Early Adolescents’ Responses Upon Witnessing Peer Victimization: A Cross-Culture Comparison Between Students in Taiwan and the United States

Abstract

To examine cross-cultural differences in behavior upon witnessing peer victimization and the reasons behind the behavior, this study evaluated the responses of early adolescents from both the United States and Taiwan. Two questions were addressed: (1) Do adolescents in Taiwan and in the United States differ in their willingness to help peer victims, and/or in the endorsement of specific help and non-help behaviors when assessed with a vignette? And (2) do adolescents in Taiwan respond and reason about their responses in the same way as do U.S. early adolescents when they witness real-life victimization? Four-hundred seventy U.S. 6th graders attending three public middle schools within the Midwest United States and 731 Taiwanese 7th graders from one middle school in southern Taiwan completed a survey that contained both open-ended questions asking about real-life reports of witnessing peer victimization and a vignette assessing responses to a hypothetical victimization event. Similarities and differences were found in witness responses from adolescents in the two cultural groups. Taiwanese adolescents were more likely to offer comfort to the victim and U.S. adolescents more likely to tell the bully to stop. Cultural values of interdependence versus independence were reflected in the reasons behind witness responses. Suggestions regarding how anti-victimization intervention and prevention programs should be tailored when they are introduced across the two cultural contexts are discussed.

Keywords

Adolescents cross-cultural peer victimization bystander help behaviors

How adolescent peer witnesses (defined as another student who also see what happens in a peer victimization event) respond to peer victimization has been studied intensively in western societies (see Banyard, 2011 for a review). This research has revealed that peer witnesses were present more so than teachers in peer victimization situations, and their intervening into victimization situations effectively stopped the bully within ten seconds in nearly two-thirds of cases (Craig & Pepler, 1997; Hawkins, Pepler, & Craig, 2001; Pepler & Craig, 1995; Salmivalli, Lagerspetz, Björkqvist, & Österman, 1996). Across all ages, peer witnesses have been shown to implement a variety of responses in victimization incidents, from standing up for the victim to ignoring a peer victimization event without offering help to the victim. Despite this surging body of research, the studies predominantly use samples within western societies. As a consequence, adolescent witnesses’ responses in non-western societies such as the Asian cultural context are currently less well understood.

Several Asian countries started paying attention to the problem of peer victimization during the last decade. Researchers reported prevalence rates and forms of victimization in Asian countries that were similar to some western countries such as United States and the United Kindom (Kanetsuna, Smith, &Morita, 2006; Schwartz, Chang, & Farver, 2001; Wei, Jonson-Reid, & Tsao, 2007). Yet similar prevalence rates do not necessarily mean similar dynamics between peer witnesses’ reactions to peer victimization given distinctive cultural characteristics within these countries. Several cultural dimensions are used to compare the Chinese culture, as one example of Asian cultures, with western countries such as the United States. The dimension of individualism-collectivism and independence-interdependence are, for instance, two common approaches to understanding culture differences. These two dimensions discuss culture differences from different angles in that the former focuses on worldviews and the later focuses on a personal mindset or self-construal (Cross, Hardin, & Gercek-Swing, 2011; Oysterman, Coon, & Kemmelmeier, 2002). Taiwanese culture, which inherits part of the Chinese cultural heritage, stresses values of interdependence and collectivism in which the development of a relational self and collective self is emphasized and group harmony is highly valued over individual independence. These values differ from those found in countries with independent cultural characteristic (e.g., United States) where the independent, separate self and individualism is strongly shaped (Kağitçibaşi, 2007; Sedikides, Gaertner, & O’Mara, 2011).

In addition to a great number of studies that revealed systematic cultural variations in aspects of cognitive, emotional, and social development of children and adolescents (see Oyserman, Coon, & Kemmelmeier, 2002 for a review), cultural values are also reflected in the ways adolescents reason about and cope when they feel they have been victimized (Ma & Bellmore, 2015). For example, Taiwanese adolescents who were victimized were more likely to seek connection with peers and adults than U.S. counterparts, which may be a reflection of the cultural value of interdependence. It is therefore plausible that Taiwanese adolescents who witness peer victimization will endorse witness responses that are more in line with similar features of interdependent cultural contexts such as offering comfort to the victim. On the other hand, adolescents from independent cultural contexts, such as the United States, who are more likely to emphasize the ability to show self-reliance may utilize direct helping responses such as telling the bully to stop when they witness a peer victimization event.

