Witnessing Violence Among High School Students in Thailand

Abstract

The goal of this study was to explore the levels of witnessing violence among Thai youths and to examine whether a Western model of witnessing violence that relates to personal and family characteristics, peer relationships, own violence, violence in the family, violence in the school, and in the community, is appropriate for the Thai reality. The random sample consisted of 2,897 youths: 1,107 (38.2%) males and 1,722 (59.4%) females. Results indicated that witnessing violence rates among the Thai youths are high, but in general, quite similar to those reported in the west. The most frequently witnessed violent act was hitting and punching in school. Peer characteristics and their advice were the most related factor to witnessing violence followed by own violence activities, violence in the family, and the surrounding subsystems, supporting the similarities of the Thai reality to the Western one. No significant differences were indicated between the rural and inner-city settings. The results were interpreted in light of the cultural characteristics of Thailand.

Keywords

witnessing violence violence peer violence family violence school violence violence in the community

The deleterious effects of witnessing violence have long been recognized, and studies designed to understand this phenomenon and tackle its effects are ongoing. From an ecological viewpoint, cross-cultural and social group differences in witnessing violence should be studied in various cultures and social groups, to better understand this phenomenon and its effects, and tailor appropriate cultural prevention and interventions to address the problem. The primary goals of this study were to explore the levels of witnessing violence among Thai youths, to examine whether a Western model of witnessing violence is appropriate for the Thai reality and to relate the witnessing violence phenomenon among Thai youth to existing violence levels among them.

The Thai Society

The traditional Thai society adhered to the Buddhist religion, which emphasizes harmony, compassion, caring for others and responsibility (Hoffmann, Demo, & Edwards, 1994; Klausner, 2002). Thai teenagers were expected to respect their parents and their religion; Thai girls were taught how to behave properly from an early age, via direct and indirect messages conveyed by parents, relatives, and significant others (e.g., role models, mass media, books, songs, or religious and other cultural practices). Girls learned to be pleasant, friendly, caring, self-sacrificing, and yielding to others’ demands as well as compliant, especially to seniors, and to men. Thai women were socialized to value submissive behavior, to be good, subordinate housewives, and less ambitious than men (Klausner, 2002).

Thailand’s increase in economic prosperity led to urbanization, westernization, and moderation of cultural and religious norms (Rigg, Veeravongs, Veeravongs, & Rohitarachoon, 2008; Vuttanont, Greenhalgh, Griffin, & Boynton, 2006). Thai society is in transition, undergoing a socioeconomic change in many aspects of life: Urbanization and industrialization is increasing, the fertility rate is in decline and the population is aging. Landlessness and the desire for advanced education are increasing, as is the need for labor in the manufacturing industry, drawing children away from their natal villages to better schools and workplaces, leading to changes in community and family life. The traditional village social network, which was based on the extended family and kinship relations, is giving way to a new form of social structures (Klausner, 2002; Yoddumnern-Attig, 1992). Traditional Thai values are undergoing dramatic changes, and Thai society is becoming more independent, egalitarian, aggressive, confrontational, litigious, and competitive. The apparent norm of aggression among the political, business, and bureaucratic elite who serve as role models, goes hand in hand with the erosion of the traditional order, which in turn, leads to higher rates of violence in society (Klausner, 2002).

The rapid increase in economic prosperity, urbanization, and westernization led to the erosion of the extended family and created a reality of lower supervision and monitoring of parents, which resulted in adolescents’ confused identity, and growing behavioral problems (Vuttanont et al., 2006). Thai youth of today are less docile and diffident in their behavior; less easily deterred and will accept a challenge (Klausner, 2002; Yoddumnern-Attig, 1992). Moreover, the coexistence in the Thai society of traditional and contemporary values and lifestyles produced a complex context for adolescents. Thai girls are becoming more economically independent through participation in the labor market and the opportunity for higher education. They adapt new ways of dating, sometimes without their parents’ consent, and find themselves confused because of their desire to become “Western” and at the same time to retain traditional submissiveness. Vuttanont et al. (2006) found that Thai youth had a dual value system and conflicting aspirations. On one hand, they aspire to Western gender roles and relationships, but at the same time, they try to retain traditional values and find themselves unprepared. Thus, Vuttanont et al. found that about 43% of Thai girls 17 years of age or below in their study reported having sexual intercourse, and about 25% of them became pregnant. Their increased participation in the labor market allows for greater socioeconomic independence and Thai girls have become freer in their sexual behavior, a fact which carries implications for youth violence.

Nevertheless, although many traditional elements of family organization are changing, the Thai family is still the source of codes of behavior and ethics; the children take care of their parents in old age and the aged are highly respected by younger people, especially in rural areas (Limanonda, 2010). The core ideology of family roles and duties has remained relatively constant due to an adherence to Theravada Buddhism (Yoddumnern-Attig, 1992). Thai culture still emphasizes patriarchal values reinforcing traditional gender roles within the context of the family (Hoffmann et al., 1994).

Witnessing Violence

The “witnessing violence” concept requires clarification; as in most studies, it is used interchangeably under different terms, definitions, and measurement tools. Whereas most scholars address witnessing violence in terms of observing a violent act, others refer to “indirect exposure” (e.g., Kerley, Xu, Sirisunyaluck, & Alley, 2010) and “community violence exposure” (e.g., Schwartz & Gorman, 2003), which includes witnessing and victimization. In many cases, the reader is unclear about the meaning of “exposure.” Does this mean being victimized, witnessing, or both, as is the case in most studies, and what weight has each of these types of violence in the term “exposure?” The time frame used also varies. The most common tools ask respondents whether they witnessed violence in the “last three months,” “the last year,” “ever,” or “throughout the lifetime.” Moreover, there is a measurement problem with witnessing violence sources that might include direct witnessing and indirect ones such as media or hearsay (Lambert, Nylund-Gibson, Copeland-Linder, & Ialongo 2010). In this study, we measured witnessing violence by asking the respondents to tell us about any violence or threats that they had ever witnessed. We define witnessing violence as acts and threats of verbal, physical, sexual, and psychological violence perpetrated against and by other people.

The Witnessing Violence Phenomenon

In a nationwide U.S. survey of children below age 17, Finkelhor, Turner, Ormrod, Hamby, and Kracke (2009) found that approximately 60% of the children surveyed had been exposed to violence in the past year, either directly or indirectly, 60.6% had experienced one or more direct or witnessed victimizations, 46.3% had been exposed to a physical assault, 6.1% had been exposed to a sexual victimization, and 25.3% had witnessed a family assault or community violence in the past year, and many had experienced more than one exposure. Lambert et al. (2010) performed a longitudinal study of urban African American early adolescents’ exposure to community violence and found that one fifth of the sample reported consistently high exposure, suggesting that the most severe and persistent experiences with community violence are concentrated in a minority of youth. Margolin et al. (2009) conducted a longitudinal study and found that more than 50% of their participants reported two to four types of violence exposure, and that over the three waves of their study, only 10% to 20% of youth were not exposed to any of the violence types.

