Teachers' views and beliefs about bullying: Influences on classroom management strategies and students' coping with peer victimization

Journal of School Psychology 46 (2008) 431 – 453

Teachers' views and beliefs about bullying: Influences on classroom management strategies and students' coping with peer victimization☆ Becky Kochenderfer-Ladd ⁎, Marie E. Pelletier Arizona State University, Division of Psychology in Education, PO Box 0611, Tempe, AZ 85287-0611, United States Received 1 July 2006; received in revised form 2 July 2007; accepted 16 July 2007

Abstract A multilevel design was used to test a model in which teachers' attitudes (beliefs) about bullying (e.g., it is normative; assertive children do not get bullied; children wouldn't be bullied if they avoided mean kids) were hypothesized to influence if and how they intervene in bullying interactions. In turn, it was hypothesized that teachers' strategies would influence how their students cope with victimization and the frequency of victimization reported by their students. Data were gathered on 34 2nd and 4th grade teachers and 363 ethnically-diverse students (188 boys; 175 girls; M age = 9 years 2 months). Results indicated that teachers were not likely to intervene if they viewed bullying as normative behavior, but were more likely to intervene if they held either assertion or avoidant beliefs. Moreover, avoidant beliefs were predictive of separating students which was then associated both directly and indirectly (via reduced revenge seeking) with lower levels of peer victimization. No grade differences emerged for teachers' views or management strategies; however, minor sex differences were detected which will be discussed. Published by Elsevier Ltd. on behalf of Society for the Study of School Psychology. Keywords: Peer victimization; Bullying; Coping; Classroom management

☆

This article was accepted under Dr. Pianta's editorship. ⁎ Corresponding author. Tel.: +1 480 965 3329; fax: +1 480 965 0300. E-mail address: [email protected] (B. Kochenderfer-Ladd).

0022-4405/$ - see front matter. Published by Elsevier Ltd. on behalf of Society for the Study of School Psychology. doi:10.1016/j.jsp.2007.07.005

432

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

The prevalence of peer victimization within the school context is an issue of central concern for educators, school psychologists and counselors, as well as developmental and educational researchers because children who are the victims of peer aggression are at risk for psychosocial and academic difficulties, including anxiety, low self-esteem, peerrejection, truancy, and school drop out (Dupper & Meyer-Adams, 2002; Hanish, Ryan, Martin, & Fabes, 2005; Kochenderfer-Ladd & Skinner, 2002). Consequently, various anti-bullying programs have been designed and implemented to reduce such harmful behaviors in schools (Olweus, 1993; Smith & Shu, 2000). A main focus of such anti-bullying efforts has been to assess the ways in which teachers approach situations of peer victimization and to encourage them to effectively manage bullying within their classrooms (e.g., Olweus, 1991). Thus, teachers are often given a central role in the management and prevention of peer victimization within schools (Craig, Henderson, & Murphy, 2000; Nicolaides, Toda, & Smith, 2002; Yoon & Kerber, 2003). Moreover, general classroom management (i.e., competent teaching, monitoring of homework and behavior, caring for students) has been directly linked to the frequency of classroom bullying, as well as indirectly linked via classroom structure (i.e., informal student relations, concentration on school activities, social norms; see Roland & Galloway, 2002). In this investigation, although we acknowledge that classroom management strategies encompass a broad range of behaviors, such as creating an orderly environment in which to implement effective teaching methods and working with students to ensure academic progress, we focus specifically on how teachers manage episodes of bullying—that is, strategies teachers employ to intervene in abusive interactions among students. Because previous findings have shown that teachers differ in what they view as bullying behaviors, as well as in what they deem serious aggressive behaviors that warrant adult intervention (Olweus, 1993; Siann, Callaghan, Lockhart, & Rawson, 1993), the premise of this study is that differences in teachers' beliefs about bullying influence their responses to victimization in their classrooms. Teacher beliefs about bullying and classroom management Although victims of bullying tend to view teacher intervention as an effective means of countering peer victimization, many students report that teachers are typically unaware of the bullying that occurs among their students (Bauman & Del Rio, 2005; Smith & Shu, 2000). Moreover, although teachers perceive they are effective at identifying incidences of bullying, researchers find that teachers tend to overestimate their abilities in detecting bullying as well as their effectiveness in intervening (Limber, 2002; Yoon & Kerber, 2003). In fact, although teachers may intervene, if their efforts are inconsistent or ineffective, their assistance may be problematic for victims. For example, Smith and Shu (2000) reported that confiding in a teacher led to more difficulties for a number of victims in their study. However, when teachers intervene in a timely, consistent, and decided manner their efforts are usually effective. Olweus (1993) maintains that the most effective intervention involves close supervision of students (e.g., on playgrounds, in lavatories, hallways and classrooms), taking action to stop harassment early and directly, and never ignoring a situation that could possibly be victimizing for a student. Thus, teachers do play an

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

433

important role in the management and prevention of peer victimization (Craig et al., 2000; Nicolaides et al., 2002; Yoon & Kerber, 2003), but their perceptions and attitudes about what constitutes bullying and what their role is in handling aggression among their students may need to be altered. Thus, the primary goals of this study were to: 1) examine teachers' beliefs about bullying, 2) determine if students' sex and age affect those views, and 3) illuminate how their beliefs relate to their management of classroom bullying behaviors. Toward this latter goal, a few studies have begun to examine how teachers' beliefs about bullying relate to their willingness to intervene as well as how they intervene. For example, Yoon and Kerber (2003) studied teachers' feelings of sympathy for victims along with their efficacy for handling bullying episodes. These investigators found that teachers believed that verbal and physical aggression were more serious forms of victimization than social exclusion, and thus, expressed more sympathy for children who were the targets of verbal and physical aggression. In turn, they were more likely to intervene in such situations. In addition, Yoon (2004) found that teachers not only intervened when they felt greater sympathy for the victim, but also when they felt greater self-efficacy in dealing with peer victimization. For this study, we adopted the paradigm forwarded by Troop and Ladd (2002) to examine teachers' beliefs about the causes of victimization as predictors of their classroom management strategies. Specifically, they compared teachers' views of bullying in regard to three general beliefs: 1) assertive beliefs; e.g., children would not be bullied or picked on if they would stand up for themselves; 2) bullying is normative behavior that helps children learn social norms; and 3) avoidant beliefs; e.g., children would not be bullied or picked on if they avoided mean kids. Troop and Ladd (2002) found that teachers not only varied in their beliefs, but that these attitudes were associated with how they handled bullying among their students. For instance, assertive beliefs hold that if children stand up for themselves, they would not be bullied; thus, such beliefs were expected to influence teachers' decisions to not intervene directly, but rather to have the children work it out themselves. Therefore, it was hypothesized that assertive beliefs would be associated with strategies, such as telling victims and bullies to work their problems out on their own (independent coping) and advising victims to stand up for themselves (advocate assertion). In comparison, they (Troop & Ladd, 2002) found that teachers who believed that peer victimization among children is normative were less likely to punish acts of aggression. Furthermore, because normative beliefs imply that there is nothing wrong, therefore intervening is not necessary, we expect that not only will such beliefs reduce the likelihood of punishment, but teachers holding this view will also be less inclined to involve the parents. Instead, normative beliefs were expected to be associated with relatively passive responses to victimization, such as telling the victims to ignore the harassing peer (i.e., advocate avoidance) or expecting children to work it out on their own (i.e., independent coping). Lastly, avoidant beliefs reflect the perspective that if children avoid abusive peers, they will not be bullied. Such beliefs were expected to influence teachers to tell victims to stay away from peers who are picking on them. Moreover, teachers may assist victims in keeping a safe distance from aggressors by separating them in the classroom. Thus, teachers endorsing avoidant beliefs were hypothesized to try to keep bullies and their victims away