In line with this expected pattern of differences, a cross-cultural examination of middle school students in Japan and England showed that more Japanese students than English students expected other witnesses to help in a peer victimization event (Kanetsuna & Smith, 2002), indicating that individuals from interdependent cultures may place more responsibilities on others to help a victim. Yet a difference in expectation may not necessarily mean a difference in the behavior. What remains unclear is if peer witnesses in these two cultural contexts really differ in what they say they would do, and in what they really do when they witness a peer being victimized. In addition, largely missing from the research on witness behavior is an exploration of the reason a peer witness chooses a specific response. Such an exploration will allow for understanding of how similar behaviors found in two cultures might bear different cultural values, emphases, and motivations. For example, witnesses in Taiwan might report that they would “ignore” a peer victimization incident because they intend to avoid being socially awkward or losing face, an important guide of Taiwanese’ behaviors (Bergmüller, 2013; Shi, 2011; Yu, 2003). However, witnesses in the United States may choose to ignore because they are pursuing self-protection from potential harm (Bellmore, Ma, You, & Hughes, 2012). Investigating the reasons behind witness responses will help clarify how a given cultural value might be associated with a witness response. It is highly likely that the reasons behind peer witnesses’ responses reflect both unique and shared values in the two cultural contexts.

This study compared witness responses and the reasons behind the responses of early adolescents from Taiwan and the United States to address two questions: First, do adolescents in Taiwan and in the United States differ on their willingness to help, and the endorsement of specific help and non-help behaviors toward a peer victim. Second, do adolescents in Taiwan respond and reason about their responses in the same way as do U.S. early adolescents when they witness a real-life victimization incident? We expected that adolescents in each culture would endorsed witness responses in line with features of their corresponding cultures (e.g.,Taiwanese adolescents adopted more comforting the victim than the U.S. adolescents and U.S. adolescents adopted more telling the bully to stop than Taiwaneseadolescents).

Method

Participants

Participants were middle school students who comprised the lowest grades within their respective schools from both Taiwan and the United States. The Taiwanese sample included 731 7th graders (56% girls, mean ages 12.8 yrs) attending one middle school in a Southern city in Taiwan. Based on self-reports of ethnicity, the sample was 94% Han Chinese (n = 722), 1% Aboriginal (n = 5), 3% New Immigrants (n = 21), and 2% Other/No response (n = 14). The U.S. sample included 470 six-grade students (52% girls, mean ages 11.5 yrs) attending three public middle schools in three small cities within the Midwest. Based on students’ self-reports of their ethnicity, the U. S. sample was 71% White (n = 333), 9% Black (n = 44), 6% Latina (n = 27), 2% Asian (n = 12), 1% American Indian (n = 5), 8% Multiethnic (n = 37), and 3% Other/No response (n = 12). The two samples are one-year apart in age because the starting grade of middle school in the two cultures is different. Having witness participants in the starting grade was prioritized over obtaining age equivalence for two reasons: (1) cross-cultural research shows a consistent spike in peer victimization during the starting grade in middle schools across different cultures despite different ages of the starting grade (Smith, Morita, Junger-Tas, Olweus, Catalano, & Slee, 1999), and (2) studies on the age differences of witness responses find consistent behavioral differences between students of elementary and middle schools but not between those within middle school years (see Rogers & Tisak, 1996; Trach, Hymel, Waterhouse, & Neale,2010).

Measures

Participants completed a survey that contained both open-ended questions and a vignette. The survey was originally designed in English and used with the U.S. sample. A back-translation procedure was used to generate the Chinese version for Taiwanesestudents.

Witness behavior vignettes. The vignettes reported in the study by Bellmore et al. (2012) were used to investigate the likelihood that a student would help upon witnessing peer victimization. In the vignettes, a short scenario depicts the participant as a witness to a student of the same gender as the participant getting threatened by another same-gender student. Each participant received one vignette of his/her own gender. All participants read a vignette in which a student of the same gender as the participant was victimized as a way to minimize cross-gender effect. The following example is a vignette version that a girl participant received: “Imagine that you arrive to class early and notice that 2 other girls are already there in the classroom. When you sit down at your desk, you notice that one girl is threatening the other girl. Would you help the girl who is being threatened?”

The vignettes were developed for the U.S. sample, and the cultural appropriateness of the scenario was verified by a focus group of several middle school teachers in Taiwan. Each participant was presented with one vignette. After reading the vignette, participants were asked to describe the likelihood of several responses. Each was first asked to rate how likely it would be that he/she would help the victim in the scenario using a four-point scale (1 = Definitely No, 2 = Probably No, 3 = Probably Yes, 4 = Definitely Yes). Those who answered no (points 1 and 2 on the scale) and those who answered yes (point 3 and 4) were directed to separate subsequent questions. Students who answered yes rated whether they would “tell a teacher or school official,” “tell the bully to stop,” and “try to comfort the victim,” using a five point Likert scale where higher scores indicated stronger endorsement of that response (1 = Definitely No, 3 = I’m Not Sure, to 5 = Definitely Yes). Students who answered no rated whether they would “just ignore it”, “keep watching it”, and “leave the classroom” using the same five-point Likert scale. These responses were derived from those assessed by Trach et al. (2010) in a study of adolescent witness behaviors. We tested the validity of the vignette by evaluating the correlations with communal social goal and empathy (Davis, 1983). As expected, both empathy and communal social goal correlated positively with the three help behaviors; communal social goal and empathy either did not correlate or they correlated negatively with the three non-help behaviors both in Taiwan and in the United States (see Ma, 2014 for specific correlation coefficients amongvariables).