Slovak and Singer (2001) found that witnessing violence was more prevalent than victimization within the past year. Students reported that experiencing violence, both as victim and witness, was most prevalent in the school setting (approximately 80% of students reported witnessing someone being threatened, slapped, hit, or punched and more than 70% witnessed someone being beaten up in this setting). The next most prevalent was witnessing violence in the neighborhood (approximately 50% of the students reported witnessing someone being threatened, slapped, hit, or punched in this setting) and in the home. Kennedy (2008) found that more than one third of African American youth studied had witnessed an adult beat up another adult within the family and approximately half reported witnessing an adult family member with an injury caused by family violence.

Only two studies related to witnessing violence were found in Asia. Hence, Kerley et al. (2010) found that long-term effects of childhood exposure to family violence (including witnessing) affect the likelihood of Thai women’s psychological and physical intimate partner perpetration, as well as subsequent psychological and physical victimization in adulthood. Chen and Astor (2010) found significant associations between school violence and direct victimization and witnessing victimization among junior high school students in Taiwan.

Witnessing Violence Effects

The effects of witnessing violence are deleterious. Overwhelming evidence shows that violence in the community affects youth violence (Karriker-Jaffe, Foshee, Ennett, & Suchindra, 2009; Kennedy, 2008; Spano, Vazsonyi, & Bolland, 2009). Cunningham (2003) found associations between witnessing interparental violence and own child abuse. Schwartz and Gorman (2003) found a negative association between community violence exposure and academic functioning of children in an urban elementary school. They suggested that children who were exposed to violence in the community might be vulnerable to academic difficulties because their functioning at school was hindered by depression symptoms. Hurt, Malmud, Brodsky, and Giannetta (2001) found that children with higher exposure to violence had lower grade point averages, more days of school absence, suffered from symptoms of anxiety and depression, and had lower self-esteem than children with less exposure. Slovak and Singer (2001) studied rural youth exposed to gun violence and found that 25% of their participants had been exposed to this at least once. The more prevalent effects of this exposure were anger, dissociation, posttraumatic stress, total trauma, higher levels of violent behaviors, and exposure to violence in other settings.

Risk Factors and Correlates of Witnessing Violence

Personal characteristics were studied regarding witnessing violence. Gender correlates with witnessing violence. Thus, Slovak, Carlson, and Helm (2007) found that males reported higher levels of violence exposure than females. Gender differences were also found to play a critical role in the study of aggression, violent behavior, and witnessing violence. Boys were found to be more exposed to community aggression and violence, to experience more threats and beatings, and to engage in more fights than girls (Farrell & Bruce, 1997). However, Lambert et al. (2010) claimed that males and females in their study may have had similar exposure to the neighborhood, and therefore experienced similar levels of community violence and Kennedy (2008) found no gender differences in the total number of types of witnessed violence.

Concerning the age variable, Slovak et al. (2007) found that younger students were more exposed to violence than older students. However, Slovak and Singer (2001) did not find age differences in the effects of exposure to firearm violence. Studies on religiosity and violence have shown lower commission of violent acts among more religious adolescents (Benda & Corwyn, 2002). A decrease in academic achievements as a result of witnessing violence was indicated in the West (Farrell & Sullivan, 2004; Schwartz & Gorman, 2003), and Ruangkanchanasetr, Hetrakul, Plitponkarnpim, and Kongsakon (2005) claimed that poor school performance was related to adolescent risk behaviors among Thai adolescents.

Family characteristics and especially violence in the family were recognized as an important factor of own violence and also as related to witnessing violence. Margolin et al. (2009) found that the longer the youth were exposed to community violence, their likelihood of also being exposed to mother-to youth and father-to-youth aggression increased. Slovak and Singer (2001) found that lower levels of parental monitoring were associated with exposure to gun violence among rural youth. Finkelhor, Ormrod, and Turner (2009) claimed that their findings support the contention that parents play a crucial role in moderating the influence of witnessing violence in the community. Laeheem, Kuning, McNeil, and Besag (2009), who studied 1,440 elementary school students in one province of Thailand, found that students who witnessed parent’s physical abuse were 4.50 times more likely to bully others than those who had never witnessed physical abuse in the family. Kerley et al. (2010) found an association between childhood exposure to family violence and adult intimate partner perpetration, mediated by partner victimization, among Thai women.

The neighborhood characteristics were indicated as one of the more influential factors of witnessing violence. The literature review of Cooley-Strickland et al. (2009) indicated a wide array of emotional, social, and behavioral influences on youth as a result of exposure to community violence. (See Margolin & Gordis, 2000, for a review of children’s reactions to community violence, interparental violence, and problems associated with exposure to violence.) Zimmerman and Messner (2010) found that as the level of disadvantage of the neighborhood increases, the exposure to violent peers increases at approximately the same rate for males and females. The economic and educational levels of the rural areas in Thailand are lower than the inner cities environments, Thai youths, especially those growing up in smaller, rural communities, experience living in a more traditional community in which the community leader (a figure with many formal and informal duties), elderly people, and older family members are highly regarded and thus relatively more influential than within communities in Western cultures. Thus, our expectation is that in rural areas, due to their more traditional nature, the violence level is lower, and thus witnessing violence will be lower than in urban neighborhoods.

The association between witnessing violence and alcohol use is acknowledged and based on extensive research. Vermeiren, Schwab-Stone, Deboutte, Leckman, and Ruchkin (2003) compared the relationships between exposure to community violence and substance use in three countries. They found that the levels of reported smoking, alcohol use, marijuana use, and hard drug use were associated with higher exposure to community violence in all three countries. Sullivan, Kung, and Farrell (2004) found that sixth graders’ witnessing of community violence was associated with subsequent initiation of cigarette, beer, wine, liquor, and advanced alcohol use. Kliewer et al. (2006) studied the effects of violence exposure among adolescents in Panama and Costa Rica and found that witnessing serious violence was associated with greater drunkenness, tobacco use, number of illicit drugs used, and problems with drugs and alcohol. A recent study of Thai adolescents revealed that 37.3% used alcohol and 37.8% had used at least one illegal drug during their lifetime (Ruangkanchanasetr et al., 2005), all of which may higher their exposure to violence.