434

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

from each other—either by advocating avoidance or by literally separating them. These hypothesized links are illustrated in the between-level section of the hypothesized model shown in Fig. 1. Classroom management strategies and children's coping with peer victimization Because teachers' strategies for managing bullying typically involve children's participation in the process, it was important to see how teachers' strategies influence their students' own handling of victimization. For instance, if teachers' strategies for handling classroom bullying influences their students' coping behaviors, we would expect that when teachers advocate assertion or independent coping (i.e., dealing with the incidences on their own), children would report more frequent use of strategies such as independent problem solving or, if prone to aggression, revenge seeking. Similarly, it was hypothesized that students whose teachers advise them to avoid bullies and/or separate them would be more likely to try to use cognitive distancing (e.g., try to forget about it or tell oneself it doesn't matter) or passive (e.g., go off by oneself or ruminate about it) coping. It was less clear what the effect of punishing or involving parents would be. On one hand, it could be argued that victimized children would turn to parents for support when punished or when parents became involved anyway. On the other hand, such strategies could

Fig. 1. Hypothesized multilevel model.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

435

conceivably be associated with bullies seeking revenge against victims for the trouble that ensued. Thus, some potential pathways were more exploratory than others. These hypotheses are illustrated in Fig. 1. Sex differences Evidence suggests that teachers use sex stereotypes when assessing and responding to student behavior (Borg, 1998; Kokkinos, Panayiotou, & Davazoglou, 2004). Thus, sex differences were anticipated in teachers' beliefs about bullying as well as their strategies for dealing with conflict between peers. For example, because boys tend to be involved more often in physical aggression and rough and tumble play, teachers may view bullying as more normative for boys than for girls. Furthermore, consistent with stereotypes that boys should be more assertive and independent, it was hypothesized that teachers would be more likely to expect boys to use confrontational coping methods, like fighting back or standing up for themselves (advocate assertion strategies). Similarly, teachers were expected to advocate more problem solving, or “working it out on their own” for boys. It was also important to examine teachers' management of bullying as a function of students' sex because strategies for coping with peer harassment do not appear to be equally effective for both sexes (Chung & Asher, 1996; Kochenderfer-Ladd & Skinner, 2002). For example, seeking social support tends to be a more adaptive and effective coping strategy for girls, whereas externalizing behaviors, like fighting back, appear to be more adaptive for boys (i.e., reduces risk for peer rejection and lessens anxious-depressed symptoms, albeit it is associated with loneliness and continued victimization; see Kochenderfer-Ladd & Skinner, 2002). In fact, Chung and Asher (1996) found that seeking social support from an adult was not only an ineffective coping strategy for fourth through sixth grade boys, but that it placed boys at greater risk for peer rejection. Thus, teachers who encourage victimized boys to seek social support may be placing them at risk for further social problems. Likewise, aggressive, externalizing methods of coping with peer conflict is associated with greater risk for peer rejection among girls, so it would be important for teachers to advocate strategies that do not encourage girls to externalize. Age differences Age was also expected to play a role in teachers' beliefs about bullying and how they respond in such situations. For example, teachers may believe that bullying is more normative among young children who are still learning how to negotiate their peer relationships in socially appropriate and productive ways. In comparison, older children may be expected to know how to conduct themselves and, thus, teachers of older students may view picking on peers as deviant rather than normative behavior. Furthermore, teachers of younger students may feel that their students need more direct guidance in learning how to handle these difficult peer interactions and, thus, intervene more directly by separating students or punishing aggressors. Because age differences were anticipated, a cross-sectional design was used in which teachers at two grade levels were compared: 2nd grade teachers whose students were 7 to 8 years old and 4th grade teachers whose students were 9 to 10 years old. Additionally, children's coping strategies have been shown to vary by age (Hunter, Boyle, & Warden, 2004; Kristensen & Smith, 2003; Salmivalli, Karhunen, & Lagerspetz,

436

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

1996), such that primary school-age children are more likely to tell an adult or teacher about acts of aggression than their secondary school-age counterparts (Whitney & Smith, 1993). In addition, Smith, Shu, and Madsen (2001) found that older children were more likely to ignore bullies and less likely to cry than younger children. Thus, not only were children expected to endorse different coping strategies based on their grade, but differences were also expected based on their teachers' beliefs and classroom management. For example, if 4th grade teachers endorse assertion and avoidance beliefs for their students, it would be reasonable for students in such classes to be more likely than their younger counterparts to use independent coping strategies as well as cognitive distancing strategies, such as trying to forget about it or acting like it doesn't bother them (e.g., nonchalance). Similarly, if 2nd grade teachers feel that aggression is normative, they may decide not to intervene at all, or if they do, it would be in minimal ways such as separating the students to control classroom disruptions. Consequently, it could be argued that younger children might report seeking help from parents or friends more frequently than their older counterparts as a way of gaining assistance that is not being provided by their teachers. Classroom management strategies and levels of peer victimization Lastly, it was of interest to examine how teachers' management of bullying relates to the degree of victimization children report within the classroom. However, because the design of this study does not allow inferences to be made about causal priority, caution will be warranted when interpreting significant associations. In other words, findings will not reveal if teachers' strategies affect the frequency of victimization in their classrooms or the other way around (i.e., extent of victimization is influencing their choice of strategies). For example, consistent with Troop and Ladd's (2002) findings, we hypothesized that teachers who separate students engaged in bullying will have lower levels of peer victimization in their classrooms, whereas those who encourage victims to assert themselves will have higher levels. However, while it may be that advocating assertion is ineffective at reducing classroom bullying, it may also be that teachers who have higher levels of aggression in their classroom are not equipped to deal with each situation on an individual basis and, thus, the easiest way to respond is to tell students to stand up for themselves. Although the direction of influence can not be directly tested without an experimental manipulation, because the interest is in the effect of teachers' classroom management on classroom levels of victimization, a model will be tested in which this direction of effect is assumed. Summary of study In sum, the present study expands the existing literature by assessing teachers' beliefs about bullying, such as bullying being normative and their views about why children get picked on (e.g., they do not stand up for themselves; they do not ignore aggressive peers), and how such beliefs influence their use of various intervention strategies to manage bullying among their students. Moreover, teacher beliefs and strategies were expected to differ as a function of the sex and grade of their students. In turn, teachers' views and management strategies are examined in relation to the coping strategies that their students use and the amount of victimization they report.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

