|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001796738
007 cr mnu|||uuuuu
008 070718s2005 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001600
Totura, Christine Marie Wienke.
Victimization and academic achievement at school :
b the role of psychosocial mediators and moderators
h [electronic resource] /
by Christine Marie Wienke Totura.
[Tampa, Fla] :
University of South Florida,
Dissertation (Ph.D.)--University of South Florida, 2005.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 145 pages.
ABSTRACT: The present study sought to examine the relationship between victimization by peers in middle school and academic outcomes. it was expected that an association between the experience of victimization and diminished academic performance would be mediated by poor psychological outcomes, as measured by moodiness, depression, anxiety, and anger. additionally, it was hypothesized that academic outcomes could be divided into two distinct constructs, motivation and achievement, with motivation and academic goal-orientation variables preceding the adequate attainment of school grades and standardized test scores. therefore, the present mediated model was tested using a structural equation modeling technique: victimization-psychological functioning-academic motivation-academic achievement. additionally, it was hypothesized that certain factors (friendship, prosocial activities and influences, school climate, aggression, and teacher-reported difficulties) would moderate the vict imization-psychological functioning pathway. victimized middle school boys and girls were expected to have varying psychological and emotional outcomes depending on proposed risk and protective factors. approximately equal numbers of males and females (n=145 and 181, respectively) were randomly selected from classrooms in 11 middle schools across 6th, 7th, and 8th grades. Students completed questionnaires that assessed hypothesized mediator and moderator variables. In addition, teachers of the selected classrooms completed a brief rating scale on each of the students, which assessed student moodiness, behavioral difficulties, and learning problems. Achievement and discipline records data were obtained. Results revealed that Psychological Functioning mediated the relationship between Victimization and Academic Motivation, which was then related directly to Academic Achievement. Only the Aggression and Climate constructs moderated the Victimization-Psychological Functioning pathway, wi th Climate factors additionally significant for boys. These results suggest that victimization is associated with poor motivation to achieve if victimized students also experience psychological difficulties. Limited motivation is then associated with poorer academic performance. Contrary to hypothesized associations, endorsing aggressive beliefs and behaviors and experiencing low levels of intervention and support at school against bullying, particularly for boys, were related to better emotional outcomes for students who are highly victimized. While statistically significant, these findings have limited effect sizes. Implications for future research and the development of school-based programming are discussed.
Adviser: Marc S. Karver, Ph.D.
t USF Electronic Theses and Dissertations.
Victimization and Academic Achievement at School: The Role of Psychosocial Mediators and Moderators by Christine Marie Wienke Totura A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Marc S. Karver, Ph.D. George Batsche, Ph.D. Michael Brannick, Ph.D. Vicky Phares, Ph.D. Kevin Thompson, Ph.D. Date of Approval October 27, 2005 Keywords: aggression, peer relationships, adolescent adjustment school functioning, gender differences Copyright 2006, Christine Marie Wienke Totura
Dedication This dissertation is dedicated to my husba nd, whose love and support helped me fulfill a dream.
Acknowledgements I would like to thank the members of my dissertation committee, Drs. George Batsche, Michael Brannick, Vicky Phares, and Kevin Thompson, for their time, support, and insight in the developmen t of this study. I am thankful to Dr. Carol MacKinnonLewis for helping chair my dissertation defe nse meeting and offering guidance as this project was conducted. I partic ularly appreciate the statis tical assistance of Qutayba Abdullatif and Dr. Michael Brannick. I also appreciate Dr. Ellis Gestens contributions during the initial stage of this project. This study would not have been possible without the backing and coordination of Ray Gadd, Amelia Van Name Larson, Sheri Dunham, Kathy Divine, and Melinda Hess in the Pasco County School District, and the administrators, teachers, and students in the Pasco County Middle Schools. Their participation during the 2003 data collection was invaluable for the present study. I would like to recognize and tha nk those who helped in the ta sks of data collection and data entry and/or verifica tion during the spring and su mmer of 2003: Vounette Deus, Gina DiPasqua, Kelly Genske, Laurel Jorgensen, Danielle Short, and Lisa Strother. I am exceptionally grateful to my husband, Nathan To tura, and my parents, Dr. H. Richard and Marianne Wienke, for their immeasurable pa tience, sensitivity, and encouragement. Additionally, I want to acknowledge Jessi ca Handelsman and Dr. Dimitra Kamboukos for their loyal friendship throughout the deve lopment of this dissertation. Finally, I would like to express interminable apprecia tion and admiration for my faculty advisor and mentor, Dr. Marc Karver, whose confidence in my abilities made this project possible.
i Table of Contents List of Tables iv List of Figures v Abstract vi Chapter 1 Introduction 1 Specific Aims 1 Background and Significance 1 Defining Bullying and Victimization 3 Assessing Bullying and Victimization 7 Self-Report, Interview, Observ ation, and Peer Nomination Techniques 7 Using Teacher Reports to Identify At-Risk Children 11 Purpose of the Present Study 12 Effects of Exposure to and Expe rience of Peer Victimization 13 Psychological Consequences 13 Behavioral Consequences 14 School-Related Consequences 15 Protective and Risk Factors 17 Peer Relationships and Prosocial Activities 17 Coping Mechanisms 19 School Climate Factors 20 Gender Differences 21 Summary and Proposed Model 23 Hypothesized Pathways and Relationships 23 Hypotheses 26 Hypothesis 1 26 Hypothesis 2 26 Hypothesis 3 26 Hypothesis 4 26 Hypothesis 5 27 Hypothesis 6 27 Hypothesis 7 28 Hypothesis 8 28 Chapter 2 Method 29 Participants 29 Measures 30
ii Student Self-Report Surveys 30 Teacher-Report Survey 34 Records Data 35 Procedure 36 Analyses 37 Chapter 3 Results 38 Descriptive Statistics 38 Principal Component Analysis 38 Correlation Analyses 39 Structural Equation Modeling (SEM) 40 Defining Latent Constructs 40 Identifying Parameters 40 Comparing Model and Observed Data Matrices 42 Methods of Estimating Parameters 43 Evaluating Goodness of Model Fit 45 Dealing with Missing Data 46 Identifying Significant Pathways 47 Tests for Mediation 47 Hypotheses 1 and 2 47 Multiple Regression Analyses 51 Tests for Moderation 51 Hypothesis 3 52 Hypothesis 4 52 Hypothesis 5 53 Hypothesis 6 53 Hypothesis 7 54 Gender Analyses 54 Hypothesis 8 54 Slope Analyses for Significant Moderators 55 Aggression 55 Climate 56 Chapter 4 Discussion 58 Mediated Model 58 Moderators of Psychological Func tioning in the Mediated Model 61 Implications of the Present Study 66 Limitations of the Present Study 68 Strengths of the Present Study 72 Future Directions 73 References 76 Appendices 113 Appendix A: Independent and Dependent Constructs 114 Appendix B: Constructs for Hypothesized Model 115
iii Appendix C: Olweus Bully/Victim Questionnaire 122 Appendix D: Center for Epidemiol ogical Studies-Depression Scale (CES-D) 130 Appendix E: State-Trait Anxiety Inventory for Children (STAIC) 131 Appendix F: State-Trait Anger Expression Inventory for Children and Adolescents (STAXI-C/A) 132 Appendix G: School Adjustment Survey (SAS) 134 Appendix H: Middle School/High School Student Survey 136 Appendix I: Adult Supervis ion at School (ASAS) 142 Appendix J: Acting-Out, Moodiness, and Learning Scale-Revised (AML-R) 143 Appendix K: Bullying Definition 145 About the Author End Page
iv List of Tables Table 1 Means and Standard Deviati ons for Constructs (Z-Scores) and Observed Variables 100 Table 2 Correlations between Va riables in the Mediated and Moderated Pathways 101 Table 3 Correlations between Mediator and Moderator Constructs 103 Table 4 Hierarchical Regression An alyses Predicting Psychological Functioning from Moderator Constructs for the Total Sample 104 Table 5 Hierarchical Regression An alyses Predicting Psychological Functioning from Moderator Constructs for Boys 105 Table 6 Hierarchical Regression An alyses Predicting Psychological Functioning from Moderator Constructs for Girls 106
v List of Figures Figure 1. Model with Mediated Pa thway between Victimization and Academic Achievement 108 Figure 2. Model with Mediated Pa thway between Victimization and Academic Motivation 109 Figure 3. Psychological Functioning M eans at High and Low Levels of Aggression for the Total Sample 110 Figure 4. Psychological Functioning M eans at High and Low Levels of Climate for the Total Sample 111 Figure 5. Psychological Functioning M eans at High and Low Levels of Climate for Boys 112
vi Victimization and Academic Achievement at School: The Role of Psychosocial Mediators and Moderators Christine Marie Wienke Totura ABSTRACT The present study sought to examine the relationship between victimization by peers in middle school and academic outcomes. It was expected that an association between the experience of victimization and diminished academic performance would be mediated by poor psychological outcomes, as measured by moodiness, depression, anxiety, and anger. Additionally, it was hypot hesized that academic outcomes could be divided into two distinct constructs, Motiv ation and Achievement, with motivation and academic goal-orientation variables preceding th e adequate attainment of school grades and standardized test scores. Therefore, th e present mediated model was tested using a Structural Equation Modeli ng technique: Victimization Psychological Functioning Academic Motivation Academic Achievement. Addi tionally, it was hypothesized that certain factors (Friendship, Prosocial Ac tivities and Influences, School Climate, Aggression, and Teacher-Reported Difficultie s) would moderate the Victimization Psychological Functioning pathway. Victimi zed middle school boys and girls were expected to have varying psychological and emotional outcomes depending on proposed risk and protective factors. Approximately equal numbers of males and females (N = 145 and 181, respectively) were randomly select ed from classrooms in 11 middle schools across 6th, 7th, and 8th grades. Students completed questionnaires that assessed
vii hypothesized mediator and modera tor variables. In additi on, teachers of the selected classrooms completed a brief rating scale on e ach of the students, which assessed student moodiness, behavioral difficulties, and learni ng problems. Achievement and discipline records data were obtained. Results revealed that Psychol ogical Functioning mediated the relationship between Victim ization and Academic Motivat ion, which was then related directly to Academic Achievement. On ly the Aggression and Climate constructs moderated the Victimization Psychological Functioning pathway, with Climate factors additionally significant for boys. These results suggest that victimiz ation is associated with poor motivation to achieve if victimized students al so experience psychological difficulties. Limited motivation is then asso ciated with poorer academic performance. Contrary to hypothesized associ ations, endorsing aggressive beliefs and behaviors and experiencing low levels of interventi on and support at school against bullying, particularly for boys, were related to be tter emotional outcomes for students who are highly victimized. While statisti cally significant, these findings have limited effect sizes. Implications for future research and the development of school-based programming are discussed.
1 Chapter One Introduction Specific Aims Aim 1. The present study will provide an overview of bullying and victimization in schools and their impact on student functioning. Aim 2. The present study will specifically e xplore the relationship between victimization and academic outcomes and evaluate a number of social and psychological factors that are hypothesized to influence the associati on between victimization and achievement. Aim 3. The present study will present a model of proposed pathways among victimization, psychological, motivational, and achievement variables, test this model for adequacy of fit with the identified sample, and examine moderators of specific pathways. Aim 4. The present study will describe si gnificant associations among model variables and propose alternatives for nonsignificant associations. Background and Significance Bullying behaviors and their contexts ha ve been assessed in several countries demonstrating that exposure to and involveme nt in bullying behavior s are significant risk factors to healthy psychologi cal and physical development (Haynie et al., 2001; Olweus, 1997a, 1997b; Roland, 2000). More specifically, involvement in bullying has resulted in negative effects on the development of fr iendships and entrance into peer groups, increased internalizing and externalizing difficulties, and potentially poor academic outcomes (Hodges et al., 1997, 1999; Juvonen, Ni shina, & Graham, 2000; Wentzel,
2 1994; Wentzel, Weinberger, Ford, & Feldman, 1990). Of particular importance for school districts, victimization has been associ ated with declines in achievement factors, whether as orientation toward academics or grades and test scores, with the mechanism through which this relationship occurs open fo r debate (Juvonen et al., 2000; Schwartz, Chang, & Farver, 2001; Schwartz & Gorman, 2003). Multiple variables influence the frequency of bullying behaviors, and the like lihood of a student becoming a bully and/or victim of bullying (Haynie et al., 2001; Nansel et al., 2001; Pellegrini, 1998; Pellegrini, Bartini, & Brooks, 1999). It is important to examine these variables and the relationships among them in order to provide schools with feasible routes that they can address to improve their environments. Estimates of bullying problems, either experienced as the perpetrator or the victim, vary across nations and studies. Percentages range from 15% in Norway (Olweus, 1997a, 1997b) to 18%-20% in E ngland (Boulton & Underwood, 1992) to 25% in Australia (Slee, 1994). Within the Unite d States, studies repor t differing frequencies of victimization, with 15% to 20% of st udents in the U.S. reporting being bullied (Batsche & Knoff, 1994). Mo re current estimates of bullying frequency report higher levels than those in past st udies, suggesting that bullying an d victimization are on the rise in certain populations. A recent study, usi ng somewhat different criteria, found much different proportions of mi ddle school population involvement in bullying situations. Seven percent of students were categorized as bullies, they had bullied others three or more times in the past year, while 31% of 6th through 8th grade students were considered victims, they had been bullied three or more times in the past year (Haynie et al., 2001). Nansel and associates (2 001) found that 30% of 6th through 10th grade students were
3 involved in moderate to freque nt bullying. Of those student s, 13% were classified as bullies, 11% were classified as victims, a nd 6% were classified as both bullies and victims. More recently, researchers have found that anywhere from 20-50% of urban school-aged children have been victimized or threatened with physical violence, which has attributed considerably to declines in academic performance (Schwartz & Gorman, 2003). The variability in bully ing and victimization estimat es is likely due to two significant issues in the field: defining wh at constitutes bullying and victimization and methodology employed to measure these behaviors. Defining Bullying and Victimization Several studies have focused primarily on defining behaviors that constitute bullying. Early definitions concentrated on individual or group violence toward an unpopular individual that begins and ends suddenly. One of the fields prominent researchers initially suggeste d that bullies are males w ho physically and emotionally harass their victims, whether the victims ar e males or females (Olweus, 1978). Olweus was the first to introduce the notion of em otional, or mental bullying, making it considerably more difficult to observe a nd agree upon all forms of definable bullying behaviors. Since his early de finition, several other defini tions of bullying also have included the notion of mental or psychological attacks in additi on to physical behaviors. Besag (1989) stressed the importance of long-term and systematic violence as integral in considering bullying behaviors. However, other researchers have not always found this element to be necessary. Arora ( 1996) argues that a single event of a physical or psychological attack or thre at delivered to a less powerful individual for the purpose of frightening and upsetting that individual is no less bullying th an long-term and sustained
4 attacks or threats. This definition also builds upon others by introducing a power differential between perp etrator and victim. Scandinavian researchers Bjorkvist, Ekman, and Lage rspetz (1982) emphasize that the long-term nature of bullying behaviors is indica tive of the social system occurring amongst students, which tends to be resistant to change. They suggest that bullying is a social form of aggression that occurs among individuals who encounter each other regularly. The emphasis in this de finition is the ongoing interaction between members in the group of students within which the bullying ta kes place. Other researchers, however, continue to consider the long-term aspect to be an important characteristic of bullying behaviors while also emphasizing the so cial and psychological aspects. For instance, Hazler (1996) defined bullying as repe ated behaviors that affect individuals physically, emotiona lly, and psychologically through wo rds, attacks, or social isolation. Some of the literatu re discusses the effect of th e long-term element of bullying on the victims, in addition to the severity a nd duration of the single bullying act. Perhaps the accumulation of bullying behaviors over time may be as relevant as or more relevant to the experience of victimization than the impact of each individual bullying behavior. Besag (1989) introduced the concept of in tentionality to bullying, which suggests a moral dimension to the behavior. Bullyi ng by this definition is intended to cause distress to others for the purpose of gratifying the aggressor. This definition suggests that it is not just the nature of the behavior that is important in determining what is bullying; the intended physical, psychological, and emotiona l impact of the behavior on others is of particular concern as well.
5 Olweus (1996) recently developed a more comprehensive definition of bullying and victimization that has been widely used in international studies. This definition identifies several concepts established in earlier definitions and reads as follows: We define or explain the word bullying. We say a student is being bullied when another student, or severa l other students: Say mean and hurtful things or make fun of him or her or call him or her mean and hurtful names Completely ignore or exclude him or her from their group of friends or leave him or her out of things on purpose Hit, kick, push, shove around, or lock him or her inside a room Tell lies or spread false rumors about hi m or her or send mean notes and try to make other students dislike him or her And other hurtful things like that. When we talk about bullying, these things happen repeatedly, and it is difficult for the student being bullied to defend himself or herself We also call it bullying, when a student is teased repeatedly in a mean and hurtful way. But we dont call it bullying when the teasing is done in a friendly and playful way. Also, it is not bullying when two students of about equa l strength or power argue or fight (Olweus, 1996, pg. 3). Olweus (1996) emphasizes that behavior is considered bullyi ng if it (1) occurs frequently either one-on-one or in a group, (2) involves a range of behaviors from physical aggressiveness to spreading rumors and (3) involves a power differential between aggressor and victim. The gender of perpetrators is no longer specified by
6 Olweus, as it had been in his and others prev ious definitions, sugge sting that both girls and boys can be bullies. Additionally, this defi nition supposes that behavior is bullying if it occurs more than once, which addresses an important distincti on between individual events and chronic victimizati on. Chronic victimization, in th is case, is associated with increased negative outcomes compared with the outcomes of students who experience infrequent bullying (Pynoos & Nader, 1988; Singer et al., 1995). Olweuss current definition has been used to guide self-re ports of behaviors for the U.S. National Blueprints Model Bullying Pr evention Program, which aims at decreasing bully and victim problems among primary and secondary school children through techniques to increase awareness of students, school admini strators, and parents of difficulties within the school environment (Olweus, Limber, & Mihalic, 1999). Of note, Olweuss definition considers te asing a form of bullying behavior as well. He indicates that repeated teasi ng, name-calling, or generally saying unpleasant things to others constitutes a form of bully ing. Pearce (1991) also developed a definition for bullying that incorporates teasing behavior Teasing could be considered bullying if it includes methods of intimidation that lead to distress in victims. Therefore, it is appropriate to suggest that racist and sexist attacks w ould be considered bullying behaviors, as long as they ha ve deliberate intent to harm others, are unprovoked, and are frequent (Swain, 1998). In other words, vic tims are not always beli eved to potentially induce bullying behavior against them. No single definition has been clearly es tablished as the gold standard for determining bullying behavior. However, several common elements emerge: physical, verbal, or psychological aggressi on intended to hurt others and cause distress in a victim,
7 the existence of a power diffe rential between bully and vi ctim, and that the bullying typically is not provoked by aggr essive acts (Swain, 1998). E ach definition is ultimately based on individual researchers opinions of what constitutes bullying behavior, thus confounding the interpretation of results between studies. In searching for a more complete definition, types of behaviors have been further categorized as direct and indirect forms of bullying (Olweus, 1996). Direct bullying behaviors are considered those overtly focused at a victim, and which tend to be easily observed. These behaviors include hitting, pushing, verbal abuse, stealing, and threats. Indirect bullying behaviors are those that are covert in their focus on th e victim. These behaviors include spreading rumors, ostracizing students, and purposefully ignoring or excluding students (Olweus, 1996). This distinction between direct and in direct behaviors has implications for how behavior is reported and observed as bullying. Assessing Bullying and Victimization Self-Report, Interview, Observat ion, and Peer Nomination Techniques In addition to the numerous ways bullyi ng and victimization have been defined, researchers have developed vari ous methods to assess bullying be havior. In general, four methodologies have been employed by past bullying studies: se lf-report surveys, interview, observation, and peer nominati on. Most commonly used, the self-report survey technique has become the method of choice for many studies. Surveys are relatively simple to administer to large num bers of students and the interpretation of responses is straightforward (Solberg & Olweus, 2003). So me studies have assessed bullying behaviors using two or three globa l items that require students to respond whether they generally bully st udents or have been bullied by students (Nansel et al.,
8 2001; Haynie et al., 2001). For these studies, the range of bullying behavior types was not assessed to the same extent that the Ol weus survey had measured them. The Revised Olweus Bully/Victim Questionnaire (Olweu s, 1996) has been accurate in assessing perceived bullying and victimization by specify ing two global items to classify general bullying and victimization with the addition of several items that iden tify various types of direct (e.g., hitting, pushing, or verbal abuse) and indir ect (e.g., social exclusion, gossiping, or spreading rumors) bullying be haviors that are engaged in and/or experienced. Bullying and victimization can be computed using the two global items and further explained using responses on the spec ific bullying type items. Because it is a brief and accurate scale to measure self -perceived victimization and bullying, many researchers choose to use the Olweus measure, or direct variations of the measure, to estimate bullying prevalence and identify stude nts with difficulties (Solberg & Olweus, 2003). As an alternate to survey techniques Wolke, Woods, Bloomfield, & Karstadt (2000) used an interview method for students in order to estimate bullying in classmates. The interview items were structured simila rly to the Olweus Bully/Victim Questionnaire and allowed students to elaborate on their ex periences rather than simply respond to items on a Likert scale. However, this me thod is time consuming making it difficult to recruit large numbers of participants. While interviews may be based on an established measure and can provide a wealth of qualitati ve information, the responses obtained from interview items are typically not scaled a nd less standardized. Using this method and considering its limitations, prevalence es timates of bullying behaviors may not be
9 comparable across schools. In addition, the information gained regarding bullying behaviors may not have equiva lent meaning across studies. Boulton (1993) employed a playgrou nd observation technique to measure bullying behaviors. This method requires independent observers to record classmate interactions and code behavi ors in accordance with Olweuss definition of bullying behaviors. An advantage of this technique is the recording of actual behavior, rather than having to rely on the accuracy, interpretation, or va lidity of child report. Disadvantages include inadequate observation of indirect bul lying and teasing, simila r to concerns about the accuracy of teacher-reported student inte rnalizing behaviors (Green, Beck, Forehand, & Vosk, 1980), and the costliness of employi ng independent observers to assess what may be relatively low base-rate behavior. However, if the emphasis of a study is not on estimating the prevalence of bullying, but on more comprehensive identification and assessment within a school of specific at -risk students, interview and observation methods may be useful (Solberg & Olweus, 2003). In addition to the survey, intervie w, and observation methods, The Peer Nomination Inventory, deve loped by Perry, Kusel, and Perry (1988), requires respondents to nominate which of their classmat es are bullies or vic tims. A benefit of this method is that students will be more likel y to validly report bullying if they have to report about others behavior. A disadvantage is that dir ect behaviors will be observed more readily than indirect, making those st udents who are physically aggressive more likely to be identified as bullies. Other st udents may not observe those who engage in such indirect bullying behaviors as isolat ion or rumor spreading, unless the reporting student experiences the bullying him or he rself. In addition, the procedures one
10 researcher uses to categorize student ratings resul ting from peer nominations are usually complex and difficult to reproduce (Solberg & Olweus, 2003). Further, the prevalence estimates obtained through peer nomination depend on fact ors within the school (e.g., number of students in the cl assroom, problem levels in th e classroom, standardization method of nominations, etc.), increasing the difficulty for other researchers to duplicate the procedures of others and extract simila r meaning from prevalence estimates (Solberg & Olweus, 2003). Olweuss paradigm for assessing bullying ha s been used in se veral international and national intervention strategies, incl uding the National Model Blueprints Bullying Prevention Program in the United States (O lweus, Limber, & Mihalic, 1999). The definition of bullying behaviors accomp anying the Revised Olweus Bully/Victim Questionnaire incorporates most components of bullying behavior that have been found important in past research. Assessment techniques have b een developed as a result of several author-formulated definitions, as previously discussed. Additionally, these assessment tools have been created to acco mplish the goal of gathering information on child behaviors via varying methods and each has pros and cons. Many have found selfreport survey techniques to be among th e easiest to administer and comprehend, especially when concerned about maximizing the accuracy of assessing perceived involvement in both easily and not-so-eas ily observed behaviors. The Olweus Bully/Victim Questionnaire is an example of a comprehensive self-r eport survey that provides distinct criteria for re porting ones behavior. This survey has been used in several studies in which information was obtained from students regarding their own behavior.
