Article

Indirect effects of bullying on school mathematics achievement in Chile

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Students who experience bullying at school present different negative outcomes, including lower academic achievement. However, the process by which bullying is connected to academic achievement is not clear. Using the Trends in International Mathematics and Science Study (TIMSS) dataset from Chilean schools in 2011, we sought to estimate the indirect effects of bullying on mathematics achievement via two key socio-motivational factors, namely school belonging and students’ engagement. Results of our multilevel latent covariate analyses showed that schools’ bullying rates were predictive of school differences in mathematics achievement, but these effects were explained by broader characteristics of the school environment such as perceived levels of safety and discipline. Crucially, the hypothesized indirect pathway was evident at the within-school level, showing that individual experiences of bullying are related to a poorer sense of belonging with the school as a whole, as well as poorer classroom engagement.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

Article
This special issue focuses on perspectives on educational effectiveness in science and mathematics and addresses the role of non-cognitive measures in Trends in International Mathematics and Science Study (TIMSS). Moving beyond a mere ranking of countries based on their average achievement by making use of the large amount of collected non-cognitive measures, brings many challenges. This special issue aims to address some of these challenges by including contributions with a more conceptual or methodological focus that encompass several non-cognitive scales, international comparative studies or multiple data points over time. The conceptual contributions are clustered around aspects of engagement and motivation, the experience of school bullying and its effects, and quality and equality. The methodological contributions study topics such as measurement invariance, sampling and teacher-centric inferences, and trend scaling. We hope that the papers in this special issue may stimulate and guide future work in the area of TIMSS.
Article
Full-text available
The author examined the relationship between student engagement and academic performance, using U.S. data of the Program for International Student Assessment 2000. The sample comprised 3,268 fifteen-year-old students from 121 U.S. schools. Multilevel analysis showed that behavioral engagement (defined as effort and perseverance in learning) and emotional engagement (defined as sense of belonging) significantly predicted reading performance. The effect of emotional engagement on reading performance was partially mediated through behavioral engagement. Findings from the present study suggest that educators, policy makers, and the research community need to pay more attention to student engagement and ways to enhance it.
Article
Full-text available
Hierarchical regression analyses conducted at the school level found that the perceived prevalence of teasing and bullying was predictive of schoolwide passing rates on state-mandated achievement testing used to meet No Child Left Behind requirements. These findings could not be attributed to the proportion of minority students in the school, student poverty, school size, or personal victimization, which were statistically controlled for. Measures of the prevalence of teasing and bullying were obtained from a statewide survey of 7,304 ninth-grade students and 2,918 teachers aggregated into school-level scores for 284 Virginia high schools. These results support the need for greater attention to the effect of teasing and bullying on high school student performance on high-stakes testing.
Article
Full-text available
Recent research highlights the role of peer victimization in students’ adjustment across a variety of domains (e.g., academic, social), but less often identifies potential mediating variables. In the current study, we tested for direct effects from peer victimization to adolescents’ academic behavior and alcohol use, as well as indirect effects through school belonging. Adolescents from two large samples (middle school: N = 2,808; high school: N = 6,821) self-reported on peer victimization, school belonging, academic outcomes (GPA, school truancy), and alcohol use (lifetime, past 30 days). Two-group structural equation models revealed (a) direct and indirect paths from peer victimization to academic functioning, (b) indirect, but not direct, effects through school belonging for lifetime drinking, and (c) direct and indirect effects from peer victimization to current drinking. Findings implicate school belonging as a mediator between peer victimization and important outcomes in adolescence.
Article
Full-text available
This prospective study of 276 Virginia public high schools found that the prevalence of teasing and bullying (PTB) as perceived by both 9th-grade students and teachers was predictive of dropout rates for this cohort 4 years later. Negative binomial regression indicated that one standard deviation increases in student- and teacher-reported PTB were associated with 16.5% and 10.8% increases in the number of dropouts, respectively, after controlling for the effects of other predictors, including school size, student body poverty and minority composition, community crime rates, and performance on standardized achievement testing. The predictive values of student and teacher perceptions of PTB were comparable in magnitude to the predictive values for other commonly recognized correlates of dropout rates. These results provide new evidence that the prevalence of peer victimization in high school is an important factor in high school academic performance. (PsycINFO Database Record (c) 2013 APA, all rights reserved)
Data
Full-text available
This article reviews the extant literature on the links between peer victimization and academic performance and engagement among children and adolescents. Although most of the research on this association is based on cross-sectional investigations, research using longitudinal designs is starting to point to the fact that peer victimization does impact changes in academic performance over time. This research also points to several mediators and moderators that explain the association between repeated victimization and academic challenges, including peer rejection, depression, and decreases in students' sense of school belonging. Teachers and administrators should address peer victimization through programs and frameworks such as positive behavior intervention supports and social-emotional learning approaches. These programs decrease aggression and victimization, increase peer acceptance and social competence, and improve academic engagement and test scores.