437

Method Participants Teachers and their students from four public elementary schools in the Southwestern United States, serving primarily low-to-middle income families (as measured by percent of students who received free lunches: 26%, 56%, 79% and 95% for the four schools), participated in the present study. Schools were selected to represent the ethnic and socioeconomic diversity of the community. All 34 (31 female; 3 male) second (n = 19) and fourth (n = 15) grade teachers agreed to allow their students to participate and completed questionnaires themselves. Parental permission was obtained for approximately 56% of all second and fourth grade children with the sample consisting of 363 students (90 2nd grade boys and 81 2nd grade girls; 98 4th grade boys and 94 4th grade girls). The average age of second graders was 8 years 1 month; average age of 4th graders was 10 years 1 month. Consistent with the community racial and ethnic background from which they were recruited, the sample was 43.50% Caucasian, 44.10% Latino, 2.5% African American, 3.60% Native American, and 6.30% other. Procedure To ensure that teachers had the opportunity to establish relationships with their students and to develop classroom management techniques for their respective classrooms, teachers were asked to complete questionnaires during the spring of the academic year. For their participation, teachers received a monetary payment based on the number of students participating in their class ($5 each child; payments ranged from $25 to $135). As there were a significant number of Spanish-speaking and bilingual students in the sample, child questionnaires were written in both English and Spanish. Moreover, Spanish speaking interviewers were available to assist students with the questions. In the spring of the school year, trained graduate and undergraduate students administered student questionnaires to groups of fourth grade students. Students were instructed not to talk to one another or share their answers with other students, in order to ensure that their answers remain private. During the group administrations of questionnaires, both English and Spanish-speaking interviewers were available to answer student questions as they arose and to assist individuals who needed extra help completing the measures. For 2nd grade students, individual interviews were conducted to assure that students understood the questions and that their answers were marked accurately. Prior to all group and individual interviews, students were informed that their answers were private and they would not be shared with other students, parents or school staff. A small schoolrelated gift was given to students for participating (e.g., small boxes of color pencils). Measures Teacher views of victimization Teachers' attitudes and beliefs were measured using Troop and Ladd's (2002) Student Social Behavior Questionnaire (SSBQ), which consists of 14 items about boys and 14

438

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

identical items regarding girls. The teachers were asked to indicate how much they agreed with the given statements using a 4-point scale: 1.00 = strongly disagree; 2.00 = disagree somewhat; 3.00 = agree somewhat; 4.00 = strongly agree. Items were averaged to create the following three scales: 1) “Assertive”; the belief that children who stand up for themselves will not be victimized (3 items, including “Kids will stop bullying a boy/girl who asserts himself/herself”; α = .82 and .74 for boys and girls, respectively); 2) “Normative”; the belief that peer victimization is normative (7 items, e.g., “For boys, teasing other children is just part of growing up,” and “Teasing helps girls learn important social norms”; α = .78 and .79 for boys and girls, respectively); and 3) “Avoidant”; the belief that children who avoid aggressors will not be victimized (2 items, such as “Students will stop picking on girls/boys who ignore them”; α = .88 and .84 for boys and girls, respectively). Two items assessing avoidance views were not used included because they reduced internal consistency estimates. Subscales were initially kept distinct by sex; thus, each teacher had six scores that were computed by averaging across each scale's respective items (range of scores = 1.00 to 4.00; see Table 1 for means and standard deviations). Teacher strategies for managing classroom bullying Teacher management strategies were measured, for boys and girls separately, using the Classroom Management Policies Questionnaire (CMPQ; Troop & Ladd, 2002). Fifty of the original 56 items from the CMPQ were used in the present study (25 duplicate items, provided once for girls and once for boys) as it was shown that adequate reliability could be obtained with a reduced scale. Teachers indicated how often they use, or would use, specific strategies for managing problem situations in their classrooms using a 4-point scale (1.00 = I never use this strategy; 2.00 = I have used this once or twice; 3.00 = I sometimes use this strategy; 4.00 = this is how I usually respond). Six scales were created corresponding to those conceptualized by Troop and Ladd (2002): 1) four items referred to separating students (e.g., “When boys are picking on a

Table 1 Means (SD) of teacher beliefs and management strategies by sex Measure

Boys

Girls

t(33)

M

SD

M

SD

Teacher views Assertive Normative Avoidance

2.51 1.32 2.40

.65 .40 .78

2.43 1.27 2.38

.72 .37 .76

1.19 2.08 a .26

Teacher strategies Punishment Advocate assertion Advocate independent Involve parents Advocate avoidance Separate students

3.48 2.19 1.75 2.82 2.26 2.85

.56 .92 .59 .77 .74 .83

3.38 2.31 1.53 2.84 2.23 2.87

.59 .94 .75 .78 .70 .89

1.79 − 1.89 2.74 a − .33 .54 − .18

a

Paired t-test results p b .05.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

439

classmate, do you change their seating in the classroom?”; α = .87 and .89, for boys and girls, respectively); 2) five items tapping advocating assertion (e.g., “How often do you tell girls in your class to assert themselves when other kids are bullying them?”; α = .77 for both sexes); 3) three items measured encouraging independent coping (e.g.,” When boys are being teased by their classmates, do you let them work it out on their own?”; α = .61 and .76, for boys and girls, respectively); 4) five items referred to advising ignoring or avoiding aggressors (e.g., “When girls are being picked on by their classmates, do you tell them to play with other kids?; α = .83 and .81 for boys and girls, respectively); 5) four items involved contacting parents (e.g., “When boys are being bullied by classmates, do you contact the parents?”; α = .93 and .94, for boys and girls, respectively); and 6) four items tapped punishment for the aggressor (e.g., “How often do you reprimand boys when they are making fun of their classmates?”; (α = .80 and .84, for boys and girls, respectively). One item was dropped from the independent scale (“…do you stay out of it?”) as it reduced estimates of internal consistency for both boys and girls. Scores for teacher strategies were computed by averaging across each scale's respective items (see Table 1 for means and standard deviations). Coping Based on Roth and Cohen's (1986) and Causey and Dubow's (1992) framework of conceptualizing coping strategies as either approach (seeking social support, problem solving) or avoidance (distancing, internalizing, and externalizing), a new 26-item measure, “What I would do” (WID) was created by modifying Causey and Dubow's (1992) 30-item Self-report Coping Scale (SRCS). The scale was adapted in five ways. First, instead of thinking about coping with a peer argument, children were directed to imagine what they would do if another child was being mean to them, through name-calling or physical harassment. Second, rather than a 5-point scale, children responded on a 3-point scale (1.00 = never; 2.00 = sometimes; 3.00 = most of the time) to indicate how often they use the various strategies. Third, whereas the SRCS does not distinguish among friends, family members or teachers in the seeking social support scale (sometimes putting them together in the same item, or referring to “someone”), we were interested in determining if children were more likely to seek support from one source over another; thus, items were written to specify parents, friends or teachers. Fourth, two of the seven SRCS internalizing items were dropped to shorten the measure and focus on “passive” coping strategies as opposed to “internalizing” as the term “internalizing” was felt to be too easily confused with internalizing outcomes, such as loneliness, depression, and so forth. Lastly, the externalizing scale was adapted to be more directly related to how children would cope with peer victimization—namely, by seeking revenge or getting even. Thus, these items were re-written to reflect that the externalizing behaviors were directed at the aggressor. For example, the item “yell to let off steam” was modified to be “yell at the kid being mean.” A list of all the items is shown in Table 2. Exploratory factor analyses (EFA) using principal component analyses with varimax rotation revealed seven factors with eigenvalues greater than 1.00. However, two items crossloaded on two factors and were thus dropped and the EFA was recomputed. Results from this second EFA indicated 6 factors with eigenvalues greater than 1.00 (ranged from 1.60 to 3.37 and accounted for 56.06% of the variance; see Table 2 for items, factor loadings and reliability alphas). EFA's revealed that the 4 items referring to adults loaded on a factor distinct from those referring to friends. Thus, a 4-item adult support scale was