Peers may affect the likelihood of witnessing violence by supporting some behaviors and avoiding others and serving as role models. The social learning theory (Akers, 1996) proposes that peer influence derives from what peers think (or say) is right or wrong and from what they do, not only in the sense of committing or not committing delinquency but also in the sense of modeling, reacting to, instructing, and supporting some behaviors. According to the theory of reasoned action, normative beliefs refer to the perceived behavioral expectations of important referent individuals or groups, such as the person’s family and friends. These normative beliefs—in combination with the person’s motivation to comply with the different referents—determine the prevailing subjective norm regarding the behavior (Ajzen, Brown, & Carvajal, 2004).

Along these lines of reasoning, we expected that association with antisocial peers would increase the likelihood of witnessing violence by the mere fact of having more opportunities for exposure to such experiences.

In the following section, we discuss theoretical underpinnings and empirical evidence of the formation of witnessing violent behavior by male and female adolescents in different subcultures.

Theoretical Background

The ecological theory (Bronfenbrenner, 1979) emphasizes the ongoing interaction of a developing child within the subsystems of the family (the microsystem), the community (the exosystem), and the society (the macrosystem), all of which influence and shape the child’s personality and behavior. Youths’ personality is shaped by the ongoing influence of different factors across the individual, family, peer, school, and community subsystems and their reactions to these influences, as well as the patterning of environmental events and life transitions that affect the child’s development and behavior.

Similarly, the behavioral theory explains violent behavior in terms of a learning process that involves modeling from the surrounding subsystems such as family, friends, community, and school (Bandura, 1986). Moreover, the theory explains human functioning as a result of reciprocal interplay of intrapersonal, behavioral, and environmental determinants. Thus, children learn about intimate relationships through direct behavioral conditioning and by imitating the behavior observed in their families (Bandura, 1986). These two theories emphasize the effects of the personal, familial, peers, school, and community subsystems, which we intend to study in our effort to learn about their relation to the phenomenon and levels of witnessing violence among Thai youth.

In addition, the differential association theory (Sutherland & Cressey, 1966) proposes that criminal motivations, attitudes, and behaviors are learned and adopted when group exposure to criminal norms exceeds exposure to anticriminal norms. Mazefsky and Farrell (2005) indicated that exposure to violence provides a powerful context for observational learning that can teach children aggressive behaviors and weaken their inhibitions against such behavior (Bandura, 1986).

Cultural explanations for violence were indicated in the literature. The behavioral theory (Bandura, 2001) proposes that sociostructural factors operate through psychological mechanisms of the self-system to produce behavioral effects. Bandura (2001) claimed that cultures differ markedly in aggressiveness, and that wide differences in aggression within the same culture might exist. A given utilization of violence in a national context might affect witnessing violence among youth. Sykes and Matza (1957) indicated that cultures in which justifications for the use of force and aggression are prevalent serve as a favorable cultural substrate for the emergence of violent behavior. Hence, a higher rate of witnessing violence can be expected in more violence-supporting cultures.

The literature review indicates multifaceted aspects related to witnessing violence. Aside from existing violence rates in a given society, witnessing violence rates among youth are rooted in chances youth have to face such realities in their ecological systems. The Western witnessing violence model relates personal characteristics (such as using alcohol, academic achievements, violence), familial characteristics (such as parents’ education and family income, violence in the family), peers characteristics (such as violence and behavior), school characteristics (such as violence in school), and neighborhood characteristics (such as violence in the community), as well as wider social circumstances, to the likelihood of youth to witness violence. Given the similarities of the Thai youth violent behavior (Pradubmook-Sherer & Sherer, 2011) to the Western model of violence, and the expected relation between violence rates and witnessing violence (Finkelhor, Turner, et al., 2009; Lee, 2012), we expect to find similar variables that shape the witnessing violence phenomenon in Thailand to Western ones, taking into account some differences resulting from differences in samples, age groups, measurement tools, and the Thai cultural reality.

This study dealt with the main subsystems expected to influence youths’ witnessing of violence following the theoretical bases, in an effort to learn about and understand the phenomenon of youth witnessing violence in Thailand. We expected to find some differences in the relative witnessing of violence as a result of the existing differences between Thai culture and Western cultures. Specifically, we expected to find lower percentages of Thai youth who witnessed violence compared with the percentage of youth who witnessed violence in Western countries. We expect to find higher levels of witnessing violence among males than females, and in inner city compared with rural areas. These relationships will be tested by Mancova tests. In this study, the effects of various risk factors as well as their interactive effects on witnessing violence will be tested in regard to the possible impact of personal characteristics, family characteristics, peer influence, violence effects (own violence, violence in the family, in the school, and in the community), and the community characteristics (rural/inner city) using a structural equation model (SEM) test. Our model predicts that the witnessing violence rates will be affected directly through students’ personal, family, and peer characteristics, and by own violence, violence in the family, in the school and in the community (see Figure 1).

Figure 1.

Expected structural equation model relationships among variables related to witnessing violence.

Method

Sample

The sample was based on the results of a previous study of risk behavior in Thailand (Ministry of Social Development and Human Security, 2006) that differentiated among three levels of violence-risk communities: warning (high violence areas), surveillance (areas with growing rates of violence), and normal (areas which have lower violence rates than the other two types); and the four regions of Thailand: northern, northeastern, central, and southern (United Nations Thailand, 2012); and Bangkok which, although included in the central region, was chosen by the ministry to represent the metropolitan area. A complex multistage sampling procedure was used. We randomly chose two districts in each region, from among the warning and surveillance risk levels as well as two districts in Bangkok, due to its size.

In each district, we randomly chose two public high schools. Basic education in Thailand is divided into 6 years of primary schooling followed by high schools which include 3 years of lower high school and 3 years of upper high school. The lower and upper high schools are within the 13 to 18 years of age. In each high school, we randomly chose five classes from among the 7th to 11th grades. Aside for the ministry of education approval, supervision, and monitoring, student consent was assured by asking them to participate in the study and inform them on their right not to participate in the study. Only few chose not to participate. All the students who were present on the day that the questionnaires were administrated were included in the study. The sample consisted of 2,897 youths: 1,107 (38.2%) males and 1,722 (59.4%) females.

Procedure

The study was based on a structured self-report questionnaire. The questionnaires were distributed in the high schools by trained research assistants. The research assistants offered help to respondents who encountered difficulties in completing the questionnaire. The study was conducted over a period of 6 months, during the summer of 2008.

Instruments

We used a Thai translation of scales originally in English. The scales were translated to Thai by three experts and then back translated to English by three other experts and verified, so as to validate the translation. When possible, our translation was compared with other Thai translations. The questionnaire was pretested with a group of 40 students and appropriate adjustments were made. The following instruments were used in the study.