Article
Full-text available
The paper draws on the Trends in International Mathematics and Science Study (TIMSS) 2007 data for nearly 60 educational systems to explore the characteristics and patterns of school violence across a number of countries and the extent to which measures of violence are associated with achievement in mathematics and science. Using a multi‐level modelling approach, we examine the relationship between school violence and mathematics achievement. Findings from the international model indicate the extent to which violence at and across different levels impacts achievement in a cross‐country, large‐scale assessment. This paper suggests significant policy‐relevant issues about school and national violence and the implications for academic achievement in an increasingly global world.
Article
Full-text available
This cross-sectional study examines differences in the frequency of suicidal ideation and suicidal behaviors across a group of verbal bullies, bully-victims, victims, physically aggressive bullies, and students not involved in bullying. A large sample of middle school students (n = 661; fifth through eighth grades; ages 10-13 years of age) completed a pencil-and-paper survey that included the University of Illinois Bully, Fight, and Victim scales. Students also self-reported how often they had thought of killing themselves or deliberating hurting themselves in past 6 months, and provided information about delinquent behaviors and symptoms of depression and anxiety. We used cluster analysis to create bully-victim subtypes: uninvolved (n = 357), victims (n = 110), verbal bullies (n = 114), bully-victims (n = 29), and physically aggressive bullies (n = 42). Approximately 32%-38% of verbal bullies and victims, 60% of bully-victims, and 43% of physically aggressive bullies reported suicidal ideation, compared with 12% of uninvolved youth. Similarly, 24%-28% of verbal bullies and victims, 44% of bully-victims, and 35% of physically aggressive bullies reported deliberately trying to hurt or kill themselves, compared with 8% of uninvolved youth. Females in the bully-victim subtype reported particularly elevated suicidal ideation and behavior. After controlling for delinquency and depression, differences in suicidal thoughts and behaviors emerged only between uninvolved youth and the victim and bully-victim groups, but these differences were minimal. Findings highlight that at a bivariate level, involvement in bullying in any capacity is linked to increased risk for suicidal ideation and behavior, and echoes previous literature documenting particularly strong mental health implications for bully-victims. Furthermore, this study points to the importance of considering delinquency and depression in conjunction with suicidal ideation and behaviors.
Article
Full-text available
Several recent studies have found an association between bullying and depression [1] or bullying and suicide-related behaviors [2], and one study found evidence consistent with a causal link, at least for girls [3,4]. These studies, in conjunction with extensive media coverage of the deaths by suicide of several young people who were victims of bullying, led the Centers for Disease Control and Prevention (CDC) to convene an expert panel focusing on the relationship between bullying involvement and suicide-related behaviors. The purpose of the panel, held in September 2010, was to synthesize the latest research about the relationship between youth involvement in bullying (youth who bully, youth who are bullied, and those who bully and are bullied) and suicide-related behaviors (attempts, fatalities, and risk factors associated with suicide, such as depression). Experts on the topics of bullying and suicide presented their research about the relationship between these two behaviors; their work is contained in this supplement. The panel and this special issue provide clarity around the complicated issues of bullying and suicide among youth. Three key themes emerged: (1) bullying among youth is a significant public health problem; it is prevalent and frequently has detrimental effects; (2) there is a strong association between bullying and suicide-related behaviors, but this relationship is often mediated by other factors, including depression and delinquency; and (3) there are public health strategies that can be applied to the prevention of bullying and suicide.
Article
Full-text available
The work done here involved estimating the extent of bullying in Latin American schools and its impact on the academic achievement of primary school students. Pupils’ socio-demographic characteristics were analysed and linked with bullying. Three- and four-multilevel models were applied to data from the Second Regional Comparative and Explanatory Study (SERCE) conducted by the United Nations Educational, Scientific and Cultural Organization (UNESCO), analysing 2,969 schools, 3,903 classrooms and 91,223 sixth-grade students in 16 Latin American countries (not including Mexico for the association between school bullying and academic performance). The study found that bullying is a serious problem throughout the region; students who suffer peer aggression yield a significantly lower performance in reading and math than those who do not; and those in classrooms with more episodes of physical or verbal violence perform worse than those in less violent classroom settings
Article
Full-text available
This meta-analysis reviewed the literature on socioeconomic status (SES) and academic achievement in journal articles published between 1990 and 2000. The sample included 101,157 students, 6,871 schools, and 128 school districts gathered from 74 independent samples. The results showed a medium to strong SES–achievement relation. This relation, however, is moderated by the unit, the source, the range of SES variable, and the type of SES–achievement measure. The relation is also contingent upon school level, minority status, and school location. The author conducted a replica of White’s (1982) meta-analysis to see whether the SES–achievement correlation had changed since White’s initial review was published. The results showed a slight decrease in the average correlation. Practical implications for future research and policy are discussed.