Communal social goals. We included a measure of communal social goals in the preliminary analysis as a way to examine whether Taiwanese students showed more interdependent cultural orientation than U.S. students. The measure of communal social goal of participants was a subscale derived from the self-report Interpersonal Goal Inventory for Children (IGI-C, Ojanen, Gronroos, & Salmivalli, 2005). It consisted of four items that asked participants how important it was for them to meet communal goals (e.g., “You feel close to your classmates”). The participants were asked to rate on a five-point Likert scale, ranging from 1 = No way to 5 = For sure Yes on how important each goal was. The four items were averaged with higher scores demonstrating a higher propensity of striving for affiliation with others (α= 0.86 for the Taiwanese sample and α= 0.79 for the U.S. sample).

Open-ended qualitative measure. Prior to completing the vignettes, participants answered open-ended questions that asked them to write a few sentences to describe “the MOST RECENT TIME” (capitals in original) that “you saw another student (or group of students) getting picked on at school.” Participants were instructed to refer to students involved only in general terms such as gender (i.e., they were instructed not to name the students). After describing the event, participants described how they responded when they witnessed this event and why they responded that way. In this study, all three responses in the open-ended questions were treated as a coherent story based on the belief that participants’ thinking was better captured by an assembly of his/her mental model of a topic across responses (Chi, 1997). An account of an early adolescent’s real-life witnessing experience of peer victimization was around 100 to 200 words in both cultures.

The first author, who is bilingual (English-Mandarin), analyzed the qualitative accounts following Strauss and Corbin’s (1990) three steps of open coding, axial coding, and selective coding until the emergence of the central category—key concepts from the analysis that have the greatest explanatory relevance of all the other concepts and about which all the other concepts will be related (see both Saldana, 2015, and Strauss & Corbin, 1990 for specific examples of open, axial, and selective coding process). The Taiwanese data were coded first. In the first cycle of coding where open coding was utilized, 100 randomly selected responses from Taiwanese adolescents were coded and memoed. Then, 100 randomly selected responses from U.S. adolescents were coded and memoed. Extensive analytic memos were written to compare codes and memos from the U.S. and Taiwanese data until important culturally comparative themes emerged. In the second cycle of the coding process, a different 100 responses of the Taiwanese adolescents and 100 responses from the U.S. adolescents were randomly chosen and used in the selective coding process to see if similar themes emerged in the new sample. During this step, analytic memos were written to elaborate the central category until the definition and examples of the central category were refined. In the third step, we coded remaining responses from both the Taiwan and U.S. samples to verify the developed central category. We also calculated the within culture prevalence of each witness response to compare response patterns across the two cultures quantitatively.

Procedure

In Taiwan, data were collected in 2012 from one public middle school located in southern Taiwan. Before data collection, we distributed a research information sheet to all seventh graders. This sheet contained similar elements to a parental consent form except it did not provide a space for a signature. The parental research information sheet, which was the method recommended by three school principals, an educational administrator, and a professor of education, was used because collecting active parental consent form is not a norm in Taiwan. After reading the information sheet, parents signed on the daily journal to indicate that they had read it and expressed their willingness for their child to participate in the study. We had no access to collect these signatures because the contact journal was private among the teacher, the student, and parents. Yet, during data collection, classroom teachers helped us identify participants whose parents refused their child’s participation in the study. Youth assent was also obtained for each participant in this study. Taiwanese participants did not receive any compensation. The participation rate of Taiwanese 7th graders in this one middle school was 98%.

In the United States, data were collected in 2011 from three public middle schools during their reading periods. Before data collection, a parental consent form was distributed to the sixth-grade students designated as potential participants by the school. Active parental consent and youth assent were obtained for each participant in the study. The consent rate across three middle schools was 87% (470 students out of 543 potential students received consent to participate and gave their own assent). The U.S. sample reflected 73% of the total sixth grade population in three schools because two schools allowed recruitment of all sixth grade students and one school allowed recruitment of selected classrooms from which teachers indicated interest in participation. In two of the schools, each student who participated in the study earned $5; in one school, a single honorarium was provided to the school to benefit all sixth-grade students within the school. All procedures in both countries followed the University-approved IRB (Institutional Review Board) protocol.