Individual Attributes

A demographic questionnaire addressed gender, age, religiosity (How often do you perform religious practices? Never = 0 to Regularly = 3), working status, working hours a day, income per month, number of siblings, average grade score last year and alcohol/drug use, which was measured by the following questions: During the last year, “How many times did you get drunk?” “How often did you use light drugs?” and “How often did you use hard drugs?” The 7-point scale answers ranged from 1 = Never to 7 Every day.

Family Characteristics

This included parents’ age, education, religiosity (How often do your parents perform religious practices? never = 0 to regularly = 3), profession, occupation and work status, family income, family economic status (very good = 4, moderate = 3, moderate to poor = 2, poor = 1), and family structure (two biological parents = 5, one biological parent = 4, single parent = 3, separated = 2, remarried = 1).

Violence in Surrounding Subsystems

Six scales were used to measure violence (Sherer & Karnieli-Miller, 2004). Each scale asked the respondent whether someone (parents, teachers, or adults) had used violent behavior toward him or her or others during the last 3 months. Each scale included four items (cursing or insulting; hitting, that is, slapping; severe hitting, that is, hitting hard to the extent of requiring medical help; sexual abuse). The response scale ranged from 0 = never to 4 = very often. The original alpha score for these scales ranged from .70 to .82.

The six scales are as follows: (a) Violent behavior of teachers toward students—We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a one-factor solution explaining 70.19% of the violence. Cronbach’s alpha score for this subscale = .84. (b) Teachers’ violence toward you in the last 3 months—We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a one-factor solution explaining 63.61% of the violence. Cronbach’s alpha score for this subscale = .76. (c). Violence between adults in the community—We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a one-factor solution explaining 69.17% of the violence. Cronbach’s alpha score for this subscale = .83. (d). Adults’ violence toward youth in the community in the last 3 months—We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a one-factor solution explaining 66.63% of the violence. Cronbach’s alpha score for this subscale = .81. (e). Parent’s violence toward you—We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a one-factor solution explaining 64.83% of the violence. Cronbach’s alpha score for this subscale = .80. (f). Father’s violence toward mother (due to cultural restrictions, we omitted the question about sexual violence between the parents)—We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a one-factor solution explaining 68.71% of the violence. Cronbach’s alpha score for this subscale = .76.

Own Violent Behavior

Own violent behavior was measured by 11 questions in the form of “Have you threatened other youth using words or a weapon?” The response scale included “number of times in the last 3 months.” Based on a principal components factor analysis with varimax rotation, we found three factors. Factor 1 explained 31.46% of the variance and included four items: sexual harassment, physical violence against date, forced dates to have sex, injured others due to fighting (sexual and physical violence). The Cronbach’s alpha score for this factor = .88. Factor 2 explained 23.61% of the variance, including four items: threw bottles or objects to hit others, participated in group fighting, carried knives or guns, used knives or guns in fighting (severe violence). The Cronbach’s alpha score for this factor = .84. Factor 3 explained 15.33% of the variance and included four items: threatened other youth using words or a weapon; hit, punched, or slapped others; bullied; and took objects or money from others (mild violence). Cronbach’s alpha score for this factor = .58.

Friends’ Behavior

This scale included six questions: “Do most of your close friends study/fight/use alcohol/use drugs/use verbal aggression/carry a weapon?” Responses were scored on a dichotomous scale (1 = Yes, 0 = No). We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a two-factor solution. Factor 1 included five variables dealing with negative behaviors (aggression, fights), explaining 44.78% of the variance (negative behavior). Cronbach’s alpha score for this subscale = .77. Factor 2 included one item dealing with friends’ positive behavior (studying), explaining 17.70% of the variance. The higher the score, the higher the rate of this type of behavior.

Friends’ Dating Violence Behavior

Friends’ dating violence behavior was measured by the attachment to abusive peers scale (DeKeseredy & Schwartz, 1998, p. 101). This scale consists of three questions, for example, “To the best of your knowledge, how many of your friends have ever made physically forceful attempts at sexual activity with women whom they were dating?” The answering scale ranged from 1 to More than 10 and Don’t know. We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a single-factor solution explaining 55.22% of the variance. Original Cronbach’s alpha score for this subscale was α = .65. Cronbach’s alpha score for this subscale = .70.

Peers’ Informational Support

This measure is based on the scale developed by DeKeseredy and Schwartz (1998) that dealt with peers’ informational support regarding dating violence. We adapted this scale to measure peers’ informational support regarding violent behavior in Thai society. This subscale includes six items, two original and four dealing with violent behavior, such as “Did any of your friends ever tell you that it is acceptable to hit someone when insulted?” The responses were scored on a dichotomous scale (1 = Yes, 0 = No). We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a two-factor solution. Factor 1 included three variables dealing with friends’ advice to fight explaining 28.19% of the variance. Cronbach’s alpha score for this subscale = .47. Factor 2 included three items dealing with friends’ advice about dating violence, explaining 24.82% of the variance. Original Cronbach’s alpha score for this subscale was α = .70. Cronbach’s alpha score for this subscale = .55. The higher the score, the higher was the level of the advice to use violence against others and against the partner.

Witnessing Violence

Witnessing violence was measured by 18 questions in the form of “Have you ever witnessed hitting, punching, slapping among youth in your community?” (Other subjects included threatening, group fights, bullying, destroying public property, carrying knives and guns, sexual harassment.) The answering scale ranged from never (0) to more than 10 times (5). We factor analyzed these variables using a factor principal components analysis with varimax rotation, which indicated a four-factor solution. Factor 1 included five variables dealing with youth carrying guns in school, physical attacks on teachers, sexual harassment, injury because of group fights (witnessing carrying guns), explaining 16.66% of the variance. Cronbach’s alpha score for this subscale = .80. Factor two included five items dealing with hitting, punching, slapping, and group fighting (witnessing hitting and punching), explaining 15.10% of the variance. Cronbach’s alpha score for this subscale = .79. Factor 3 included five items dealing with threatening, and carrying knives in school, explaining 14.55% of the variance (witnessing threats). Cronbach’s alpha score for this subscale = .82. Factor 4 included three items dealing with bullying and destroying public property (witnessing bullying), explaining 13.19% of the variance. Cronbach’s alpha score for this subscale = .70.

Results

The demographic characteristics of the sample are described in Table 1. From Table 1, we acknowledge that most of our participants were females—an expected result, as the dropout rate from school is higher among males. Most of our participants were not religious, did not work, and had an average of three brothers and sisters. Most of the parents were married, and approximately 16% were divorced or remarried. Family income was low; fathers had a higher education level than mothers (see Table 1).