11 Using Teacher Reports to Identify At-Risk Children Assessment of child behavior can incorpor ate information from several sources. Ideally, a comprehensive assessment of child ad justment within the schools should utilize multiple informants. Oftentimes, obtaining re liable information from several raters in a single environment proves too costly and time consuming. Many studies, for that reason, rely on a single rater, commonly, teacher re port. Such assessment of large student populations is more cost effective and e fficient method for obtaining information on school environments and individual students. Conversely, the multi-informant literature suggests that a single means of assessment of ten lacks information that could be obtained through additional methods and perspectives. Therefore, studies that include multiple raters and techniques are in the best positi on to provide a well-r ounded picture of child outcomes (Holmbeck et al., 2002, 2003). Past research has shown the stability of teacher report of behavioral difficulties in school-aged children. For example, modera te stability was found for the Achenbach Teacher Report Form (TRF), specifically for scales related to externalizing behaviors (Achenbach, 1991; Verhulst & Van der Ende, 19 91). Of specific interest for the current study, the AML-R measure has displayed high re liability and validity for screening and evaluation purposes (Cowen, Dorr, Clarfiel d, Kreling, McWilliams, Pokracki, Pratt, Terrell, & Wilson, 1973). Multiple studies have confirmed the AML-Rs ability to identify children at-risk for subsequent adjustment and academic problems (Carberry & Handal, 1980; Durlak & Jason, 1984). In a ddition, teacher completed AML-R scores were consistent with independent observati ons of disruptive behavi ors and psychological and attention difficulties (Durlak, Stein, & Mannarino, 1980). The Teacher Checklist
12 (Dodge & Coie, 1987), which measures student involvement in aggression and bullying, has also been shown to adequately asse ss student behavior in comparison with observational techniques (Pel legrini & Bartini, 2000). While studies have shown teacher report to be a stable and effective method of assessment, reliability of teacher reports may differ between externalizing and internalizing difficulties (Gree n, Beck, Forehand, & Vosk, 1980). In situations with large numbers of students, teachers may have mo re difficulty identifying internalizing behaviors, such as anxiety, depression, and withdrawal, th an externalizing behaviors, such as aggression and inattention (Gillespie & Durlak, 1995). In reporting troublesome behaviors in the classroom, teachers are likely more concerned with students who present with very overt and aggressive behaviors rather than the students who are exceedingly quiet and withdrawn. In relati on to the present stu dy, teachers are more likely to identify direct bullying and victimization than indirect. This is typical of the observation and peer nomination methods previously discussed. Additionally, teacher report methods may not reflect the most accurate frequency of bully ing incidents since victimization generally occurs in places on the school campus that ar e not readily supervised by school personnel (Pellegrini & Bartini, 2000). Therefore, it is important to consider teacher report surveys as a part of a larger constellation of m easurement tools in order to explain child behaviors. Purpose of the Present Study It has been demonstrated that bullying and victimizat ion are significant problems in schools and that there is considerable deba te over how to define and measure bullying behaviors. Despite the lack of an identif ied assessment gold standa rd, the present study
13 attempts to maximize the identification of bot h direct and indirect victimization through the use of teacher and self -report methods. Regardless of the assessment methods employed, it can be speculated that victims generally represent a greater proportion of those students involved in bu llying situations, perhaps beca use bullies target several students at a time and/or victims are more lik ely to report behavior to which they do not attribute personal responsibility. The pres ent study seeks to ex amine psychological and school-related outcomes of victimization and f actors that may increase or decrease the likelihood of poor academic and psychological functioning as a result of victimization. Effects of Exposure to and Expe rience of Peer Victimization Psychological Consequences In examining the relationship between victimization and functioning, it has been documented that students experiences with pe er victimization or exposure to violence have consistently been associated with em otional maladjustment (Boivin & Hymel, 1997; Crick & Grotpeter, 1996; Egan & Perry, 1998; Kochenderfer & Ladd, 1996a; Olweus, 1978, 1994a, 1994b). Psychologically, they are more anxious, depressed, withdrawn, and have lower self-esteem (Craig, 1998; Hayni e et al., 2001; Olweus, 1995; Rigby & Slee, 1991). Victimized students may cry easily, e xhibit anxiety, appear withdrawn, and lack self-esteem and confidence (Boivin & Hyme l, 1997; Olweus, 1978; Perry et al., 1988; Perry, Williard, & Perry, 1990; Schwartz, D odge, & Coie, 1993). Additionally, strong predictors of reported anger include exposure to and experience of violent victimization, such as threats, hitting, or beatings (Singer, Anglin, Song, & Lunghofer, 1995). Some victims may blame themselves for their social status, which contribu tes to feelings of loneliness and depression (Graham & J uvonen, 1998; Nolen-Hoeksema et al., 1992;
14 Toner & Munro, 1996; Renshaw & Brown, 1993; Nolen-Hoeksema, Girgus, & Seligman, 1986). It is prudent to be concerned about the psyc hological consequences of victimization at school as th ey are often associated with decreased daily functioning, including school avoidance and difficu lty concentrating on schoolwork (Juvonen, Nishina, & Graham, 2000). Behavioral Consequences With regards to school behavior, some vi ctimized students exhibit externalizing difficulties and find themselves in situations where they become involved in disrupting classroom discipline and displaying aggr ession (Boivin & Hymel, 1997; Olweus, 1978; Perry et al., 1988; Perry, Perr y, & Kennedy, 1992). Victims also report more behavioral difficulties and acceptance of misconduct than students uninvolved in bullying, although not to the degree as those w ho perpetrate bullying (Haynie et al., 2001). This acceptance of aggression is notable cons idering that victimized stude nts are the targets of other students misconduct. Furthermore, childhood aggression is related to development of internalizing and externaliz ing difficulties (Coie, Loch man, Terry, & Hyman, 1992). For instance, students who are victimized and also take part in bullyi ng other students have poorer emotional adjustment than those w ho do not engage in aggressive practices (Haynie et al., 2001). Some victimized student s can have lower self-concepts and poorer perceptions of their competency to appropr iately interact with others (Callaghan & Stephen, 1995; Hawker & Boulton, 2000; Neary & Joseph, 1994; Rigby & Cox, 1996). More specifically, aggression and beha vioral misconduct are associated with psychological symptomatology, which in turn, is associated with students perceptions of
15 poor self-concept (Coie, Lochman, Terry, & Hyman, 1992; Edens, Cavell, & Hughes, 1999; Hay, 2000; Marsh, Parada, Yeung, & Healey, 2001). Aggressive and noncompliant behavior at school can al so interfere with school performance (Coie & Krehbiel, 1984; Schwar tz & Gorman, 2003). Researchers have found that students who are aggressive at schoo l are typically rejected by their peers, are more poorly adjusted to school and educationa l goals, and tend to perform more poorly academically (Coie, Dodge, & Kupersmidt, 1990; Hay, 2000). Involvement in aggressive and disruptive behavior at school can increa se the likelihood of academic failure (Wells & Rankin, 1983), which may impact the educa tional exchange with teachers (Wentzel, 1993a) and distract students from learning (Doyle, 1986). Students who engage in high levels of disruptive and noncompliant behavi or in school require teachers to focus on classroom management rather than instructi on (Wentzel, 1993a). Thus, it is suggested that the relationship between behavioral difficulties at school and poor academic outcomes may be more powerful for those who are victimized and exhibit poorer psychological functioning than thos e who are victimized alone. School-Related Consequences While past literature has well establishe d the relationship between victimization and resulting internalizing and externalizing difficulties, peer victimization has also been associated with school-related factors (Boulton & Underwood, 1992; Graham & Juvonen, 1998; Hawker & Boulton, 2000; Kochenderf er & Ladd, 1996a, 1996b; Reid, 1989; Slee, 1994). Victims are less popular in school th an other students, including bullies (Pellegrini, Bartini, & Brooks, 1999). Co mpared to students uninvolved in bullying, victims bond and adjust more poorly to sc hool and classmates (H aynie et al., 2001).
16 Generally, dislike by and rejection from peers can be viewed as a stressful situation (Albee, 1984) in which students who are less readily accepted by thei r peer group are less involved in peer activitie s at school (Dodge, 1983; Dodge, Coie, & Brakke, 1982). Victimization is associated, as well, w ith school avoidance (Kochenderfer & Ladd, 1996a). Students who are victimized by peer s have a greater inci dence of truancy or avoiding school activities in order to avoid bullying s ituations. School avoidance generally has a negative impact on student s motivation at and interest in school (Wentzel, 1998) and their academic performance (Juvonen et al., 2000). The literature discussing the associati on between victimization and academic achievement is less confirmatory. Many pr evious studies operationalized achievement using grades or single subject exam scores, teacher-reported learning difficulty ratings, or student reports of perceived scholastic performance and commitment to educational goals (Austin & Joseph, 1996; Juvonen et al. 2000; Na nsel et al., 2001; Schwartz & Gorman, 2003). Additionally, some researchers only confirmed associations between victimization and achievement in specific ethnic groups (McCall, Beach, & Lau, 2000; Schwartz, Chang, & Farver, 2001). It is believed that poor academic and school outcomes are due to psychological maladjustm ent and emotional distress that follows experiences with victimization. Specifically, researchers have investigated victimization by peers at school and psychological difficulties, such as depression, anxiety, and anger, and have found mixed results in their a ssociation with academic outcomes, both academic processes and achievement (Juvonen et al., 2000; Nansel et al., 2001; Schwartz, Chang, & Farver, 2001; Schwartz & Gorm an, 2003; Wentzel, 1994; 1998; Wentzel & Caldwell, 1997; Wentzel, Weinberger, Ford, & Feldman, 1990). In fact, researchers
17 reported that the possible mechanism in whic h achievement outcomes may be related to bullying is through the moderating and media ting effects of psychosocial and emotional factors such that, when these factors are not analyzed, the direct relationship between academic achievement and involvement in bullying is reduced to nonsignificance (Juvonen et al., 2000; Kochenderfer & Ladd, 1996a; Wentzel & Caldwell, 1997). Other studies found that valuing educational goals and decision-making competency are among the best predictors of academic behavior, su ch as engagement in learning and time spent on academic tasks, which are arguably rela ted to overall achievement (Ames, 1992; Miller & Byrnes, 2001). Barriers to valui ng educational goals and optimal academic performance can take the form of emotional a nd peer difficulties, such as victimization and related psychological dysfunction, which ca n have a detrimental impact on grades and test scores. Further research examin ing the relationships of various types of academic outcomes with bullying and victimi zation may clarify some of the current inconsistencies in the literature. Protective and Risk Factors Peer Relationships and Prosocial Activities While the association between victimi zation and psychological dysfunction has been established in the literature, studies s uggest that some experiences and aspects of childrens lives may interact with the relations hip and alter outcomes. Previous research has recognized that quality friendship moderates the experi ence of victimization and harsh home environments and suggested that frie ndship effects be assessed in the context of other related factor s with victimization (i.e., emoti onal regulation; Schwartz, Dodge, Pettit, & Bates, 2000). While student victims tend to have more interpersonal difficulties
18 and poorer social skills than other student s (Besag, 1989; Haynie et al., 2001) and tend to be disliked by peers (Graham & Juvone n, 1998), poorer psychological adjustment, as related to peer victimization, may be attenuated by greater student and teacher support at school (Wentzel, 1998). It is believed that so cial relationships are related to adjustment because they can temper the negative effect s of stressful situations, such as peer harassment and victimization (Cohen & Wills, 1985). In fact, research has found that peer social support is a nega tive predictor of ps ychological distress, with girls reporting greater levels of distress and friendship salience than boys (Wentzel, 1998). Friendship has previously been found to act as a modera tor between victimization and emotional and behavioral difficultiesthose with fewer frie nds have greater diffi culties (Hodges et al., 1997). Victimized students fair better psychol ogically when they have supportive peer relationships that protect th em from negative interactions with others (Hodges et al., 1997; Hodges et al., 1999). There is variability in the degree to which friendships protect against victimization and the negative effects thereof; if friends e xhibited characteristics (e.g., internalizing or weakness) that make it difficult for them to provide protection, internalizing and externalizi ng outcomes of victimized students increased (Hodges et al., 1997, 1999). Similarly, victimized students who report spending an above average amount of time with their friends had increase s in internalizing difficulties. This may be a result of overly involved and enmeshed relationships, which serve to enhance moodiness and other internalizi ng problems rather than de-escalate them (Hodges et al., 1999). Positive peer relationships, in particular, are believed to influence students emotional well-being, which has implications for general adjustment and involvement in
19 prosocial activities (Wentzel 1998). Peer relationships and involvement in prosocial activities can influence student social responsibility and involvement in positive behaviors at school, which have a positive asso ciation with desire to do well in school, both behaviorally and academically (DeRos ier, Kupersmidt, & Patterson, 1994; Muma, 1965; Wentzel, 1991, 1993a, 1993b). This relationship, however, is influenced by a variety of factors. For example, soci al support was found to improve student bonding with school by alleviating the negative impact of psychological distress from victimization (Cohen & Wills, 1985; Wentzel, 19 98). Prosocial interactions with peers (Green, Forehand, Beck, & Vosk, 1980) and co mpliant classroom behavior (Wentzel, Weinberger, Ford, & Feldman, 1990) have been related to positive educational outcomes. Additionally, the desire to behave in prosoc ial ways is related to academic motivation (Wentzel, 1991; 1993b). Social re latedness contributes to the adaptation of socially and institutionally sanctioned goals, whereas la ck of bonding with others could lead to rejection of such goals (Connell & Wellborn, 1991). Coping Mechanisms Research has found that a large proportion of studen ts report experiencing victimization, roughly 75%, while a sma ller number (15%) of students experience significant distress and maladjustment related to victimization (Hoove r, Oliver, & Hazler, 1992). Some studies have suggested that this difference in student experiences following victimization may be due to the way that st udents cope with negativ e peer interactions and distressful situations (F ields & Prinz, 1997; Kochende rfer & Ladd, 1997; Smith et al., 2001). Coping strategies in response to victimization that are characterized by problem solving and seeking the support of others attenuated symptoms of anxiety and
20 depression and buffered peer relationships (Kochenderfer-Ladd & Skinner, 2002). However, those coping strategies that employ the use of aggressive responses or selfblaming strategies have a tendency to exacerba te internalizing difficulties, particularly with girls (Dempsey, 2002; Koch enderfer-Ladd & Skinner, 2002). Coping strategies are believed to temp er the relationship between psychological functioning and academics in that those w ho use positive methods for dealing with distress will function better at school. This is important cons idering that past research has found that the lack of peer and teacher relationships at sch ool puts students at risk for academic difficulties (Austin & Draper, 1984; Coie, Dodge, & Kupersmidt, 1990; DeRosier, Kupersmidt, & Patterson, 1994; Goodenow, 1993; Kochenderfer-Ladd & Skinner, 2002; Li, 1985; Midgley, Feldlauf er, & Eccles, 1989; Phelan, Davidson, & Cao, 1991; Parker & Asher, 1987; Wentzel, 1991). School Climate Factors Development of a supportive school climate on the part of teachers and staff is uniquely predictive of student classroom pe rformance, interest in education, and goalorientation (Wentzel, 1998). Studies have shown that classroom and school climate characteristics are important in understanding individual student char acteristics (Barth, Dunlap, Dane, Lochman, & Wells, 2004; Wang, Haertal, & Walberg, 1990). Students in better school environments rate their teacher s as organized and supportive (Barth et al., 2004). Furthermore, students will be motivated to engage in school ac tivities if there is a sense that teachers care for and support st udents, particularly those students who generally view their classmat es as threatening (Barth et al., 2004; Cohen & Wills, 1985; Wentzel, 1997). Overall, re searchers have documented that children who do not have
21 supportive relationships with peer s and adults at school, or ar e less accepted by peers, are at greater risk for academic failure (Austin & Draper, 1984; Coie, Dodge, & Kupersmidt, 1990; Goodenow, 1993; Kochenderfer & Ladd, 1996a, 1996b; Li, 1985; Midgley, Feldlaufer, & Eccles, 1989; Parker & Ashe r, 1987; Phelan, Davidson, & Cao, 1991). It has been found that students who were better bonded with school exhibited higher academic achievement (Wentzel, 1994; 1998). More specifically, supportive peer and teacher relationships at school were found to predict increases in interest in school, and therefore academic performance, often by way of psychological adjustment (Felner, Aber, Primavera, & Cauce, 1985; Goodenow 1993; Wentzel, 1998; Wentzel & Asher, 1995; Wentzel & Caldwell, 1997). Notably, there is little evidence that interv ention variables have been specifically examined in the type of pathway proposed in this study. However, researchers have documented that interventions designed to ta rget attributions about aggression are less effective in environments in which aggressi on is viewed as an appropriate behavior and/or response to provocation (Aber, J ones, Brown, Chaundry, & Samples, 1998). Based on this research and the findings th at suggest social support and structured classroom environments at school improve students experiences with victimization (Springer & Padgett, 2000), it is probable that certain school climate variables, such as intervention on the part of teachers and classm ates, may mitigate the relationship between victimization and development of psychological symptomatology. Gender Differences Haynie and colleagues (2001) suggest that boys and girls may engage in and experience different types of bullying behavior. Girls tend to organize their bullying in a
22 more social manner, around rumor spreading and manipulation of friendships, while boys exhibit more physically aggressive activities (verbal abuse, phys ical attacks, and threats). The only form of bullying that is more prevalent among girls is that of social intimidation, or relational aggression (Crick & Grotpeter, 1996; Batsche & Knoff, 1994; Carney & Merrell, 2001). Additionally, female victims are more concerned with being ignored at school and negatively evaluated by peers than male victims (Slee, 1995). Girls exposed to victimiza tion reported greater levels of psychological distress, such as anxiety, depression, and anger (Si nger et al., 1995, Springe r & Padgett, 2000). Girls coping styles te nd to be more prosocial than that of boys (Hausman, Spivak, & Prothrow-Stith, 1994; Schwab-Stone, Ayers, Kasprow, Voyce, Barone, Shriver, & Weissberg, 1995). The use of problem-solvi ng and seeking adult intervention may be more beneficial in terms of follow up psyc hological functioning fo r girls and those who are infrequently victimized (Carver, Scheier, & Wein traub, 1989; Chung & Asher, 1996; Endler & Parker, 1990; Maccoby & Jacklin, 1974) This pattern of response toward aggression decreases th e likelihood that girls would become a perpet rator of aggressive behavior as a result of victimization (Slaby, 1998). Given girl s concerns about maintaining peer relationships and status in social networks, victim ization may place girls at greater risk for developing symptoms such as depression and anxiety (Horowitz, Weine, & Jekel, 1995). Interestingly, studies have found that higher levels of perceived social support were associated with greater reported psychol ogical distress for girl s (Springer & Padgett, 2000). This finding could be explained as a contagion e ffect (Springer & Padgett, 2000, pp. 377), in that during times of stress, gi rls may have a tendency to seek increased
23 connectedness with others in similar situa tions, potentially hei ghtening their distress (Belle, 1987). Therefore, it is unclear fo r the present study whether social support and friendships will intensify psychological dysfunction resulting from victimization for girls or protect against the developm ent of psychological symptoms. Summary and Proposed Model Hypothesized Pathways and Relationships Based on a review of the relevant lite rature, several hypothesized associations were examined. Victimization was expected to relate to poorer academic outcomes, by way of psychological functioning. More spec ifically, the expected victimizationpsychological functioning relationship was expe cted to relate with achievement through the direct association with academic mo tivation processes (i.e., Victimization Psychological Functioning Academic Motivation Academic Achievement). There is considerable evidence to suggest that acad emic processes are precursors to objective academic achievement outcomes. Researchers have found that academic outcomes are more directly related to academically oriented attitudes, interest in school, and motivation to earn high grades (Ames, 1992; Corno & Mandinach, 1983; Pressley, Borkowski, & Schneider, 1987; Sivan, 1986; Wentzel, 1993b, 1994, 1997; Wentzel, Weinberger, Ford, & Feldman, 1990; Zimmerman & Schunk, 1989). Therefore, it was hypothesized that victimization would be associated with psyc hological functioning, which in turn, would be associated with motivation to achieve and then ultimately academic achievement measures. Additionally, specific related constructs (Friendship, School Climate, Prosocial Activities and Influences, Aggre ssion, and Teacher-Reported Difficulties) are each expected to moderate the hypothesized Victimization Psychological Functioning
24 path, either positively or negatively. Disruptio ns in psychological functioning related to victimization are believed to vary depending on students friendships and levels of involvement in positive activit ies with family and peers, experiences with support and intervention on the part of adults or peers at school, and engagement in aggressive behaviors. Past research has supported the present hypothesized victimization psychological factors academic outcomes temporal pathway (Juvonen, Nishina, & Graham, 2000; Kochenderfer & Ladd, 1996a ; Schwartz & Gorman, 2003; Wentzel, 1991). The present study hypothesized that the pathway between experience of victimization and psychological difficul ties will be moderated by involvement in prosocial activities and positive influences, su ch as spending time in family activities and with friends who do not engage in deviant beha viors. Illustrativel y, it was expected that victimized students would exhibit fewer ps ychological difficulties the greater their involvement in positive and socially sancti oned peer and family-based activities. The degree to which prosocial influences and activities moderate this relationship was expected to vary by gender given the positiv e correlation between p eer relationships and engagement in prosocial activit ies, such that the relations hip between victimization and psychological functioning may be attenuated mo re strongly for girls than for boys. It was also expected for the present study that aggressive coping beliefs and engagement in aggressive behaviors would moderate the relationship between victimization and psychological outcomes, with aggression associated with poorer functioning. It was hypothesi zed for the present study that relationships involving the belief in and use of aggressive behaviors w ould vary by gender, in that aggression would
25 influence the relationship between victim ization and psychological functioning more strongly for boys than for girls. For the present study, it was expected that Friendship, as defined by the number of friends students have, as well as School C limate, defined by level of intervention and supervision at school, would m oderate the relationship betw een student experiences with victimization and psychological functioning. For instance, those victimized students who have strong peer relationships and have sc hool environments in which there was an emphasis on intervention against poor peer relationships and student misconduct would have better psychological f unctioning in terms of depr ession, moodiness, anxiety, and anger. It is unclear, however, whether victim ization and peer harassment have an independent direct relationship with acad emic outcomes (Juvonen, Nishina, & Graham, 2000). It was hypothesized in the present st udy that psychological functioning would mediate the relationship between vict imization and academic motivation and achievement. Past research has suggested that the indirect pa thway is the most parsimonious in predicting outcomes from peer harassment (Juvonen, Nishina, & Graham, 2000). The following section outlines the specific hypotheses for the present study and proposed model (see Figure 1). See also Appendix A for a list of the independent and dependent constructs and A ppendix B for a descrip tion of the variables and items that defined each construct.
26 Hypotheses Hypothesis 1 It was expected that Victimization w ould be related negatively to Academic Motivation, as defined by orientation towa rd educational goals, school efficacy, and teacher-reported learning ratings, by way of Psychological Functioning, as defined by self-reported depression, anger, and anxi ety ratings and teacher-reported moodiness ratings. Specifically, Psychol ogical Functioning was believed to mediate the relationship between Victimization and Academic Motiv ation, such that Motivation would be negatively influenced by Victimization if students have poor Psychological outcomes. Hypothesis 2 It was expected that Victimization would be related to Academic Achievement, as defined by standardized test scores and grades, by way of Psychological Functioning and Academic Motivation (Victimization Psychological Functioning Academic Motivation Academic Achievement). Hypothesis 3 The relationship between Victimizat ion and Psychological Functioning was expected to be moderated by Friendship, in th at victimized students who have more close friends that they spend time with at school would have better Psychological Functioning than students who have a limited number of friends. Hypothesis 4 Prosocial Activities and Influences were hypothesized to moderate the relationship between Victimization and Psyc hological Functioning. Victimized students who reported greater involvement in community activities (e.g., youth groups or clubs),
27 activities at school (e.g., special interest clubs), sports, fa mily activities (e.g., spending regular time with family members), and have friends who devalue aggressive and deviant behaviors were more likely to report bette r Psychological Functioni ng than students who were not involved with such indi viduals and in pr osocial activities. Hypothesis 5 It was expected that reported engage ment in Aggression and holding aggressive beliefs, as assessed by self-reported bullyi ng and attitudes toward negative coping strategies, would influence the relationship between Vict imization and Psychological Functioning. Victimized students were hypothesized to report greater psychological symptoms if involved in aggr essive behaviors than if th ey are not. This may seem counterintuitive, in that it could be expected that vi ctimized students fare better psychosocially when engaged in rule-breaking and aggressive behavior because they are somehow showing an ability to fit in with those who victimi ze others. However, studies have found that students who are victimized and victimize others are at risk for the poorest outcomes (Haynie et al ., 2001; Nansel et al., 2001). Speculatively, this finding may be due to aggressive victims inability to adequately and consistently fit in with any peer group (Pellegrini, 1998). Hypothesis 6 It was expected that Teacher-Reported Difficulties at school would moderate the relationship between Victimization and Psyc hological Functioning, in that victimized students who are identified by teachers as engaging in poor peer interactions and behavioral misconduct at school, would have poorer psychological outcomes compared
28 with those students who do not engage in problem behaviors or are not bullied by classmates. Hypothesis 7 It was hypothesized that School Climate, defined by intervention and supervision at school, would moderate the relationsh ip between Victimization and Psychological Functioning. Victimized students who have supportive school envi ronments, in that teachers and students intervene to stop bullyi ng and disruptive behavior, were expected to have better psychological outcomes as a resu lt of their victimizat ion than students who are not supported by teachers and classmates. Hypothesis 8 The hypothesized moderator influences on the mediated pathways were expected to differ by gender. For instance, Friendshi ps and Prosocial Activ ities and Influences were expected to play a stronger role in explaining the relationship between Victimization and Psychological Functioning for girls than for boys. Because boys engage in problem behaviors at school more frequently than girls, it is expected that Teacher Reported Difficulties and Aggression would play a stronger role in victimized boys psychological outcomes.
29 Chapter 2 Method Participants Four thousand two hundred and seventy-two (4272) students were recruited from all eleven middle schools (student age range 11 14 years) in a la rge school district (approximately 55,000 students), which included urban, suburban, and rural areas, during the 2002-2003 academic year. Two sets of surv eys were administered: Form A assessed psychosocial, family, and academic functio ning, and Form B assessed school climate factors. Three classes out of the seven re cruited per grade completed Form A, three classes completed Form B, and one class per grade completed both Form A and Form B. Teachers completed behavior rating scales on a subgroup of 2053 students. Only those participants (N = 327) who re ported some degree of victimization by peers on both Forms A and B were included in the present study. The majority of the sample was Caucasian (N = 241, 73.7%), while 12.2% were Latino/La tina/Hispanic (N = 40), 3.4% AfricanAmerican (N = 11), 2.4% Asian/Indian (N = 8) and 8.3% as other (N = 27). Fewer boys (N = 145, 44.3%) than girls (N = 181, 55.4%) participated, 2 (1, N = 327) = 3.98, p < .05.