Results

Preliminary analysis

Before proceeding with the main analyses, we evaluated the amount of missing data on each variable used to address questions one and two as well as the distributions of these variables in both U.S. and Taiwan samples. No variables had missing data for more than 5% of the participants in both U.S. and Taiwan samples, except for the likelihood of helping in Taiwan, which had 24.6% of the missing data. One possible reason for the high number of missing data for the likelihood of helping in Taiwan may be due to the survey format. We used the same survey format in both cultures that asked students to circle the response for this specific question. According to the feedback from Taiwan’s teachers in the post data collection stage, it appeared that Taiwanese students were not used to answering questions in this format. We compared Taiwanese students who answered this question versus those who did not on the real life reports of witnessed events and found that they were no more or less likely to (1) describe a witness event, χ² (728) = 0.054, p = 0.82, and (2) help or not help in the real life, χ² (558) = 0.283, p = 0.60. We felt that the vignette data is therefore not biased in terms of possible behavioral responses when witnessing peer victimization. All skew and kurtosis values were below 2.5, indicating all variables were not non-normally distributed. To examine if Taiwanese students were more aligned with an interdependent culture orientation compared to the U.S. students, we compared the mean of communal social goal orientation (i.e., the degree of importance to feel close to other classmates) across the two groups. The result showed that Taiwanese students reported a higher communal social goal orientation than the U.S. students, M = 4.28, SD = .080, M = 3.85, SD = 0.77, Taiwan and U.S. respectively,t (1174) = 9.16, p = 0.00.

In the open-ended questions of real-life witness experiences of peer victimization, 211 (29%) students out of the 731 participants in Taiwan reported a valid victimization event that they had witnessed. The remaining students reported they had not witnessed an event (n = 210, 29%), described a personal experience where they had been victimized or aggressed against someone (n = 19, 0.0001%), or provided no response (n = 291, 40%). We coded all the valid accounts (n = 211, 29%) in Taiwan until reaching a saturation point as discussed in the Method section. In the U.S., 379 (81%) of the 470 students reported a valid victimization event; the rest reported they had not witnessed an event (n = 57, 12%), described a personal experience where they had been victimized or aggressed against someone (n = 21, 4%), or provided no responses (n = 13, 3%).

Question 1: Do witness responses differ between cultures for what they say they would do?

To answer whether there are cultural differences in witness responses to a vignette, we conducted seven t-tests to examine mean differences between the two cultures on likelihood of helping, the three help behaviors including telling the teacher, confronting the bully, and comforting the victim, and the three non-help behaviors including ignoring, watching, and leaving the event. The nested structure of the observations within classroom was accounted for by bias-correcting the standard error estimates provided by the model proposed by Asparouhov and Muthén (2006). Table 1 showed all the results of t-tests of this question. With a Bonferroni adjustment, Taiwanese adolescents, in comparison to their U.S. counterparts, reported higher endorsement of telling the teacher, t (1015) = 4.85, p = 0.00, higher endorsement of comforting the victim, t (1011) = 4.81, p = 0.00, lower endorsement of telling the bully to stop,t (1025) = – 5.65, p = 0.00, and higher endorsement of leaving the situation, t (106) = – 3.53, p = 0.00. There were no significant differences across the two culture groups in witness endorsement of likelihood of helping, t (1011) = – 0.94, p = 0.03 (after the Bonferroni adjustment, only p-value lower than 0.007 was considered significant), ignoring the situation, t (106) = – 0.46, p = 0.06, and watching the situation, t (105) = – 0.14, p = 0.06.

Question 2: Do witness responses and reasons for responses differ between cultures for what they say they have done?

Witness responses to real-life incidents. The results showed mostly similarities in terms of the types of witnesses’ responses that were described by adolescents in Taiwan and the United States. General help responses included telling the teacher, telling the bully to stop/defending the victim, comforting and befriending the victim. Witnesses in both cultures also demonstrated similar non-help behaviors such as ignoring the event or the victim, just watching, walking away, doing nothing, and joining/reinforcing the bully. One interesting cultural difference appeared in adolescents’ help responses. In Taiwan, when adolescents decided to intervene in a victimization incident, some tried telling the bully to stop in a socially polite way such that they phrased the action as “suggested the bully to stop” (e.g., “Taiwanese student: There was one boy who hit another boy with fists. What I did is to suggest him (the bully) not to do that so often”). The Chinese term of “Chan-Sho” (a close translation is “to suggest”) was frequently used under these circumstances as a way to express the friendliness and politeness toward the bully.