Table 1.

Sample Characteristics.

Variables	Numbers	M or percent	SD	Minimum	Maximum
Gender
Male	1,722	60.9%
Female	1,107	39.1%
Age		15.66	1.83	12	20.00
Religiousness (performs religious duties?)
Never	42	1.5%
Seldom	1,541	53.7%
Often	954	33.2%
Regularly	333	11.6%
Total	2,897	100%
Working status
Works	151	5.2%
Don’t work	2,743	94.8%
Working hours a day		.22	1.33	0	12
Income per month		140	1135	0	30,000
Number of siblings		2.42	1.47	0	15
Parents’ family structure
Married	2,162	74.6%
Divorced	376	13%
Widow	194	6.7%
Remarried	78	2.7%
Single parent	20	0.7%
Parents’ religiousness
Never	30	1.0%
Seldom	1,129	39.5%
Often	1,159	40.5%
Regularly	542	18.9%
Family income
Less than 6,000	850	30.1%
6,001 to 10,000	947	33.6%
10,001 to 20,000	526	18.6%
20,001 to 30,000	243	8.6%
More than 30,000	255	9.0%
Fathers’ education
Primary or lower	1,389	53.1%
High school	809	30.9%
Part of bachelor’s degree	167	6.4%
Bachelor degree	196	7.5%
Higher than bachelor	57	2.25%
Mothers’ education
Primary or lower	1,662	61.7%
High school	666	24.7%
Part of bachelor’s degree	133	4.9%
Bachelor’s degree	190	7.1%
Higher than bachelor’s	42	1.6%
Family structure
Only father is alive	147	5.1%
Only mother is alive	47	1.6%
Separated	376	13%
Married and live together	2,182	75.3%
Single-parent family	78	2.7%

Witnessing violence was indicated by the vast majority of the participants. The most frequently witnessed violent act was “hitting, punching, and slapping in school,” which was indicated by 61.8% of the participants (M = .87, SD = .93), followed by “hitting, punching, and slapping in the community” witnessed by 46.6% of the participants (M = .61, SD = .85), “destroying school property” by 40.3% (M = .57, SD = .86), “injuries by fights in school” by 35% of the participants (M = .48, SD = .79), “threatening” by 34.2% (M = .51, SD = .88), and “bullying in school,” which was witnessed by 28.1% (M = .41, SD = .80). As indicated in the West, witnessing violence among the Thai youths was highly observed in the schools and neighborhoods. As expected, these results are lower than the ones observed in Western societies.

We used a Mancova test to study the differences between community type (inner city = 0, rural = 1) and gender (females = 0, males = 1) on the witnessing violence scales. Given that rural areas are socially, economically, and educationally inferior to the metropolitan areas (McGregor, Camfield, Masae, & Promphakping, 2008), we controlled for father and mother education, and family income effects. The Mancova test indicated a significant overall effect: Wilks’s Lambda = .896, F(4, 2402) = 95.41, p < .001. Significant differences were indicated on gender: Wilks’s Lambda = .992, F(4, 2402) = 5.04, p < .001, n = .008. Contrary to our expectation, no significant differences appeared on the community type variable. Significant univariate main effect of witnessing violence by gender was indicated on all four witnessing violence measures. F(1, 2405) ranged from 4.41 to 19.69, p < .03. On all four variables, males had higher mean scores than females (on witnessing carrying guns, males M = .33, SD = .56; females M = .27, SD = .44. On witnessing hitting and punching, males M = .60, SD = .63; females M = .56, SD = .60. On witnessing threatening, males M = .54, SD = .72; females M = .42, SD = .59. On witnessing bullying, males M = .37, SD = .54; females M = .30, SD = .46). No interaction effects were indicated. These results support our expectations of finding higher mean scores of witnessing violence among males.

To examine the structural relations between the risk factors and witnessing violence, we used the SEM procedure with the AMOS program (Arbuckle, 1999). The model produced the following fit results: Sample size ≈2,897; χ²(df = 267) = 19,369.85, p < .001. NFI = 0.90; IFI = 0.92; CFI = 0.92; root mean square error of approximation (RMSEA) = 0.046 (0.045 > RMSEA < 0.048; see Figure 2). The significant chi-square statistic might result from the large sample size. In such cases, fit indices offer an estimation of the fit of the model (Kline, 2005). The fit indices indicate that the model fits the data well.

Figure 2.

A structural equation model of youth violence.

Except for the path from family characteristics to witnessing violence, all other specified paths were significant. As can be seen from Figure 1, the strongest predictor of witnessing violence was peers’ effects (β = .64, p < .001), followed by violence effects (β = .38, p < .001) and personal characteristics (β = −.07, p < .001). Family characteristics were not significant.

Discussion

This study dealt with witnessing violence among Thai youth. Following theoretical expectations regarding risk factors of witnessing violence, we examined the levels of witnessing violence in the family, the community, and school along with peers’ personal and familial characteristics to find the relations that exist among these subsystems and witnessing violence among youth in Thailand.

We found that witnessing violence rates among the Thai youths are high, but lower than the ones of youth in the Western countries (Finkelhor, Turner, et al., 2009; Margolin et al., 2009; Slovak & Singer, 2001). It should be noted, however, that the comparison is problematic due to different measurement times: Our study asked about “ever” witnessing violence, comparative studies asked about the “past year” and other cases reported the percentages of several types of violent acts simultaneously (Slovak & Singer, 2001).

Reported rates of witnessing violence by youth in the States and in Thailand may be different due to differences of samples, age groups, measurement tools, culture, and more. Thus, although we found some expected differences in rates of witnessing violence between the United States and Thailand, there are some similarities between them as well. The most frequently witnessed violent act among Thai youth was hitting and punching in school (about 61%), as was the case in the United States (about 80%) (Slovak & Singer, 2001). Thai youth experienced other acts to a similar extent (such as violence in the community—about 50% of the youth in the United States and 46.6% in Thailand). These similarities reflect the common experience of youth in various countries. In both cultures, youth spend many hours a day in school in which they have an opportunity to experience and witness a wide variety of violent acts (Chen & Astor, 2010).

The reported rates of witnessing violence in Thailand are alarming, reflect an escalating phenomenon of violence among Thai youth (Klausner, 2002; Pradubmook-Sherer & Sherer, 2011), and call for wide social interventions at the family, school, community, and peers levels, using educational, cultural, and social measurements to tackle the phenomenon of witnessing violence among the Thai youth.

