence estimation of school bullying with the lying prevention in schools: A teacher's manual]. Kuala Lumpur: Utusan Publications.
Olweus Bully/Victim Questionnaire. Aggressive Behavior, 29, 239268. Yahaya, A., Ramli, J., Hashim, S., Ibrahim, M. A., & Rahman, R. R. R. A. (2009). Teachers and
Songsiri, N., & Musikaphan, W. (2011). Cyber-bullying among secondary and vocational students perception towards bullying in Batu Pahat District secondary school.
students in Bangkok. Journal of Population and Social Studies, 19, 235242. European Journal of Social Sciences, 11, 643658.
Tapanya, S. (2006). A survey of bullying among students. Chiangmai: Psychiatric Depart- Ybarra, M., Boyd, D., Korchmaros, J., & Oppenheim, J. (2012). Dening and measuring
ment, Faculty of Medicine, Chiangmai University (in Thai). cyberbullying within the larger context of bullying victimization. Journal of Adolescent
Terry, A. A. (1998). Teachers as targets of bullying by their pupils: A study to investigate Health, 51, 5358.
incidence. British Journal of Educational Psychology, 68, 255268. Yodprang, B., Kuning, M., & McNeil, N. (2009). Bullying among lower secondary school
Tokunaga, R. S. (2010). Following you home from school: A critical review and synthesis of students in Pattani Province, Southern Thailand. Asian Social Science, 5, 4652.
research on cyberbullying victimization. Computers in Human Behavior, 26, 277287.
Tom, S. R., Schwartz, D., Chang, L., Farver, J. M., & Xu, Y. (2010). Correlates of victimization Website references
in Hong Kong children's peer groups. Journal of Applied Developmental Psychology, 31,
Tto, M. M., & Farrington, D. P. (2011). Effectiveness of school-based programs to reduce GSHS (a).
bullying: A systematic and meta-analytic review. Journal of Experimental Criminology, Singapore Children's Society (a).
7, 2756. TIMMS (a).
Tudkuea, T., & Laeheem, K. (2014). Development of indicators of cyberbullying among html
youths in Songkla Province. Asian Social Science, 10, 7480.
Uba, I. U. (2009). Relationship between bullying, victimization, pro-social behaviour and de-
pression among teenagers in Selangor Malaysia. Unpublished Master of Science thesis
University Putra Malaysia.