We further calculated the frequency of common help/non-help responses across the two cultural groups and conducted chi-square analyses to determine whether the distribution of witness responses to real-life incidents was consistent with that for responses to the vignettes. The percentage of each witness response relative to all witness responses used within each culture is summarized in Table 2. Similar rates of Taiwanese and U.S. witnesses helped in the real-life peer victimization incident (51.2% vs. 51.8%). Yet the distribution of witness responses differed significantly across the two cultural groups, χ² (558) = 50.7, p = 0.000. We inspected the standardized residuals (Z scores) to determine when the observed frequencies significantly differed from the expected frequencies in each cell (Wickens, 1989). When adolescents indicated a help response, Taiwanese students, in comparison to U.S. students, were more likely to comfort the victims (16.9% vs. 10.2%) and advise the bully to stop (2.6% vs. 0%), and less likely to directly tell the bully to stop (23.1% vs. 34.7%). When adolescents indicated a non-help response, Taiwanese students were more likely to ignore the situation (8.7% vs. 3.9%) and join in/reinforce the bully (7.2% vs. 1.4%). U.S. adolescents were on the other hand more likely to leave the situation (8.8% vs. 2.6%) and do nothing (24.2% vs. 16.9%). See Table 2 for Z scores.

Reasons behind peer witness responses. Adolescents in both cultures expressed several similar reasons across responses including feeling empathy for the victim, personal justice for the peer victimization situation, or avoiding being a target of the bully. Yet three themes emerged that depicted different reasons behind similar witness responses in the two cultures. First, when adolescents decided to help, Taiwanese adolescents emphasized that it was due to a personal belief that peers should be friendly and helping (e.g., Taiwanese student: “I feel like that all the classmates should be helping each other, friendly to each other, and not getting into fight with each other”). U.S. adolescents, in contrast, expressed a personal belief that a person should be treated fairly (e.g., U.S. student: “I stepped in the middle and stop the whole fight. I did this because it was not a fair fight and I don’t want any one to get in trouble for something stupid”).

Second, when witnesses chose to offer comfort to the victim, Taiwanese witnesses emphasized showing connection to the victim through showing understanding (e.g., Taiwanese students: “I comforted her, because I knew what it felt like when being socially excluded. I think this could make her feel much better … ” and “I told her that I also shared similar situation with her. So you are not alone. I did this because I felt that she is very much alike me, and I hoped this could help her being optimistic”). U.S. adolescents instead focused on assuring the self-confidence of the victim (U.S. student: “I told the kid who’s art project had been picked on that it was good and not to think about what things had been said that hurt him”).

Third, when adolescent witnesses in the two cultures chose to stand aside and not help the victim, Taiwanese adolescents emphasized doing so to abide with social conformity whereas U.S. adolescents focused on not overstepping boundaries. U.S. witnesses who did not help the victim gave the reason of not stepping in other’s business, rather than from a sense of conformity to the social group (e.g., U.S. student: “I didn’t say anything she is 13 so she knows what to do … Because I didn’t want to be rude and be one part of her business”). Taiwanese adolescents who did not help used the term “we” to group themselves with the bullies, so as to justify their non-help behavior as a response to complying with the social group’s decision to exclude the victim (e.g., Taiwanese student: “The girl who excluded other people in many activities is socially excluded by other classmates because we think that everyone thinks the reason of her exclusion of others is very weird.”). Some Taiwanese adolescents seemed to place their own choice of action subordinate to the larger group’s judgment and action. They pointed out that bullying was definitely not a right thing; yet they also experienced a sense of helplessness to act against the large group’s agreement to ostracize the victim:

Taiwanese student: “There is a girl in our classroom who gets socially excluded by most classmates, and it is because her behaviors that make other people feel uncomfortable. I don’t want to socially exclude her, yet I can’t help with the situation because there are too many people choose to act against her … .”

Discussion

Our results showed both common features in types of witness responses of Taiwan and U.S. adolescents, and a number of differences in the prevalence of these responses. The coherence between both quantitative and qualitative results supported our hypotheses that both Taiwanese and U.S. adolescents respond to peer victimization in ways parallel to features in their respective cultures.

Our results revealed a consistency of cultural differences of peer witnesses’ responses to peer victimization both across what adolescents said they would do (i.e., hypothetical vignette) and what they really did in real life (i.e., real-life witnessed episodes). Situated in an interdependent cultural context where the development of the relational and collective self is emphasized over individual independence, Taiwanese adolescents who helped the victim were more likely to comfort the victim but not tell the bully to stop compared to how U.S. adolescents responded. This pattern might exist not only because Asian cultures place more focus on maintaining social harmony, but also because Asian adolescents may be prone to utilize interdependent connection as a source of response. That is, providing comfort served as a means to share a social connection with the victim. The demonstration of shared understanding, reassurance of social connection was repeated themes mentioned in Taiwanese adolescents’ account of comforting the victim, which echoes Kağitçibaşi’s (2007) notion of relational self that is stressed in the interdependent culturalcontext.