30 Measures Student Self-Report Surveys The Demographic portion of the student surv ey packet was administered in order to obtain information on age, gender, race/e thnicity, family consistency, and friends. Students were asked how many close friends th ey have at school, which was used as an indicator of the Friendship construct. The Revised Olweus Bully/Victim Questionnaire (Olweus, 1996) is a 39-item scale covering aspects of bullying problems (s ee Appendix C). Participants responded to 9 items about direct and indire ct victimization (i.e., I was called mean names) and 9 items about direct and indirect involvement in bullying others (i.e., I spread false rumors about another student and tried to make others dislike him/her) on a 5-point scale: 1 = I havent been bullied at school in the past couple of months , 2 = it has only happened once or twice, 3 = 2 or 3 times a month, 4 = about once a week, and 5 = several times a week. Participants also completed 4 items related to involvement on the part of students and adults at school to stop bully ing as indicators of the School Climate construct (i.e., How often do teachers or othe r adults at school try to put a stop to it when a student is being bulli ed?). Cronbachs alpha for the bullying items scale from this sample is .67, while alpha for the vict imization items scale is .72. Victimization items will be used as an indicator of the Vi ctimization latent construct. Bullying items will be used as one of the Aggression cons truct indicators. The Olweus Bully/Victim Questionnaire self-report responses has b een found to correlat e significantly ( r = .60-.70 range) with student nominations of vict imized classmates (Olweus, 1991a, 1991b).
31 The Center for Epidemiologica l Studies-Depression Scale ( CES-D ; Radloff, 1977) is a 20-item scale (I felt depressed, Cronbachs alpha = .87) used to measure self-reported depressive symptomatology on a 4point scale, 0 = nev er to 3 = most of the time, and will be aggregated as an indicator of Psychological Functioning (see Appendix D). The CES-D scale showed concor dant validity in identifying depressive symptomatology compared with the Beck De pression Inventory, with an 88% agreement between the two scales (Robert, Lewinsohn, & Seeley, 1991). A corresponding version of the scale items was developed for use with children and has shown adequate ability to assess depressive symptomatology (Faulsti ch et al., 1986; Weissman, Orvaschel, & Padian, 1980). The State/Trait Anxiety Inventory for Children ( STAIC ; Spielberger, 1973) is a 40-item questionnaire (see Appendi x E) that assesses self-report ed anxiety as an indicator of Psychological Functioning, on a 3-point scale, 1 = hardly ev er to 3 = often. Two 20-item scales comprise the questionnaire: Stat e, related to current estimated levels of anxiety, and Trait, related to c onsistent and cross-s ituational levels of anxiety. Only the Trait Anxiety subscale (e.g., I worry too much and I notice my heart beats fast) was collected during survey administra tion in order to remain consis tent with past literatures assessment of typical child mood in relation to behavior (Cronbach s alpha = .91). The STAIC was found to correlate .75 with the Ch ildrens Manifest Anxiety Scale (CMAS; Castaneda, McCandless, & Palermo, 1956) a nd .63 with the General Anxiety Scale for Children (GASC; Sarason, Davidson, Lighthall, Waite, & Ruebush, 1960). The State/Trait Anger Expression Inve ntory for Children and Adolescents ( STAXI-C/A ; Spielberger, Jacobs, Brunner, & Lunsford, 2002) is a 53-item survey that
32 assesses self-reported anger (see Appendix F) The STAXI-C/A was developed based on the adult version of the Revised State/Tra it Anger Expression I nventory (STAXI-2), which contains six major scales: State A nger, Trait Anger, and Anger Expression, comprised of the Anger Out, Anger In, Anger-Control/Out, and Anger-Control/In subscales (Spielberger, 1998). For the surv ey administration this study is based on, the Trait Anger (Cronbachs alpha = .83) and Anger Expressi on (Cronbachs alpha = .72) scales will be used as indicators of Psychological Functioning and Aggression, respectively. The STAXI has s hown to relate significantly to self-report of intensity and frequency of daily anger (Deffenbacher, 1992). Pilot data is currently being collected on this scale and further informa tion regarding reliability and validity must be obtained. A sample item on the Trait Anger scale includes I feel angry; while a sample item for the Anger Expression scale is I ge t into arguments. Items are rated on a 3-point scale (1 = Hardly Ever, 2 = Sometimes, and 3 = Often). The School Adjustment Survey ( SAS ; Santa Lucia & Gesten, 2000) is a 34-item scale assessing self-reported student bonding and adjustment to school, classmates, and teachers (see Appendix G). The survey consis ts of five scales: School Spirit (I like school, Cronbachs alpha = .85), Goal-Orien tation (Education is important for success in life, Cronbachs alpha = .74), Child-Peer Relations (Most student s at school like to include me in their activities, Cronbachs al pha = .63), Child-Teacher Relations (I think my teachers care about me, Cronbachs alpha = .84), and Alienation (I dont feel safe at school, Cronbachs alpha = .63). For the present study, only the Goal-Orientation scale will be used as an indicator of the Academ ic Motivation latent variable. The SAS has
33 been shown to adequately discriminate am ong students engaged in school and those atrisk for failure (Santa Lucia & Gesten, 2000). The Middle School/High School Student Survey (Safe Community-Safe School Project, 2002) is a 131-item questionnaire that measures various components of parental influences, peer relationships, exposure to school violence, teacher relations, beliefs about aggression and substance use, risk-t aking behaviors, and school bonding (see Appendix H). Eight items measured peer harassment (Cronbachs alpha = .82) on a 4point scale (0 = no, to 3 = more than 6 tim es), which will be used as an indicator of the Victimization latent construct (e.g., Another student pushed, shoved, slapped, or kicked me), 2 items measured aspects of Fr iendships (Friendship index, I spend most of my free time at school w ith my friends), 12 items measured Friends Attitudes toward Aggression (My friends think its wrong to hit other people , Cronbachs alpha = .86) and Friends Attitudes toward Deviant Behavior (My friends think using drugs is a dumb idea, Cronbachs alpha = .87) and 6 ite ms measured Family Involvement (e.g., I like to do things with my family, Cronbachs alpha = .75) and Peer Involvement (e.g., I am involved in clubs at my school, Cronbachs alpha = .62) as indicators of Prosocial Activities and Influences, 6 ite ms measured Bullying Others (e.g., I harassed another student, Cronbachs alpha = .86) and 8 items measured Aggressive Coping Behaviors (e.g., I deliberately kept someone out of my group because I was angry at them) and Beliefs about Aggressive Copi ng (e.g., It is OK to push or shove other people around if youre mad, Cronbachs alpha = .67) as in dicators of Aggression, 4 items measured school Climate by way of intervention efforts at school (e.g., Adults at my school teach us not to pick on other students, Cronbach s alpha = .74), and 4 items measured School
34 Efficacy, an indicator of Academic Motiv ation (e.g., If I stu dy hard, I will get good grades, Cronbachs alpha = .65). Items th at comprise the scales showed convergent validity with items from established scales, such as the Individual Protective Factors Index (Springer & Phillips, 1997), The M onitoring the Future Survey (Johnson & Bachman, 1980), and the Youth Risk Behavior Surveillance System (Center for Disease Control and Prevention, 2002). The Adult Supervision at School index consists of six items developed for the 2003 data collection by the author (Totura et al., 2005) to asse ss adult supervision within schools (see Appendix I), as an indicator of Climate (e.g., In my sc hool teachers are in the hall when we change classes and In my classroom teachers walk around while students are working). While the inde x may have limited internal consistency (Cronbachs alpha = .60), it is likely that the items represent separate factors that have validity in terms of m easuring aspects of adult presence on school campuses. In addition, researchers (Bickman, Lambert, Karver, & Andrade, 1998; Clark & Watson, 1995) point out that there is often a trade off between internal consistency, breadth in validly measuring a construct, and test length. Th is measure was developed optimizing two of these areas. Teacher-Report Survey The AML Behavior Rating Scale Revised (AML-R) is a 12-item teacher-report survey (Cowen et al., 1973) used to asse ss student maladjustment (see Appendix J). Three scales comprise the AML-R: Acting-Ou t (Disrupts class di scipline, Cronbachs alpha = .90), Moodiness (Is unhappy, Cronbach s alpha = .83), and Learning (Gets off task, Cronbachs alpha = .92). The Acting-Out scale will be used as an indicator of
35 Teacher-Reported Difficulties, the Moodiness scale will be used as an indicator of Psychological Functioning, and the Learning scale will be used as an indicator of the Academic Motivation latent construct. The AML-R has adequate validity and internal consistency ranges over a two week peri od, alpha = .80 to alpha = .86 (Carberry & Handal, 1980; Cowen et al., 1973; Durla k, Stein, & Mannarin o, 1980; Gillespie & Durlak, 1995). Scores on the AML-R have al so been correlated with personality and academic achievement measures (Dorr, Stephens, Pozner, & Klodt, 1980), and have distinguished between children who were referr ed for mental health services and those who were not (Cowen et al., 1973). One item was added to the AML-R survey that assessed global levels of bullying by others (i .e., This child has been bullied at school in the past couple of months), as an indicator of the Teacher-Reported Difficulties latent construct. Records Data The standardized Florida Comprehensive Achievement Tests (FCAT) is a statewide measure of academic achievement in reading, writing, and mathematics. Internal reliabilities for the total test battery ra nge from .86 to .91 for grades 4 through 10 (Florida Department of Education, 2002). Fi eld-test items for the FCAT were developed by Harcourt Educational Measurement (see Analysis of the FCAT Test Item Review Conducted by the Florida Department of Education and Harcourt Educational Measurement: 1999 for more information on the de velopment of the FCAT items; Harcourt Educational Measurement, 2000). The Developmental Scale Scores for Reading Comprehension and Math Problem-Solvi ng tests were used as an assessment of academic achievement, and as an indicator of Academic Achievement, that could be
36 compared across school years. The range of FCAT Developmental Scale Scores is 863008. Validity of the FCAT scales is dete rmined by scoring directors and Florida Department of Education represen tatives who evaluate scores to be sure they fall within a range of accuracy (Florida De partment of Education, 2002). Student Grades were obtained as a measure of Academic Achievement. Grades are defined on a 5-point scal e: A = 4, B = 3, C = 2, D = 1 and F = 0. Grades were aggregated and averaged for each student. The total number of Discipline Referrals for the 2002-2003 academic year was obtained for each participant and averaged as an indicator of Teacher-Reported Difficulties. Referrals are disciplinary re ports completed by teachers and staff for individual student behaviors, such as disobedience and trua ncy, and are aggregated by the district. Procedure This study was developed in collaboration with the school distri ct as part of a broader assessment of school environment. W ithin each school, seven classes per grade, approximately 20-30 students each class, were randomly selected to complete student and teacher surveys. Students completed survey packets that addressed individual, family, and school-related factors and were administ ered by teachers with the help of school psychologists, guidance counsel ors, and study research assi stants in a group format within selected classes during the second half of the school year. Students and teachers were provided with a standard definition of bullying behavior to guide responses (see Appendix K). Student and teacher surveys were coded to maintain ch ild confidentiality. Since this survey was part of a district mandated needs a ssessment, consent procedures
37 were determined by the school administration co nsistent with district policy. A letter was sent to students parents or guardians inform ing them that their child would be involved in a survey to improve school climate. T hose who chose to declin e participation were asked to contact the school and were not assessed. Analyses In order to examine the proposed hypotheses, five levels of analyses were utilized. Descriptive statistics were first obtained fo r each variable and construct proposed. These were obtained following a Pr incipal Component Analysis with Varimax rotation that elucidated which observed variables load ed together on the hypothesized latent constructs. Pearson Product-Moment correl ations were run to assess the simple associations between each of the observed variables and late nt constructs. Structural Equation Modeling was used to test the validity of the proposed Victimization Psychological Functioning Academic Motivation Academic Achievement mediated model. Hierarchical regression analyses were conducted to examine hypothesized moderator effects, by gende r, on the Victimization Psychological Functioning portion of the model. Finally, simple slope analyses were run for significant moderators to assess at which levels of the moderator constructs the relationship between Victimization and Psychological Functioning differed.
38 Chapter 3 Results Descriptive Statistics Table 1 presents the means and standard deviations for each observed variable and latent construct in the pr esent study for the total samp le and by gender. Means for observed variables are unstandardized and m eans for the constructs are composed of composites of variable standardized z-scores Overall according to possible scale score ranges, rates of victimization, aggression, a nd behavioral and psyc hological difficulties were by and large low. In general, boys ha d higher scores on meas ures of victimization, involvement and beliefs in aggression, and beha vioral difficulties as reported by teachers. Girls, however, typically reported higher scores on measures of support at school, orientation toward academic goals, and involvement with family and peers. Principal Component Analysis Prior to analyses to investigate th e hypothesized associ ations, a Principal Components Analysis (PCA) with a Varimax rotation was conducted in order to assess observed variable loadings on the proposed constructs. An eigenvalue analysis (e.g., eigenvalues > 1) suggested that an eight-fact or solution was ideal, with goal-orientation, school efficacy, and classroom learning variab les (i.e., Academic Motivation) and adult and peer intervention and supe rvision variables (s chool Climate) loading together on one factor. In order to preserve the theoretic al foundations of the proposed constructs and maximize the merits of the underlying meas urement according to PCA results (Nunnelly
39 & Bernstein, 1994), nine factors were concep tualized and assessed in modeling analyses as follows: Victimization, Psychological Functioning, Academic Motivation, Academic Achievement, Friendship, Prosocial Activi ties and Influences, Aggression, TeacherReported Difficulties, and Climate (see Appendix B for a description of variables and constructs). Academic Motivation and Climate were conceptualized as separate factors those variables that identified Climate loaded more strongly (average loading .69) on the PCA identified factor than Academic Motivation variable s (average loading .40). Researchers have demonstrated over multiple iterations that the appropriate number of components to retain falls between 1/2 and 1/3 the number of obser ved variables (29 for the present study; Zwick & Velicer, 1986). Additionall y, components that contain complex variables with lower loadings (e.g., .40) and unique variable s with high loadings (e.g., .70) affect the decision rules for determ ining the number of components to retain; therefore, the difference in average loadi ngs between Motivation and Climate variables on the identified component suggests that two underlying constructs may be present (Zwick & Velicer, 1986). Overall, structural equation modeling is a technique that is robust to measurement and rater error in obser ved variables and will, therefore, correct for deviations in the relationships between indicator variables and latent constructs (DeShon, 1998). Correlation Analyses Pearson Product-Moment correlations were conducted between each indicator variable (Table 2) and between latent constr ucts composed of the mean of standardized (z-scores) variable scores (Table 3). Examination of correlations found that Victimization had limited association with academic variables ( r s = -.02 to -.14).
40 However, Psychological Functioning variable s were generally si gnificantly correlated with Motivation measures ( r s = -.01 to -.67), which were, in turn, significantly associated with Achievement variables ( r s = -.02 to .53). This patter n of correlations was evident for both boys and girls. Structural Equation Modeling (SEM) Defining Latent Constructs The latent constructs for the presen t study are as follows: Victimization, Psychological Functioning, Academic Motiva tion, Academic Achievement, Friendship, Prosocial Activities and Influence, Aggr ession, Teacher-Reported Difficulties, and Climate. Latent constructs are factors that are not able to be directly measured or observed, therefore, they must be defined by variables that can act as indicators for the constructs. Indicator variables are directly observed a nd measured variables that load onto latent constructs in a si milar way that items would load onto specific factors in the confirmatory factors analysis method (see A ppendices A and B for a detailed account of the independent and dependent constructs a nd their indicator vari ables and underlying items with principal component factor loadings). Identifying Parameters Model parameters are aspects of the proposed model that are unknown prior to analysis. The parameters ar e characteristics of the samp le population related to the distribution of the variables in the model. They are estimat ed, typically from the sample correlation and/or covariance matrices specifi ed by statistical programming methods. Model parameters in SEM are similar to parameters estimated in regression analyses,
41 such as the standard error of estimate and regression weights. The following are parameters in each of the hypothesized models: 1) The variances of each independent variable are model parameters. All residual error terms, whether assigned to observed or latent constructs or unobserved because they cannot be measured, are considered inde pendent variables an d, therefore, model parameters. Residual error is variation in the observed variables due to measurement error or variance that remains unexplained by the observed variab le loadings on each latent construct. The unexplained vari ance is the amount of indicator variance unshared with the other indicator s defining a latent construct. 2) The covariances between independent va riables are model parameters, unless otherwise stated by theory that some are equal to zero or another constraint. 3) The factor loadings that are attached to th e latent constructs a nd their indicators are model parameters, unless otherwise hypothesized. 4) All regression coefficients between observed variables or latent constructs are model parameters. The regression coefficients ar e represented by pathways that originate from some latent construc ts and end at others. 5) The variances and covariances between de pendent variables and the covariances between dependent and independent va riables are never considered model parameters. This is because these varian ces and covariances are explained by other estimated model parameters. 6) The metric, or scaling, for each latent constr uct needs to be set. Unlike the individual indicator variables, there is no natural metric that unde rlies the constructs. The purpose of the metric is to standardize i ndicator variables that may otherwise have
42 very different distributions and variances in order to compute a construct score. For the present study, the pathway for each indica tor that had the strongest association with its respective underlying construct, according to the principal components analysis, was fixed to 1.0 in order to st andardize the construc t scale. Fixing an indicator pathway to 1.0 is conventiona l practice in SEM (Raykov & Marcoulides, 2000). In SEM, there are three t ypes of model parameters that are of interest free, fixed, and constrained. The parameters determ ined by the previously discussed rules are considered free parameters. These are esti mated by the SEM statistical program. Fixed parameters have their value set to a specific constant and do not change their value when the model is tested against the sample data. In the present study, each construct variance set to one is a fixed parameter. Comparing Model and Observed Data Matrices Statistical SEM programs attempt to create linear combinations of every available variable indicated in a proposed model, which would determine every element of a variance-covariance matrix. This symmetric ma trix is referred to as the reproduced, or model-implied, covariance matrix and can be denoted by Each element of is a function of the model parameters, which each has a numeric counterpart originating from the observed sample covariance matrix S When S and are set equal to each other, the SEM procedure attempts to solve a system of equations, with model parameters acting as unknown variables, to determine the fit of the proposed model with the observed data. If the difference between S and is small, then it is appropriate to assume that the proposed model fits the actual sample data well. If the differen ce is large, then the
43 proposed model does not fit ade quately with the observed data There are two reasons for inconsistencies between the model and the data: 1) the proposed model may not be adequate enough to explain relationships among the observed variables or, 2) the observed data is not good in some sense. In order to assess how good the proposed model is, it is important to assess the distance between S and by subtracting the two matrices from each other, thereby creating a separate matrix of difference values. Solving the matrix of difference values ta kes into account the m odel parameters and elements of the observed variances and cova riances and can be referred to as a fit function, F If F equals zero, the S and matrices are identical. Methods of Estimating Parameters There are four main methods fo r measuring the fit between the S and matrices. The unweighted least squares (ULS) method us es the simple unweighted sum of squared differences between the corre sponding elements of the S and matrices as a fit function. ULS is typically employed when similar scales were used to measure variables analyzed in the model. The maximum likelihood (M L) and generalized least squares (GLS) methods are used when the data is normally distributed, however, the ML method can be used with some deviations from multivar iate normality. Maximum likelihood procedures determine estimates for the model parameters that increase the lik elihood of observing the analyzed data if it were to be collected from the same population again. This is done by scanning all possible numeric model paramete rs and selecting those that minimize the fit function, F Additionally, maximum likelihood is an unbiased estimation for samples with missing data (Wothke, 2000). For seriou s deviations from normality, the weighted least squares, or asymptotically distribu tion free (ADF), method can be used if the
44 observed sample size is large. Another pot ential solution for nonnormal data is to use a transformation technique with the raw data (e.g., squaring data points, square root transformations, reciprocal transformations, and logarithmic transformations). Each parameter estimation results in consistent estimates. Additionally, the ML, GLS, and ADF methods ensure that the estimates matc h population parameter values as the sample size increases. For the present study, the ML proc edure will be used as it is effective in dealing with missing data with deviations from normality. Parameter estimates are determined thr ough an iterative procedure, in which the statistical SEM program starts with initial estimates and continues to derive estimates over and over and terminates at the final st ep when the resulting fit function changes by a very small amount. The parameters in this last step are considered the final solution values and represent the required parameter estimates. The only way these parameters are meaningful is if the iterative process te rminates at a final solution. If termination does not occur, it is possible that the proposed model is inadequate for the observed data or may contain unidentified parameters (parameters in which there are not enough empirical data to provide a unique estimate). Generally, models that contain unidentified parameters are not reliable a model must be fully identified in order to compute an adequate estimation of fit. In addition, a model is considered identified when the number of equations to be solved in the comparison between the S and matrices is greater than the number of unknown elements. This condi tion can be determined by counting the number of model parameters and subtracting this value fr om the number of nonredundant elements in the sample matrix S (i.e., p(p+1)/2, where p = the number of observed
45 variables). The difference is labeled the degrees of freedom of the proposed model and if it is positive, then the model is identified. The final converged solution provides a measure of sampling variability, or standard error, for each parameter estimate. The magnitude of the standard error is an indication of how stable the parameter estima te would be if repeated samplings were conducted. The standard errors are then us ed to compute t-values and evaluated for significance. For example, if a free parameters t-value is greater than +2 or less than -2, then the parameter is significant and is consid ered distinct from the null in the population. The parameter estimates must also have the hypothesized direction and magnitude to consider the model f it for evaluation. Evaluating Goodness of Model Fit The model goodness-of-fit can be estimated using the inferential statistic chisquare (T = (N -1) F where N is the sample size and F is the computed minimal value of the fit function for the parameter estimation pr ocedure utilized, ML, GLS, or ADF). The SEM statistical program will compare the chisquare T value in relation to the models degrees of freedom and produce a corresponding p value for significance determination. The model is considered appropriate for data estimation if the resulting p value is greater than the preset significance level, typically p = .05. While the chi-square index is generally used most frequentl y, there are tendencies for T and p values to become biased based on sample size. Therefore, it is prude nt to examine other goodness-of-fit indices to fully evaluate model fit. Another plausible i ndex that can be examined is the root mean square error of approximation (RMSEA). It has been argued that an RMSEA value less than .05 is indicative of a model that ha s appropriate fit with the data (Raykov &
46 Marcoulides, 2000). The RMSEA is also not sample-dependent, a relative strength compared to the chi-square index. Both methods will be utilized to assess goodness-of-fit for the present proposed model. For th e present study, both the reduced and full mediated models were identified, with degr ees of freedom of 37 and 36, respectfully. Dealing with Missing Data Current literature suggests that the best, most unbiased method for working with missing data in structural equation mode ling is the maximum likelihood estimation (Wothke, 2000). Traditional methods, such as mean-imputation and listwise and pairwise deletion, typically provide ine fficient estimates for missing values. For example, listwise deletion drops all cases with missing data fr om computations and equations are applied only to those cases with complete data acro ss all variables. This process discards a significant amount of available da ta. Pairwise deletion, while le ss restrictive than listwise deletion, also discards a considerable amount of observed data. Pairwise deletion computes estimates for each variable using completed data, but will not provide estimates for cases in which there is missing data on one or more variables of interest. This method uses more of the observed data than li stwise deletion, but imposes statistical complications when each variable analysis can depend on different sample sizes based on missing data. Analyses are essentially run on different portions of the observed data. Mean imputation involves replacing missing vari able data with the mean value of the same variable. This method attempts to comp lete the raw dataset, although estimates are typically negatively biased, meaning that es timates can be systematically larger or smaller than those calculated through listwise or pairwise deletion. Alternately, the fullinformation maximum likelihood (FIML) method uses all of the information that is
47 available in the raw dataset along with info rmation about missing data points based on the information available from the observe d data. The FIML technique is based on theory and maximizes the likelihood of the proposed model fit given what is available with the observed data. This technique w ill be used for the present study because it provides less biased estimates compared w ith listwise and pairwise deletion and mean imputation and uses all available data points, rather than discarding information that does not meet analysis criteria (Wothke, 2000). Identifying Significant Pathways If the proposed model is adequate to explain the observed data, then the significance of individual hypothe sized pathways can be evaluated. The weights of the linear equations computed between latent constructs will be examined for significance of magnitude and direction of a ssociation in the same degree as beta weights would be examined in regression analyses. Tests for Mediation Hypotheses 1 and 2. The Lisrel 8.7 (Joreskog & So rbom, 2004) statistical structural equation modeling program was us ed to evaluate the hypothesized mediated model: Victimization Psychological Functioning Academic Motivation Academic Achievement. In order for a variable to be considered a mediator, four conditions must be present: 1) the independe nt predictor variable (i.e ., Victimization) must be significantly associated with the proposed mediator variable (i.e., Psychological Functioning and Academic Motivation), 2) the predictor must be significantly associated with the dependent variable (i.e., Academic Achievement), 3) the mediator must be significantly associated with th e dependent variable, and 4) th e effect of the predictor on
48 the dependent variable is decreased afte r accounting for the mediator (Baron & Kenny, 1986; Holmbeck, 1997). In order to satisfy a ll four conditions in structural equation modeling, one would first examine the signifi cance of the direct pathway between the predictor and dependent variable. Once th at fit has been satisfied, the predictor mediator dependent variable model is tested and the predictor mediator and mediator dependent variable pathways are examined. Each pathway should be significant in the hypothesized direction. In th e final step, one woul d assess the fit of a full model with the direct predictor to depe ndent variable pathway, and then examine the fit of a model with the reduced predictor mediator dependent variable path. If the full model with the direct pathway does not significantly improve fit over the reduced model without the direct path, th en there is a mediational eff ect. It is important to note that due to the cross-secti onal nature of the data, causal inferences cannot be made among the constructs examined in the present model. Using a Maximum Likelihood fit functi on, the proposed full model (with the direct Victimization to Academic Achievement path) Victimization Psychological Functioning Academic Motivation Academic Achievement was found to adequately fit the data from the total sample of 327, 2 (36, N = 327) = 48.32, p = .08, Root Mean Square Error of Approximation ( RMSEA ) = .032, p = .90. The reduced model without the direct Victimization Academic Achievement path was also of adequate fit, 2 (37, N = 327) = 49.11, p = .09, RMSEA = .032, p = .91. The addition of the direct pathway did not significantly improve model fit ( p >.10 for 2 difference .79), ther efore, it appears that Psychological Functioning and Academ ic Motivation are reasonable mediating factors for the relationship between Victimization and Academic Achievement.