Thai people view violence as a normal phenomenon among youth (Sherer et al., 2008), and given the theoretical connection between violence and witnessing violence, we may expect them to accept current levels of witnessing violence as normal as well. This may lead to lower emphasis on the importance of this phenomenon, and even to higher rates of witnessing violence because of less emphasis on measures to tackle this phenomenon.

Regarding risk factors, in general, our results indicated similar effects to those indicated in the West. In terms of personal characteristics, our results indicate higher age to be related to higher witnessing violence, which supports findings by Finkelhor, Turner, et al. (2009) but contradicts the finding of Slovak et al. (2007). This outcome may result from a cultural impact. In the Thai society, the older the Thai youth get, the more independent they are to wonder and the more they become attached to their friends and regard peers’ importance, as indicated in our results. As associating with peers is related to higher violence rates, it follows that it results in higher witnessing violence as well.

We found academic achievements in the preceding year related to a lower degree of witnessing of violence among the participants, supporting Farrell and Sullivan’s (2004) findings of a decrease in academic achievements as a result of witnessing violence.

Although the loadings on alcohol consumption were low, they relate to a higher degree of witnessing violence. This result is widely supported by the literature (Voisin et al., 2007) and might reflect the increasing rate of alcohol consumption among Thai youths (Assunta, 2001; Ruangkanchanasetr et al., 2005), which might be a point of concern for policy makers in Thailand. Religiousness impact witnessing violence seems that being more religious leads to behaviors that limit the exposure to violence activities.

Peer characteristics, their behavior, and their advice were the most related risk factor to witnessing violence. This result is supported by the literature to some extent (Chen & Astor, 2010; Quinn, Bell-Ellison, Loomis, & Tucci, 2007; Ward, Martin, Theron, & Distiller, 2007). This indicates that peer support is an important component of the adolescent culture that underlies influences, often creates violence, and lends support to the theory of reasoned action (Fishbein & Ajzen, 1975), which predicts that one’s attitudes, norms, and behaviors are influenced by the approval of significant others. However, Lambert et al. (2010) indicated that deviant peer affiliation was not a predictor of community violence exposure. Our result may possibly reflect the relative important role of peers among youth in Thailand.

No significant differences of witnessing violence were indicated between the rural and inner-city settings. This result can be explained on two grounds. One possible argument is that the Thai culture is strong and influences violence rates among Thai people to the same extent, regardless of community type, and thus witnessing violence is the same in both settings. An alternative explanation is based on an argument that although rural communities are more traditional in some aspects, they are becoming increasingly confrontational and aggressive—differences of violence between rural and inner-city settings are becoming blurred (Klausner, 2002). This explanation is supported by the social learning theory, which emphasizes that deviant models are reinforced by various socializing institutions, by the media and among peers (Akers, 1996). Moreover, because we studied youth, we might expect them to have higher exposure to the media and higher adaptation to changing realities. Thus, the rates of violence and witnessing violence show no difference among Thai youth from inner-city or rural settings.

As expected, we found higher rates of witnessing violence among males than among females (Slovak et al., 2007). However, studies in the West indicated mixed results as well (Farrell & Bruce, 1997; Kennedy, 2008; Lambert et al., 2010). The differences between the Thai males and females may partly be explained by the witnessing violence rates in the community. Regarding the difference in witnessing violence in the community, Farrell and Bruce (1997) and Kennedy (2008) indicated that boys are more likely to witness severe community violence. Although males witnessed more violent acts than females, it should be noted that the female participants witnessed all types of violence. This result could be explained by behavioral differences between the genders, such as associating with troubled peers leading to exposure to higher rates of community violence. The gender differences could also be explained by gender roles that cut across cultural boundaries. Even though the Thai reality is changing, cultural restrictions still expect girls to be modest, and less exposed to their surroundings. Thus, the females have a lower likelihood of witnessing violence in their communities, which may emphasize that the witnessing violence differences between males and females in Thailand may partly be explained by cultural expectations of the Thai society. This should not be the case regarding witnessing violence in school or at home, and should be the same for males and females.

We found only one study in the United States that indicated a difference in gender reports of witnessing parents’ nasty fights; 44. 2% of the females reported witnessing nasty fights between parents, whereas only 38% of the males attested to this violence (Ghanizadeh & Tavassoli, 2007). This might indicate a cultural effect and further studies are needed to clarify this point. It seems that the difference in witnessing and reporting violence in the family might result from gender differences in the exposure to family violence, interpretation given to the same phenomenon, or from gender differences in the readiness to reveal such information. Thus, Miller and White (2003) explained that the meanings and consequences of violence by females are strikingly different from those of males and that both are grounded in gender inequalities. These explanations coincide with the social role theory (Archer, 2006; Eagly & Wood, 1999) proposing that one reason that women and men conform to gender stereotypes is because they act in accordance with their social roles, which are often segregated along gender lines, and that these roles produce expectancies that lead to different patterns of behavior in men and women.

In terms of own violence and violence in the surrounding subsystems, we found own violence, family violence, violence in school, and in the community related to higher rates of witnessing violence. These results are widely supported by the literature and indicate the similarities of risk factors in the West to the Thai realities (Cunningham, 2003; Kennedy, 2008; Spano, Vazsonyi, & Bolland, 2009).

We think that the home is probably the first place in which violence is witnessed, followed by school and then the community. However, as primary high school education is compulsory in Thailand, and considering the higher rates of violence among youth, the likelihood of witnessing violence during adolescent years at school is probably higher than that in the community (Slovak & Singer, 2001). Moreover, we think that witnessing might have additive effects. Thus, children who witnessed violence at school after witnessing it at home might be far more influenced than children who were not exposed to violence at home but who became exposed to violence at school. The same goes for witnessing violence in the community. This line of reasoning is supported by studies that indicated that violence in one place reverberates into other environments (Slovak & Singer, 2001). Moreover, the importance of witnessing violence might be indicated by the behavioral theory, which explains that violence is learned from modeling of the surrounding subsystems such as the family, friends, community, and school (Bandura, 1986), and the ecological theory (Bronfenbrenner, 1979), which emphasizes the influences of the surrounding subsystem of the family, the community, and the society on children’s personality and behavior. We should emphasize that even if youth report no direct witnessing of violence, they might be exposed to community violence via hearsay or media (Lambert et al., 2010). The risk in higher levels of witnessing violence is that youth may think that such behaviors are “normal” and accepted, thus leading to adaptation of such behaviors among the youth.