In contrast, U.S. adolescents’ responses were in line with features of independent cultural contexts—where the independent and separate self and individualism is strongly shaped—as they were more likely to directly tell the bully to stop. Even when they comforted the victim, the goal was to reassure the victim’s self-confidence and a sense of independence instead of sharing connection and a sense of understanding. The individual-centered perspectives, such as the importance of individual rights, the reassurance of personal confidence and personal value, and assuring personal safety, were repetitive themes in U.S. adolescents’ accounts of reasons behind their help responses to the victim. The investigations of why a witness helped a victim allowed us to track possible discrepancies in the reasons behind similar types of witness reactions in the two cultures.

It is noteworthy that more Taiwanese adolescents joined-in and reinforced the bully than U.S. adolescents. Matsunaga (2010) pointed out that victims in Asian societies are normally viewed as “black sheep” of the group, and that “individuals draw on the collectives they belong to in forming and maintaining their identities” (p. 332). The group-centered orientation in the interdependent culture further makes these victims less welcomed as witnesses may easily identify with the collective’s reaction however inappropriate it seemed to be (Morita, Soeda, Soeda, & Taki, 1999). Our qualitative results echoed these arguments in that Taiwanese early adolescents were found to easily align themselves with the group. The perceived “we”ness created a strong pressure of social conformity that inhibited their willingness to help in a victimization situation. The perceived “we”ness was so strong that it made a witness more often chose to blame the victim rather than the bully for disrupting the social harmony because of victim’s personal idiosyncrasy that was considered non-conforming to the large social context.

These results highlight the importance of tailoring elements in anti-victimization intervention and prevention program when they are introduced to different cultural contexts. For example, it may be particularly important to teach peer witnesses in interdependent culture intervening strategies that permit witnesses to maintain a sense of interpersonal connection and group belongingness. Kanetsauna, Smith, & Morita (2006) suggested that peer support systems might be more relevant for Asian students since they are generally more sensitive to others’ views and reactions toward them. On the other hand, intervention programs that enable maximum personal safety of the witnesses and that emphasize personal responsibilities of helping the victims may be more relevant in an independent cultural context. Despite cultural differences, some elements of intervention programs might be equally important across cultural contexts. Matsunaga (2010) showed that emotional support to the victim provided by peer witnesses has been effective in improving victims’ long-term psychological well-being both in independent (i.e., U.S.) and interdependent (i.e., Japan) cultural contexts. In our sample only 17% ofTaiwanese adolescents and 10% of U.S. adolescents offered comfort to the victim in the real-life witnessing a peer victimization episode. More attention is needed cross-culturally to motivate witnesses forming a peer support system to the victim.

How to effectively motivate peer witness to report a victimization incident also needs researchers’ attention in the two cultural contexts (i.e., the united States and Taiwan). Teachers’ intervening in the victimization was found to be effective in stopping the victimization episode (Novick & Isaacs, 2010). Yet only 8.2% of Taiwanese witnesses and 6.9% of U.S. witnesses in our sample told a teacher when peer victimization happened in real life. Kanetsuna & colleagues (2006) reported a much higher percentage of victims—both in interdependent and independent cultures—expecting peer witnesses to seek help from an adult. Lack of trust toward teachers was the major reason reported both in an independent and interdependent culture when peer victimization is concerned (Morita, 2001; Novick & Isaacs, 2010). It may be top priorities in both cultural contexts to enhance teacher awareness of peer victimization and empower teacher intervening strategies.

Even with a focus on two cultures that uses two methods, we also note some limitations of the study. First, our samples were merely from one middle school in Taiwan and three middle schools in the United States. We do not assume that our sample could be totally comparable nor could we assume generalizability to the overall population in the two cultural groups. We also did not test the direct associations between cultural orientations and witness responses, so it is not clear to what extent cultural orientations played a role in directing help responses. Second, the vignette used in this study was originally developed in the United States. The specific peer victimization condition described in the vignette may not have the same meaning in Taiwan. For example, the vignette used the term “threaten” to describe a school bullying event. The extent to which students perceived the seriousness of “threaten” or other victimization-related terms may be different in the two cultures (Smith & Monks, 2008). It was possible that Taiwanese students, who witness more relational victimization tended to perceive “threaten” as a much more serious event than did a student in the United States. This may result in Taiwanese students’ more often resorting to teacher authority to handle the perceived relative serious situation. Future cross-cultural studies need to examine to what degree participants from two cultures evaluate the seriousness of specified peer victimization, as cultural comparisons on witness responses will be more reliable when the specified peer victimization is considered serious to a similar degree.