Article
Full-text available
This study examined associations among depression, suicidal behaviors, and bullying and victimization experiences in 1491 high school students using data from the 2009 Youth Risk Behavior Survey. Results demonstrated that depression mediated the association between bullying/victimization and suicide attempts, but differently for males and females. Specifically, depression mediated the link between traditional victimization and suicide attempts similarly across gender, whereas depression mediated the link between cyber victimization and suicide attempts only for females. Similarly, depression mediated the link between traditional bullying and suicide attempts for females only. Depression did not mediate the link between cyberbullying and suicide attempts for either gender. Implications of the findings are discussed, including the importance of greater detection of depression among students involved in bullying, and the need for a suicide prevention and intervention component in anti-bullying programs. Findings suggest that bullying prevention efforts be extended from middle school students to include high school students.
Article
Full-text available
Longitudinal data from a study of kindergarten through 5th graders were used to estimate a structural model in which chronic peer exclusion and chronic peer abuse were hypothesized to mediate the link between children's early peer rejection, later classroom engagement, and achievement. Peer exclusion and abuse were expected to predict changes in 2 forms of school engagement (classroom participation and school avoidance), and changes in both forms of engagement were expected to predict changes in achievement. The model fit the data well and lent support to the premise that distinct forms of peer maltreatment and classroom engagement mediate the link between early peer rejection and changes in children's achievement. Early peer rejection was associated with declining classroom participation and increasing school avoidance, but different forms of chronic peer maltreatment mediated these relations. Whereas chronic peer exclusion principally mediated the link between peer rejection and classroom participation, chronic peer abuse primarily mediated the link between rejection and school avoidance. Children's reduced classroom participation, more than gains in school avoidance, anteceded decrements in children's achievement. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
The aim of this study was to explore the variation of general subjective well-being on individual and school levels. The classroom survey data of the School Health Promotion Survey (SHPS) were gathered in 1998 ( N = 39 886) and in 1999 ( N = 47 455) in different parts of Finland. The subjects were 8th and 9th graders from 458 secondary schools. Well-being was measured using the General Subjective Well-being Indicator (GSWI). The variation was studied using linear multilevel modeling. One percent of the variation in GSW occurred at the school level. The great majority of subjective well-being variation occurred between pupils.
Article
Full-text available
Background Bullying and victimization are widespread phenomena in childhood and can have a serious impact on well-being. Children from families with a low socioeconomic background have an increased risk of this behaviour, but it is unknown whether socioeconomic status (SES) of school neighbourhoods is also related to bullying behaviour. Furthermore, as previous bullying research mainly focused on older children and adolescents, it remains unclear to what extent bullying and victimization affects the lives of younger children. The aim of this study is to examine the prevalence and socioeconomic disparities in bullying behaviour among young elementary school children. Methods The study was part of a population-based survey in the Netherlands. Teacher reports of bullying behaviour and indicators of SES of families and schools were available for 6379 children aged 5–6 years. Results One-third of the children were involved in bullying, most of them as bullies (17%) or bully-victims (13%), and less as pure victims (4%). All indicators of low family SES and poor school neighbourhood SES were associated with an increased risk of being a bully or bully-victim. Parental educational level was the only indicator of SES related with victimization. The influence of school neighbourhood SES on bullying attenuated to statistical non-significance once adjusted for family SES. Conclusions Bullying and victimization are already common problems in early elementary school. Children from socioeconomically disadvantaged families, rather than children visiting schools in disadvantaged neighbourhoods, have a particularly high risk of involvement in bullying. These findings suggest the need of timely bullying preventions and interventions that should have a special focus on children of families with a low socioeconomic background. Future studies are necessary to evaluate the effectiveness of such programs.
Article
Full-text available
The Division of Violence Prevention (DVP) at the Centers for Disease Control and Prevention (CDC) is committed to preventing youth violence. For the past decade, DVP has supported research on risk and protective factors associated with youth interpersonal violence and the development and evaluation of prevention programs. This commentary addresses the findings, relevant themes, and implications of the five empirical articles included within this special issue that come from several CDC-funded Academic Centers for Excellence (ACEs) on Youth Violence Prevention. This commentary also focuses on the social context of bullying and peer victimization with particular emphasis on the influence of contextual and school-level indicators. Finally, it summarizes how the findings presented are consistent with, and can advance, DVP's research and prevention priorities in youth violence.
Chapter
Full-text available
The present chapter addresses bullying and victimization from an ecological perspective. The assumption of a multilevel approach allows for identifying several layers of complexity and, consequently, different levels for intervention strategies. First we introduce bullying and victimization, addressing three main topics: a critical review of what these phenomena involve, the notions of prevention and promotion as intervention goals, and the contextualization of bullying and victimization within the emerging peer culture of early adolescence. Then, the next section presents a layered analysis identifying four levels of complexity: the individual, the dyad, the peer group, and the institution, incorporating a developmental perspective; implications for interventions at each of these levels are included. Finally, the last section presents some guidelines for intervention based on literature on implementation and our own work in school-based bullying prevention.