Moreover, we believe that risk factors are related. Living in a violent neighborhood might be related to greater likelihood of witnessing violence in the family, victimization in the neighborhood and school, and own violence perpetration. Kennedy (2008) found a linkage between witnessing an adult being beaten within the family, being punched, or hit by a parent or adult caregiver, witnessing adult family members injured due to family violence and being beaten up in the community, for male and female students. Similar results were indicated in a Thai context. Laeheem et al. (2009) found among primary school students that witnessing physical abuse within the family was the strongest associated determinant of bullying. The behavioral theory (Bandura, 2001) would explain this association, arguing that children witnessing family violence might copy the parents’ behavior and might then use violence to achieve success in their own social interactions, by that be more exposed to witnessing and experiencing violence in their life.

Given the gradual westernization of Thai culture, it is imperative to continue the study of youth violence and witnessing violence by Thai youth. We think that cultural differences must be studied specifically for each individual behavior. Regarding the traditional nature of the Thai culture, we believe that some traditional cultures might support violence because of their values and norms, whereas others might not. It appears that the Thai culture, which endorses male supremacy, also lends some support to youth violence, and thus we find high rates of witnessing violence by youth in Thailand.

Limitations, Implications, and Recommendations for Further Study

Aside from the possibility that the participants did not reveal their true experiences because of social desirability, or their unwillingness to report such behaviors, it is possible that males and females interpret violent acts differently (Edelen, McCaffrey, Marshall, & Jaycox, 2009). Moreover, despite our careful measures to validate our tools, it is possible that some cultural or linguistic errors occurred. Another caveat concerns a possible misunderstanding of the violence measures or differences in the general interpretation of phenomena, such as witnessing violence. Generalizations from this study should be made with caution due to the nature of the sample. Readers should be cautious in interpreting the similarities and differences we found between the Western and the Thai rates of witnessing violence. Aside for the similarities, many variables contribute to these differences, among them questions of validity, such as different measurement tools and samples, and finally, culture. The study results should be taken as a first attempt to study witnessing violence among youth in Thailand, and more studies should be taken to validate these findings.

Given the devastating nature of witnessing violence, more needs to be done in Thailand to tackle this phenomenon. Educating parents about this risk might help to lower rates of witnessing violence among youth (Luthar & Goldstein, 2004). Future research should also study the ongoing phenomenon of witnessing violence using a wider array of independent variables and involving qualitative methodology to reveal the covert meanings of witnessing violence. A longitudinal study of witnessing violence by Thai youth might reveal some unique experiences.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ajzen

Brown

T. C.

Carvajal

(2004). Explaining the discrepancy between intentions and actions: The case of hypothetical bias in contingent evaluation. Personality and Social Psychology Bulletin, 30, 1108-1121.

Akers

R. L.

(1996). Is differential association/social learning cultural deviance theory? Criminology, 34, 229-247.

Arbuckle

J. L.

(1999). AMOS (Version 4.0) [Computer software]. Chicago, IL: Small Waters Corporation.

Archer

(2006). Cross-cultural differences in physical aggression between partners: A social-role analysis. Personality and Social Psychology Review, 10, 133-153.

Assunta

(2001). Impact of alcohol consumption on Asia. Consumers Association of Penang, Malaysia. Retrieved from http://www.ias.org.uk/resources/publications/theglobe/globe200103-04/gl200103-04_p4.html

Bandura

(1986). The social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura

(2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1-26.

Benda

B. B.

Corwyn

R. F.

(2002). The effect of abuse in childhood and in adolescence on violence among adolescents. Youth Society, 33, 339-365.

Bronfenbrenner

(1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.

10.

Chen

J. K.

Astor

R. A.

(2010). School violence in Taiwan: Examining how western risk factors predict school violence in an Asian culture. Journal of Interpersonal Violence, 25, 1388-1410.

11.

Cooley-Strickland

Quille

T. J.

Griffin

R. S.

Stuart

E. A.

Bradshaw

C. P.

Furr-Holden

(2009). Community violence and youth: Affect, behavior, substance use, and academics. Clinical Child and Family Psychology Review, 12, 127-156.

12.

Cunningham

M. S.

(2003). The joint contribution of experiencing and witnessing violence during childhood on child abuse in the parent role. Violence and Victims, 18, 619-639.

13.

DeKeseredy

W. S.

Schwartz

M. D.

(1998). Woman abuse on campus: Results from the Canadian national survey. London, England: Sage.

14.

Eagly

A. H.

Wood

(1999). The origins of sex differences in human behavior: Evolved dispositions versus social roles. American Psychologist, 54, 408-423.

15.

Edelen

M. O.

McCaffrey

D. F.

Marshall

G. N.

Jaycox

L. H.

(2009). Measurement of teen dating violence attitudes: An item response theory evaluation of differential item functioning according to gender. Journal of Interpersonal Violence, 24, 1243-1263.

16.

Farrell

A. D.

Bruce

S. E.

(1997). Impact of exposure to community violence on violent behavior and emotional distress among urban adolescents. Journal of Clinical Child Psychology, 26, 2-14.

17.

Farrell

A. D.

Sullivan

T. N.

(2004). Impact of witnessing violence on growth curves for problem behaviors among early adolescents in urban and rural settings. Journal of Community Psychology, 32, 505-525.

18.

Finkelhor

Ormrod

K. R.

Turner

A. H.

(2009). Lifetime assessment of poly-victimization in a national sample of children and youth. Child Abuse & Neglect, 33, 403-411.

19.

Finkelhor

Turner

Ormrod

Hamby

Kracke

(2009). Children’s exposure to violence: A comprehensive national survey (NCJ bulletin, publication number 227744). Retrieved from https://www.ncjrs.gov/pdffiles1/ojjdp/227744.pdf

20.

Fishbein

Ajzen

(1975). Belief, attitude, intention, and behavior. Reading, MA: Addison-Wesley.

21.

Ghanizadeh

A. M.

Tavassoli

M. M.

(2007). Gender comparison of exposed trauma and posttraumatic stress disorder in a community sample of adolescents. Journal of Trauma Nursing, 14, 165-169.

22.

Hoffman

K. L.

Demo

D. H.

Edwards

J. N.

(1994). Physical wife abuse in a non-western society: An integrated theoretical approach. Journal of Marriage and Family, 56, 131-146.

23.

Hurt

Malmud

Brodsky

N. L.

Giannetta

(2001). Exposure to violence—Psychological and academic correlates in child witnesses. Archives of Pediatrics & Adolescent Medicine, 155, 1351-1356.

24.

Karriker-Jaffe

K. J.

Foshee

V. A.

Ennett

S. T.

Suchindran

(2009). Disadvantage and social organization on rural adolescents’ aggression trajectories. American Journal of Community Psychology, 43, 189-203.

25.

Kennedy

A. C.

(2008). An ecological approach to examining cumulative violence exposure among urban, African American adolescents. Child Adolescent Social Work Journal, 25, 25-41.