Third, wording of the stimuli is critically important. Some adolescents of Taiwanese culture did not consider it socially polite to directly “tell” the bully to stop probably because it may disrupt group harmony. Rather, they termed their action as “suggestion”, or “advice” for the bully to stop to show their social politeness. Future researchers should be careful in phrasing the specific help response in the vignettes when Asian participants are involved because they may not endorse the response they perceive as socially impolite even if the underlying intention in the behavior (e.g., interact with the perpetrator) is similar. Fourth, both the discrepancy between results of vignettes and real-life reports and cultural differences in witness responses could be a function of differences in prevalence of victimization types in the two cultures. Bergmüller (2013) found that physical and verbal victimization were more prevalent in individualistic cultures. Wang & Cheng (2010) pointed out that relational form of aggression was much more prevalent in Taiwan. This sample also showed that Taiwanese adolescents reported more relational victimization episodes (41% vs. 10%, z = 8.5) whereas U.S. adolescents reported more verbal episodes (42% vs. 29%, z = 3.0) (Ma, 2014). Witnesses’ help behaviors and recommended strategies may vary with the type of victimization (Kanetsuna, Smith, & Morita, 2006). This may explain why Taiwanese students were less likely to leave the situation in real life witnessing peer victimization episodes since nearly half of the reported episodes were long-term social exclusion of a victim where the witness was not likely to simply leave the immediate context. It would be valuable for future study to scrutinize witness responses relative to types of victimization and compare these patterns across cultural groups.

In sum, this study showed that adolescents in the two cultures used a similar behavior repertoire when they witnessed another a victimization event, yet the prevalence of each witness behavior and the reason behind similar behavior could be different from culture to culture. Educators should be sensitive to the potential cultural values that may be associated with witness behavior when they adapt a given peer victimization intervention and prevention program from another cultural context.

References

Asparouhov

, & Muthén

B. O.

(2006). Multilevel modeling of complex survey data. Proceedings of the Joint Statistical Meeting in Seattle, August 2006. ASA section on Survey Research Methods, 2718–2726.

Banyard

V. L.

(2011). Who will help prevent sexual violence: Creating an ecological model of bystander intervention. Psychology of Violence, 1, 216–229. doi: 10.1037/a0023739

Bellmore

, Ma

, You

, & Hughes

(2012). A two-method investigation of early adolescents’ responses upon witnessing peer victimization in school. Journal of Adolescence, 35, 1265–1276. doi: 10.1016/j.adolescence.2012.04.012 .

Bergmüller

(2013). The relationship between cultural individualism– collectivism and student aggression across 62 countries. Aggressive Behavior, 39, 182–200. doi: 10.1002/ab.21472

Chi

M. H.

(1997). Quantifying qualitative analyses of verbal data: A practical guide. Journal of the Learning Sciences, 6, 271–315. doi: 10.1207/s15327809jls0603_1

Craig

W. M.

, & Pepler

D. J.

(1997). Observations of bullying and victimization in the school yard. Canadian Journal of School Psychology, 13, 41–59. doi: 10.1177/082957359801300205

Davis

M. H.

(1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113–126. doi: 10.1037/0022-3514.44.1.113

Hawkins

D. L.

, Pepler

, & Craig

(2001). Naturalistic observations of peer interventions in bullying. Social Development, 10, 512–527. doi: 10.1111/1467-9507.00178

Kağitçibaşi

(2007). Family, self, and human development across culture (2nd edition). Mahwah, NJ:Lawrence. Erlbaum.

10.

Kanetsuna

, & Smith

P. K.

(2002). Pupil insights into bullying, and coping with bullying: A bi- national study in Japan and England. Journal of School Violence, 1, 5–29. doi: 10.1300/J202v01n03_02

11.

Kanetsuna

, Smith

P. K.

, & Morita

(2006). Coping with bullying at school: Children’s recommended strategies and attitudes to school-based interventions in England and Japan. Aggressive Behavior, 32, 570–580. doi: 10.1002/ab.20156

12.

T.L.

(2014). Connection or independence: A cross-cultural comparison of students’ responses upon witnessing and experiencing peer victimization in school: A mixed-method study. (Doctoral dissertation). Retrieved from ProQuest. (Accession 814 no. 3624948)

13.

, & Bellmore

(2015). Connection or independence: Cross cultural comparisons of adolescents’ coping with peer victimization using mixed-method. Journal of Cross-Cultural Psychology, 47, 1–22. doi: 10.1177/0022022115605386

14.