49 Upon examination of parameter estimates in the full model (see Figure 1 for path coefficients), the level of Victimization is significantly relate d to poor Psychological Functioning; maximum likelihood estimate ( mle ) = 14.14, R2 = .30, t (326) = 5.37, p < .05. The pathway between poor Psychological Functioning and Academic Motivation was also significant; mle = -.011, R2 = .11, t (326) = -2.59, p < .05. Additionally, the path between Academic Motivation and Acad emic Achievement was significant; mle = 141.22, R2 = .025, t (326) = 2.10, p < .05; although, this effect is small. However, the pathway between Victimization and Academ ic Achievement was not significant; mle = 72.37, R2 = .007, t (326) = -1.23, p > .05. This was the case as well when this pathway was examined initially in a model that did not include the mediators; mle = -75.18, R2 = .002, t (326) = -1.19, p > .05. That step violated th e second condition for mediation (Baron & Kenny, 1986; Holmbeck, 1997) and s uggests that Psychological Functioning and Academic Motivation may not actually ac t as mediators in this design. Rather, Victimization may have an indirect effect on Academic Achievement via Psychological Functioning and Academic Motivation. The indirect effect was confirmed by examination of the reduced model that does not account for the di rect pathway between Victimization and Academic Achievement. In this model, the association between Victimization and Psychologica l Functioning was significant; mle = 14.10, R2 = .31, t (326) = 5.37, p < .05. As in the full model, the pathways between Psychological Functioning and Academic Motivation ( mle = -.011, R2 = .11, t (326) = -2.60, p < .05) and between Academic Motivati on and Academic Achievement ( mle = 146.52, R2 = .022, t (326) = 2.23, p < .05) were significant. Again, this effect size was small, suggesting that the relationship between motiv ation variables and achievement variables
50 may differ by student experien ces of victimization. When examined further, it was found that the correlation between Motiva tion and Achievement was much lower for those students who were more severely victimized (incidents occur two times a month or more; r = .019) than for those who were infreque ntly victimized (incidents occur less than once a month; r = .358). The relationship for the sample as a whole was also much higher ( r = .317). Examination of the correlations among standardized constructs (Table 3) also provides further evidence of an indirect rather than me diated effect. Violations of the Baron & Kenny (1986) mediat or conditions 2 and 3 were indicated in the simple associations between Victimiza tion and Academic Achievement ( r = -.06, ns ) and Psychological Functioning and Academic Achievement ( r = -.093, ns ). Although the proposed model does not appear to represent a mediated effect between Victimization and Academic Achievem ent, simple correlations suggest that the relationship between Victimization and A cademic Motivation ma y be mediated by Psychological Functioning (see Figure 2). The pathway between Victimization and Academic Motivation was significant when in itially examined alone without the other model constructs, mle = -0.36, R2 = .008, t (326) = -2.79, p < .05, satisfying condition 2 in Baron & Kennys (1986) mediation criteria The full model (with the direct Victimization to Academic Motivation path) Victimization Psychological Functioning Academic Motivation Academic Achievement was found to adequately fit the data from the total sample of 327, 2 (36, N = 327) = 45.58, p = .16, RMSEA = .029, p = .95. Although the addition of this path way only marginally improved the fit of the model compared with the previously examined Victimization Psychological Functioning Academic Motivation Academic Achievement reduced pathway ( p =
51 .05 for 2 difference 3.53), the pathway between Vi ctimization and Academic Motivation was reduced to nonsignificance; mle = -0.046, R2 = .004, t (326) = -0.94, p > .05. These results suggest that Psychol ogical Functioning is an ad equate mediator between Victimization and Academic Motivation, which has a subsequent dir ect relationship with Achievement. In order for the proposed models to fit th e data and to take into account shared variance among construct variables, the fo llowing covariances were specified and estimated as free parameters: Anxiety and Goal-Orientation, Grade Point Average (GPA) and FCAT Reading score, Moodiness and Depression, Moodiness and GPA, Learning Difficulties and GPA, Learni ng and Moodiness, School E fficacy and Depression, GPA and Depression, Learning and FCAT Readi ng, Learning and FCAT Math scores, GPA and Efficacy, and Moodiness and Goal-Orientation. Multiple Regression Analyses Tests for Moderation Multiple regression analyses, with a Bonfe rroni correction to control for familywise error, were conducted to examine hypothe sized moderator effects. In order to assess the same constructs used in the structural equation m odeling and to avoid problems of multicollinearity be tween main effects and interaction terms, each observed variable was centrally standa rdized (z-scores) and aggreg ated (mean scores) in their respective constructs (s ee Aiken & West, 1991, and Holmbeck, 1997, for more information). A moderator effect is defined as an interaction between the predictor (e.g., Victimization) and moderator (e.g., Friendship) that is significantly associated with the dependent variable (e.g., Psychological Functi oning) once the varian ce of the predictor
52 and moderator main effects have been accounted for (Baron & Kenny, 1986; Holmbeck, 1997; 2002). Hypothesis 3. It was expected that the relationship between Victimization and Psychological Functioning would be modera ted by Friendship, as measured by the number and availability of good friends at sc hool. According to hierarchical multiple regression analyses (in a procedure specifi ed in Holmbeck, 1997), the addition of the Friendship construct did not contribute a main effect in predicting Psychological Functioning (see Table 4 for R2 after each step and variable beta weights). The interaction of Victimization and Friendship also did not account for a significant portion of variance in Psychol ogical Functioning ( = .093, t (323) = 1.75, p = .08), indicating that having close friends and quality time to spend with them at school does not moderate the relationship betw een level of victimization a nd psychological difficulties. Hypothesis 4. It was expected that Pros ocial Activities and Influences, as measured by engagement in positive activit ies with family and friends who do not condone problem behaviors, would moderate the relationship between Victimization and Psychological Functioning. Both Victimizati on and Prosocial Activities and Influences (see Table 4) were significantly asso ciated with Psychological Functioning; = .348, t (323) = 6.91, p < .05 and = -.228, t (323) = -4.48, p < .05, respectively. However, the interaction of Victimization and Prosocial Activities and Influences was only marginally related to Psychological Functioning, = .097, t (323) = 1.92, p = .056, suggesting that engagement in positive activities does not ad equately moderate th e association between level of victimization and psychological problems.
53 Hypothesis 5. It was hypothesized in the present study that engagement and belief in aggressive behaviors w ould moderate the associati on between Victimization and Psychological Functioning. Both Victimization ( = .336; t (323) = 6.52, p < .05) and Aggression ( = .273; t (323) = 5.19, p < .05) were significan tly associated with Psychological Functioning (Table 4). Additiona lly, the interaction of Victimization and Aggression was significant ( = -.118; t (323) = -2.26, p = .024). However, with a Bonferroni correction for multiple tests of significance (5 moderator analyses; p = .05/5 = .01), the interaction is reduced to nonsigni ficance. These findings partly supported Aggression as a moderator; although, caution should be exercised in interpretation because the moderator effects disappeared once the significance leve l was corrected and the magnitude of the effect size is relatively small. In general, results suggested that victimized students who also engage in and advocate aggressive be haviors have better psychological functioning. As levels of concurrent victimization and aggression increase, psychological difficulties decrease. Hypothesis 6. It was expected that Teacher -Reported Difficulties, such as disruptive behavior and poor peer relationships at school, would moderate the relationship between Victimization and Psychological Functioning. While both Victimization ( = .384; t (323) = 6.84, p < .05) and Teacher-Reported Difficulties ( = .130; t (323) = 2.50, p < .05) were significant in expl aining variance in Psychological Functioning in this final step, the interact ion between the two constructs was not ( = .074; t (323) = -1.35, p = .18; Table 4). Exhibiting dis obedient behavior and experiencing bullying at school, as reported by their teache rs, does not appear to have the proposed moderator effects on victimizati on and psychological difficulties.
54 Hypothesis 7. It was proposed that positive school Climate factors, such as intervention against rule-breaking behavior and supervision by adults, would moderate the relationship between Victimizatio n and Psychological Functioning. Both Victimization ( = .399; t (323) = 7.51, p < .05) and Climate ( = -.104; t (323) = -2.01, p < .05; ns after the Bonferroni correct ion) were significant in explaining the variance in Psychological Functioning (Table 4). As exp ected, even with the Bonferroni correction ( p = .01) for multiple tests, the interaction of Victimization and Climate was also significant ( = .141; t (323) = 2.67, p = .008), indicating that having a school climate characterized by structure, intervention, and supervision moderates the magnitude of the association between Victimization and Psychological Functioning. Again, caution should be exercised in interpre ting these results. While signifi cant, the effect size of the moderator relationship is small indicating that the clinical relevance of such a relationship is limited. Gender Analyses Hypothesis 8. It was hypothesized that certain moderator effects would differ by gender. Specifically, it was expected that social support factors, such as Friendship, Prosocial Activities and Influe nces, and school Climate, would be more important in explaining the relationship between Victim ization and Psychological Functioning for girls than for boys. Alternatively, Aggre ssion and Teacher-Reporte d Difficulties were expected to relate more strongly for boys. For boys, the only significant moderated e ffect was Climate. The third step of the overall model with Victimization, Clim ate, and the interaction of both was significant, F (3, 140) = 13.56, p < .05 (see Table 5). The Victimization x Climate
55 interaction was also significant ( = .275; t (141) = 3.28, p = .001) even after the Bonferroni p = .01 correction, suggesting that inte rvention and supervision at school moderates the level of psyc hological difficulties for victimized boys. There were no significant moderator effects for girls (see Table 6). Slope Analyses for Significant Moderators In order to examine the nature of the moderator effects on Psychological Functioning, slope analyses for each signi ficant moderator (Aggression and Climate) were conducted according to the procedure discussed by Aiken and West (1991) and Holmbeck (2002). Previously discussed regr ession analyses tested for the presence of moderation to explain the conditions under which Victim ization is related to Psychological Functioning. However, a slope an alysis is needed to further explain at which levels of the predictor and modera tor the dependent construct will vary (Holmbeck, 2002). For the present study, it is impor tant to note that z-scores were used to compute construct parameters; therefore, it is statistically possi ble to have scores below zero when, conceptually, negativ e values would be impossible. Aggression. According to a slope analysis fo r the moderator Aggression, results indicate that at both high (+ 1 SD a bove Aggression mean) and low (1 SD below Aggression mean) levels of Aggression, th e relationship between Victimization and Psychological Functioning is significant, b = .362; t (323) = 6.16, p = .000 for high levels and b = .211; t (323) = 4.09, p = .000 for low levels (see Figure 3). Note that scores above zero indicate poorer Psychological Functioning. What this suggests is that students who are highly victimized and engage in hi gher levels of Aggression experience fewer psychological difficulties than highly victim ized students who do not believe and engage
56 in such behaviors. Students with low leve ls of both Victimization and Aggression have better Psychological Functioning than other students. Based on previous moderator analyses, caution should be ex ercised in interpreting these results since the effect size magnitude for Aggression is small. Climate. The relationship between Victimi zation and Psychological Functioning varies at different levels of school Climate. At both high and low levels of Climate, Victimization is significantly associated with changes in psychological difficulties, b = .449; t (323) = 6.60, p = .000 for high levels and b = .232; t (323) = 4.40, p = .000 for low levels. Upon examination of the plotted regre ssion lines (see Figure 4; note that scores above zero indicate poorer Psychological Functi oning), it appears that at low levels of Victimization, psychological difficulties va ry as a function of Climate. Highly victimized students who experience either high or low levels of support and intervention in their schools have greater reported psychol ogical problems than students who are not victimized as frequently or severely. However, those w ho experience lower levels of Victimization and low inte rvention in their schools have considerably poorer psychological functioning than those that experience higher levels of support and intervention. Based on previous moderato r analyses, caution should be exercised in interpreting these results si nce the effect size magnitude for Climate is small. Climate also served as a m oderator for boys, specifically. Again, at both high and low levels of Climate, the relationshi p between Victimization and Psychological Functioning is significant, b = .566; t (323) = 5.60, p = .000 for high levels and b = .209; t (323) = 3.33, p = .001 for low levels (see Figure 5; note that scores above zero indicate poorer Psychological Functioning). These fi ndings suggest that boys who experience
57 high levels of Victimization and high levels of intervention on the part of teachers and peers at school have greater psychological difficulties than highly victimized boys who have lower levels of support against beha vioral misconduct at school. However, boys who are infrequently victimized and experien ce high levels of intervention in their school climates have better psychological functioning than infrequently victimized students who report low levels of Climate.
58 Chapter 4 Discussion The present study sought to examine th e relationship among the experience of victimization at school, psychological difficu lties, and academic outcomes for students in middle school. Specifically, the present study expanded on previous research by proposing that the relationship between victim ization and academic outcomes is mediated by a students psychological functioning. Ev en more explicitly, it was suggested that academic outcomes come in two forms, acad emic motivation processes and academic achievement variables, with academic pro cesses preceding and mediating the association between psychological functioning and achieveme nt. Furthermore, certain behavioral, peer and family related, and school environmen tal factors were expected to moderate the relationship between student-re ported experiences of peer vi ctimization and experienced psychological difficulties, such that positiv e influences were believed to protect victimized adolescent boys and girls from emotional problems and negative experiences were believed to put them at greater risk. The following se ctions discuss the findings of the present study. Mediated Model Through modeling techniques in the present study, it was found that Psychological Functioning mediated the re lationship between Victimization and Academic Motivation, but Acad emic Achievement was only indirectly associated with Victimization by means of its di rect link with Academic Moti vation. Thus, experience of
59 victimization at school was only related to academic outcomes in the present study by way of psychological and motivatio nal properties. Because this pathway is indirect, this suggests that not all students who are vi ctimized will ultimately have troubles academically. In addition, many students with academic motivation problems have poor academic outcomes without having experienced victimization. Victimized students presented with an interesting profile based on the current findings. As the level of self-reported victim ization increased, stude nts experienced more symptoms of depression, anxiety, anger, a nd general moodiness. In addition, students who were victimized and had poor psychol ogical outcomes were less oriented and motivated toward academic goals. However, the association between experience of victimization and academic performance is tenuous according to the present findings, peer harassment is related to academic outcome s via the indirect (but not direct) influence of motivation to achieve. The explanation for this result may be th at victimization at school is recognized by teachers and school ad ministrators as a risk factor for poor emotional functioning and subsequently poor academic goal-orientations and performance. Therefore, students who e xperience victimization may also receive substantially more academic support and pe rform acceptably despite difficulties in motivation related to victimization. In f act, this hypothesized m echanism was supported via a post hoc examination of the difference in Motivation Achievement correlations for students severely victimized versus thos e who reported little to no victimization. The relationship between academic motivation pr ocesses and academic achievement was nonsignificant for the severely victimized students, arguably those who are receiving more scholastic support, while the relationship between academic motivation and
60 academic achievement was strongest for thos e who experience the least victimization, potentially those who are not targeted for academic programming. While it appears that studies have not specifically examined this hypothesis, re searchers have found that in grade school classrooms where teachers address bullying and provide a learning environment that is exceptiona lly achievement-oriented, vict imized students are likely to be more satisfied with school and have bett er academic outcomes (Verkuyten & Thijs, 2002). Previous literature supports several of the findings that emerged in the mediation analyses. Researchers have found that victim ization is not directly related to academic outcomes. Instead, a victimized students psychological functioning is the mechanism through which negative experiences with aggr essive students can have an impact on achievement at school (Juvonen et al., 2000; Kochenderfer & Ladd, 1996a; Wentzel & Caldwell, 1997). However, each of these studies varied in how they assessed academic outcomes, with some reporting processing t ype variables (e.g., orientation toward academic goals; Nansel et al., 2001; Sc hwartz et al., 2001; Wentzel, 1998) as achievement and others reporting standardized test scores or grades as measures of achievement (Juvonen et al., 2000; Schwartz & Gorman, 2003). While past research looked at several types of achievement variable s, each finding added to the larger picture of child outcomes subsequent to victimiza tion. Although not discu ssed as an indirect effect, other researchers have described similar findings in th eir own studies, such that the association between victimization and academic outcomes was not significant unless psychosocial variables were taken into account (Juvonen et al., 2000; Kochenderfer & Ladd, 1996a; Wentzel & Caldwell, 1997). The pr esent study built upon this past research
61 by examining the fit of a model that incorpor ated previously studied academic process and achievement variables in a temporal path way and finding that psychological factors only mediate the associati on between victimization a nd motivation to achieve. Moderators of Psychological F unctioning in the Mediated Model It was expected as well that the relationship between Victimization and Psychological Functioning would be modera ted by a number of factors, including friendships at school, prosoc ial involvement with and infl uences of peers and family, school difficulties reported by teachers, engage ment and beliefs in aggressive behaviors and coping strategies, and the level of support and intervention in the school climate, thereby having either a positive or deleterious influence on academic outcomes. However, only aggressive behaviors and coping beliefs and school climate factors emerged as significant moderators of the Victimization Psychological Functioning relationship. When examin ing moderators by gender, only for victimized boys did increases in reported support and interventi on at school become associated with high levels of psychological dysfunction. For victim ized girls, none of the proposed variables moderated their experiences with depression, anxiety, and anger. Notably, most of the proposed moderator factors did not significantly modify the relationship between victimization and psyc hological outcomes. Alone, the factors Friendship, Prosocial Activities and Influen ces, and Teacher-Reported Difficulties were related to Psychological Functioning for victim ized students. However, as levels of victimization increased, these factors did very little to change th e relationship between student experience of victimi zation and psychological distress. Recent studies have also demonstrated that merely having friendships does little to impact vi ctimization, but that
62 being exposed to aggressive friends may put students at greater risk for being bullied (Hanish, Ryan, Martin, & Fabes, 2005). Thus, friendship, activities, and peer influences may only impact victimization based on more sp ecific qualities that should be explored in future studies. It is also possible that some of these proposed moderator effects may actually act in statistically diffe rent ways outside of the scope of this study; for instance, as mediators, or to moderate relationships at alternative pathways, such as between Psychological Functioning and Academic Motivation. Of those factors that did signifi cantly moderate the Victimization Psychological Functioning pathway, they moderated the relati onship in the direc tion opposite of what was expected. Contrary to research that s uggests victims who bully others have increased emotional difficulties (Haynie et al., 2001; Na nsel et al., 2001), the present findings suggested that beliefs and engagement in aggressive behaviors buffer the negative psychological effects of peer victimization. Studies have recently found that all victims are not as shy and withdrawn as previously beli eved. In fact, some vi ctims are as able to engage in aggressive behaviors as their bully ing counterparts, partic ularly in response to aggression by others (Kochenderfer & Ladd, 1997; Schwartz, McFadyen-Ketchum, Dodge, Petit, & Bates, 1998; Camodeca, Goossens, Meerum Terwogt, & Schuengel, 2002). Specifically, it is susp ected that victimized student s feel powerless when bullied and their desire to retaliate in aggressive ways is fueled by frustration and anger over a sense of helplessness (Camodeca & Goossens 2005a). Although victims may be readily aware of assertive and prosocia l problem-solving strategies fo r dealing with negative peer interactions, research has shown that st udents typically choose retaliatory methods (Camodeca & Goossens, 2005a; Futrell, 1996). Because a considerable proportion of
63 victims experience frustration as a result of bullying, it is plausible to suspect that using counter-aggression against bullies is the pref erred means for alleviating the harassment and frustration and circumvent any additiona l negative emotions related to the bullying experience. However, this effect may only be temporary as research has documented that bullies may not be deterred by retaliation a nd victims who use aggressive coping tend to have high levels of anger and emotiona l distress (Perry, Williard, & Perry, 1990; Camodeca & Goossens, 2005b). All in all, the effect si ze for Aggression effects was small, suggesting that aggression may only be a good coping mechanis m for some victims and indicating the need for caution in interpreting these results. Further research shoul d try to identify those students who have better outcomes as a result of using aggressive coping to victimization. To our knowledge, no studies have demonstrated the protective function of aggressive coping behaviors and beliefs for child and adolescent emotionality; however, research with adults suggests that aggr essive and antisocial coping in situations of high stress (as may be the case with peer victimization) act ually prevents subseque nt angry and anxious feelings (Monnier, Hobfall, Dunahoo, Hu lsizer, & Johnson, 1998). Similarly, in situations where family honor is at stake, r eacting in an aggressive manner is normative and protective of self-esteem in some cu ltures (Mosquera, Manst ead, & Fischer, 2002). Again divergent from hypothesized asso ciations, supervisi on and intervention against behavioral misconduct at school did not protect highly victimized students against poor psychological outcomes. It is possible that students ma y feel more anxious at the prospect of seeking the help of adults and peers because tattling may incite further victimization (Pepler, Craig, Ziegler, & Ch arach, 1994). Yet, students who experienced
64 minimal victimization and high levels of support and intervention in their school environments fell well below the mean on m easures of depression, anxiety, moodiness, and anger. These findings were espe cially salient for middle school boys. Psychologically, highly victimized boys fare d worse when adults and peers actively intervened in their schools through discussi ons and dissemination of school anti-bullying policies, while those who were minimally vi ctimized responded better to high levels of support and intervention. This finding is re markable to consider. Upon examination of the items that speak to school climate issu es, many address students understanding of school rules against aggression and perceptions of positive adult and peer influences. None of the items address specific structured intervention programs, those of which may be effective in improving the psychological functioning of students, particularly boys, who are highly victimized. Studies on effec tive school-based interventions suggest that school-wide support of intervention efforts, rather than individual and inconsistent actions taken by teachers or peers, is crucia l to the success of pr ogramming (Olweus et al., 1999; Vernberg & Gamm, 2003). The fact is that the belief victims are somehow deserving of bullying or that peer harassmen t is a normal part of growing up is still prevalent in many school environments (Mon tada & Lerner, 1998; Vernberg & Gamm, 2003). Some school personnel also see their role as purely educational, even though schools are currently being cal led on to address a number of student psychosocial problems (Astor, Pitner, & Duncan, 1996). The only way that in tervention programming can be effective is for school administrators and teachers to trul y believe that schools must do what is necessary to provide safe learning environments and act accordingly together (Farrington, 1993; Morr ison, Furlong, & Morrison, 1997).