440

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

created by averaging across the four items (M = 2.23, SD = .61). Additionally, five items reflected revenge seeking strategies (M = 1.31, SD = .48), four could be construed as cognitive distancing (M = 1.81, SD = .51), and six items referred to passive coping (M = 1.62, SD = .40). Lastly, a 5-item problem solving scale was created by combining the three items that loaded directly on this factor with the two friend items that loaded on their own factor (M = 2.14, SD = .47). By combining these two factors, a more internally consistent scale was created without loss of conceptual meaningfulness. Peer victimization Frequency and type of peer victimization was assessed using a modified version of the self-report measure of victimization developed by Kochenderfer and Ladd (1996). Their Table 2 Highest factor loading of each item Item

I

Adult support (α = .81) 8. Tell your mom or dad what happened 18. Tell the teacher what happened 23. Ask Mom or Dad what to do 26. Ask the teacher what to do

.69 .79 .75 .80

Distancing (α = .62) 1. Act like nothing happened 6. Forget the whole thing 11. Tell yourself it doesn't matter 21. Tell the mean kids you do not care Problem solving (α = .68) 2. Try to think of ways to stop it 17. Try to find out why it happened 22. Change things to keep it from happening again 3. Tell a friend what happened a 7. Ask a friend what you should do a Revenge seeking (α = .86) 4. Do something mean right back to them 9. Hurt the kid who was mean to you 14. Throw or hit something because you get angry 19. Yell at kid who is being mean 24. Hurt the kid back Passive coping (α = .59) 5. Go off by yourself 10. Become so upset you cannot talk to anyone 15. Feel like crying 16. Think about it for a long time 20. Blame yourself for doing something wrong 25. Worry that other kids would not like you a

II

III

IV

V

VI

.69 .52 .79 .66

.73 .60 .59 b.30 b.30

.78 .58

.83 .83 .70 .77 .84

.54 .59 .66 .40 .38 .66

Although these items loaded on their own factor, they were conceptually relevant to problem solving and including them in the problem solving scale increased the internal reliability from .63 to .68.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

441

four item measure tapped four different types of peer victimization: physical (i.e., hit or pushed), verbal (called names), indirect or relational (kids say mean things about you behind your back), and general (pick on or tease you). For this study, an indirect item was added to distinguish between indirect and relational victimization; specifically, an item reflecting damage to personal property was included to convey bullying that is not aimed directly at the child, but indirectly hurts the child through destroying possessions that are important to him or her. Children rated the degree to which kids in their class victimized them using a 4-point scale (1.00 = never; 2.00 = a little/once or twice; 3.00 = sometimes; 4.00 = a lot). Children's individual levels of peer victimization were calculated by averaging across the five items (range of scores = 1.00 to 4.00; M = 1.61, SD = .73, α = .84). Results Preliminary analyses were first conducted to test hypotheses regarding sex and grade mean group differences among the variables. Next, correlations between teachers' views regarding peer victimization and their use of specific strategies for managing bullying in their classroom were computed separately for boys and girls. Lastly, multilevel modeling using Mplus 3.1 was used to test the model that teachers' beliefs of bullying influence their use of management strategies, which in turn, is associated with their students' self-reports of how often they are victimized and how they cope with peer harassment. Sex and grade mean group comparisons Differences in teachers' views of bullying and their management strategies To determine if teachers view bullying differently for boys and girls and to examine if they report using different strategies for managing bullying based on sex, paired t-tests of the means reported in Table 1 were computed. Consistent with our hypotheses, sex differences were detected for normative views, such that teachers were more likely to view peer victimization as normative behavior among boys than girls. Moreover, teachers advocated independent coping more frequently for boys than girls. No other sex differences were detected for teachers' views or their management strategies. Grade effects of teachers' views and strategies were analyzed using a one-way multivariate analysis of variance (MANOVA) in which all teacher views and classroom management strategies were examined simultaneously with grade entered as the independent variable. Omnibus tests indicated that there were no significant effects, Wilks' Lambda F(18,15) = 2.01, p N .05. Thus, findings did not support the hypothesis that teachers' views and strategies were dependent upon the grade of their students. Differences in children's peer victimization and coping strategies To test hypotheses related to sex and grade differences in children's self-reports of coping and peer victimization, a 2 (sex) × 2 (grade) MANOVA was conducted. Omnibus tests indicated significant main effects of sex (Wilks' Lambda F(6,354) = 8.01, p b .001) and grade (Wilks' Lambda F(6,354) = 10.07, p b .001) as well as a sex by grade interaction (Wilks' Lambda F(6,354) = 3.46, p b .001). Univariate F-tests showed that the sex effects were for problem solving, revenge seeking and passive coping, and the grade effects were

442

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

for telling an adult, problem solving, revenge-seeking and peer victimization. However, sex by grade interactions qualified the main effects for problem solving and revenge seeking, so only the unqualified sex and grade effects are interpreted. Specifically, girls reported using passive coping more frequently than boys (F(1,359) = 10.66, p b .001; Ms = 1.70 and 1.56 for girls and boys, respectively) and younger children reported seeking adult assistance more often than older children (F(1,359) = 32.56, p b .001; Ms = 2.41 and 2.08 for 2nd and 4th graders, respectively). Lastly, 4th graders reported higher levels of peer victimization than 2nd grade children (F(1,359) = 4.51, p b .05; Ms = 1.53 and 1.70 for 2nd and 4th graders, respectively). Breakdown of the interaction effects revealed that 4th grade boys were less likely to endorse problem solving strategies (M = 1.96) than second boys (M = 2.20), 2nd grade girls (M = 2.192) and 4th grade girls (M = 2.22). In contrast, 4th grade boys more likely to report using revenge strategies (M = 1.63) compared to second boys (M = 1.21), 2nd grade girls (M = 1.11) and 4th grade girls (M = 1.23). Correlational analyses Relations between teachers' beliefs of bullying and their management strategies Because items were posed specifically in terms of sex (i.e., “boys who…”, “girls would not be…”) as well as to examine potential sex differences in the relations among constructs, correlations were computed separately for male and female students (see Table 3). The low to moderate correlations that emerged among teacher views indicated the scales were tapping three distinct constructs. Likewise, the moderate correlations indicated that six distinct management strategies were being assessed. The pattern of correlations between teachers' views of peer victimization and their selfreported strategies for managing classroom bullying was remarkably similar for boys and girls (see Table 3). For example, for both boys and girls, assertive beliefs were positively associated with advocating assertion whereas normative beliefs were associated with less likelihood of involving parents. Moreover, avoidant beliefs were positively related to advocating avoidance as well as separating students.