Article
Full-text available
The Variance Inflation Factor (VIF) and tolerance are both widely used measures of the degree of multi-collinearity of the ith independent variable with the other independent variables in a regression model. Unfortunately, several rules of thumb – most commonly the rule of 10 – associated with VIF are regarded by many practitioners as a sign of severe or serious multi-collinearity (this rule appears in both scholarly articles and advanced statistical textbooks). When VIF reaches these threshold values researchers often attempt to reduce the collinearity by eliminating one or more variables from their analysis; using Ridge Regression to analyze their data; or combining two or more independent variables into a single index. These techniques for curing problems associated with multi-collinearity can create problems more serious than those they solve. Because of this, we examine these rules of thumb and find that threshold values of the VIF (and tolerance) need to be evaluated in the context of several other factors that influence the variance of regression coefficients. Values of the VIF of 10, 20, 40, or even higher do not, by themselves, discount the results of regression analyses, call for the elimination of one or more independent variables from the analysis, suggest the use of ridge regression, or require combining of independent variable into a single index.
Article
Full-text available
The association between bullying behaviour and academic achievement was investigated in 1016 children from primary schools (6–7-year-olds/year 2: 480; 8–9-year-olds/year 4: 536). Children were individually interviewed about their bullying experiences using a standard interview. Key Stage I National Curriculum results (assessed at the end of year 2) were collected from class teachers, and parents completed a behaviour and health questionnaire. Results revealed no relationship between direct bullying behaviour and decrements in academic achievement. Conversely, higher academic achievement at year 2 predicted bullying others relationally (e.g. social exclusion at year 4). Relational victimisation, Special Educational Needs (SEN), being a pupil from a rural school or small classes and low socioeconomic status (SES) predicted low academic achievement for year 2 children. Findings discount the theory that underachievement and frustration at school leads to direct, physical bullying behaviour.
Article
Full-text available
To examine the association between income inequality and school bullying in an international sample of preadolescents and to test for mediation of this association by the availability of social support from families, peers, and schools. The study used economic data from the 2006 United Nations Development Program Human Development Report and survey data from the 2005/2006 Health Behavior in School-aged Children (HBSC) study which included 66,910 11-year-olds in 37 countries. Ecological correlations tested associations between income inequality and bullying among countries. Multilevel linear and ordinal regression analyses tested the effects of income inequality on perceived social support and bullying others at school. Income inequality was associated with rates of bullying among the 37 countries (r = .62). Multilevel analyses indicated that each standard deviation increase in income inequality corresponded with more frequent bullying by males (odds ratio = 1.17) and females (odds ratio = 1.24), less family support and school support but more peer support. Social support from families and schools was associated with less bullying after differences in wealth were taken into account; however, social support did not account for the association between income inequality and bullying. Countries with high income inequality have more school bullying among preadolescents than countries with low income inequality. Further study is needed to understand the mechanisms that account for this association. Findings suggest that adolescents in areas of wide income inequality-not only those in deprived schools and neighborhoods- should be a focus of anti-bullying campaigns.
Article
Full-text available
In multilevel modeling (MLM), group-level (L2) characteristics are often measured by aggregating individual-level (L1) characteristics within each group so as to assess contextual effects (e.g., group-average effects of socioeconomic status, achievement, climate). Most previous applications have used a multilevel manifest covariate (MMC) approach, in which the observed (manifest) group mean is assumed to be perfectly reliable. This article demonstrates mathematically and with simulation results that this MMC approach can result in substantially biased estimates of contextual effects and can substantially underestimate the associated standard errors, depending on the number of L1 individuals per group, the number of groups, the intraclass correlation, the sampling ratio (the percentage of cases within each group sampled), and the nature of the data. To address this pervasive problem, the authors introduce a new multilevel latent covariate (MLC) approach that corrects for unreliability at L2 and results in unbiased estimates of L2 constructs under appropriate conditions. However, under some circumstances when the sampling ratio approaches 100%, the MMC approach provides more accurate estimates. Based on 3 simulations and 2 real-data applications, the authors evaluate the MMC and MLC approaches and suggest when researchers should most appropriately use one, the other, or a combination of both approaches.
Article
Full-text available
Peer group contextual effects of aggressive behavior among middle school students (6th-8th graders) were examined using a short-term longitudinal design. More specifically, the homophily hypothesis that peer group membership influences individual-level bullying and fighting was evaluated with multilevel sex-specific models of individual- and peer-level aggression scores. Peer groups were identified via social network analysis. Intraclass correlation coefficients yielded through hierarchical linear modeling demonstrated substantial within-group similarity on self-reported bullying and fighting, suggesting that students affiliate with individuals who bully and fight at the same frequency. Peer group bullying and fighting was associated with individual-level behavior, even after controlling individual baseline levels for males and females. However, peer contextual effects explained more variance in individual bullying than individual fighting. This differential impact of peer group membership suggests that future studies consider peer relations across subtypes of aggression.