65 Few intervention techniques take the cri tical step of incorporating student perspectives and suggestions (Camodeca & Goo ssens, 2005a). This could be considered a significant failure on the part of interven tions against bullying gi ven that researchers have discovered some students are less likely to seek the help of others and are more comfortable retaliating or trying to problem-sol ve themselves (Lightner, Bollmer, Harris, Milich, & Scambler, 2000; Mooney, Creeser, & Blatchford, 1991; Salmivalli, Kurhunen, & Lagerspetz, 1996; Shapiro, Baumeister, & Kessler, 1991; Smith, Shu, & Madsen, 2001; Warm, 1997). For example, one research er demonstrated that students believed boys experience increased teasing by cla ssmates when provided a peer-support intervention to protect agains t victimization, potentially decreasing the likelihood that male victims would be satisfied with this type of intervention (C owie, 2000). For those who did not experience a considerable amount of peer victimization in the present study, the mere propagation of rules against misconduct, efforts on the part of school personnel and students to put a stop to bullying, and s upervision by adults at school contributed to positive school environments and healthier student psychological functioning. The disparities in these findings related to sc hool climate factors highlight the need to examine differing groups of victimized student s, those who are victimized infrequently versus frequently and boys versus girls. A dditionally, these findings suggest that efforts on the part of teachers and peers to put a stop to bullying that are not part of an empirically tested intervention program may actually backfire and contribute to the stigmatization of victimized students a nd subsequent psychosocial dysfunction. Markedly, victimized girls did not ha ve moderating factors that improved their psychological functioning. Because the type of bullying girls experience tends to be
66 indirect and based on manipula tion of friendships (i.e., rela tional aggression), it may be harder for friendships and peer activities to consistently serve as a source of positive support for victimized girls (e.g., Horowitz et al., 1995). Theref ore, it is comprehensible that factors such as Friendship and Pros ocial Activities and Influences would not significantly improve the psychosocial circum stances of victimized girls in middle school. A growing body of literature has begun to identify aggressive students, or bullies, as some of the most popular p eers at school (Hawley & Vaughan, 2003; Rose, Swenson, & Waller, 2004). This is especially true for female a dolescent aggressors (Rose et al., 2004). With this dynamic occurri ng in schools, it is qui te possible that many peers and school staff would not categorize rela tional forms of aggression as bullying and therefore not intervene on behalf of victims. In fact, the case may be that aggressive students, whether boys or girl s, are somewhat recognized as leaders in their peer groups and that any intervention by adults and students could be to the detriment of a victim who must continue to co-exist with their peer group at school. Implications of the Present Study The present study has important implicati ons for understanding peer victimization at school and its association w ith psychosocial and environmenta l factors. Simply stated, not all victimized students will have probl ems academically on exams and with class grades. Rather, adolescent boys and girls w ho are harassed, teased, threatened, and/or attacked at school by classmates and have increased depressive, anxious, or angry symptomatology will more likely have difficu lties learning and being oriented toward educational goals, which has a negative influence on academic achievement. The relationship between motivation toward acad emic goals and achievement at school
67 appears to be significant for students who are less severely or infrequently victimized. Speculatively, those who are highly victimized may more readily come to the attention if teachers and school staff and, therefore, receive extra assistance to ensure their academic performance does not suffer, despite the possi bility of co-occurring psychological and motivational difficulties. Given the curre nt emphasis on achievement and academic performance in American public schools (i.e., No Child Left Behind Act, 2001), this is a highly probable hypothesis that needs to be examined in future research. Moreover, involvement in positive activiti es with friends and family, limited engagement in aggressive behaviors in genera l and to cope with this victimization, and supervision and interven tion of adults and students at sc hool against bullying, particularly for boys, do not help alleviate some of th e emotional and psychological difficulties experienced by victimized students. It is suggested from the present study that many students feel that aggressive techniques are appropriate in response to bullying and anger, probably in lieu of seeking the support of stude nts, teachers, or school policies. This is especially true for victimized boys, who may be more sensitive to the implications of having teachers or school staff drawing atten tion to their difficulties with peers. Although these effects are small, future re search should furthe r investigate these associations between student e xperiences of victimization and their school environments. School-based prevention and intervention effo rts should be modified to address these relationships. Specifically, more anti -bullying programming should incorporate continuous feedback from students to be certa in that the strategies taught to prevent victimization and ramifications thereof are effective and feasible to implement for students. Training for teachers, counselors and school administrators should focus on
68 preventing adjustment and academic difficultie s at school while making certain not to undermine student self-esteem and confiden ce to appropriately problem-solve peer relationships on their own. Recent findings s uggest the need for an ecological approach to addressing the deleterious effects of victim ization such that child experiences at home should inform experiences at school and vi ce versa (Henrich, Schwab-Stone, Fanti, Jones, Ruchkin, 2004). Adolescents likely firs t learn skills to c ope with stressful situations from family members. It is impor tant to have a strong parent-school personnel communication to ensure that the practices modeled for children at home are appropriate for problem-solving at school. Additionally, researchers should be cauti ous about the conclusions drawn in the present study. Some of the findi ngs may be tied to the level of victimization that students experienced in the study. Associations among constructs could be very different depending on the level of victimization, and perhaps types of vict imization. Future research should examine various levels and types of victimization to guarantee that the relationships in the presen t study are generalizable. Limitations of the Present Study Implications of the present study must be considered in light of several limitations. Overall, the study dealt with cros s-sectional data, which limits the ability to make causal conclusions about construct rela tionships. In particular, the lack of significant mediator and moderator findings ma y be due to a number of assessment and statistical issues. Overall, the majority of the sample was victimized infrequently, indicating that experiences of overt and indirect aggressi on typically occu rred only once or twice in the last month. According to re searchers such as Olwe us (1996), the level of
69 reported incidents identified as victimization in the present study would not necessarily constitute true victimizati on status. As was found in th e present study, frequency of victimization had an impact on the rela tionship between academic motivation and achievement. Previous studies reported that children who are chr onically and severely victimized are at the greatest risk for in creased psychological difficulties compared with children who do not experience victimization at such levels (Olweus, 1993; Pynoos & Nader, 1988; Singer et al., 1995). Additionall y, the Victimization construct was assessed via student-report, which has implicati ons for how victimization by peers is conceptualized in the present study. While th e findings are likely generalizable to other samples, what was actually examined was student perceptions that they were attacked, threatened, or harassed by their classmates. Much research has been devoted to the inspection of students involved in problem p eer situations who viewed themselves as victimized and acted aggressively, but were ac tually subject to a hostile attribution bias that altered their in terpretation of interpersonal re lationships (Crick & Dodge, 1994; Dodge & Crick, 1990). Peer, teacher, and even parent report may indicate very different levels of victimization among students. Therefor e, it is important for assessments of such behaviors to use a combination of methods a nd informants, such as self-report and peer nomination, to obtain a more accurate pi cture of student difficulties (Ladd & Kochenderfer-Ladd, 2002). Furthermore, the Olweus self-report measure requires students to recall the frequency with which th ey have been bullied over several months during the academic year. Students may ha ve trouble accurately remembering bullying incidents, as well as little desire to label th eir difficulties with peers as victimization,
70 despite being provided with a detailed definiti on that indicates which behaviors constitute bullying. Although the sample of students used in the analyses (N = 327) was selected randomly from the middle school population, it is considerably smaller than the total sample from the entire 2002-2003 district -wide assessment (N = 4272). Because the assessment typically spanned two class period s, some classrooms and teachers did not have the resources to participate. Moreover, the sample consisted primarily of Caucasian students in rural and suburban communities, raising questions about the generalizability of the present findings. In order to examin e hypothesized relations hips in structural equation modeling, it was necessary to have multiple measures and indicators of each factor, which significantly limite d which participants and data were included. While the remaining sample that had completed data on each measure was adequate enough to assess the mediator hypotheses in structur al equation modeling, it was limited in its power to assess moderator and gender differe nces in which more variables would be included and the models would be examined across samples split by gender. Therefore, a second step of hierarchical regression analyses was conduc ted. While this two-step method is sufficient to test moderator effect s, it is less preferable in comparison with structural equation modeling, which has the ability to account for multiple variances simultaneously rather than serially in a number of step-wise re gression models. Despite compelling theoretical findings, some researchers have been concerned about the minimized power multiple regressi on may have to detect moderator effects (McClelland & Judd, 1993). This is particularly true of nonexperimental field research designs in which distributions of some m oderator and predictor variables and their
71 combined interaction residual variance in the model are limited. In field studies, researchers have little contro l over the distributions of obs erved variables, which can be skewed or restricted in some way. As a consequence, the imp act of the moderator variance is likely to be small since the inte raction residual variance, which tends to be lower in these cases, plays a cr itical role in calculating eff ect size. Likewise, undetected and unevaluated covariances may also be responsible for lowe ring the interaction residual. Because these modera tor effects are typically diffi cult to detect, even minimal effect sizes are worthy of further exploration. As indicators of the proposed constructs, some of the measures included were not psychometrically strong. Many measures were simply indices aggregated from a series of items on global surveys that assessed certain behaviors of interest in the present study. It is possible that several of these indices, while assessing sp ecific variables, were not sufficiently reliable or valid in measuring the constructs and th eir associations as proposed. For example, the Academic Motiv ation construct was si gnificantly, but not strongly, related to Academic Achievement. Ac cording to prior resear ch, it is likely that the indicators included in the Motivation construct were not equally predictive of achievement outcomes (Ames, 1992). Master y goals (i.e., School Efficacy) are typically better indicators of achievement than Performance goals (i.e., Learning Difficulties; Ames, 1992; Miller & Byrnes, 2001). Future research could expand upon the present model by examining the predictive differences between types of goal-orientation and motivation variables. Additionally, some variables, such as frequency of discipline referrals or behavioral misconduct, may be appropriately skewed because they do not readily occur often with the average student. In an effort to preserve the natural
72 experiences of victimized students at school, the ability to detect moderator effects may have been sacrificed for some constructs. Strengths of the Present Study Notwithstanding some of the concerns a bout the present stud y, there are several strengths. Although there were sample size issues with regard to the moderator analyses, it is notable that for some portion of the investigation enough power existed to examine multiple associations among student and school related constructs using the sophistication of structural equation modeling. Again, struct ural equation modeling has the ability to account for and measure error variances simultaneously that are associated with the measurement of constructs and the va lidity of pathway associations that other statistical procedures cannot. While r oughly 10% of the data was missing from participants in the model, the analyses were likely not significantly impacted given the use of maximum likelihood procedures in stru ctural equation modeling. The ability to measure error and deal well with missing data are considerable stre ngths for structural equation modeling compared with other te chniques used to predict mediational relationships. Additionally, information was collected using multiple informants (student, teacher, and records data) within multiple child domains (individual beliefs and practices, peer relationships, family involve ment, school adjustment, and achievement). Using this rich dataset in which multi-informant and multi-domain information was compiled as indicators of constructs in a pow erful statistical design, several associations between student victimization and school outcome s that were inconsistent in the previous literature were clarified. The prior finding th at victimization is indirectly related to achievement through psychological variable s was confirmed, while the relationship
73 between psychological functioning in vict imized students and specific academic constructs (i.e., motivational process and ach ievement) was revealed. In light of the present findings, it is no longer enough to assume that peer victimization results in poor academic functioning; the mechanism through whic h grades and standardized test scores are affected for victimized students is by way of negative emoti onal experiences and limited motivation and focus on academic goals. Future Directions Given the scope of the present study a nd the previous li terature, several recommendations for future research can be made. Primarily, because this study used cross-sectional data, the conclusions draw n from the findings are limited. Further research should examine the probability of the relationships among peer victimization and academic, psychological, and behavioral mediators and moderators demonstrated here over time and with other developmen tal groups. Studies have found that the importance of some moderators in explaini ng associations between victimization and emotional variables is differential by age (Camodeca & Goossens, 2005a; Hanish et al., 2004). Similarly, gender and culture need to be explored further to understand how victimization and related difficulties may differ between boys and girls and by such factors as ethnicity, community make-up, or economic status. In addition, students experience varying levels, types, and seve rity of victimizatio n. The present study generally represented a sample of students who are modestly victimized, according to specific criteria in the literature (Olweus, 1996). Post hoc analyses suggest that the academic outcomes of students differ by the frequency of the victimization they experience. More research needs to be conduc ted in this area to fully disclose these
74 effects. The low levels of reported victim ization may be a function of the method of assessment, primarily student report. Adoles cents may have more difficulty accurately defining their relationships with peers, whether as a result of misperception or unwillingness to label themselves a victim. Researchers suggest that information should be gathered from multiple individuals, such as students, school personnel, peers, and parents, and through various methods (e.g., self-report, observation, and peer nomination) in order to get the more complete picture of child and adolescent adjustment (Holmbeck et al., 2002; 2003; La dd & Kochenderfer-Ladd, 2002). The moderator variables th at speak to involvement with family members and friends/peers were not signifi cant in explaining the relatio nship between victimization and psychological outcomes in the present study. It is possible that these variables may serve a different function than as modera tors. Further work should re-examine the conceptual and statistical mechanism by wh ich these constructs may be related to victimization, psychological difficulties, and academic outcomes. Because families serve as an important context for adolescent social ization (Parke, 2004), fu ture studies should systematically examine the association of parenting practices, family functioning, and sibling relationships with victimization at school and academic outcomes. Addressing family contexts also provides vital information for the development of intervention programming that incorporates both ho me and school environmental factors and optimizes the functioning of children a nd adolescents (Vernberg & Gamm, 2003). Interestingly, school climate variables, as defined by adult supervision and intervention and peer support against bully ing, had a negative association with the emotional experiences of frequently victimized students. It is im possible to tell from
75 these climate variables whether structured school-wide interventi on programming or an accumulation of individual unstructured res ponses to peer aggression and misconduct are assessed. It is probable that the school climate construct in the present study is tapping into generally well-intentioned, yet ineffective, strategies that thusly have a detrimental impact of the psychological f unctioning of highly victimized st udents. The results related to beliefs about and engagement in aggressi on behaviors and coping provide insight into which intervention techniques st udents find effective, predom inantly retaliation versus support-seeking. Since there is considerable literature demonstrating that child and adolescent victims of peer harassment who ar e aggressive and fight back have poorer emotional outcomes (e.g., Dill, Vernberg, Fonagy, Twemlow, & Gamm, 2004), the unconventional results in the pr esent study must be confirme d with other samples. Taken as a whole, it is evident that victim ization is related to difficulties at school, but not every victimized student will have poor academic performance. Factors such as psychological symptoms, orientation to achieve and school and indivi dual responses to bullying have significant interactions with adolescents peer relati onships and functioning at school. The current findings advan ce the literature forward by presenting the differential experiences that students have at school with their peers and highlighting aspects that need continuing e xploration in order for the unde rstanding of peer aggression and school adjustment to evolve.
76 References Aber, J. L., Jones, S. M., Brown, J. L., Chaundry, N., & Samples, F. (1998). Resolving conflict creatively: Evaluating the deve lopmental effects of a school-based violence prevention program in ne ighborhood and classroom contexts. Development and Psychopathology, 10, 187-213. Achenbach, T. M. (1991). Manual for the Teachers Report Form and 1991 profile. Burlington, VT: University of Verm ont, Department of Psychiatry. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Albee, G. W. (1984). Prologue: A model for cl assifying prevention programs. In J. M. Joffe, G. W. Albee, & L. D. Kelly (Eds.), Readings in primary prevention of psychopathology: Basic concepts (pp. 228-245). Hanover, NH: University Press of New England. Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84 (3), 261-271. Arora, C. M. J. (1996). Defining Bullying. School Psychology International, 17, 317329. Astor, R. A., Pitner, R. O., & Duncan, B. B. (1996). Ecological approaches to mental health consultation with t eachers on issues related to youth and school violence. Journal of Negro Education, 65, 336-355.
77 Austin, A. M., & Draper, D. C. (1984). Th e relationship among peer acceptance, social impact, and academic achievement in middle childhood. American Educational Research Journal, 21 (3), 597-604. Austin, S. & Joseph, S. (1996). Assessment of bully/victim problems in 8 to 11 yearolds. British Journal of Educ ational Psychology, 66, 447-456. Baron, R. M., & Kenny, D. A. (1986). The mode rator-mediator variab le distinction in social psychological research: Conc eptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Barth, J. M., Dunlap, S. T., Dane, H., Lochma n, J. E., & Wells, K. C. (2004). Classroom environment influences on aggression, peer relations, and academic focus. Journal of School Psychology, 42, 115-133. Batsche, G. M. & Knoff, H. M. (1994). Bullies and their victims: Understanding a pervasive problem in the schools. School Psychology Review, 23 (2), 165-174. Belle, D. (1987). Gender differences in the social moderators of stress. In R. C. Barnett, L. Biener, & G. K. Baruch (Eds.), Gender and stress (pp. 257-277). New York: Free Press. Besag, V. E. (1989). Bullies and victims in schools. Milton Keynes: Open University Press. Bickman, L., Lambert, E. W., Karver, M. S ., Andrade, A. R. (1998). Two low-cost measures of child and adolescent f unctioning for services research. Evaluation & Program Planning, 21 (3) 263-275.
78 Bjorkvist, K., Ekman, K., & Lage rspetz, K. (1982). Bullies and victims. Their ego picture, ideal ego picture, and normative ego picture. Scandinavian Journal of Psychology, 23, 307-313. Boivin, M., & Hymel, S. (1997). Peer e xperiences and social self-perceptions: A sequential model. Developmental Psychology, 33 (1), 135-145. Boulton, M. J. (1993). A comparison of adults' and child ren's abilities to distinguish between aggressive and playful fighting in middle school pupils: Implications for playground supervision and behavior management. Educational Studies, 19 (2), 193-203. Boulton, M. J. & Underwood, K. (1992). Bully/victim problems among middle school children. British Journal of Educational Psychology, 62, 73-87. Camodeca, M., & Goossens, F. A. (2005a). Childrens opinions on effective strategies to cope with bullying: The importance of bullying role and perspective. Educational Research, 47 (1), 93-105. Camodeca, M., & Goossens, F. A. (2005b). Aggression, social cognitions, anger, and sadness in bullies and victims. Journal of Child Psychology and Psychiatry, 46 (2), 186-197. Camodeca, M., Goossens, F. A., Meerum Te rwogt, M., & Schuengel, C. (2002). Bullying and victimization among school-age children: Stability and links to proactive and reactive aggression. Social Development, 11 (3), 332-345. Carberry, A., & Handel, P.J. (1980). The use of the AML Scale with a Headstart population: Normative a nd validation studies. American Journal of Community Psychology, 8 (3), 353-363
79 Carney, A. G. & Merrell, K. W. (2001). Bullying in schools: Perspectives on understanding and preventing an international problem. School Psychology International, 22 (3), 364-382. Carver, C. S., Scheier, M. E., & Weintra ub, J. K. (1989). Assessi ng coping strategies: A theoretically ba sed approach. Journal of Personality and Social Psychology, 56, 267-283. Castaneda, A., McCandless, B. R., & Palermo, D. S. (1956). The childrens form of the manifest anxiety scale. Child Development, 27, 317-326. Center for Disease Control and Prevention. ( 2002). Youth Risk Behavior Surveillance United States, 2001. Morbidity and Mortalit y Weekly Report, 51 (Suppl.), SS-4. Chung, T., & Asher, S. R. (1996). Children s goals and strategies in peer conflict situations. Merrill-Palmer Quarterly, 42, 125-147. Clark, L. A., & Watson, D. (1995). Constructi ng Validity: Basic I ssues in Objective Scale Development. Psychological Assessment, 7 (3), 309-313. Cohen, S., & Wills, T. A. (1985). Stress, so cial support, and the buffering hypothesis. Psychological Bulletin, 98, 310-357. Coie, J. D., Dodge, K. A., & Kupersmidt, J. B. (1990). Peer group behavior and social status. In S. R. Asher & J. D. Coie (Eds.), Peer rejection in childhood (pp. 1759). Cambridge, England: Cambridge. Coie, J. D., & Krehbiel, G. (1984). Effects of academic tutoring on the social status of low-achieving, socially rejected children. Child Development, 55, 1465-1478.
80 Coie, J. D., Lochman, J. E., Terry, R., & Hy man, C. (1992). Predicting early adolescent disorder from childhood aggr ession and peer rejection. Journal of Consulting and Clinical Psychology, 60, 783-792. Connell, J. P., & Wellborn, J. G. (1991). Competence, autonomy, and relatedness: A motivational analysis of self-esteem proce sses. In M. R. Gunnar & L. A. Sroufe (Eds.), Self processes and development: The Minnesota symposia on child development (Vol. 23, pp. 43-78). Hillsdale, NJ: Erlbaum. Corno, L., & Mandinach, E. B. (1983). The role of cognitive engagement in classroom learning and motivation. Educational Psychologist, 18, 88-108. Cowen, E. L., Dorr, D., Clarfield, S. P., Kr eling, B., McWilliams, S. A., Pokracki, F., Pratt, D. M., Terrell, D. L., & Wilson, A. B. (1973). The AML: A quick screening device for early identifi cation of school maladaptation. American Journal of Community Psychology, 1, 12-35. Cowie, H. (2000). Bystanding of standing by: Gender issues in coping with bullying in English schools. Aggressive Behavior, 26, 85-97. Craig, W. M. (1998). The relationship among bullying, victimization, depression, anxiety, and aggression in el ementary school children. Personality & Individual Differences, 24 (1), 123-130. Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social informationprocessing mechanisms in childrens social adjustment. Psychological Bulletin, 115 (1), 74-101. Crick, N. R., & Grotpeter, J. K. (1996). Relational aggression, gender, and socialpsychological adjustment. Child Development, 66, 710-722.
81 Deffenbacher, J. L. (1992). Trait anger: Th eory, findings, and implications. In C. D. Spielberger & J. N. Butcher (Eds.), Advances in personality assessment (Vo. 9, pp. 177-201). Hillsdale, NJ: Erlbaum. Dempsey, M. (2002). Negative coping as medi ator in the relation between violence and outcomes: Inner-city African American youth. American Journal of Orthopsychiatry, 72 (1), 102-109. DeRosier, M. E., Kupersmidt, J. B., & Patter son, C. J. (1994). Childrens academic and behavioral adjustment as a function of the chronicity and proximity of peer rejection. Child Development, 65, 1799-1813. DeShon, R. P. (1998). A cautionary note on meas urement error corrections in structural equation models. Psychological Methods, 3 (4), 412-423. Dill, E. J., Vernberg, E. M., Fonagy, P., Twemlow, S. W., & Gamm, B. K. (2004). Negative affect in victimized children: The roles of social withdrawal, peer rejection, and attit udes toward bullying. Journal of Abnormal Child Psychology, 32 (2), 159-173. Dodge, K. A. (1983). Behavioral ant ecedents of peer social status. Child Development, 54, 1386-1399. Dodge, K. A., & Coie, J. D. ( 1987). Social information-proce ssing factors in reactive and proactive aggression in childrens peer groups. Journal of Personality and Social Psychology, 53, 1146-1158. Dodge, K. A., Coie, J. D., & Brakke, N. P. ( 1982). Behavior patterns of socially rejected and neglected preadolescents: The roles of social approach and aggression. Journal of Abnormal Child Psychology, 10, 389-410.
82 Dodge, K. A., & Crick, N. R. (1990). Social information-processing bases of aggressive behavior in children. Personality and Social Psychology Bulletin, 16 (1), 8-22. Dorr, D., Stephens, J., Pozner, R., & Klodt, W. (1980). Use of the AML scale to identify adjustment problems in fourth-, fifth-, and sixth-grade children. American Journal of Community Psychology, 8, 341-352. Doyle, W. (1986). Classroom organization a nd management. In M. C. Wittrock (Ed.), Handbook of research on teaching (pp. 392-431). New York: Macmillian. Durlak, J. A. & Jason, L. A. (1984). Prev entive programs for school-aged children and adolescents. In M. C. R oberts & L. Peterson (Eds.), Prevention of problems in childhood: Psychological research and applications (pp. 103-132). New York: Wiley. Durlak, J. A., Stein, M. A., & Mannarino, A. P. (1980). Behavioral validity of a brief teacher rating scale (the AML) in iden tifying high-risk acting-out schoolchildren. American Journal of Community Psychology, 8 (1), 101-115. Edens, J. F., Cavell, T. A., & Hughes, J. N. (1999). The self-system of aggressive children: A cluster-anal ytic investigation. Journal of Child Psychology and Psychiatry, 40, 441-453. Egan, S. K., & Perry, D. G. (1998). Does low self-regard invite victimization? Developmental Psychology, 34, 299-309. Endler, N. S., & Parker, J. D. A. (1990). Multidimensional assessment of coping: A critical evaluation. Journal of Personality and Social Psychology, 58, 844-854.
83 Farrington, D. P. (1993). Understanding a nd preventing bullying. In M. Tonry (Ed.), Crime and justice: Vol. 17. A review of research (pp. 381-458). Chicago, IL: University of Chicago Press. Faulstich, M. E., Carey, M. P., Ruggiero, L., et al. (1986). Assessment of depression in childhood and adolescence: An evaluation of the center for epidemiological studies depression scale for children (CES-DC). American Journal of Psychiatry, 143 (8), 1024-1027. Felner, R. D., Aber, M. S., Primavera, J., & Cauce, A. M. (1985). Adaptation and vulnerability in high-risk adolescent s: An examination of environmental mediators. American Journal of Community Psychology, 13, 365-379. Fields, L., & Prinz, R. J. (1997). Coping and adjustment during childhood and adolescence. Clinical Psychology Review, 17, 937-976. Florida Department of Education. (2000). Analysis of the FCAT Test Item Review: 1999 Harcourt Educational Measurement. Florida Department of Education (2002). Curriculum, Instruction and Assessment. Retrieved September 1, 2003, from the Your Florida Department of Education Web site: http://www.firn.edu/doe/sas/fcat/ Futrell, M. H. (1996). Violence in the clas sroom: A teachers perspective. In A. M. Hoffman (Ed.), Schools, violence, and society (pp. 3-20). Westport, CT: Praeger. Gillespie, J. F. & Durlak, J. A. (1995). Screening children at risk for school maladjustment: Further evidence for the value of the AML. Journal of Community Psychology, 23 18-27.
84 Goodenow, C. (1993). Classroom bel onging among early adolescent students: Relationships to motivation and achievement. Journal of Early Adolescence, 13, 21-43. Graham, S., & Juvonen, J. (1998). Self-blame and peer victimization in middle school: An attributional analysis. Developmental Psychology, 34 (3), 587-599. Green, S. B., & Babyak, M. A. (1997). Control of type I errors with multiple tests of constraints in struct ural equation modeling, Multivariate Behavioral Research, 32 (1), 39-51. Green, K. D., Beck, S. J., Forehand, R., & Vosk, B. (1980). Validity of teacher nominations of child behavior problems. Journal of Abnormal Child Psychology, 8 (3), 397-404. Hanish, L. D., Ryan, P., Martin, C. L., & Fa bes, R. A. (2005). The social context of young childrens peer victimization. Social Development, 14 (1), 2-19. Hausman, A. J., Spivak, H., & Prothrow-Stit h, D. (1994). Adoles cents knowledge and attitudes about an expe rience with violence. Journal of Adolescent Health, 15, 400-406. Hawker, D. S. J., & Boulton, M. J. (2000). Tw enty years research on peer victimization and psychosocial maladjustment: A meta-a nalytic review of cross-sectional studies. Journal of Child Psychology and Ps ychiatry and Allied Disciplines, 41, 441-455. Hawley, P. H., & Vaughn, B. E. (2003). Aggres sion and adaptive functioning: The bright side to bad behavior. Merrill-Palmer Quarterly, 49 (3), 239-242.