Table 3 Correlations among teacher views and management strategies (boys above diagonal; girls below) Scale 1. Assertion views 2. Normative views 3. Avoidance views 4. Punishment 5. Advocate assertion 6. Independent 7. Involve parents 8. Advocate avoidance 9. Separate students

1 .85 .19 .08 .02 .61⁎ .03 .29 .07 .20

2 .02 .93 .51⁎ −.06 −.14 .01 −.35⁎ .18 .27

3 − .10 .39⁎ .90 − .21 − .04 − .14 − .04 .48⁎ .56⁎

4

5

6

7

8

9

− .03 .07 − .01 .83 .33 − .13 .33 − .08 .28

.54⁎ − .33 − .17 .17 .91 .02 .54⁎ .12 .31

−.12 .04 .23 −.24 −.21 .79 −.25 .29 −.20

.33 − .39⁎ − .11 .34 .62⁎ − .44⁎ .94 .04 .41⁎

−.01 .13 .46⁎ −.12 .13 .50⁎ −.13 .83 .32

− .08 .28 .51⁎ .42 .27 − .25 .27 .31 .84

Note. ⁎p b .05. Correlations for boys and girls on the same construct are shown along the diagonal.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

443

Because the pattern of correlations was similar across sex, the data were further examined to determine if distinguishing between boys and girls in this way was necessary. Thus correlations between measures of the same construct for boys and girls were calculated; high correlations (ranging from .79 to .94; see diagonal of Table 3) indicated that teachers appear to hold similar beliefs about bullying and report using the same strategy regardless if they are thinking about boys or girls. Thus, it was determined to combine teacher beliefs and strategies about boys and girls into single scores, thereby reducing the number of teacher variables from 18 to 9 (range remained 1.00 to 4.00). Additional analyses were conducted to determine if teachers endorsed particular beliefs and strategies to a greater or lesser extent than others. Specifically, repeated measures contrasts were used to compare the means of the three teacher beliefs and the six strategies (new variables with items for boys and girls combined; see previous paragraph). Results from simple contrasts in which all means were compared to the lowest mean revealed that teachers were less likely to endorse normative beliefs than either assertion or avoidant views (Ms = 1.30, 2.47, and 2.39, respectively; all ps b .001). Two sets of simple contrasts were then conducted to determine the least and most frequently used strategies—first comparisons of all means to the lowest mean and then all means compared to the highest. Findings indicated that teachers were less likely to tell children to handle the victimization on their own (independent M = 1.78) than to use any other strategy (ps for all contrasts b .001: advocate avoidance M = 2.36; involve parents M = 2.83; separate students M = 2.90; and punish M = 3.39), except advocate assertion (M = 2.01; p N .05). Additionally, teachers were more likely to report using punishment than any other strategy (ps for all contrasts b .001). Correlations among children's peer victimization and coping strategies The relation between children's self-reports of peer victimization and coping strategies are reported in Table 4. The low to moderate correlations indicate five distinct coping strategies; albeit, children who indicated using problem solving strategies also tended to seek adult assistance. For boys, revenge seeking and passive coping were positively associated with victimization while, for girls, only passive coping was related to victimization.

Table 4 Correlations among children's coping strategies and peer victimization (boys above the diagonal; girls below) Coping scale Adult support (AS) Distancing (DIST) Problem solving (PS) Revenge seeking (REV) Passive coping (PAS) Peer victimization (PV) ⁎p b .05.

AS

DIST .22⁎

.12 .49⁎ −.01 .24⁎ −.11

.20⁎ .08 .08 .00

PS .55⁎ .23⁎ −.02 .26⁎ .01

REV

PAS

PV

− .35⁎ − .13 − .14⁎

.30⁎ .11 .40⁎ .01

− .08 − .09 .01 .39⁎ .32⁎

.12 .11

.40⁎

444

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

Structural equation modeling Multilevel modeling procedures were used to test the model shown in Fig. 1. First, the between model was tested to examine the link between teachers' (n = 34) beliefs and their self-reported strategies for handling bullying. Then, the within-level variables of the model were fit to the data (n = 363). Lastly, the between and within levels of the model were joined taking into account that children are nested with classrooms. The data were analyzed using Mplus 3.1, and the comparative fit index (CFI ≥ .95) and root mean square of approximation (RMSEA b .08; Browne & Cudeck, 1993; Hu & Bentler, 1998) were used to evaluate model fit. When suggested by modification indices, errors were allowed to correlate within, but not across, domains (i.e., teacher views, teacher strategies, coping). The initial test of the between-level model shown in Fig. 1 revealed a poor fit to the data, CFI = .87, RMSEA = .20 (90% CI = .10 to .31). Consistent with correlations, none of the paths from teachers' beliefs to advocating independent coping or punishment were significant; thus, those two strategies were dropped from the model. Next, the recommended modification indices were used to obtain a good, parsimonious fit to the data. Standardized path weights for the between-level model that evidenced a good fit to the data are shown in Fig. 2 (note: no errors were allowed to correlate in this model): CFI = 1.00, RMSEA = .00 (90% CI = .00 to .06). Consistent with our hypotheses, endorsing assertion beliefs was associated with teachers' reports that they would tell students to stick up for themselves (i.e., advocating assertion) while holding avoidant beliefs was associated with the teachers indicating they would separate students or tell victims to stay away from aggressors (i.e., advocating avoidance). Moreover, teachers who believed bullying was normative behavior were less likely to involve the children's parents. In addition, pathways

Fig. 2. Final between-level model of paths from teachers' beliefs about bullying to their management strategies. All standardized pathweights shown are significant at p b .05.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

445

that had not been hypothesized were detected: assertive beliefs were positively associated with involving parents and normative beliefs were negatively predictive of advocating assertion. Next, the within-level variables were added to the model. Because no within level hypotheses were posited, the initial test of the multilevel model constrained all within variables to the within level and no pathways among variables were specified; however, peer victimization was allowed to correlate with each coping strategy. This model was a poor fit to the data, CFI = .22, RMSEA = .27. Following modification recommendations, specific pathways were added. Results from this model generating approach are shown Fig. 3 (CFI = .99; RMSEA = .04; standardized path weights are shown). It should be noted that cognitive distancing was not significantly associated with either peer victimization or any of the other coping strategies and was thus omitted from further analyses. The within model showed that problem solving strategies predicted greater levels of seeking adult support. In turn, seeking adult assistance predicted lower levels of revenge seeking as well as less frequent peer victimization. In contrast, passive coping was related to greater victimization. Peer victimization and revenge coping correlated positively; but specifying a direction of effect did not improve model fit.

Fig. 3. Final within-level model of relations among children's coping strategies and peer victimization. All standardized pathweights shown are significant at p b .05.