Article
Purpose: The purpose of the study was to examine the association between income inequality and school violence and between the performance inequality and school violence in two international samples. Methods: The study used data from Trends in International Mathematics and Science Study 2011 and from the Central Intelligence Agency of United States which combined information about academic performance and students' victimization (physical and social) for 269,456 fourth-grade students and 261,747 eighth-grade students, with gross domestic product and income inequality data in 52 countries. Ecological correlations tested associations between income inequality and victimization and between school performance inequality and victimization among countries. Multilevel ordinal regression and multilevel regression analyses tested the strength of these associations when controlling for socioeconomic and academic performance inequality at school level and family socioeconomic status and academic achievement at student level. Results: Income inequality was associated with victimization rates in both fourth and eighth grade (r ≈ .60). Performance inequality shows stronger association with victimization among eighth graders (r ≈ .46) compared with fourth graders (r ≈ .30). Multilevel analyses indicate that both an increase in the income inequality in the country and school corresponds with more frequent physical and social victimization. On the other hand, an increase in the performance inequality at the system level shows no consistent association to victimization. However, school performance inequality seems related to an increase in both types of victimizations. Conclusions: Our results contribute to the finding that income inequality is a determinant of school violence. This result holds regardless of the national performance inequality between students.
Article
The present study examined the effects of teacher attunement to victimization on student perceptions of the bullying culture of their schools as a means of fostering a sense of belonging among early adolescents. Participants (n = 1,264) in sixth grade reported on the frequency that they had been bullied, and teachers were asked to report students who were “picked on.” Teacher attunement represented the correspondence between self-identified and teacher-identified victims. Attunement at the beginning of the school year was related to positive changes in student reports that their peers would intervene in bullying; in turn, sense of belonging was greater when students perceived that their peers would intervene in bullying. Teacher attunement was indirectly related to greater belonging through its impact on student perceptions of the bullying context.
Article
This article examined the role of centering in estimating interaction effects in multilevel structural equation models. Interactions are typically represented by product term of 2 variables that are hypothesized to interact. In multilevel structural equation modeling (MSEM), the product term involving Level 1 variables is decomposed into within-cluster and between-cluster random components. The choice of centering affects the decomposition of the product term, and therefore affects the sample variance and covariance associated with the product term used in the maximum likelihood fitting function. The simulation study showed that for an interaction between a Level 1 variable and a Level 2 variable, the product term of uncentered variables or the product term of grand mean centered variables produced unbiased estimates in both Level 1 and Level 2 models. The product term of cluster mean centered variables produced biased estimates in the Level 1 model. For an interaction between 2 Level 1 variables, the product term of cluster mean centered variables produced unbiased estimates in the Level 1 model, whereas the product term of grand mean centered variables produced unbiased estimates for the Level 1 model. Recommendations for researchers who wish to estimate interactions in MSEM are provided.
Article
I describe a test of linear moderated mediation in path analysis based on an interval estimate of the parameter of a function linking the indirect effect to values of a moderator—a parameter that I call the index of moderated mediation. This test can be used for models that integrate moderation and mediation in which the relationship between the indirect effect and the moderator is estimated as linear, including many of the models described by Edwards and Lambert (200710. Edwards, J.R., & Lambert, L.S. (2007). Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis. Psychological Methods, 12, 1–22.[CrossRef], [PubMed], [Web of Science ®]View all references) and Preacher, Rucker, and Hayes (200743. Preacher, K.J., Rucker, D.D., & Hayes, A.F. (2007). Assessing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42, 185–227.[Taylor & Francis Online], [Web of Science ®]View all references) as well as extensions of these models to processes involving multiple mediators operating in parallel or in serial. Generalization of the method to latent variable models is straightforward. Three empirical examples describe the computation of the index and the test, and its implementation is illustrated using Mplus and the PROCESS macro for SPSS and SAS.
Article
Over the past decade and a half, methodologists working with structural equation modeling (SEM) have developed approaches for accommodating multilevel data. These approaches are particularly helpful when modeling data that come from complex sampling designs. However, most data sets that are associated with complex sampling designs also include observation weights, and methods to incorporate these sampling weights into multilevel SEM analyses have not been addressed. This article investigates the use of different weighting techniques and finds, through a simulation study, that the use of an effective sample size weight provides unbiased estimates of key parameters and their sampling variances. Also, a popular normalization technique of scaling weights to reflect the actual sample size is shown to produce negatively biased sampling variance estimates, as well as negatively biased within-group variance parameter estimates in the small group size case.