85 Hay, I. (2000). Gender self-c oncept profiles of adolescents suspended from high school. Journal of Child Psychology and Psychiatry, 41, 345-352. Hayduk, L. A. (1987). Structural equation modeling wi th LISREL: Essentials and advances. Baltimore, MD: Johns Hopkins Press. Haynie, D. L., Nansel, T., Eitel, P., Crump, A. D., Saylor, K., Yu, K., & Simons-Morton, B. (2001). Bullies, victims, and bully/victims: Distinct groups of at-risk youth. Journal of Early Adolescence, 21 (1), 29-49. Hazler, R. J. (1996). Breaking the Cycle of Violence: Interventions for Bullying and Victimization. Athens, OH: Ohio University. Henrich, C. C., Schwab-Stone, M., Fanti, K ., Jones, S. M., & Ruchkin, V. (2004). The association of community violence expos ure with middle-school achievement: A prospective study. Applied Developmental Psychology, 25, 327-348. Hodges, E. V. E., Malone, M. J., & Perry, D. G. (1997). Individual risk and social risk as interacting determinants of vi ctimization in the peer group. Developmental Psychology, 33, 1032-1039. Hodges, E. V. E., Boivin, M., Vitaro, F., & Bukowski, W. M. ( 1999). The power of friendship: Protecting agai nst an escalating cycle of peer victimization. Developmental Psychology, 35, 94-101. Holmbeck, G. N. (1997). Toward terminological conceptual, and statistical clarity in the study of mediators and moderators: Ex amples from the child-clinical and pediatric psychology literatures. Journal of Consulting an d Clinical Psychology, 65 (4), 599-610.
86 Holmbeck, G. N. (2002). Post-hoc probing of significant moderational and mediational effects in studies of pediatric populations. Journal of Pediatric Psychology, 27 (1), 87-96. Holmbeck, G. N., Li, S. T., Schurman, J. V., Friedman, D., & Coakley, R. M. (2002). Collecting and managing multisource and mu ltimethod data in studies of pediatric populations. Journal of Pediat ric Psychology, 27 (1), 5-18. Holmbeck, G. N., Westhoven, V. C., Shapera Phillips, W., Bowers, R., Gruse, C., Nikolopoulos, T., Totura, C. M. Wienke & Davison, K. (2003). A multimethod, multi-informant, and multi-dimensional perspective on psychosocial adjustment in preadolescents with spina bifida. Journal of Consulting and Clinical Psychology, 71 (4), 782-796. Hoover, J. H., Oliver, R., & Hazler, R. J. (1992). Bullying: Percep tions of adolescent victims in the Midwestern USA. School Psychology International, 13, 5-16. Horowitz, K., Weine, S., & Jekel, J. (1995). PTSD symptoms in ur ban adolescent girls: Compounded community trauma. Journal of the Americ an Academy of Child and Adolescent Psychiatry, 34, 1353-1361. Johnston, L. D., & Bachman, J. G. (1980). Monitoring the Future: Questionnaire responses from the nation' s high school seniors, 1975. Ann Arbor, MI: Institute for Social Research. Joreskog, K. G., & Sorbom, D. (2004). Interactive Lisrel 8.7 Lincolnwood, IL: Scientific Software International, Inc.
87 Juvonen, J., Nishina, A., Graham, S. (2000). Peer harassment, psychological adjustment, and school functioning in early adolescence. Journal of Educational Psychology, 92 (2), 349-359. Kenny, D., & Judd, C. M. (1984). Estimating the nonlinear and interactive effects of latent variables. Psychological Bulletin, 96, 201-210. Kochenderfer, B. J., & Ladd, G. W. (1996a). Peer victimization: Cause or consequence of school maladjustment? Child Development, 67, 1305-1317. Kochenderfer, B. J., & Ladd, G. W. (1996b). Peer victimization: Manifestations and relations to school adjust ment in kindergarten. Journal of School Psychology, 34, 267-283. Kochenderfer, B. J., & Ladd, G. W. (1997). Vi ctimized childrens responses to peers aggression: Behaviors associ ated with reduced versus continued victimization. Development and Psychopathology, 9, 59-73. Kochenderfer-Ladd, B., & Skinner, K. (2002). Childrens coping strategies: Moderators of the effects of peer victimization. Developmental Psychology, 38 (2), 267-278. Ladd, G. W., & Kochenderfer-Ladd, B. (2002) Identifying victims of peer aggression from early to middle childhood: Analysis of cross-informant data for concordance, estimation of relational adju stment, prevalence of victimization, and characteristics of identified victims. Psychological Assessment, 14 (1), 74-95. Li, A. K. F. (1985). Early rejected status a nd later social adjustme nt: 3-year follow-up. Journal of Abnormal Child Psychology, 13, 567-577.
88 Lightner, R. B., Bollmer, J. M., Harris, M. J ., Milich, R., & Scambler, D. J. (2000). What do you say to teasers? Parent and child evaluations of responses to teasing. Journal of Applied Developmental Psychology, 21 (4), 403-427. Maccoby, E., & Jacklin, C. (1974). The psychology of sex differences. Stanford, CA: Stanford University Press. Marsh, H. W., Parada, R. H., Yeung, A. S ., & Healey, J. (2001). Aggressive school troublemakers and victims: A longitudinal model examining the pivotal role of self-concept. Journal of Educational Psychology, 93 (2), 411-419. McCall, R. B., Beach, S. R., & Lau, S. (2000). The nature and correlates of underachievement among elementary school children in Hong Kong. Child Development, 71, 785-801. McClelland, G. H. & Judd, C. M. (1993). Statis tical difficulties of de tecting interactions and moderator effects. Psychological Bulletin, 114 (2), 376-390. Midgley, C., Feldlaufer, H., & Eccles, J. ( 1989). Student/teacher re lations and attitudes toward mathematics before and after the transition to junior high school. Child Development, 60, 981-992. Miller, D. C., & Byrnes, J. P. (2001). To ach ieve or not to achie ve: A self-regulation perspective on adolescents academic decision making. Journal of Educational Psychology, 93 (4), 677-685. Monnier, J., Hobfoll, S. E., Dunahoo, C. L., Hulsizer, M. R., & Johnson, R. (1998). Theres more than rugged individualism in coping. Part 2: Construct validity and further model testing. Anxiety, Stress, and Coping, 11 247-272.
89 Montada, L., & Lerner, M. (1998). Responses to victimization and beliefs in a just world. Critical issues in social justice. New York: Plenum. Mooney, A. M., Creeser, R., & Blatchford, P. (1991). Childrens views on teasing and fighting in junior schools. Educational Research, 33 (2), 103-112. Morrison, G. M., Furlong, M. J., & Morrison, R. L. (1997). The safe school: Moving beyond crime prevention to school empowerme nt. In A. P. Goldstein & J. C. Conoley (Eds.), School violence intervention: A practical handbook (pp. 236264). New York, NY: Guildford. Mosquera, P. M. R., Manstead, A. S. R., & Fischer, A. H. (2002). The role of honour concerns in emotional reactions to offenses. Cognition & Emotion, 16 (1), 143163. Muma, J. R. (1965). Peer evalua tion and academic performance. Personnel and Guidance Journal, 44, 405-409. Nansel, T. R., Overpeck, M., Pilla, R. S., Ruan, W. J., Simons-Morton, B., & Scheidt, P. (2001). Bullying behaviors among US yout h: Prevalence and association with psychosocial adjustment. Journal of the American Medical Association, 285 (16), 2094-2100. Neary, A., & Joseph, S. (1994). Peer victimi zation and its relations hip to self-concept and depression among schoolgirls. Personality and Individual Differences, 16, 183-186. Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1986). Learned helplessness in children: A longitudinal study of depressi on, achievement, and explanatory style. Journal of Personality and Social Psychology, 51, 435-442.
90 Nolen-Hoeksema, S., Girgus, J. S., & Selig man, M. E. P. (1992). Predictors and consequences of childhood depressive symptoms: A 5-year longitudinal study. Journal of Abnormal Psychology, 101 (3), 405-422. Nunnelly, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York, NY: McGraw-Hill, Inc. Olweus, D. (1978). Aggression in the schools: Bullies and whipping boys. Washington, DC: Hemisphere (Wiley). Olweus, D. (1991a). Bully/victim probl ems among schoolchildren: Basic facts and effects of a school based intervention pr ogram. In D. Pepler & K.H. Rubin (Eds.), The development and treatment of childhood aggression (pp. 411-448). Hillsdale, NJ: Lawrence Erlbaum. Olweus, D. (1991b). Victimization among sc hool children. In R. Baenniger (Ed.), Targets of violence and aggression (pp. 45-102). Holland: Elsevier Science Publishers B.V. Olweus, D. (1993). Victimizati on by peers: Antecedents and long-term outcomes. In K. H. Rubin & J. B. Asendorpf (Eds.), Social withdrawal, inhibition, and shyness in childhood (pp. 315-341). Hillsdale, NJ: L. Erlbaum Associates. Olweus, D. (1994a). Annotati on: Bullying at school: Basic facts and effects of a schoolbased intervention program. Journal of Child Psychology and Psychiatry, 35 (7), 1171-1190. Olweus, D. (1994b). Bullying at school: L ong-term outcomes for the victims and an effective school-based intervention program. In L.R. Huesmann (Ed.), Aggressive behavior: Current perspectives. New York: Plenum Press.
91 Olweus, D. (1995). Bullying or peer abuse at school: Facts and intervention. Current Directions in Psychological Science, 4 (6), 196-200. Olweus, D. (1996). The Revised Olweus Bully/Victim Questionnaire. Bergen, Norway: Research Center for Health Promotion (HEMIL), University of Bergen, N-5015 Bergen, Norway Olweus, D. (1997a). Bully/victim problems in school: Facts and intervention. European Journal of Psychology of Education, 12 (4), 495-510. Olweus, D. (1997b). Bully/victim problems in school: Knowledge base and effective intervention program. The Irish Journal of Psychology, 18 (2), 170-190. Olweus, D., Limber, S. & Mihalic, S. F. (1999). Blueprints for violence prevention, book nine: Bullying prevention program. Boulder, CO: Center for the Study and Prevention of Violence. Parke, R. (2004). Development in the family. Annual Review of Psychology, 55, 365-399. Parker, J. G., & Asher, S. R. (1987). Peer relations and later pe rsonal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102, 357-389. Pearce, J. (1991). What can be done about the bully? In M. Elliot (Ed.), Bullying: A practical guide to coping for schools. Harlow: Longman. Pellegrini, A. D. (1998). Bullies and victims in school: A review and call for research. Journal of Applied Developmental Psychology, 19 (2), 165-176. Pellegrini, A. D., & Bartini, M. (2000). An empirical comparison of methods of sampling aggression and victimi zation in school settings. Journal of Educational Psychology, 92 (2), 360-366.
92 Pellegrini, A. D., Bartini, M., & Brooks F. (1999). School bullies, victims, and aggressive victims: Factors relating to group affiliation and victimization in early adolescence. Journal of Educational Psychology, 91 (2), 216-224. Pepler, D. J., Craig, W. M., Ziegler, S., & Charach, A. (1994). An evaluation of an antibullying intervention in Toronto schools. Canadian Journal of Community Mental Health, 13, 95-110. Perry, D.G., Kusel, S.J., & Perry, L.C. (1988). Victims of peer aggression. Developmental Psychology, 24 (6), 807-814. Perry, D. G., Perry, L. C., & Kennedy, E. (1992) Conflict and development of antisocial behavior. In C. U. Shan tz & W. W. Hartup (Eds.), Conflict in child and adolescent development (pp. 301-329). New York: Camb ridge University Press. Perry, D. G., Williard, J. C., & Perry, L. C. (1990). Peers perceptions of the consequences that victimized children provide aggressors. Child Development, 61, 1310-1325. Phelan, P., Davidson, A. L., & Cao, H. T. ( 1991). Students multiple worlds: Negotiating the boundaries of family, peer, and school cultures. Anthropology and Education Quarterly, 22, 224-250. Ping, R. A. (1996). Latent variable interac tion and quadratic eff ect estimation: A twostep technique using struct ural equation analysis. Psychological Bulletin, 119 (1), 166-175. Pressley, M., Borkowski, J. G., & Schneid er, W. (1987). Cognitive strategies: Good strategy users coordinate metacognition and knowledge. Annals of Child Development, 4, 89-129.
93 Pynoos, R. S., & Nader, K. (1988). Psychologi cal first aid and trea tment approach to children who are exposed to community violence: Research implications. Journal of Traumatic Stress, 1, 445-473. Radloff, L. S. (1977). A CES-D scale: A self -report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401. Raykov, T., & Marcoulides, G. A. (2000). A first course in struct ural equation modeling. Mahwah, NJ: Lawrence Erlbaum Associates. Reid, K. (1989). Bullying and persistent schoo l absenteeism. In D. P. Tattum & D. A. Lane (Eds.), Bullying in schools (pp. 89-94). London: Trentham Books. Renshaw, P. D., & Brown, P. (1993). Lone liness in middle childhood: Concurrent and longitudinal predictors. Child Development, 64, 1271-1284. Rigby, K., & Cox, I. (1996). The contribution of bullying at school and low self-esteem to acts of delinquency among Australian teenagers. Personality & Individual Differences, 21, 609-612. Rigby, K., & Slee, P. (1991). Bullying am ong Australian school children: Reported behavior and attitude s toward victims. Journal of Social Psychology, 131 (5), 615-627. Robert, R. E., Lewinsohn, P. M., & Seeley, J. R. (1991). Screening for adolescent depression: A comparison of depression scales. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 58-66. Roland, E. (2000). Bullying in school: Thre e national innovations in Norwegian schools in 15 years. Aggressive Behavior, 26 135-143.
94 Rose, A. J., Swenson, L. P., & Waller, E. M. (2004). Overt and re lational aggression and perceived popularity: Developmental differe nces in concurrent and prospective relations. Developmental Psychology, 40 (3), 378-387. Safe Community-Safe School Project (2002). The middle school/high school survey revised The Center for the Study and Preven tion of Violence, University of Colorado, Boulder. Salmivalli, C., Kurhunen, J., & Lagerspetz, K. (1996). How do the victims respond to bullying? Aggressive Behavior, 22, 99-109. Santa Lucia, R. C. & Gesten E. L. (2000). Adjustment across the middle school transition: Gender and race considerations. Poster presentation at the annual meeting of the American Psycholog ical Association, Washington D.C. Sarason, S. B., Davidson, K. S ., Lighthall, F. F., Waite, R. R., & Ruebush, B. K. (1960). Anxiety in elementary school children. New York: Wiley. Schwab-Stone, M. E., Ayers, T. S., Kasprow, W., Voyce, C., Barone, C., Shriver, T., & Weissberg, R. P. (1995). No safe have n: A study of violence exposure in an urban community. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 1343-1352. Schwartz, D., Chang, L., & Farver, J. M. ( 2001). Correlates of vic timization in Chinese childrens peer groups. Developmental Psychology, 37 (4), 520-532. Schwartz, D., Dodge, K. A., & Coie, J. D. (1993). The emergence of chronic peer victimization. Child Development, 64, 1755-1772.
95 Schwartz, D., Dodge, K. A., Pettit, G. S., & Bates, J. E. (2000). Friendship as a moderating factor in the pathway betw een early harsh home environment and later victimization in the peer group. Developmental Psychology, 36 (5), 646-662. Schwartz, D., & Gorman, A. H. (2003). Community violence e xposure and childrens academic functioning. Journal of Educational Psychology, 95 (1), 163-173. Schwartz, D., McFadyen-Ketchum, S. A., Dodge, K. A., Petit, G. S., & Bates, J. E. (1998). Peer group victimizati on as a predictor of childre ns behavior problems at home and in school. Development and Psychopathology, 10, 87-99. Shapiro, J. P., Baumeister, R. F., & Kessler J. W. (1991). A three component model of childrens teasing: Aggre ssion, humor, and ambiguity. Journal of Social and Clinical Psychology, 10, 459-472. Singer, M. I., Anglin, T. M., Song, L. Y., & Lunghofer, L. (1995). Adolescents exposure to violence and associated symptoms of psychological trauma. Journal of the American Medical Association, 273, 477-482. Sivan, E. (1986). Motivation in so cial constructivist theory. Educational Psychologist, 21, 209-233. Slaby, R. G. (1998). Preventing youth violence through research-guided intervention. In P. K. Trickett & C. J. Schellenback (Eds.), Violence against children in the family and the community (pp. 371-399). Washington, DC : American Psychological Association. Slee, P. T. (1994). Situational and interper sonal correlates of anxiety associated with peer victimization. Child Psychiatry & Human Development, 25 (2), 97-107.
96 Slee, P. T. (1995). Peer victimization and its relationship to de pression among Australian primary school students. Personality & Individual Differences, 18, 57-62. Smith, P. K., Shu, S., Madsen, K. (2001). Char acteristics of victim s 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. Solberg, M. E., & Olweus, D. (2003). Preval ence estimation of sc hool bullying with the Olweus bully/victim questionnaire. Aggressive Behavior 29 (3), 239-268. Spielberger, C.D. (1973). Preliminary test manual for the state-trait anxiety inventory for children Palo Alto, CA: Consulti ng Psychologists Press, Inc. Spielberger, C.D. (1998). The revised state-trait anger exp ression inventory-2 (STAXI2). Psychological Assessment Resources, Inc. Spielberger, C.D., Jacobs, G.A., Brunne r, T.M., & Lunsford, G.D. (2002). The state-trait anger expression inventory for childre n and adolescents (STAXI-C/A). Psychological Assessment Resources, Inc. Springer, C., & Padgett, D. K. (2000). Gender differences in young adolescents; exposure to violence and rate s of PTSD symptomatology. American Journal of Orthopsychiatry, 70, 370-379. Springer, J. F., & Phillips, J. L. (1997). Individual Protective Fa ctors Index (IPFI): A measure of adolescent resiliency. EMT Associates, Inc. Swain, J. (1998). What doe s bullying really mean? Educational Research, 40 (3), 358364.
97 Toner, M. & Munro, D. (1996). Peer-social attr ibutions and self-efficacy of peer-rejected preadolescents. Merrill-Palmer Quarterly, 42, 339-357. Totura, C. M. Wienke, MacKinnon-Lewis, C ., Gesten, E. L., Gadd, R., Divine, K. P., Dunham, S., & Kamboukos, D. (2005). Bullying and victimization among boys and girls in middle school: The influ ence of family and school contexts. Manuscript submitted for publication. U.S. Department of Education. (2001). No child left behind act of 2001 (Public Law PL 107-110). Washington, DC. Education Publications Center. Verhulst, F. C. & Van Der Ende, J. (1991). Four-year follow-up of teacher-reported problem behaviours. Psychological Medicine, 21 965-977. Verkuyten, M., & Thijs, J. (2002). School satis faction of elementary school children: The role of performance, peer relations, ethnicity, and gender. Social Indicators Research, 59 203-228. Vernberg, E. M., & Gamm, B. K. (2003). Re sistance to violence prevention interventions in schools: Barriers and solutions. Journal of Applied Psychoanalytic Studies, 5 (2), 125-137. Wang, M. C., Haertal, G. D., & Walberg, H. J. (1990). What influences learning? A content analysis of review literature. Journal of Educational Research, 84, 30-43. Warm, T. R. (1997). The role of teasing in development and vice versa. Developmental and Behavioral Pediatrics, 18 (2), 97-101. Weissman, M. M., Orvaschel, H., & Padian, N. (1980). Childrens symptom and social functioning self-report scales : Comparison of mothers and childrens reports. Journal of Nervous Mental Disorders, 168 (12), 736-740.
98 Wells, E., & Rankin, J. (1983). Self-concept as a mediating factor in delinquency. Social Psychology Quarterly, 46, 11-22. Wentzel, K. R. (1991). Relations between social competence and academic achievement in early adolescence. Child Development, 62, 1066-1078 Wentzel, K. R. (1993a). Do es being good make the grade? Social behavior and academic competence in middle school. Journal of Educational Psychology, 85 (2), 357-364. Wentzel, K. R. (1993b). Motivation and achieve ment in early adolescence: The role of multiple classroom goals. Journal of Early Adolescence, 13, 4-20. Wentzel, K. R. (1994). Relations of social goal pursuit to social acceptance, classroom behavior, and perceive d social support. Journal of Educational Psychology, 86 (2), 173-182. Wentzel, K. R. (1998). Social relationships and motivation in middle school: The role of parents, teachers, and peers. Journal of Educational Psychology, 90 (2), 202-209. Wentzel, K. R., & Asher, S. R. (1995). A cademic lives of neglected, rejected, popular, and controversial children. Child Development, 62, 1066-1078. Wentzel, K. R., & Caldwell, K. (1997). Friendships, peer acceptance, and group membership: Relations to academic achievement in middle school. Child Development, 68 (6), 1198-1209. Wentzel, K. R., Weinberger, D. A., Ford, M. E., & Feldman, S. S. (1990). Academic achievement in preadolescence: The role of motivational, affective, and selfregulatory processes. Journal of Applied Developmental, 11, 179-193.
99 Wolke, D., Woods, S., Bloomfield, L., & Kars tadt, L. (2000). The association between direct and relational bullying and be havior problems among primary school children. Journal of Child Psychology & Psychiatry & Allied Disciplines, 41 (8), 989-1002. Wothke, W. (2000). Longitudinal and multi-grou p modeling with missing data. In T. D. Little, K. U. Schnabel, & J. Baumert (Eds.), Modeling longitudinal and multilevel data: Practical issues, applie d approaches, and specific examples (pp. 1-24) Mahwah, NJ: Lawrence Erlbaum Associates. Zimmerman, B. J., & Schunk, D. H. (1989). Self-regulated learning and academic achievement: Theory, research, and practice. New York, NY: Springer-Verlang. Zwick, W. R., & Velicer, W. F. (1986). Comp arison of five rules for determining the number of components to retain. Psychological Bulletin, 99 (3), 432-442.
100 Table 1 Means and Standard Deviations for Constructs (z-scores) and Observed Variables Construct/Variable Total Boys Girls M SD M SD M SD Victimization .0003 .849 .111 .967 -.084 .731 Olweus scale 1.47 .474 1.52 .580 1.44 .368 MSHS scale .458 .524 .527 .579 .406 .472 Psychological Functioning .005 .724 -.001 -.031 .014 .721 Depression 14.93 10.52 14.85 10.57 15.02 10.52 Anxiety 33.40 8.82 32.16 8.90 34.35 8.66 Anger 21.84 5.01 22.07 5.89 21.66 5.22 Moodiness 7.50 3.10 7.80 3.38 7.26 2.85 Academic Motivation -.014 .796 -.217 .771 .148 .783 School Efficacy 2.90 .525 2.81 .522 2.97 .519 Goal Orientation 4.07 .724 3.89 .747 4.20 .681 Learning Difficulties (reversed) 3.91 .919 3.79 .982 4.00 .858 Academic Achievement -.008 .811 -.169 .872 .122 .737 GPA 2.77 .961 2.45 1.04 3.03 .812 FCAT Reading 1766.1 271.95 1735.2 275.36 1791.2 268.08 FCAT Math 1802.6 189.04 1803.6 192.63 1803.5 185.92 Friendship .0003 .781 -.204 .809 .166 .721 # of Friends 3.10 1.08 3.03 1.15 3.17 1.03 Quality 3.35 .660 3.13 .661 3.54 .603 Prosocial Activities and Influences -.002 .678 -.103 .649 .080 .693 Family Involvement 3.29 .718 3.31 .706 3.28 .731 Peer Involvement 1.33 .381 1.30 .377 1.35 .383 Friends Attitudes against Aggression 2.81 .621 2.64 .591 2.95 .612 Friends Attitudes against Deviant Behavior 3.35 .708 3.29 .717 3.40 .699 Aggression .006 .746 .163 .826 -.118 .653 Olweus Bullying 1.16 .236 1.20 .297 1.13 .169 MSHS Bullying .254 .458 .342 .527 .186 .384 Anger Expression 43.71 8.90 44.78 9.33 42.89 8.48 Aggression Coping Beliefs 2.10 .652 2.32 .649 1.93 .606 Aggressive Coping Behavior s .846 .429 .836 .428 .854 .432 Teacher-Reported Difficulties -.009 .855 .238 .921 -.203 .747 Acting-Out 7.25 3.46 8.32 4.00 6.39 2.69 Referrals 1.92 3.41 2.81 3.74 1.23 2.95 Bullied by Others 1.30 .685 1.48 .850 1.17 .481 Climate -.003 .781 -.171 .803 .134 .739 Olweus Intervention 2.31 .781 2.22 .806 2.38 .754 MSHS Intervention 2.81 .611 2.72 .631 2.88 .588 Supervision 3.15 .667 2.99 .683 3.28 .628 Note. N = 327 for Total sample, 145 for Boys, and 181 for Girls.