Matsunaga

(2010). Testing a mediational model of bullied victims’ evaluation of received support and post-bullying adaptation: A Japan-US cross-cultural comparison. Communication Monographs, 77, 312–340. doi: 1080/03637751003758235

15.

Morita

, (2001). Ijime no kokusai hikaku kenkyu [Cross-national comparative study of bullying]Japan: Kaneko Shobo.

16.

Morita

, Soeda

, & Taki

(1999). Japan. In Smith

P. K.

, Morita,Y., Junger-Tas,J., Olweus,D., Catalano,R., & Slee,P. (Eds), The nature of school bullying: A cross-national perspective (pp. 309–323). New York: Routledge.

17.

Novick

R. M.

, Isaacs

(2010). Telling is compelling: The impact of student reports of bullying on teacher intervention. Educational Psychology, 30, 283–296. doi: 10.1080/01443410903573123

18.

Ojanen

, Grönroos

, & Salmivalli

(2005). An interpersonal circumplex model of children’s social goals: Links with peer-reported behavior and sociometric status. Developmental Psychology, 41, 699–710. doi: 10.1037/0012-1649.41.5.699

19.

Oyserman

, Coon

, & Kemmelmeier

(2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3–73. doi: 10.1037//0033-2909.128.1.3

20.

Pepler

D. J.

, & Craig

W. M.

(1995). A peek behind the fence: Naturalistic observations of aggressive children with remote audiovisual recording. Developmental Psychology, 31, 548–553. doi: 10.1037/0012-1649.31.4.548

21.

Rogers

M. J.

, & Tisak

M. S.

(1996). Children’s reasoning about responses to peer aggression: Victim’s and witness’s expected and prescribed behaviors. Aggressive Behavior, 22, 259–269.

22.

Saldaña

(2015). The coding manual for qualitative researchers. CA:Sage.

23.

Salmivalli

, Lagerspetz

, Björkqvist

, Österman

, & Kaukiainen

(1996). Bullying as a group process: Participant roles and their relations to social status within the grou. Aggressive Behavior, 22, 1–15. doi: 10.1002/(SICI)1098-2337(1996)22:1<1::AID-AB1>3.0.CO;2-T

24.

Salmivalli

, Voeten

, & Poskiparta

(2011). Bystanders matter: Associations between reinforcing, defending, and the frequency of bullying behavior in classrooms. Journal of Clinical Child and Adolescent Psychology, 40, 668–676. doi: 10.1080/15374416.2011.597090.

25.

Schwartz

, Chang

, & Farver

J. M.

(2001). Correlates of victimization in Chinese children’s peer groups. Developmental Psychology, 37, 520–532. doi: 10.1037/0012-1649.37.4.520

26.

Sedikides

, Gaertner

, & O’Mara

E. M.

(2011). Individual self, relational self, collective self: Hierarchical ordering of the tripartite self. Psychological Studies, 56, 98–107. doi: 10.1007/s12646-011-0059-0

27.

Shi

(2011). The impact of face on Chinese students’ simulated negotiation practices with Americans. Language and Intercultural Communication, 11, 26–40. doi: 10.1080/14708477.2010.517846.

28.

Sittichai

, & Smith

P. K.

(2015). Bullying in south-east Asian countries: A review. Aggression and Violent Behavior, 23, 23–25. doi: 10.1016/j.avb.2015.06.002

29.

Smith

P. K.

, & Monks

C. P.

(2008). Concepts of bullying: Developmental and cultural aspects. International Journal of Adolescent Medicine and Health, 20, 101–112. doi: 10.1515/IJAMH.2008.20.2.101

30.

Smith

P. K.

, Morita

, Junger-Tas

, Olweus

, Catalano

, & Slee

, (1999) The nature of school bullying: A cross-national perspective. London: Routledge.

31.

Strauss

, & Corbin

J. M.

(1990). Basics of qualitative research: Grounded theory procedures and techniques. Thousand Oaks, CA US:: Sage Publications, Inc..

32.

Trach

, Hymel

, Waterhouse

, & Neale

(2010). Bystander responses to school bullying: A cross-sectional investigation of grade and sex differences. Canadian Journal of School Psychology, 25, 114–130. doi: 10.1177/0829573509357553

33.

Wei

, Jonson-Reid

, & Tsao

(2007). Bullying and victimization among Taiwanese 7th graders: A multi-method assessment. School Psychology International, 28, 479–500. doi: 10.1177/0143034307084137

34.

Wickens

, (1989) Multiway contingency tables analysis for the social sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

35.

(2003). On the university of face: Evidence from Chinese compliment response behavior. Journal of Pragmatics, 35, 1679–1710. doi: 10.1016/S0378-2166(03)00074-2.