Article
We tested whether gender-specific vs. common classroom norms were more powerful moderators of the association between a risk factor (rejection) and peer victimization among girls and boys. The participants were 1220 elementary schoolchildren from grades 4–6 (with 10–13 years of age).We compared different multilevel models including combined vs. separate regressions for boys and girls, as well as the effects of norms of the whole class, same-sex classmates, and cross-sex classmates. Among girls, the association between rejection and victimization was strongest in classes where bullying behavior was common, and anti-bullying attitudes were rare among girls. Among boys, the strength of the slope of victimization on rejection could not be explained by either common or gender-specific classroom norms, but boys’ level of bullying behavior was related to overall classroom level of victimization. The findings suggest that contextual factors may contribute to victimization especially among high-risk girls. The importance of exploring multiple levels of influence on children’s social development is discussed.
Article
Bullying is generally defined as largely unprovoked, negative physical or psychological actions perpetrated repeatedly over time between bully/ies and victims. Bullying can lead to fear of school, absenteeism, and stunted academic progress, which in turn are precursors to dropping out of school. This paper's aim is to report rates of bullying behaviour, and to investigate whether bullying behaviour predicts high school dropout in Cape Town, South Africa. Stratified, proportional sampling yielded 39 from a total of 214 schools, from which 40 learners were randomly selected from the combined class list of two, randomly chosen, Grade 8 classes in each of the 39 participating schools. Thus 1 470 learners (from a total of 181 018) completed a self-report questionnaire in 1997, and were followed-up in 2001. This report focuses on those learners who had dropped out of school between 1997 and 2001 (n = 776; 55.2%). Univariate and multiple logistic regression models were used to investigate the relationship between bullying behaviours and dropout, controlling for factors known to be strongly related to high school dropout, namely age, socioeconomic status, race, or ethnicity, being raised by a single parent, repeating a grade, and substance use. Odds ratios and 95 per cent confidence intervals were calculated, taking the clustering of schools into account. In 1997, 52% of the boys and 37% of the girls had been involved in bullying behaviours. Of the three bullying categories (bully, victim, and bully—victim), girls but not boys in the ‘bully—victim’ category were significantly more likely to drop out of school (OR 1.82; CI 1.09–3.04, and when controlling for confounders OR 2.60; CI 1.32–5.10). The pervasiveness of both high school dropout and bullying behaviour points to an urgent need for future research, and intervention in these areas.
Article
This paper presents results from a thorough systematic review on the efficacy of school bullying (perpetration and victimization) in predicting aggression and violence later in life. Results are based on prospective longitudinal studies. Two meta-analyses are presented examining whether: a) school bullying (perpetration and victimization) is a significant predictor of later aggression and violence, and b) whether each effect remains significant after controlling for other major childhood risk factors which were significantly related to both the predictors and the outcomes. Results are based on extensive searches of the literature. Nineteen electronic databases and 63 journals were searched from the inception of each database or journal through the end of March, 2012. Bullying perpetration at school was a significant predictor of violence (Adjusted OR = 2.04; 95% CI: 1.69–2.45) an average of six years later in life. This value of OR means that bullying perpetration increased the risk of later violence by about two-thirds. The summary effect size for bullying victimization versus violence was markedly smaller but still highly significant (Adjusted OR = 1.42; 95% CI: 1.25–1.62). This value of OR means that victimization increased the risk of later violence by about one-third. Analyses are presented of various potential moderators (such as the number of risk factors controlled for and the length of the follow-up period) in an attempt to explain the significant heterogeneity in effect sizes. Sensitivity analyses are performed on both meta-analyses, and they indicate that overall there is no evidence of publication bias. The overall findings favor the existence of a more general long-term underlying antisocial tendency rather than a more specific underlying violent tendency. Implications of our research for policy and practice are highlighted and future needs in this area of research are indicated.
Article
Commitment and attachment to school and perception of school norms were examined in a sample of sixth, seventh, and eighth graders to determine whether bullies, victims, bully victims, and students who reported no or low levels of bullying and victimization differed in their level of bonding to school and their perceptions of standards and expectations for behavior in the school environment (protective factors). Risk factors for bullying were also examined. Results of a discriminant analysis demonstrated differences among the groups on the measures of risk and protective factors and perception of school norms. The grouping of variables differentiated between the comparison group and the bully, victim, and bully victim groups on a dimension of healthy functioning indicated by low risk for bullying and an investment in prosocial behaviors and beliefs. The results have implications for schools in promoting prosocial bonding through the development of academic, emotional, and social competence.
Article
In extending our understanding of how the social climate of schools can affect academic outcomes, this study examined the relationship between school bullying, student— teacher (S-T) connectedness, and academic performance. Using data collected in Canada as part of a larger international study conducted by the Organisation for Economic Co-operation and Development, participants included 27,217 students aged 15 years and 1,087 school principals. Results of multilevel analyses revealed that math achievement was negatively related to school bullying and positively related to S-T connectedness. For boys, there was a significant interaction between bullying and S-T connectedness, suggesting a buffering effect of S-T connectedness on the relationship between school bullying and math achievement. Similar results were evident for reading achievement.