101 Table 2 Correlations between Variables in th e Mediated and Moderated Pathways Note. Ns range from 268 to 327. MSHS = Middle School/High School Student Survey. AML-R = Acting-Out, Moodiness, and Learning Scale Revised. GPA = Grade Po int Average. FCAT = Florid a Comprehensive Achievement Tests. Bold correlations are significant at p < .05. Underlined correlations are si gnificant at p < .01. Scale 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 1.Olweus Victim .45 .01 .21 .35 .23 -.02 .03 .01 -.05 -.03 .03 -.15 -.05 .04 2. MSHS Victim .07 .29 .34 .27 -.13 -.08 -.14 -.13 -.05 -.03 -.01 -.10 .20 3. Moodiness .05 .10 .12 -.24 -.67 -.20 -.45 -.07 -.02 -.004 -.08 .20 4. Depression .53 .40 -.30 -.09 -.30 -.22 -.07 -.07 .03 -.03 .09 5. Anxiety .42 -.03 -.01 -.14 -.03 .08 .03 -.06 .10 .02 6. Anger -.21 -.06 -.13 -.12 .01 .03 .10 .02 .19 7. Goal Orientation .21 .46 .40 .16 .12 -.05 .25 -.19 8. Learning (reversed) .22 .53 .36 .25 -.04 .14 -.21 9. School Efficacy .23 -.02 -.04 .04 .25 -.15 10. GPA .37 .36 -.01 .15 -.19 11. FCAT Reading .65 .03 .09 -.04 12. FCAT Math -.002 -.01 -.01 13. # of Friends .22 .03 14. Friendship Quality -.10 15. Olweus Bullying 16. MSHS Bullying .09 .47 .20 .13 .11 .22 -.23 -.19 -.23 -.20 -.08 -.01 .02 -.18 .48 17. Anger Expression -.04 .13 .29 .18 .05 .40 .35 -.33 -.24 -.29 -.18 -.12 .06 -.14 .41 18. Aggressive Coping Beliefs -.02 .24 .26 .11 .03 .23 -.33 -.27 -.43 -.30 -.02 .08 .07 -.29 .36 19. Aggressive Coping Behaviors .06 .38 .11 .16 .20 .30 -.20 -.14 -.27 -.10 .05 .05 .09 -.05 .41 20. Olweus Intervention -.05 -.07 -.07 -.11 .03 .02 .29 .01 .27 .07 -.01 -.04 .08 .15 -.06 21. MSHS Intervention -.08 -.27 -.21 -.28 -.10 -.09 .46 .12 .55 .19 -.04 -.14 .06 -.18 -.15 22. Adult Supervision -.08 -.12 .01 -.22 .05 -.06 .32 .00 .31 .04 .02 -.10 .03 .21 -.11 23. AML-R Bullied .19 .11 .19 .07 .04 .07 -.10 -.13 -.02 -.22 .06 .05 -.08 -.15 .05 24. Referrals .07 .20 .23 .06 -.03 .09 -.08 -.35 -.10 -.33 -.10 -.08 .04 -.06 .08 25. Acting-Out .05 .17 .70 -.04 -.05 .15 -.19 -.63 -.15 -.42 -.16 .09 .06 -.10 .31 26. Friends Aggressive Attitudes (rev.) -.07 -.25 -.18 -.22 -.11 -.16 .37 .21 .50 .24 .02 -.04 .09 .36 -.31 27. Friends Deviant Behavior Attitudes (rev.) -.06 -.17 -.18 -.19 -.04 -.14 .30 .24 .40 .19 -.01 -.08 .02 .25 -.18 28. Family Involvement -.01 -.18 -.15 -.31 -.16 -.24 .35 .16 .49 .20 -.05 -.04 -.06 .18 -.12 29. Peer Involvement .10 .03 -.04 .01 .04 .03 .17 .06 .27 .15 .07 .12 .04 .14 -.03
102 Table 2 (continued) Correlations between Variables in th e Mediated and Moderated Pathways Note. Ns range from 268 to 327. MSHS = Middle School/High School Student Survey. AML-R = Acting-Out, Moodiness, and Learning Scale Revised. GPA = Grade Point Average. FCAT = Florida Comprehensive Achievement Tests. Bold correlations are significant at p < .05. Underlined correlations are significant at p < .01. Scale 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 1.Olweus Victim .09 -.04 -.02 .06 -.05 -.08 -.08 .19 .07 .05 -.07 -.06 -.01 .10 2. MSHS Victim .47 .13 .24 .38 -.07 -27 -.12 .11 .20 .17 -.25 -.17 -.18 .03 3. Moodiness .20 .29 .26 .11 -.07 -21 .01 .19 .28 .70 -.18 -.18 -.15 -.04 4. Depression .13 .18 .11 .16 -.11 -28 -.22 .07 .06 -.04 -.22 -.19 -.31 .01 5. Anxiety .11 .05 .03 .20 .03 -10 .05 .04 -.03 -.05 -.11 -.04 -.16 .04 6. Anger .22 .40 .23 .30 .02 -.09 -.06 .07 .09 .15 -.16 -.14 -.24 .03 7. Goal Orientation -.23 .35 -.33 -.20 .29 .46 -.32 -.10 -.08 -.19 .39 .30 .35 .17 8. Learning (reversed) -.19 -.33 -.27 -.14 .01 .12 .00 -.13 -.35 -.63 .21 .24 .16 .06 9. School Efficacy -.23 -.24 -.43 -.27 .27 .55 .31 -.02 -.10 -.15 .50 .40 .49 .27 10. GPA -.20 -.29 -.30 -.10 .07 .19 .04 -.22 -.33 -.42 .24 .19 .20 .15 11. FCAT Reading .08 -.18 -.02 .05 -.01 -.04 .02 .06 -.10 -.16 .02 -.01 -.05 .07 12. FCAT Math -.01 -.12 .08 -.05 -.04 -.14 -.10 .05 -.08 -.09 -.04 -.08 -.04 .12 13. # of Friends .02 .06 .07 .09 .08 .06 .03 -.08 .04 .06 .09 .02 -.06 .04 14. Friendship Quality -.18 -.14 -.29 -.05 .15 .29 .21 -.15 -.06 -.10 .36 .25 .18 .14 15. Olweus Bullying .48 .41 .36 .41 -.06 -15 -.11 .05 .08 .31 -.31 -.18 -.12 -.03 16. MSHS Bullying .35 .51 .67 -.07 -15 -.11 .19 .16 .31 -.44 -.40 -.22 -.11 17. Anger Expression .41 .32 -.10 -22 -.11 .20 .13 .33 -.29 -.23 -.26 -.08 18. Aggressive Coping Beliefs .47 -.17 -.46 -.23 .23 .27 .37 -.65 -.48 -.35 -.13 19. Aggressive Coping Behaviors -.05 -.25 -.07 .08 .07 .19 -.39 -.28 -.29 .03 20. Olweus Intervention .48 .33 -.06 .05 -.10 .28 .17 .11 .11 21. MSHS Intervention .45 -.05 -.13 -.16 .53 .41 .43 .16 22. Adult Supervision .09 .01 .01 .23 .18 .16 .11 23. AML-R Bullied .21 .29 -.10 -.18 -.06 -.00 24. Referrals .45 -.05 -.21 -.10 -.03 25. Acting-Out -.21 .23 -.08 .02 26. Friends Aggressive Attitudes (rev.) .53 .37 .11 27. Friends Deviant Behavior Attitudes (rev.) .39 .10 28. Family Involvement .16 29. Peer Involvement 1.0
Table 3 Correlations between Mediator and Moderator Constructs Construct Victimization Psychological Functioning Academic Motivation Academic Achievement Aggression Climate (Intervention/ Supervision) Friendship Teacher Reported Difficulties Prosocial Activities & Influences Victimization Boys Girls .379** -.092 -.060 .248** -.165** -.114* .193** -.123* Psychological Functioning .402** .366** -.347** -.093 .322** -.156* .029 .195** -.256** Academic Motivation -.110 -.027 -.369** -.352** .317** -.424** .482** .195** -.268** .606** Academic Achievement -.072 -.006 -.130 -.068 .280** .298** -.170** .036 .070 -.240** .146** Aggression .260** .190* .346** .314** -.321** -.482** -.179* -.099 -.279** -.094 .280** -.511** Climate (Intervention/ Supervision) -.199* -.091 -.143 -.180* .474** .445** .051 -.051 -.196* -.317** .225** -.084 .453** Friendship -.191* .025 .121 -.060 .103 .191 .004 .056 -.095 -.006 .225** .152* -.055 .244** Teacher Reported Difficulties .169* .171* .287** .121 -.229** -.220** -.288** -.108 .329** .137 .042 -.123 .037 -.031 -.188** Prosocial Activities & Influences -.105 -.117 -.204* -.303** .507** .662** .130 .122 -.500** -.511** .470** .417** .157 .276** -.106 -.215** Note. Ns range from 268 to 327. significant at p < .05; ** signifi cant at p < .01. Correlations for the total sample are on the upper diagonal and correlations split by gender (Girls/Boys) are on the lower diagonal. 103
104 Table 4 Hierarchical Regression Analyses Predic ting Psychological Functioning from Moderator Constructs for the Total Sample Construct R2 B SE B Model: Friendship Step 1: Victimization .144** .346 .045 .406** Step 2: Friendship .005 .006 .048 .061 Step 3: Victimization x Friendship .008 .008 .045 .093 Model: Prosocial Activities and Influences Step 1: Victimization .144** .297 .043 .348** Step 2: Involvement with Family and Peers .045** -.243 .054 -.228** Step 3: Victimization x Involvement with Family & Peers .009 .107 .056 .097 Model: Aggression Step 1: Victimization .144** .286 .044 .336** Step 2: Aggression .056** .266 .051 .273** Step 3: Victimization x Aggression .013* -.102 .045 -.118* Model: Teacher-Reported Difficulties Step 1: Victimization .144** .328 .048 .130** Step 2: Teacher-Reported Difficulties .015* .110 .044 .130* Step 3: Victimization x Teacher-Reported Difficulties .005 -.006 .048 -.075 Model: Climate Step 1: Victimization .144** .341 .045 .399** Step 2: Climate .009 -.010 .048 -.103* Step 3: Victimization x Climate .018** .139 .052 .141** Note. N = 327. = significant at p = .05. ** = significant at p = .01. indicates ma rginal significance, p < .06. Beta weights are reported from Step 3 in the regression models.
105 Table 5 Hierarchical Regression Analyses Predic ting Psychological Functioning from Moderator Constructs for Boys Construct R2 B SE B Model: Friendship Step 1: Victimization .161** .381 .065 .506** Step 2: Friendship .040** .161 .070 .178* Step 3: Victimization x Friendship .015 .009 .057 .143 Model: Prosocial Activities and Influences Step 1: Victimization .161** .307 .058 .407** Step 2: Involvement with Family and Peers .027* -.218 .087 -.194* Step 3: Victimization x Involvement with Family & Peers .019 .149 .081 .143 Model: Aggression Step 1: Victimization .161** .280 .060 .371** Step 2: Aggression .063** .256 .070 .288** Step 3: Victimization x Aggression .012 -.008 .052 -.122 Model: Teacher-Reported Difficulties Step 1: Victimization .161** .307 .065 .407** Step 2: Teacher-Reported Difficulties .049** .187 .061 .235** Step 3: Victimization x Teacher-Reported Difficulties .007 -.006 .060 -.091 Model: Climate Step 1: Victimization .161** .389 .064 .514** Step 2: Climate .004 -.007 .069 -.076 Step 3: Victimization x Climate -.059** .223 .068 .275** Note. N = 145. = significant at p = .05. ** = significant at p = .01. Beta weights are reported from Step 3 in the regression models.
106 Table 6 Hierarchical Regression Analyses Predic ting Psychological Functioning from Moderator Constructs for Girls Construct R2 B SE B Model: Friendship Step 1: Victimization .134** .374 .076 .380** Step 2: Friendship .005 -.007 .071 -.073 Step 3: Victimization x Friendship .001 -.004 .105 -.027 Model: Prosocial Activities and Influences Step 1: Victimization .134** .316 .070 .320** Step 2: Involvement with Family and Peers .069** -.281 .071 -.270** Step 3: Victimization x Involvement with Family & Peers .002 .006 .082 .050 Model: Aggression Step 1: Victimization .134** .284 .070 .288** Step 2: Aggression .062** .316 .079 .286** Step 3: Victimization x Aggression .012 -.175 .108 -.116 Model: Teacher-Reported Difficulties Step 1: Victimization .134** .367 .072 .372** Step 2: Teacher-Reported Difficulties .003 .006 .068 .059 Step 3: Victimization x Teacher-Reported Difficulties .003 -.007 .084 -.061 Model: Climate Step 1: Victimization .134** .348 .069 .353** Step 2: Climate .022* -.144 .068 -.148* Step 3: Victimization x Climate .000 -.0002 .089 -.002 Note. N = 181. = significant at p = .05. ** = significant at p = .01. Beta weights are reported from Step 3 in the regression models.
107 Figure Caption Figure 1. Model with mediated pathway betw een victimization and academic achievement. Figure 2. Model with mediated pa thway between victimizati on and academic motivation. Figure 3. Psychological functioning means at high a nd low levels of aggression for the total sample. Figure 4. Psychological functioning means at high and low levels of climate for the total sample. Figure 5. Psychological functioning means at high and low levels of climate for boys.
Olweus Scale MSHS Scale Victimization Psychological Functioning Academic Motivation Academic Achievement Depression Anxiety Moodiness Anger Goal Orientation Efficacy Learning GPA FCAT Reading FCAT Math .61 .73 .71 .55* .55 -.08 .73 .10 -.33* .96 .46 .16 .16* .51 .92 .70 108 Note. Dotted pathways indicate observed variable estimates ( mle ) fixed at 1.0. The full model includes the dashed pathway, 2 = 48.37, p > .05. The model absent the dashed pathway is the proposed mediated model, 2 = 49.11, p > .05. Standardized regression coefficients ( R2) were indicated for variable loadings and pathway estimates.
Olweus Scale MSHS Scale Victimization Psychological Functioning Academic Motivation Academic Achievement Depression Anxiety Moodiness Anger Goal Orientation Efficacy Learning GPA FCAT Reading FCAT Math .63 .69 .71 .59* .58 -.06 .74 .10 -.36* .99 .41 .14 .14* .48 .86 .75 109 Note. Dotted pathways indicate observed variable estimates ( mle ) fixed at 1.0. The full model includes the dashed pathway, 2 = 45.58, p > .05. The model absent the dashed pathway is the proposed mediated model, 2 = 49.11, p > .05. Standardized regression coefficients ( R2) were indicated for variable loadings and pathway estimates.
110 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 HighLow VictimizationPsychological Functioning High Aggression Low Aggression
111 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 High Low VictimizationPsychological Functioning High Climate Low Climate
112 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 High Low VictimizationPsychological Functioning High Climate Low Climate
114 Appendix A: Independent and Dependent Constructs Independent Constructs In the present study, independent variable s are defined as those constructs that only initiate a pathway. The following construc ts are considered independent variables: Independent Constructs in the Mediated Pathways Victimization Independent Constructs as Hypothesized Moderators Aggression Friendship Climate (Intervention/Supervision) Involvement with Family and Peers Teacher Reported Difficulties Dependent Constructs Dependent variables are defined as thos e constructs that are a result of a directional pathway. These variables may al so initiate pathways, as is the case with mediator variables. Dependent Constructs in the Mediated Pathways Academic Achievement Dependent Constructs as Hypothesized Mediators Psychological Functioning Academic Motivation
115 Appendix B: Constructs for Hypothesized Model VICTIMIZATION Indicators: Olweus Bully/Victim Questionnaire Victim Items (factor loading = .764) I was called mean names, was made fun of, or teased in a hurtful way Other student left me out of things on purpose, excluded me from their group of friends, or completely ignored me I was hit, kicked, pushed, shoved around, or locked indoors Other students lied or spread false rumo rs about me and tried to make others dislike me I had money or other things taken from me or damaged I was threatened or forced to do things I didnt want to do I was bullied with mean names or comments about my race or color I was bullied with mean names, commen ts, or gestures with a sexual meaning I was bullied in another way Middle School/High School Student Survey Victim Items (factor loading = .648) Another student encouraged me to fight Another student pushed, shoved, slapped, or kicked me I was harassed by another student Another student threatened to hit or hurt me A classmate acted cold towards me or gave me the silent treatment A classmate deliberately kept me out of their group because they were angry with me A classmate said bad things about me to hurt my reputation or my friendships with others Other students ganged up against me and were mean to me as a group PSYCHOLOGICAL FUNCTIONING Indicators: AML-R Teacher Form Moodiness Scale (factor loading = .122) Has to be coaxed to play with others Is unhappy Feels hurt when criticized Is moody CES-Depression Scale (factor loading = .827) I was bothered by things that usually dont bother me I did not feel like eating; my appetite was poor I felt that I was just as good as other people I had trouble keeping my mind on what I was doing I felt depressed
116 Appendix B (Continued) I felt that everything I did was an effort I felt hopeful about the future I thought my life had been a failure I felt fearful My sleep was restless I was happy I talked less than usual I felt lonely People were unfriendly I enjoyed life I had crying spells I felt sad I felt that people disliked me I could not get going I felt that I could not shake off the blue s even with help from my family or friends Trait Anxiety Scale (factor loading = .763) I worry about making mistakes I feel like crying I feel unhappy I have trouble making up my mind It is difficult for me to face my problems I worry too much I get upset at home I am shy I feel troubled Unimportant thoughts run through my mind and bother me I worry about school I have trouble deciding what to do I notice my heart beats fast I am secretly afraid I worry about my parents My hands get sweaty I worry about things that may happen It is hard for me to fall asleep at night I get a funny feeling in my stomach I worry about what others think of me Trait Anger Scale (f actor loading = .665) I am annoyed I feel annoyed when I do a good job and no one notices me
117 Appendix B (Continued) I get mad when I am punished unfairly I feel grouchy I get mad I get angry when I do well and am told I did something wrong I feel angry when Im blamed for something I did not do I am hotheaded I get angry quickly I feel like yelling when I do someth ing good and someone says I did bad I get furious when scolded in front of others I feel angry ACADEMIC MOTIVATION Indicators: School Adjustment Scale Goal-Ori entation (factor loading = .508) I try as hard as I can to do my best at school It bothers me when I dont do something well Education is important for success in life I feel prepared for middle school I think I will go to college AML-R Teacher Form Learning Scale (re verse scored; factor loading < .10) is confused with schoolwork gets off-task needs help with schoolwork has difficulty learning School Efficacy (factor loading = .597) Middle School/High School Survey o I do things that make a difference at my school o At school, I help decide things like class activities and rules o If I study hard, I will get good grade o If I really want to achieve some thing at school, I know I can do it ACADEMIC ACHIEVEMENT Indicators: Grade Point Average (factor loading = .413) Middle School Course Grades Standardized Test Scores FCAT Reading Scores (factor loading = .864) FCAT Math Scores (factor loading = .838)
118 Appendix B (Continued) FRIENDSHIP Indicators: Number of Friends (factor loading = .820) Demographic Inventory o How many good friends do you have at school? Friendship Quality (factor loading = .591) Middle School/High School Student Survey o I have a friend my age who cares about me o I spend most of my free time at school with my friends AGGRESSION Indicators: Olweus Bully/Victim Ques tionnaire Bullying Items (factor loading = .759) I called another student mean names, made fun of or teased him/her in a hurtful way I kept him or her out of things on pur pose, excluded him or her from my group of friends, or completely ignored him or her I hit, kicked, pushed, shoved around, or locked him/her indoors I spread false rumors about him/her and tried to make others dislike him/her I took money or other things from hi m/her or damaged his/her belongings I threatened or forced him/her to do things he/she didnt want to do I bullied him/her with mean names or comments about his/her race or color I bullied him/her with mean names, comments, or gestures with a sexual meaning I bullied him/her in another way Middle School/High School Survey Bu llying Items (factor loading = .731) I encouraged other students to fight I pushed, shoved, or kicked other students I harassed another student I threatened to hit or hurt another student I said bad things about someone to hur t their reputation or their friendships with others I ganged up with other students and we did mean things to another kid Anger Expression Scale (factor loading = .577) I am patient with others If I dont like someone, I keep it a secret I try to calm my angry feelings I keep my cool I hide my anger
119 Appendix B (Continued) I try to relax I dont tell anyone Im angry I keep my anger in I try to calm down I control my temper I have more anger than I show I take a deep breath I control my angry feelings I am afraid to show my anger I try to reduce my anger I stop myself from losing my temper I get mad inside, but I dont show it I do something to relax and calm down I show my anger I say mean things I lose my temper I get into arguments I get into fights I do things like slam doors Beliefs about Aggressive Coping (factor loading = .443) Middle School/High School Survey o It is OK to push or shove other people around if youre mad o It is OK to take your anger out on others by using physical force o I think it is OK to hit someone back if they hit you first Aggressive Coping Behavior s (factor loading = .722) Middle School/High School Survey o I walked away from a fight o I got into a physical fight to get something I wanted from another student o I was mean to someone when I was angry o I acted cold toward someone or gave them the silent treatment when I was angry at them o I deliberately kept someone out of my group because I was angry at them PROSOCIAL ACTIVITIES AND INFLUENCES Indicators: Family Involvement (factor loading = .588) Middle School/High School Survey o I like to do things with my family o I have dinner with my family
120 Appendix B (Continued) Peer Involvement (factor loading < .10) Middle School/High School Survey o I am involved in clubs at my school o I am involved in sports teams at my school o I am involved in othe r activities at school o I am involved in clubs (like Boy Sc outs/Girl Scouts), sports teams, church groups or other ac tivities outside of school Friends Attitudes toward Aggression (re verse scored; factor loading = .579) Middle School/High School Student Survey o My friends think it is wrong to hit other people o My friends think it is OK to yell at others and say mean things o My friends think it is OK to push or shove other people if you are mad o My friends think it is OK to physic ally fight to get what you want o My friends think it is wrong to call other people mean names o My friends think it is wrong to get into physical fights (like hitting or pushing) with others o My friends think it is OK to hit someone back when they hit you first o My friends think it is OK to take your anger out on others by using physical force (like hitting or pushing) Friends Attitudes toward Deviant Behavi ors (reverse scored; factor loading = .721) Middle School/High School Student Survey o My friends think it is OK to drink alcohol o My friends drink to get drunk o My friends think using drugs is a dumb idea o My friends think it is OK to smoke cigarettes CLIMATE (INTERVENTION/SUPERVISION) Indicators: Olweus Bully/Victim Questionn aire (factor loading = .758) How often do the teachers or other adults at school try to put a stop to it when a student is being bullied at school How often do other students try to put a stop to it when a student is being bullied at school Has your classroom teacher or any teacher talked with you about your bullying other students at school in the past couple of months Overall, how much do you think your class teacher has done to counteract bullying in the past couple of months Middle School/High School Su rvey (factor loading = .747) Adults at my school teach us not to pick on other students
121 Appendix B (Continued) Adults at my school try hard to keep students from bullying or picking on each other All students at my school who break the rules are treated the same, no matter who they are When someone breaks the rules here, ad ministrators take appropriate action Students in my school obey the rules Adult Supervision at School (factor loading = .558) In my school, teachers and administra tors are in the hall when we change classes In my school, teachers and administrato rs are in the halls when we are in classes In my classroom, teachers walk around while students are working In my school, teachers and administrato rs supervise open areas where students gather In my school, teachers and administrato rs supervise the places where students can hide TEACHER-REPORTED DIFFICULTIES Indicators: Teacher Form Global Bullied Item (factor loading = .582) This child has been bullied at sc hool in the past couple of months Records Data Referrals (factor loading = .548) Discipline infractions averaged acro ss the school year for each student AML-R Teacher Form Acting-Out Scale (factor loading = .810) Gets into fights or quarrels with classmates Is restless Disrupts class discipline Is impulsive
122 Appendix C: Olweus Bully/Victim Questionnaire You will find questions about your life in school. There are several answers next to each question. Each answer has a number by it. Darken in the circle on the scantron form that matches the answer that best describes you for each statement. Here are some questions about being bullied by other students. Fi rst, we define or explain the word bullying. We say a student is being bullied when another student, or several other students: Say mean and hurtful things or make fun of him or her or call him or her hurtful names Completely ignore or exclude him or her from their gr oup of friends or leave him or her out of things on purpose Hit, kick, push, shove around, or lock him or her inside a room Tell lies or spread false rumors about hi m or her or send mean notes and try to make other students dislike him or her And other hurtful things lik e that, including be ing teased in a mean and hurtful way. When we talk about bullying, these things ha ppen repeatedly, and it is difficult for the student being bullied to defend himself or herself. Note that we also call it bullying when a student is teased repeatedly in a mean and hurtful way. But, we dont call it bullying when the teasing is done in a friendly and playful way. Also, it is not bullying when students of about equal st rength or power argue or fight. ABOUT BEING BULLIED BY OTHER STUDENTS Have you been bullied at school in the past couple of months in one or more of the following ways? Please answ er all of the questions. I havent been bullied in the past couple of months It has only happened once or twice 2 or 3 times a month About once a week Several times a week 4. How often have you been bullied at school in the past couple of months? 1 2 3 4 5
123 Appendix C (Continued) 5. I was called mean names, was made fun of, or teased in a hurtful way. 1 2 3 4 5 7. I was hit, kicked, pushed, shoved around, or locked indoors. 1 2 3 4 5 8. Other students told lies or spread false rumors about me and tried to make others dislike me. 1 2 3 4 5 9. I had money or other things taken away from me or damaged. 1 2 3 4 5 10. I was threatened or forced to do things I didnt want to do. 1 2 3 4 5 11. I was bullied with mean names or comments about my race or color. 1 2 3 4 5 12. I was bullied with mean names, comments, or gestures with a sexual meaning. 1 2 3 4 5 13. I was bullied in another way. In this case, please write where:_________________ 1 2 3 4 5 14. In which classes is the stude nt or students who bully you? I havent been bullied in the last couple of months In my class In a different class but same grade In a higher grade In a lower grade In different grades 1 2 3 4 5 6
124 Appendix C (continued) 15. Have you been bullied by boys or girls? I havent been bullied in the last couple of months Mainly by one girl By several girls Mainly by one boy By several boys By both boys and girls 1 2 3 4 5 6 16. By how many students have you usually b een bullied? I havent been bullied in the last couple of months Mainly by one student By a group of 2-3 students By a group of 4-9 students By a group of more than 9 students By several different students of groups 1 2 3 4 5 6 17. How long has the bullying lasted? I havent been bullied in the last couple of months It lasted one or two weeks It lasted about a month It has lasted about 6 months It has lasted about a year It has gone on for several years 1 2 3 4 5 6 I havent been bullied in the last couple of months I have been bullied in one or more of the following places in the past couple of months 18. Where have you been bullied? 1 2 Continue here if you have been bul lied in the past couple of months: Have you been bullied: No Yes 18a. on the playground/athletic field ( during recess or break times)? 1 2 18b. in the hallways/stairwells? 1 2