446

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

Finally, the within and between levels of the model were linked by testing the hypotheses that teachers' strategies for handling classroom bullying would be associated with the amount of student-reported peer victimization as well as the strategies their students report using to cope with victimization. To limit the number of parameters to be estimated and, thus, conserve statistical power, analyses were conducted in the following manner: First, pathways from the four teacher-reported strategies to peer victimization were entered and weak or nonsignificant paths were removed. Results at this stage revealed that separating students was associated with lower levels of peer victimization whereas advocating avoidance was associated with higher levels. Next, pathways from teacherreported strategies to student-reported coping strategies were examined beginning with revenge, followed by problem solving, then adult seeking, and finally passive coping. After the introduction of each coping strategy, weak or nonsignificant pathways were removed. The final model indicated an adequate fit to the data and is illustrated in Fig. 4 (CFI = .92; RMSEA = .06; standardized path weights are shown). For this model, errors were allowed to correlate for advocating avoidance and separating students (standardized weight = .25) and advocating avoidance and advocating assertion (standardized weight = .07). The only hypothesis supported was that separating students would be associated with lower levels of peer victimization. However, three pathways that had not been hypothesized were detected: 1) advocating avoidance was associated with higher levels of peer victimization, 2) separating students reduced children's tendency to seek revenge, and 3) advocating avoidance was associated with higher levels of revenge seeking. These associations will be discussed below.

Fig. 4. Multilevel model of children's peer victimization and coping nested within classrooms. All standardized pathweights shown are significant at p b .05.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

447

Discussion The present study was designed to address three primary issues. First, we examined if the beliefs teachers hold about bullying and the strategies they use to handle such incidences vary as a function of the sex and grade of their students. Second, we investigated how such beliefs may influence teachers' responses to bullying. Third, it was of interest to determine if the management strategies teachers used to handle victimization were predictive of student-reported levels of peer victimization as well as how their students cope with victimization. As will be discussed, in general, findings support the contention that teachers' beliefs are powerful predictors of if and how they decided to intervene in bullying interactions. Moreover, although teachers' strategies were not typically associated with how their students indicated they coped, evidence was found to argue that some management strategies are more effective than others. Sex and grade differences in beliefs and strategies Although very few sex differences emerged, there was some support for the contention that sex stereotypes influence teachers' attitudes about bullying and disruptive behaviors (Borg, 1998; Kokkinos et al., 2004). Specifically, consistent with our hypotheses, teachers tended to rate bullying as more normative among boys, and they were more likely to expect boys to deal with such problems on their own (independent coping). Based on previous findings showing that boys are involved more often in overt forms of bullying, such as hitting and pushing than girls (Olweus, 1993; Rigby & Slee, 1999; Whitney & Smith, 1993), it could be speculated that these forms of victimization are more consonant with what teachers conceive of as bullying behavior (Boulton, 1997; Yoon & Kerber, 2003). Moreover, overt aggression is more likely to come to teachers' attention because of its disruptive nature. In contrast, girls are more likely to be involved in covert forms of aggression (e.g., relational aggression, such as rumor spreading) that teachers either do not notice or do not construe as bullying. Thus, because teachers may be witnesses to overt victimization more frequently for boys, they may conclude that “boys will be boys” and that such behavior is, hence, normative. Moreover, because teachers perceive boys as more aggressive, they may then also believe that boys are able to handle bullying without the need for teacher intervention. Interestingly though, findings failed to show a correlation between holding normative beliefs for boys and teachers telling boys to handle bullying on their own. These potentially contradictory findings could be reconciled by considering the possibility that because boys are engaged more often in fighting, teachers may assume that boys do not need any additional encouragement to assert or defend themselves. No other sex differences were found to suggest that teachers view or respond to bullying differently for boys or girls. In fact, despite the sex differences that were found for normative views and independent coping, both of these were endorsed by teachers with the least frequency. In other words, across sex, teachers in this study were least likely to see bullying as relatively harmless behavior that helps children learn important social norms. Furthermore, they were least likely to expect their students (both boys and girls) to cope with bullying independent of any assistance or intervention.

448

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

Contrary to our expectations, no grade differences were detected for either teacher beliefs or in management strategies for handling bullying. The most likely explanation for these null findings is that the age group studied was not significantly different to warrant differential treatment. In other words, teachers may not perceive a significant developmental difference between second graders and fourth graders that necessitates using different strategies. It is possible that investigators comparing elementary school teachers to their middle or junior high school counterparts would identify significant differences in the ways teachers deal with victimization based on grade level. Relations between beliefs and management strategies Consistent with previous research (Boulton, 1997; Troop & Ladd, 2002), variability in attitudes and beliefs about peer victimization were related to the strategies teachers use to manage classroom bullying. For example, when faced with a bullying situation, teachers who believe that children would not be bullied if they assert themselves were most likely to encourage victimized students to stand up for themselves. Interestingly, however, teachers appear to make a clear distinction between assertion and independent coping. That is, not only were these two strategies uncorrelated, but while teachers holding assertive beliefs indicated that they would advise victimized children to assert themselves, they do not just tell, or expect, them to “handle it on their own.” Perhaps teachers feel that elementaryschool-aged children are still young enough to require some guidance in coping with victimization; thus, they may be taking time to teach children how to stand up for themselves (or making recommendations about how to do so) rather than quickly dismissing them to go handle it on their own. Some support for this speculation can be drawn from the positive path from assertive beliefs to involving parents. It is possible that teachers believe that parents can help teach their children more assertive ways of responding to aggressive peers. Future investigations illuminating if this is the case could prove very informative and beneficial. In comparison to assertive beliefs, teachers who endorsed normative beliefs were least likely to intervene. Specifically, normative beliefs were either uncorrelated with teacher strategies or negatively predictive of specific strategies, such as being significantly less likely to advocate assertion or involve parents. These findings offer support for Yoon and Kerber's (2003) contention that teachers do not intervene in bullying if they do not feel sympathy for the victim or if they believe that getting involved is unnecessary. On one hand, this suggests that challenging normative views may be a promising component of intervention programs designed to, at least in part, raise teachers' awareness of the harmfulness of bullying behavior and how to distinguish between fighting among equals and bullying behavior. On the other hand, the good news is that teachers in this study did not endorse normative views as strongly as assertive and avoidant beliefs. It is possible that this finding reflects a general increase in teachers' awareness of the harmful effects of bullying as a result of media attention and societal pressures to reduce school bullying. However, we can not rule out the possibility that this may reflect a selection bias as the teachers who participated in this study did so voluntarily. In other words, it is possible that our sample of teachers were already sensitive to bullying issues, and it is because they do not view bullying behavior as