Article
Background: Many authorities agree that bullying has a widespread impact on school climate, affecting bystanders as well as victims. This study tested the contention that a climate of bullying can have a schoolwide impact on student engagement in school. Methods: Hierarchical linear modeling assessed the relations between student perception of bullying climate and student engagement at the individual and school level in a statewide sample of 7058 ninth graders randomly selected from 289 schools participating in the Virginia High School Safety Study. Student engagement was assessed by self-report scales measuring commitment to school and involvement in school activities. Results: Individual differences in perception of school climate characterized by bullying were associated with lower commitment to school, but not less involvement in school activities. School-level differences in student perceptions of bullying climate were associated with both lower commitment to school and less involvement in school activities, after controlling for the effects of gender, race, school size, proportion of ethnic minority students in the school, and individual-level perception of bullying climate. Conclusion: Efforts to improve student engagement should consider the schoolwide impact of bullying on all students.
Article
In this article we study the approximately unbiased multi-level pseudo maximum likelihood (MPML) estimation method for general multi-level modeling with sampling weights. We conduct a simulation study to determine the effect various factors have on the estimation method. The factors we included in this study are scaling method, size of clusters, invariance of selection, informativeness of selection, intraclass correlation, and variability of standardized weights. The scaling method is an indicator of how the weights are normalized on each level. The invariance of the selection is an indicator of whether or not the same selection mechanism is applied across clusters. The informativeness of the selection is an indicator of how biased the selection is. We summarize our findings and recommend a multi-stage procedure based on the MPML method that can be used in practical applications.
Article
This paper presents a meta-analytic review of 33 studies, with a total of 29 552 participants, that examined the concurrent association between peer victimization and academic achievement. The results revealed a small but significant negative correlation between peer victimization and academic achievement under both the random-effects model (r=−.12, p < .001) and the fixed-effects model (r=−.10, p < .001). Factors that moderated the strength of this association across studies include the peer victimization informant, the indicator of academic achievement, whether there was shared method variance, and the national setting of the study. An exploratory analysis revealed that the strength of the correlations did not differ for boys and girls. The results help resolve the conflicting findings in the existing literature and suggest the need for further investigation into the association between peer victimization and academic achievement.
Article
In this paper, using data from the Progress in International Reading Literacy Study (2006-PIRLS) and the Trends in International Mathematics and Science Study (2007-TIMSS), we investigate the impact of being a victim of school bullying on educational achievement for Italian students enrolled at the fourth and eighth grade levels. Firstly, we apply an OLS estimator controlling for a number of individual characteristics and school fixed effects. Secondly, in order to attenuate the impact of confounding factors, we use propensity score matching techniques. Our empirical findings based on average treatment effects suggest that being a victim of school bullying has a considerable negative effect on student performance at both the fourth and the eighth grade level. Importantly, the adverse effect of bullying on educational achievement is larger at age 13 than at age 9. Hence, school violence seems to constitute a relevant factor in explaining student performance.
Article
Multilevel modeling (MLM) is a popular way of assessing mediation effects with clustered data. Two important limitations of this approach have been identified in prior research and a theoretical rationale has been provided for why multilevel structural equation modeling (MSEM) should be preferred. However, to date, no empirical evidence of MSEM's advantages relative to MLM approaches for multilevel mediation analysis has been provided. Nor has it been demonstrated that MSEM performs adequately for mediation analysis in an absolute sense. This study addresses these gaps and finds that the MSEM method outperforms 2 MLM-based techniques in 2-level models in terms of bias and confidence interval coverage while displaying adequate efficiency, convergence rates, and power under a variety of conditions. Simulation results support prior theoretical work regarding the advantages of MSEM over MLM for mediation in clustered data.
Article
Background: The goal of this study was to assess the association between bullying and symptoms of depression among middle school students in Chile. Methods: Secondary data analysis of Chile's 2004 Global School-Based Health Survey. Results: A total of 8131 middle school students participated in the study. Forty-seven percent of students reported having been bullied in the past month and 30% reported having been sad and hopeless for 2 or more weeks in the past year. Students in the seventh and eighth grades were more likely to report having been bullied in the past month than students in ninth grade. Ninth grade students reported higher levels of loneliness, difficulty sleeping, and suicidal thoughts than students in the seventh and eighth grades. Boys were more likely than girls to report being bullied in the past month, but girls were more likely than boys to report symptoms of depression, such as prolonged feelings of sadness and hopelessness, loneliness, difficulty sleeping, and suicidal thoughts. Students who reported being bullied in the past month were more likely than non-bullied students to report symptoms of depression. A higher number of days of being bullied in the past month was associated with a statistically significant increase in reported rates of sadness and hopelessness (p < .001). Conclusions: Bullying is common among middle school children in Chile, and bullying and symptoms of depression are strongly linked. This finding is consistent with studies of bullying and depression in adolescents from other parts of the world.