26.

Kerley

K. R.

X. H.

Sirisunyaluck

Alley

J. M.

(2010). Exposure to family violence in childhood and intimate partner perpetration or victimization in adulthood: Exploring intergenerational transmission in urban Thailand. Journal of Family Violence, 25, 337-347.

27.

Klausner

W. J.

(2002). Thai culture in transition. Bangkok, Thailand: The Siam Society.

28.

Kliewer

Murrelle

Prom

Ramirez

Obando

Sandi

Karenkeris

M. D. C.

(2006). Violence exposure and drug use in Central American youth: Family cohesion and parental monitoring as protective factors. Journal of Research on Adolescence, 16, 455-477.

29.

Kline

R. B.

(2005). Principles and practice of structural equation modeling. New York, NY: Guilford.

30.

Laeheem

Kuning

McNeil

Besag

V. E.

(2009). Bullying in Pattani primary schools in southern Thailand. Child: Care, Health and Development, 35, 178-183.

31.

Lambert

Nylund-Gibson

Copeland-Linder

Ialongo

(2010). Patterns of community violence exposure during adolescence. American Journal of Community Psychology, 46, 289-302.

32.

Lee

(2012). Community violence exposure and adolescent substance use: Does monitoring and positive parenting moderate risk in urban communities? Journal of Community Psychology, 40, 406-421.

33.

Limanonda

(2010). Families in Thailand: Beliefs and realities. Journal of Comparative Family Studies, 26, 67-82.

34.

Luthar

S. S.

Goldstein

(2004). Children’s exposure to community violence: Implications for understanding risk and resilience. Journal of Clinical Child and Adolescent Psychology, 33, 499-505.

35.

Margolin

Gordis

B. E.

(2000). The effects of family and community violence on children. Annual Review of Psychology, 51, 445-479.

36.

Margolin

Vickerman

Ramos

Serrano

Gordis

Iturralde

. . .Spies

L. A.

(2009). Youth exposed to violence: Stability, co-occurrence, and context. Clinical Child and Family Psychology Review, 12, 39-54.

37.

Mazefsky

C. A.

Farrell

A. D.

(2005). The role of witnessing violence, peer provocation, family support, and parenting practices in the aggressive behavior of rural adolescents. Journal of Child and Family Studies, 14, 71-85.

38.

McGregor

J. A.

Camfield

Masae

Promphakping

(2008). Wellbeing, development and social change in Thailand. Thammasat Economic Journal, 26(2), 1-27.

39.

Miller

White

N. A.

(2003). Gender and adolescent relationship violence: A contextual examination. Criminology, 41, 1207-1248.

40.

Ministry of Social Development and Human Security. (2006). Social warning and surveillance center social quality: The analysis of risk behavior among Thai children. Nakorn Pathom, Thailand: Fasai publishing.

41.

Pradubmook-Sherer

Sherer

(2011). Violence among high school students in Thailand: Cultural perspectives. International Journal of Intercultural Relations, 35, 867-880.

42.

Quinn

G. P.

Bell-Ellison

Loomis

Tucci

(2007). Adolescent perceptions of violence: Formative research findings from a social marketing campaign to reduce violence among middle school youth. Public Health, 121, 357-366.

43.

Rigg

Veeravongs

Rohitarachoon

(2008). Reconfiguring rural spaces and remaking rural lives in central Thailand. Journal of Southeast Asian Studies, 39, 355-381.

44.

Ruangkanchanasetr

Hetrakul

Plitponkarnpim

Kongsakon

(2005). Youth risk behavior survey: Bangkok, Thailand. Journal of Adolescent Health, 36, 227-235.

45.

Schwartz

Gorman

A. H.

(2003). Community violence exposure and children’s academic functioning. Journal of Educational Psychology, 95, 163-173.

46.

Sherer

Karnieli-Miller

(2004). Aggression and violence among Jewish and Arab Youth in Israel. International Journal of Intercultural Relations, 28, 93-109.

47.

Sherer

. (2008). Social and cultural dimension influencing violence in children and youth in Thailand. Research report submitted to the Ministry of Social Development and Human Security Thailand (In Thai).

48.

Slovak

Carlson

Helm

(2007). The influence of family violence on youth attitudes. Child and Adolescent Social Work Journal, 24, 77-99.

49.

Slovak

Singer

M. I.

(2001). Gun violence exposure and trauma among rural youth. Violence and Victims, 16, 389-400.

50.

Spano

Vazsonyi

A. T.

Bolland

(2009). Does parenting mediate the effects of exposure to violence on violent behavior? An ecological-transactional model of community violence. Journal of Adolescence, 32, 1321-1341.

51.

Sullivan

T. N.

Kung

E. M.

Farrell

A. D.

(2004). Relation between witnessing violence and drug use initiation among rural adolescents: Parental monitoring and family support as protective factors. Journal of Clinical Child and Adolescent Psychology, 33, 488-498.

52.

Sutherland

E. H.

Cressey

D. R.

(1966). Principles of criminology. Philadelphia, PA: Lippincott Williams & Wilkins.

53.

Sykes

G. M.

Matza

(1957). Techniques of neutralization: A theory of delinquency. American Sociological Review, 22, 664-670.

54.

Vermeiren

Schwab-Stone

Deboutte

Leckman

P. E.

Ruchkin

(2003). Violence exposure and substance use in adolescents: Findings from three countries. Pediatrics, 111, 535-540.

55.

Voisin

D. R.

Crosby

Yarber

W. L.

Salazar

L. F.

DiClemente

R. J.

Staples-Home

(2007). Witnessing community violence and health-risk behaviors among detained adolescents. American Journal of Orthopsychiatry, 77, 506-513.

56.

Vuttanont

Greenhalgh

Griffin

Boynton

(2006). “Smart boys” and “sweet girls”—Sex education needs in Thai teenagers: A mixed-method study. Lancet, 368, 2068-2080.

57.

Ward

C. L.

Martin

Theron

Distiller

G. B.

(2007). Factors affecting resilience in children exposed to violence. South African Journal of Psychology, 37, 165-187.

58.

Yoddumnern-Attig

(1992). Conjugal and parental roles: A behavioral look into the past and present. In Yoddumnern-Attig

Richter

Soonthorndhada

Sethaput

Pramualratana

(Eds.), Changing roles and statuses of women in Thailand: A documentary assessment (pp. 25–35). Salaya, Thailand: Institute for Population and Social Research, Mahidol University.

59.

Zimmerman

M. G.

Messner

F. S.

(2010). Neighborhood context and the gender gap in adolescent violent crime. American Sociological Review, 75, 958-980.