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

449

normative and have concerns about how to handle classroom bullying that they agreed to take part in the project. Thus, while cautiously optimistic that teacher attitudes about bullying may be changing (e.g., viewing it less as a normal part of growing up), additional research is still needed to determine if teachers, in general, hold this belief or if efforts are still needed to raise awareness that bullying has harmful long-term consequences. Compared to assertive and normative views, avoidant beliefs appear to be the most beneficial as they influence teachers to intervene in effective ways. Specifically, teachers who believe that children would not be victimized if they avoided aggressive or mean kids were more likely to take action directed at helping children attain this goal, such as recommending that victimized students stay away from aggressors and find other kids to play with. Moreover, they were also more to become actively involved by helping victims stay clear of bullies by separating them in the classroom. Such teachers may be aware that bullied students are not always able to avoid their aggressors, so they need teacher assistance to help keep them at a safe distance. As we turn to next, this appears to be the best combination of beliefs and strategies, as separating students appears to be related to lower levels of peer victimization in the classroom as well as reduced opportunity for children to seek revenge. Linking classroom management strategies with children's, coping and victimization In general, findings did not reveal associations between how teachers handle bullying and how their students do. However, separating students appeared to be the most effective management strategy as it was both directly associated with lower levels of peer victimization, most likely through reduced opportunities to harass peers, and indirectly via reduced likelihood of seeking revenge. This latter finding is consistent with previous reports showing that externalizing behavior not only places children at risk for victimization, but it also maintains the vicious cycle of aggression (Hanish et al., 2004; Kochenderfer & Ladd, 1997; Kochenderfer-Ladd & Skinner, 2002; Perry, Perry, & Kennedy, 1993; Schwartz, Proctor, & Chien, 2001). Separating students may be one way to break this cycle; that is, by thwarting children's ability to engage in retaliatory aggression, teachers may be providing a cooling off period that reduces the likelihood of continuing the cycle of violence through revenge seeking behaviors. In contrast, advocating avoidance seemed to have the opposite effect; namely, higher levels of peer victimization and revenge seeking. In other words, just telling children to avoid bullies without providing a way to do so was not effective. It is possible that children may find this suggestion unhelpful and interpret it as a sign that they will not get assistance from the teacher. Consequently, they may conclude that they have to take matters into their own hands and fight back. Again, future investigations would be needed to support such speculations; but findings clearly indicate a need to understand the underlying processes involved so that teachers can be prevented from inadvertently engaging in counterproductive practices. Although not associated with any teacher strategies, it is worth noting that children who sought adult support reported lower levels of peer victimization and revenge seeking. Relatedly, problem solving was also an effective coping strategy as it was more likely to lead children to seek advice from parents and teachers. Such findings suggest that it would behoove researchers to identify ways that teachers may encourage students to engage in

450

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

these types of coping—or to determine if they are already doing so and, if so, how. Clearly, there are additional ways teachers handle bullying among their students that were not currently measured; future investigations will want to consider what those may be. Grade and sex considerations in intervention designs Although teachers' beliefs and strategies for dealing with bullying were of primary interest, some notable grade and sex differences in the ways children cope with victimization were found that deserve consideration when intervening with young children. For example, consistent with Smith and colleagues' work (Kristensen & Smith, 2003; Smith & Shu, 2000; Smith et al., 2001) showing that some strategies are more common at specific ages (i.e., running away and crying are more common among younger children), grade differences in children's coping were found. For example, younger children reported seeking adult assistance more frequently than older children. This finding is not too surprising given than adults tend to encourage younger children to seek help. Moreover, as children get older, they may find that adults aren't as willing to help as parents and teachers may convey to older children that they should learn to work social problems out on their own. Alternatively, they may find that adults are not effective at intervening and stopping peer victimization—and, in fact, may discover that involving adults is a violation of peer social norms. For example, Chung and Asher (1996) found that, for 4th through 6th grade boys, seeking social support from adults was not only ineffective at reducing risk for victimization, but it also increased their risk for peer rejection. Thus, although caution should be used in encouraging older children to involve adults, given findings that children who sought adult assistance were reporting lower levels of peer victimization and revenge seeking, intervention efforts need to consider how older children can turn to adults when necessary with the confidence that adults will be decisive and effective. Additional grade effects were qualified by sex, such that 4th grade boys were more likely to endorse using revenge strategies and less likely to report using problem solving than girls or their 2nd grade counterparts. Both of these findings suggest that 4th grade boys may benefit from intervention programs targeted at teaching them effective problem solving skills as well as how to control their anger and curb their tendencies to retaliate. Implications and limitations This study showed that teachers' perceptions and views about peer victimization influence how they manage bullying among their students. Moreover, evidence was obtained to support the contention that how teachers respond to bullying affects the level of peer victimization in the classroom, both directly as well as through how their students cope with aggressive peers. In light of these findings, it is important to consider teacher views about bullying when planning prevention and intervention programs aimed at reducing school bullying, as teachers' strategies play an important role countering bullying within schools (Craig et al., 2000; Nicolaides et al., 2002; Yoon & Kerber, 2003). Moreover, the effectiveness of teachers' management strategies in lessening the incidence of bullying must be considered in relation to how such strategies influence children's use of coping strategies, with some attention give to sex and grade differences. Hence, evaluating

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

451

children's coping responses is an essential part of research focusing on the impact of teacher views and management strategies on incidences of peer victimization. While such findings are illuminating, future investigations are warranted to further our understanding of the factors involved in intervening effectively in bullying. For instance, one potential limitation of this study was that no data were gathered regarding the teachers themselves; but it would be important to know if teachers' beliefs and strategies are influenced by their age, racial or ethnic background, the amount of education they have, the number of years they have been teaching, and so forth. It is possible that by studying master teachers who have had years of experience with children, we can learn even more effective ways of dealing with school bullying. It is likely that teachers become more knowledgeable, effective and flexible at preventing and intervening in bullying as they gain experience responding to victimization and have discovered what works well and what does not. Additionally, experimental studies would be needed to test if the strategies teachers use do indeed influence children's coping and subsequent victimization in their classrooms. In other words, because results thus far have been based on correlational studies, it is possible that teachers' beliefs are based on the amount of victimization they witness (i.e., seeing more bullying may lead to normative views) and the strategies they implement are in response to how their students cope with aggression (i.e., retaliatory coping may lead teachers to punish or separate students). Nevertheless, although bidirectional paths are conceivable, the need for effective interventions implemented by capable and knowledgeable adults necessitates tests of the paths of influence from teachers' intervention strategies to subsequent reductions in bullying. Thus, experimental manipulations are needed to see if changes in teachers' strategies alter students' bullying behavior. Additionally, because teachers were asked about the strategies they use to respond to bullying in general (i.e., children getting picked on), we do not know to what degree their strategies may differ depending upon the chronicity or nature of the victimization. In other words, teachers may alter their strategies depending upon if it is the first time bullying has occurred between two children, or if the nature of the aggression was physical, verbal, indirect (property destroyed or damaged) or relational. Earlier studies showed that teachers tend to view relational aggression (i.e., social exclusion, rumor spreading) with less seriousness than overt aggression (i.e., hitting, pushing; Boulton, 1997; Yoon & Kerber, 2003), thus it seems likely that they would use different strategies for dealing with relational versus overt victimization. However, studies are still needed to confirm this hypothesis. In conclusion, the results of this study further our understanding of the nature of teacher involvement and intervention in children's peer victimization. Although teacher intervention has consistently been shown to play a significant role in countering bullying within schools, few studies have investigated how individual teachers choose to manage bullying among their students (Olweus, 1993; Smith & Shu, 2000; Yoon, 2004). The use of teacher management strategies is of central concern, as bullying has been shown to occur most often in the classroom and on the playground, and has been found to place children at risk for a number of psychosocial and academic difficulties (Hanish et al., 2005; Kochenderfer-Ladd & Skinner, 2002; Smith & Shu, 2000; Wolke, Woods, Stanford, & Schulz, 2001). Further, findings suggest valuable information that can guide school psychologists, counselors and educators in the development and evaluation of anti-bullying programs.