125 Appendix C (Continued) 18c. in class (with the teacher present)? 1 2 18d. in the classroom (without the teacher present)? 1 2 18e. in the bathroom? 1 2 18f. in gym class or the gym locker room/shower? 1 2 18g. in the lunch room? 1 2 18h. on the way to and from school? 1 2 18i. at the school bus stop? 1 2 18j. on the school bus? 1 2 18k. somewhere else in school? In this case, please write where:_________________ 1 2 Have you told (that you have been bullied): No Yes 19a. your class (homeroom) teacher? 1 2 19b. another adult at school (a diffe rent teacher, the principal, the school nurse, the custodian, the school psychologist, etc.)? 1 2 19c. your parents/guardians? 1 2 19d. your brothers or sisters? 1 2 19e. your friends? 1 2 19f. somebody else? In this case, please write who:_______________ 1 2 Almost Never Once in a while Sometimes Often Almost Always 20. How often do the teachers or other adults try to put a st op to it when a student is being bullied at school? 1 2 3 4 5 21. How often do other students try to put a stop to it when a student is being bullied at school? 1 2 3 4 5 I havent been bullied in the last couple of months (skip the next 6 questions) I have been bullied but I have not told anyone (skip the next 6 questions) I have been bullied and I have told somebody 19. Have you told anyone that you have been bullied at school in the past couple of months? 1 2 3
126 Appendix C (Continued) I havent been bullied in the last couple of months No, they havent contacted the school Yes, they have contacted the school once Yes they have contacted the school several times 22. Has any adult at home contacted the school to try to stop your being bullied at school in the past couple of months? 1 2 3 4 That is probably what he or she deserves I dont feel much I feel a bit sorry for him or her I feel sorry for him or her and want to help him or her 23. When you see a student your age being bullied at school, what do you feel or think? 1 2 3 4 ABOUT BULLYING OTHER STUDENTS I havent bullied another student(s) in the past couple of months It has only happened once or twice 2 or 3 times a month About once a week Several times a week 24. How often have you taken part in bullying another student(s) at school in the past couple of months? 1 2 3 4 5
127 Appendix C (Continued) Have you bullied another student(s) at school in th e past couple of months in one or more of the following ways? Please answer all of the questions. I havent bullied another student(s) in the past couple of months It has only happened once or twice 2 or 3 times a month About once a week Several times a week 25. I called another student mean names, made fun of or teased him or her in a hurtful way. 1 2 3 4 5 26. I kept him or her out of things on purpose, excluded him or her from their group of friends, or completely ignored him or her. 1 2 3 4 5 27. I hit, kicked, pushed, shoved him or her around or locked him or her indoors. 1 2 3 4 5 28. I spread false rumors about him or her and tried to make others dislike him or her. 1 2 3 4 5 29. I took money or other things from him or her or damaged his or her belongings. 1 2 3 4 5 30. I threatened or forced him or her to do things he or she didnt want to do. 1 2 3 4 5 31. I bullied him or her with mean names or comments about his or her race or color. 1 2 3 4 5 32. I bullied him or her with mean names, comments, or gestures with a sexual meaning. 1 2 3 4 5
128 Appendix C (Continued) 33. I bullied him or her in another way. In this case, please write in what way:_____________ 1 2 3 4 5 I havent bullied other student(s) at school in the past couple of months No, they havent walked with me about it Yes, they have talked with me about it once Yes, they have talked with me about it several times 34. Has your class (homeroom) teacher talked with you about your bullying other students at school in the past couple of months? 1 2 3 4 35. Has any adult at home talked with you about your bullying other students at school in the past couple of months? 1 2 3 4 Yes Yes, maybe I dont know No, I dont think so No Definitely No 36. Do you think you could join in bullying a student whom you didnt like? 1 2 3 4 5 6 I have never noticed that students my age are bullied I take part in the bullying I dont do anything but I think the bullying is OK I just watch what goes on I dont do anything but I think I ought to help the bullied student I try to help the bullied student in one way or another 37. How do you usually react if you see or understand that a student your age is being bullied by other students? 1 2 3 4 5 6
129 Appendix C (Continued) Never Seldom Sometimes Fairly Often Often Very Often 38. How often are you afraid of being bullied by other students in your school? 1 2 3 4 5 6 Little or Nothing Fairly Little Somewhat A good deal Much 39. Overall, how much do you think your class teacher has done to counteract bullying in the past couple of months? 1 2 3 4 5
130 Appendix D: Center for Epidemiologi cal Studies-Depression Scale (CES-D) DIRECTIONS: For each statement below, darken in the circle on the scantron form for the number that best describes how often you felt or behaved this way for each following statementDURING THE PAST WEEK Rarely or none of the time (Less than 1 day) Some or a little of the time (1-2 Days) Occasionally or a moderate amount of time (3-4 Days) Most or all of the time (5-7 Days) DURING THE PAST WEEK: 1. I was bothered by things that usually don't bother me 0 1 2 3 2. I did not feel like eating; my appetite was poor 0 1 2 3 3. I felt that I could not shake off the blues even with help from my family or friends 0 1 2 3 4. I felt that I was just as good as other people 0 1 2 3 5. I had trouble keeping my mind on what I was doing 0 1 2 3 6. I felt depressed 0 1 2 3 7. I felt that everything I did was an effort 0 1 2 3 8. I felt hopeful about the future 0 1 2 3 9. I thought my life had been a failure 0 1 2 3 10. I felt fearful 0 1 2 3 11. My sleep was restless 0 1 2 3 12. I was happy 0 1 2 3 13. I talked less than usual 0 1 2 3 14. I felt lonely 0 1 2 3 15. People were unfriendly 0 1 2 3 16. I enjoyed life 0 1 2 3 17. I had crying spells 0 1 2 3 18. I felt sad 0 1 2 3 19. I felt that people disliked me 0 1 2 3 20. I could not get "going" 0 1 2 3
131 Appendix E: State-Trait Anxiety Inventory for Children (STAIC) Trait Anxiety Scale items: DIRECTIONS: A number of statements that boys a nd girls use to describe themselves are given below. Read each statement caref ully and decide if it is hardly ever, sometimes, or often true for you. Then darken the scantron circle with the same number as the statement that describes you best. Th ere are no right or wr ong answers. Do not spend too much time on any one statement. Remember to darken the circle for each statement that best describes how you usually feel Hardly Ever Sometimes Often 1. I worry about making mistakes. 1 2 3 2. I feel like crying. 1 2 3 3. I feel unhappy. 1 2 3 4. I have trouble making up my mind. 1 2 3 5. It is difficult for me to face my problems. 1 2 3 6. I worry too much. 1 2 3 7. I get upset at home. 1 2 3 8. I am shy. 1 2 3 9. I feel troubled. 1 2 3 10. Unimportant thoughts run through my mind and bother me. 1 2 3 11. I worry about school. 1 2 3 12. I have trouble deciding what to do. 1 2 3 13. I notice my heart beats fast. 1 2 3 14. I am secretly afraid. 1 2 3 15. I worry about my parents. 1 2 3 16. My hands get sweaty. 1 2 3 17. I worry about things that may happen. 1 2 3 18. It is hard for me to fall asleep at night. 1 2 3 19. I get a funny feeling in my stomach. 1 2 3 20. I worry about what others think of me. 1 2 3
132 Appendix F: State-Trait Anger Expression Inventory for Children and Adolescents (STAXI-C/A) DIRECTIONS: A number of statements that boys a nd girls use to describe themselves are given below. Read each statement caref ully and decide if it is hardly ever, sometimes, or often true for you. Then darken the scantron circle with the same number as the statement that describes you best. Th ere are no right or wr ong answers. Do not spend too much time on any one statement. Remember to darken the circle for each statement that best describes how you usually feel Hardly Ever Sometimes Often 1. I am annoyed. 1 2 3 2. I feel annoyed when I do a good job and no one notices me. 1 2 3 3. I get mad when I am punished unfairly. 1 2 3 4. I feel grouchy. 1 2 3 5. I get mad. 1 2 3 6. I get angry when I do well and am told I did something wrong. 1 2 3 7. I feel angry when Im blamed for something I did not do. 1 2 3 8. I am hotheaded. 1 2 3 9. I get angry quickly. 1 2 3 10. I feel like yelling when I do something good and someone says I did bad. 1 2 3 11. I get furious when scolded in front of others. 1 2 3 12. I feel angry. 1 2 3 DIRECTIONS: A number of statements that boys a nd girls use to describe themselves are given below. Read each statement caref ully and decide if it is hardly ever, sometimes, or often true for you. Then darken the scantron circle with the same number as the statement which describes how you res pond or behave when you are angry or very angry. There are no right or wrong answers. Do not spend too much time on any one statement. Remember to darken the circle on the scantron form for the answer that best describes how you usually respond or behave when angry or very angry. Hardly Ever Sometimes Often 13. I am patient with others. 1 2 3 14. I show my anger. 1 2 3 15. If I dont like someone, I keep it a secret. 1 2 3 16. I try to calm my angry feelings. 1 2 3 17. I keep cool. 1 2 3 18. I say mean things. 1 2 3 19. I hide my anger. 1 2 3
133 Appendix F (Continued) 20. I try to relax. 1 2 3 21. I dont tell anyone I am angry. 1 2 3 22. I lose my temper. 1 2 3 23. I keep my anger in. 1 2 3 24. I try to calm down. 1 2 3 25. I control my temper. 1 2 3 26. I get into arguments. 1 2 3 27. I have more anger than I show. 1 2 3 28. I take a deep breath. 1 2 3 29. I control my angry feelings. 1 2 3 30. I get into fights. 1 2 3 31. I am afraid to show my anger. 1 2 3 32. I try to reduce my anger. 1 2 3 33. I stop myself from losing my temper. 1 2 3 34. I do things like slam doors. 1 2 3 35. I get mad inside, but dont show it. 1 2 3 36. I do something to relax and calm down. 1 2 3
134 Appendix G: School Adjustment Survey (SAS) DIRECTIONS: Read each sentence carefully and dark en the circle on the scantron form for the number that sounds most like you for each statement. Strongly Disagree Disagree I dont know Agree Strongly Agree 1. Students usually get along well with each other in this school. 1 2 3 4 5 2. Making friends is very difficult in this school. 1 2 3 4 5 3. I am in the wrong group to feel a part of this school. 1 2 3 4 5 4. A student can be himself/herself and still be accepted by other students in this school. 1 2 3 4 5 5. Most students at school like to include me in their activities. 1 2 3 4 5 6. I always seem to be left out of important school activities. 1 2 3 4 5 7. I think my teachers care about me. 1 2 3 4 5 8. Teachers are not usually available before class to talk with students. 1 2 3 4 5 9. My teachers often get to know me well. 1 2 3 4 5 10. Most teachers like my friends and me. 1 2 3 4 5 11. I care what most of my teachers think about me. 1 2 3 4 5 12. Some teachers would choose me as one of their favorite students. 1 2 3 4 5 13. I like school. 1 2 3 4 5 14. My teachers dont pay much attention to me. 1 2 3 4 5 15. I get a lot of encouragement at my school. 1 2 3 4 5 16. Other kids in my class have more friends than I do. 1 2 3 4 5 17. I feel a sense of school spirit. 1 2 3 4 5 18. I dont feel safe at this school. 1 2 3 4 5
135 Appendix G (Continued) Strongly Disagree Disagree I dont know Agree Strongly Agree 19. I have friends who are of different racial and ethnic backgrounds at this school. 1 2 3 4 5 20. Discipline is fair at this school. 1 2 3 4 5 21. I feel like Im learning a lot in school. 1 2 3 4 5 22. School is important to me. 1 2 3 4 5 23. I believe that Im learning important things in school. 1 2 3 4 5 24. I liked school more last year than I do this year. 1 2 3 4 5 25. I feel that I can go to my teacher for advice or help with schoolwork. 1 2 3 4 5 26. I feel that I can go to my teacher for advice or help with non-school related problems. 1 2 3 4 5 27. Most of my teachers dont really expect very good work from me. 1 2 3 4 5 28. I dont care how well I do in school. 1 2 3 4 5 29. I try as hard as I can to do my best at school. 1 2 3 4 5 30. I am an important member of this school. 1 2 3 4 5 31. It bothers me when I dont do something well. 1 2 3 4 5 32. Education is important for success in life. 1 2 3 4 5 33. I feel prepared for middle school. 1 2 3 4 5 34. I think I will go to college. 1 2 3 4 5
136 Appendix H: Middle School/High School Student Survey Questions regarding how students feel about their school and people in school Each item is rated on a 4-point scale ( YES! = always or almost always true for you, yes = usually true for you, no = not usually true for you, and NO! = never or almost never true for you) 1. I like school 2. I look forward to going to school 3. I try hard at school 4. I cant wait to drop out of school 5. I do things that make a difference at my school 6. My school tells my parents when I do a good job 7. My parents come to activities at my school 8. My parents make sure I do my homework 9. My teachers tell me when I do a good job 10. There is gang activity at my school 11. My teachers listen when I have something to say 12. I have a teacher who really cares about me 13. Adults at my school teach us not to pick on other students 14. Adults at my school try hard to keep students from bullying or picking on each other 15. I like my teachers 16. People in my school respect students of all races 17. People of my race can succeed in my school 18. I feel lonely at school 19. There is graffiti at my school 20. There is pressure to join gangs at my school 21. My school building is clean 22. I like the way my school looks 23. Students in my school obey the rules 24. There are gang fights at my school 25. All students at my school who break the ru les are treated the same, no matter who they are 26. When someone breaks the rules here, ad ministrators take appropriate action 27. At school, I help decide things like class activities and rules 28. Finishing high school is important to me 29. School is a waste of time 30. If I study hard, I will get better grades 31. If I really want to achieve some thing at school, I know I can do it 32. I care what my teachers think of me 33. I respect the teachers in my school
137 Appendix H (Continued) 34. I respect the principal in my school Questions regarding how safe students feel at school Each item is rated on a 4-point scale ( YES! = always or almost always true for you, yes = usually true for you, no = not usually true for you, and NO! = never or almost never true for you) 35. I feel safe at my school 36. I feel safe on my school bus 37. I feel safe walking to school 38. During the past school year, did you ever stay away from school because you were afraid you would not be safe at school this item is rated as yes or no and then asks how many times st udents stayed away from school 39. During the past school year, did you ever stay away from school because you were afraid you would not be safe traveling to school this item is rated as yes or no and then asks how many tim es students stayed away from school Questions regarding students friends Each item is rated on a 4-point scale ( YES! = always or almost always true for you, yes = usually true for you, no = not usually true for you, and NO! = never or almost never true for you) 40. I have a friend my age who cares about me 41. I spend most of my free time at school with my friends 42. My friends think it is wrong to hit other people 43. My friends think it is OK to yell at others and say mean things 44. My friends think it is OK to push or shove others when (?) you are mad 45. My friends think it is OK to physic ally fight to get what you want 46. My friends think it is wrong to call other people mean names 47. My friends think it is wrong to get into physical fights (like hitting or pushing) with others 48. My friends think it is OK to hit someone back when they hit you first 49. My friends think it is OK to take your anger out on others by using physical force (like hitting or pushing) 50. My friends think it is OK to drink alcohol 51. My friends drink to get drunk 52. My friends think that using drugs is a dumb idea 53. My friends think that it is OK to smoke cigarettes
138 Appendix H (Continued) Questions regarding things that go on an students schools The following items are rated as yes or no and then ask how many times students witnessed the activity in the past month. 54. I saw other students in a fight 55. I saw another student get pushed, shoved, slapped, or kicked 56. I saw another student get harassed 57. I saw a student threaten to hit or hurt another student at school 58. I saw a student with a gun at school 59. I saw a student with anothe r weapon (besides a gun) The following items are rated as yes or no and then ask how many times students participated in the activ ity in the past month. 60. I encourage other students to fight 61. I pushed, shoved, slapped, or kicked other students 62. I got into a physical fight to get so mething I want from another student 63. I walked away from a fight 64. I acted cold toward someone or gave them the silent treatment when I was angry at them 65. I harassed another student 66. I deliberately kept someone out of my group because I was angry at them 67. I threatened to hit or hurt another student 68. I was mean to someone when I was angry 69. I said bad things about someone to hurt th eir reputation or their friendships with others 70. I carried a gun to school 71. I ganged up with other students and we did mean things to another kid The following items are rated as yes or no and then ask how many times students experienced the activity in the past month. 72. Another student encouraged me to fight 73. Another student pushed, shoved, slapped, or kicked me 74. I was harassed by another student 75. Another student threatened to hit or hurt me 76. A classmate acted cold towards me or gave me the silent treatment 77. A classmate deliberately kept me out of th eir group because they were angry with me 78. A classmate said bad things about me to hurt my reputation or my friendships with others 79. Other students ganged up against me and were mean to me as a group
139 Appendix H (Continued) Questions regarding things that happened at students schools during the school year The following items are rated as yes or no and then ask how many times students experienced or participated in th e activity during the school year. 80. I was in a physical fight on school property 81. I was threatened or injured with a we apon such as a gun, knife, or club on school property 82. I was in a physical fight on school propert y in which I was injured and had to be treated by a doctor or nurse Questions regarding bullying and intimidation at students schools The following items are rated on a 4-point sc ale (None, Few, Some, or Many) during the school year 83. How many students in your school often get picked on in a mean way by other students? 84. How many students in your school often pi ck on other students in a mean way? 85. How many students at your school are afra id of you because they think you are mean? 86. How many students at your sc hool do you pick on often? 87. How many students at your school of ten pick on you in a mean way? 88. How many students at your school are you afraid of because they are mean? Questions regarding drug/alc ohol use at students schools The following items are rated as yes or no and then ask how many times students witnessed/ participated in the activity during the past month. 89. I saw a student smoking on school grounds 90. I saw a student using alcohol at school 91. I saw a student using i llegal drugs at school 92. I saw another student se lling drugs at school 93. I smoked cigarettes 94. I drank wine, beer, or other alcohol 95. I used marijuana 96. I used another illegal drug 97. I smoked on school grounds 98. I went to class drunk 99. I went to class high
140 Appendix H (Continued) Questions regarding students feelings toward hurting others Each item is rated on a 4-point scale ( YES! = always or almost always true for you, yes = usually true for you, no = not usually true for you, and NO! = never or almost never true for you) 100. I think it is wrong to hit other people 101. It is OK to yell at others and say mean things to them 102. It is OK to push or shove other people around if youre mad 103. It is wrong to call other people mean names 104. It is OK to take your anger out on others by using physical force (like hitting or pushing) 105. You have to physically fight to get what you want 106. I think it is OK to hit someone back if they hit you first Questions regarding students families Each item is rated on a 4-point scale ( YES! = always or almost always true for you, yes = usually true for you, no = not usually true for you, and NO! = never or almost never true for you) 107. My parents want me to get good grades 108. I can tell my parents the way I feel about things 109. I like to do things with my family 110. I have dinner with my family 111. My family has rules about where I can go and what I can do 112. When Im not home, one of parents knows where I am and who I am with 113. My parents limit the amount of TV I watch 114. My parents know who my friends are 115. My parents notice when I do a good job and let me know 116. There will always be people in my life I can count on 117. Besides my family, there is an adult who I can trust 118. I believe there is some good in everybody Questions regarding guns The following items are rated on a 3-point scale (Yes, No, or I Dont Know) 119. Do you know where you could get a gun? 120. Would it be hard for you to get a gun if you wanted to?
141 Appendix H (Continued) Questions regarding students activities The following items are rated on a 3-point scale (Yes, No, or I Dont Know) 121. There are clubs at my school 122. I am involved in clubs at my school 123. There are sports teams at my school 124. I am involved in sports teams at my school 125. There are other activities at my school 126. I am involved in other activities at my school 127. I am involved in clubs (like Boy Scouts/ Girl Scouts), sports teams, church groups, or other activit ies outside of school 128. I go to church or other religious or faith-based activities regularly Questions regarding students school attendance The following items are rated as yes or no and then ask how many times students participated in the ac tivity during the past month. 129. I missed school because I was sick 130. I missed school because I cut or skipped 131. I missed school for other reasons
142 Appendix I: Adult Superv ision at School (ASAS) DIRECTIONS: Read each sentence carefully and dark en the circle on the scantron form for the number that sounds most like you for each statement. Strongly Disagree Disagree I dont know Agree Strongly Agree 1. In my school, teachers and administrators are in the hall when we change classes. 1 2 3 4 5 2. In my school, teachers and administrators are in the halls when we are in class. 1 2 3 4 5 3. In my school, there are lots of places where teachers and administrators cannot see what is going on. 1 2 3 4 5 4. In my classroom, teachers walk around while students are working. 1 2 3 4 5 5. In my school, teachers and administrators supervise the open areas where students gather. 1 2 3 4 5 6. In my school, teachers and administrators supervise the places where students can hide. 1 2 3 4 5
143 Appendix J: Acting-Out, Moodiness, a nd Learning Scale-Revised (AML-R) Childs Name: _____________________ D.O.B.: _______________ Childs Gender: ___ Male ___ Female Is this child in Exceptional Education? : ___ Yes ___ No If yes, please specify ________________________ This child is in a: ___ Self-Contained ___ Continuous Progress -classroom Instructions: Please rate the childs behavior, as you have obs erved and experienced it since the beginning of school according to the following scale, by circling the appropriate number: (1) Never You have literally never observe d this behavior in this child. (2) Seldom You have observed this behavior once or twice. (3) Moderately often You have seen this behavior more often than once a month but less often than once a week. (4) Often You have seen this behavior more ofte n than once a week but less often than daily. (5) Most or all of the time You have seen this behavi or with great frequency, averaging once a day or more often. This child: 1. gets into fights or quarrels with classmates 1 2 3 4 5 2. has to be coaxed to play or work with peers 1 2 3 4 5 3. is confused with school work 1 2 3 4 5 4. is restless 1 2 3 4 5 5. is unhappy 1 2 3 4 5 6. gets off-task 1 2 3 4 5 7. disrupts class discipline 1 2 3 4 5 8. feels hurt when criticized 1 2 3 4 5 9. needs help with school work 1 2 3 4 5 10. is impulsive 1 2 3 4 5 11. is moody 1 2 3 4 5 12. has difficulty learning 1 2 3 4 5
144 Appendix J (Continued) This Child: Not in the past couple of months It has only happened once or twice 2 or 3 times a month About once a week Several times a week 13. has been bullied at school in the past couple of months? 1 2 3 4 5 14. has taken part in bullying another student(s) at school in the past couple of months? 1 2 3 4 5
145 Appendix K: Bullying Definition Per protocol for the Olweus Bully/Victim Questionnaire, the following definition of bullying was read at each group administration to guide responses and provided in student questionnaire packets: We define or explain the word bullying. We say a student is being bullied when another student, or several other students Say mean and hurtful things or make fun of him or her or call him or her mean and hurtful names Completely ignore or exclude him or her from their gr oup of friends or leave him or her out of things on purpose Hit, kick, push, shove around, or lock him or her inside a room Tell lies or spread false rumors about hi m or her or send mean notes and try to make other students dislike him or her And other hurtful things like that. When we talk about bullying, these things happen repeatedly, and it is difficult for the student being bullied to defend himself or herself We also call it bullying when a student is teased repeatedly in a mean and hurtful way. But we dont call it bullying when the teasing is done in a friendly a nd playful way. Also, it is not bullying when two students of about equal strength or po wer argue or fight (Olweus, 1996, p. 3).
About the Author Christine M. Wienke Totura graduated from Loyola University Chicago in 2000 with Departmental Honors and a Bachelor of Science degree in Psychology and Criminal Justice. During her undergraduate career Dr. Totura conducted research on the developmental trajectories of families with ch ildren who have pediatric illnesses. Dr. Totura received her Master of Arts degree in Clinical Psychology from the University of South Florida in 2003. While earning her Ph.D. in Clinical Ps ychology at the University of South Florida, Dr. Totura was actively involved in research examining peer aggression and victimization in school settings. She pres ented at several national conferences, coauthored publications, and assisted in the preparation of federall y-funded grants. In addition, Dr. Totura provided assessment, therapeutic, and consu ltative services for children and families in school and community settings. Dr. Totura completed a one-year APA-approved clinical internship in 2006 in which she developed and evaluated behavioral interventions and edu cational programming in schools.