Article
The main aim of this paper is to investigate to what extent self-reported bullying at age 14 predicts later offending, violence and other life outcomes. In the Cambridge Study in Delinquent Development, 411 South London males were followed up from age 8-10 to age 48-50, using repeated face-to-face interviews and searches of criminal records. Bullying at age 14 predicted violent convictions between ages 15 and 20, self-reported violence at age 15-18, low job status at age 18, drug use at age 27-32, and an unsuccessful life at age 48. These results held up after controlling for explanatory and behavioural childhood risk factors at age 8-10. Bullying might increase the likelihood of these later outcomes. Interventions that decrease bullying would most likely be followed by decreases in violent offending, drug use, and unsuccessful lives.
Article
Several methods for testing mediation hypotheses with 2-level nested data have been proposed by researchers using a multilevel modeling (MLM) paradigm. However, these MLM approaches do not accommodate mediation pathways with Level-2 outcomes and may produce conflated estimates of between- and within-level components of indirect effects. Moreover, these methods have each appeared in isolation, so a unified framework that integrates the existing methods, as well as new multilevel mediation models, is lacking. Here we show that a multilevel structural equation modeling (MSEM) paradigm can overcome these 2 limitations of mediation analysis with MLM. We present an integrative 2-level MSEM mathematical framework that subsumes new and existing multilevel mediation approaches as special cases. We use several applied examples and accompanying software code to illustrate the flexibility of this framework and to show that different substantive conclusions can be drawn using MSEM versus MLM.
Article
A model-based framework, due originally to R. A. Fisher, and a design-based framework, due originally to J. Neyman, offer alternative mechanisms for inference from samples to populations. We show how these frameworks can utilize different types of samples (nonrandom or random vs. only random) and allow different kinds of inference (descriptive vs. analytic) to different kinds of populations (finite vs. infinite). We describe the extent of each framework's implementation in observational psychology research. After clarifying some important limitations of each framework, we describe how these limitations are overcome by a newer hybrid model/design-based inferential framework. This hybrid framework allows both kinds of inference to both kinds of populations, given a random sample. We illustrate implementation of the hybrid framework using the High School and Beyond data set.
Article
This article provides researchers with a guide to properly construe and conduct analyses of conditional indirect effects, commonly known as moderated mediation effects. We disentangle conflicting definitions of moderated mediation and describe approaches for estimating and testing a variety of hypotheses involving conditional indirect effects. We introduce standard errors for hypothesis testing and construction of confidence intervals in large samples but advocate that researchers use bootstrapping whenever possible. We also describe methods for probing significant conditional indirect effects by employing direct extensions of the simple slopes method and Johnson-Neyman technique for probing significant interactions. Finally, we provide an SPSS macro to facilitate the implementation of the recommended asymptotic and bootstrapping methods. We illustrate the application of these methods with an example drawn from the Michigan Study of Adolescent Life Transitions, showing that the indirect effect of intrinsic student interest on mathematics performance through teacher perceptions of talent is moderated by student math self-concept.
Book
This text is a Stata-specific treatment of generalized linear mixed models, also known as multilevel or hierarchical models. These models are "mixed" in the sense that they allow fixed and random effects and are "generalized" in the sense that they are appropriate not only for continuous Gaussian responses but also for binary, count, and other types of limited dependent variables.
Article
The prefix svr stands for SurVey Replication, and refers to commands that analyze complex survey data using replication methods. The available methods are balanced repeated replication (BRR), and several versions of survey jackknife (JK1, JK2, and JKn). These methods are an alternative to the Taylor series linearization methods available in Stata's {help svy} commands. See help survwgt for details on replication methods.
Article
Multilevel modelling is sometimes used for data from complex surveys involving multistage sampling, unequal sampling probabilities and stratification. We consider generalized linear mixed models and particularly the case of dichotomous responses. A pseudolikelihood approach for accommodating inverse probability weights in multilevel models with an arbitrary number of levels is implemented by using adaptive quadrature. A sandwich estimator is used to obtain standard errors that account for stratification and clustering. When level 1 weights are used that vary between elementary units in clusters, the scaling of the weights becomes important. We point out that not only variance components but also regression coefficients can be severely biased when the response is dichotomous. The pseudolikelihood methodology is applied to complex survey data on reading proficiency from the American sample of the 'Program for international student assessment' 2000 study, using the Stata program gllamm which can estimate a wide range of multilevel and latent variable models. Performance of pseudo-maximum-likelihood with different methods for handling level 1 weights is investigated in a Monte Carlo experiment. Pseudo-maximum-likelihood estimators of (conditional) regression coefficients perform well for large cluster sizes but are biased for small cluster sizes. In contrast, estimators of marginal effects perform well in both situations. We conclude that caution must be exercised in pseudo-maximum-likelihood estimation for small cluster sizes when level 1 weights are used. Copyright 2006 Royal Statistical Society.
Article
pv estimates statistics when there are multiple estimates of the dependent variable referred to as plausible values. It is specially (but not exclusively) designed to be used with the PISA, TIMSS and PIRLS student achievement datasets.
Victimization and exclusion: Links to peer rejection, classroom engagement, and achievement
  • Buhs