452

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

Acknowledgements This paper is based on a Master's thesis completed by the second author under the advisement of the first author. Preparation of this project was supported by National Science Foundation Grant #0318462 awarded to Becky Kochenderfer-Ladd, Arizona State University. We would like to thank the principals, teachers, students and parents of Keller, Loma Linda, Mendoza, and Orangedale Elementary Schools for their participation. References Bauman, S., & Del Rio, A. (2005). Knowledge and beliefs about bullying in schools: Comparing pre-service teachers in the Unites States and the United Kingdom. School Psychology International, 26(4), 428−442. Borg, M. G. (1998). Secondary school teachers' perception of pupils' undesirable behaviors. British Journal of Educational Psychology, 68(1), 67−79. Boulton, M. J. (1997). Teachers' views on bullying: Definitions, attitudes and ability to cope. British Journal of Educational Psychology, 67(2), 223−233. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136−162). Newbury Park, CA: Sage. Causey, D. L., & Dubow, E. F. (1992). Development of a self-report coping measure for elementary school children. Journal of Clinical Child Psychology, 21(1), 47−59. Chung, T., & Asher, S. R. (1996). Children's goals and strategies in peer conflict situations. Merrill-Palmer Quarterly, 42, 125−147. Craig, W. M., Henderson, K., & Murphy, J. G. (2000). Prospective teachers' attitudes toward bullying and victimization. School Psychology International, 21(1), 5−21. Dupper, D. R., & Meyer-Adams, N. (2002). Low-level violence: A neglected aspect of school culture. Urban Education, 37(3), 350−364. Hanish, L. D., Eisenberg, N., Fabes, R. A., Spinrad, T. L., Ryan, P., & Schmidt, S. (2004). The expression and regulation of negative emotions: Risks factors for young children's peer victimization. Development and Psychopathology, 16, 335−353. Hanish, L. D., Ryan, P., Martin, C. L., & Fabes, R. A. (2005). The social context of young children's peer victimization. Social Development, 14(1), 2−19. Hu, L. T., & Bentler, P. M. (1998). Cut-off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1−55. Hunter, S. C., Boyle, J. M. E., & Warden, D. (2004). Help seeking amongst child and adolescent victims of peeraggression and bullying: The influence of school-stage, gender, victimization, appraisal, and emotion. British Journal of Educational Psychology, 74, 375−390. Kochenderfer, B. J., & Ladd, G. W. (1996). Peer victimization: Cause or consequence of school maladjustment. Child Development, 67, 1305−1317. Kochenderfer, B. J., & Ladd, G. W. (1997). Victimized children's responses to peers' aggression: Behaviors associated with reduced versus continued victimization. Development and Psychopathology, 9, 59−73. Kochenderfer-Ladd, B. J., & Skinner, K. (2002). Children's coping strategies: Moderators of the effects of peer victimization. Developmental Psychology, 38(2), 267−278. Kokkinos, C. M., Panayiotou, G., & Davazoglou, A. M. (2004). Perceived seriousness of pupils' undesirable behaviors: The student teachers' perspective. Educational Psychology, 24(1), 109−120. Kristensen, S. M., & Smith, P. K. (2003). The use of coping strategies by Danish children classed as bullies, victims, bully/victims, and not involved, in response to different (hypothetical) types of bullying. Scandinavian Journal of Psychology, 44, 479−488. Limber, S. (2002, May 3). Addressing youth bullying behaviors. Paper presented at the Educational Forum on Adolescent Health: Youth bullying. Retrieved January, 13, 2006, from http://www.ama-assn.org/ama1/pub/ upload/mm/39/youthbullying.pdf Nicolaides, S., Toda, Y., & Smith, P. K. (2002). Knowledge and attitudes about school bullying in trainee teachers. British Journal of Educational Psychology, 72, 105−118.

B. Kochenderfer-Ladd, M.E. Pelletier / Journal of School Psychology 46 (2008) 431–453

453

Olweus, D. (1991). Bully/victim problems among school children: Basic facts and effects of a school-based intervention program. In D. Pepler & K. Rubin (Eds.), The development and treatment of childhood aggression. Hillsdale, NJ: Lawrence Erlbaum Associates. Olweus, D. (1993). Bullying at school: What we know and what we can do. Oxford: Blackwell. Perry, D. G., Perry, L. C., & Kennedy, E. (1993). Conflict and the development of antisocial behavior. In C. U. Shantz & W. W. Hartup (Eds.), Conflict in child and adolescent development (pp. 301−329). New York: Cambridge University Press. Rigby, K., & Slee, P. T. (1999). Australia. In P. K. Smith (Ed.), The nature of school bullying: A cross-national perspective (pp. 324−339). New York: Routledge. Roland, E., & Galloway, D. (2002). Classroom influences on bullying. Educational Research, 44(3), 299−312. Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with stress. American Psychologist, 41, 813−819. Salmivalli, C., Karhunen, J., & Lagerspetz, K. M. J. (1996). How do the victims respond to bullying? Aggressive Behavior, 22(2), 99−109. Schwartz, D., Proctor, L. J., & Chien, D. H. (2001). The aggressive victim of bullying: Emotional and behavioral dysregulation as a pathway to victimization by peers. In J. Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 147−174). New York: The Guilford Press. Siann, G., Callaghan, M., Lockhart, R., & Rawson, L. (1993). Bullying: Teachers' views and school effects. Educational Studies, 19(4), 307−321. Smith, P. K., & Shu, S. (2000). What good schools can do about bullying: Findings from a survey in English schools after a decade of research and action. Childhood, 7(2), 193−212. Smith, P. K., Shu, S., & Madsen, K. (2001). Characteristics of victims of school bullying: Developmental changes in coping strategies and skills. In J. Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 332−351). New York: Guilford Press. Troop, W. P., & Ladd, G. W. (2002). Teachers' beliefs regarding peer victimization and their intervention practices. Poster presented at the Conference on Human Development, Charlotte, NC. Whitney, I., & Smith, P. K. (1993). A survey of the nature and extent of bully/victim problems in junior/middle and secondary schools. Educational Research, 35, 3−25. Wolke, D., Woods, S., Stanford, K., & Schulz, H. (2001). Bullying and victimization of primary school children in England and Germany: Prevalence and school factors. British Journal of Psychology, 92(4), 673−696. Yoon, J. S. (2004). Predicting teacher interventions in bullying situations. Education and Treatment of Children, 27(1), 37−45. Yoon, J. S., & Kerber, K. (2003). Bullying: Elementary teachers' attitudes and intervention strategies. Research in Education, 69, 27−35.